FLUOXETINE FOR THE TREATMENT OF SELECTIVE MUTISM WITH ELEVATED SOCIAL ANXIETY SYMPTOMS : A NONCONCURRENT MULT I PLE BASELINE DESIGN ACROSS FIVE CASES By Justin A . Barterian A DISSERTATION S ubmitted to Michigan State Unive rsity in partial fulfillment of the requirements for the degree of School Psychology Doctor of Philosophy 201 5 A BSTRACT FLUOXETINE FOR THE TREATMENT OF SELECTIVE MUTISM WITH ELEVATED SOCIAL ANXIETY SYMPTOMS: A NONCONCURRENT MULTIPLE BASELINE DESIGN ACROSS FIVE CASES By Justin A . Barterian Selective mutism is a debilitating disorder with academic and social consequences , yet little research is available regarding psychopharmaco lo g i cal intervention s (Carlson et al., 2008) . This dissert ation study exami ne d the utility of fluoxetine (Prozac) for the treatment of five children , ages 5 to 1 4 , diagnosed with selective mutism , who also demonstrated elevated levels of social anxiety symptomology . A non - concurrent multiple - baseline desig n with a sin gle - blind placebo - controlled procedure was used to examine treatment outcomes . E ffectiveness was evaluated by a multi - gate analysis process , including: a) visual analysis (Kratochwill et al., 2010) ; b) the Wampold and Worsham (1986) randomization test ; and c) Mann - Whitney U (Tau - U; Parker, Vannest, Davis, & Sauber, 2011) effect size calculation . M ultiple methods of assessment including Direct Behavior Ratings ( DBR ; Chafouleas, Riley - Tillman, & Christ, 2009) and standardized measures [e.g ., the Multidimensional Anxiety Scale for Children Second Edition ( MASC - 2; March, 2012 ) ] were used . Information regarding a dverse effects with an emphasis on behavioral disinhib i tion and ratings of parental acceptance of the intervention were also gather ed . Visual analysis of all five cases indicated f luoxetine did not demonstrate utility for the reduction of social anxiety symptoms and was ineffective in increasing spontaneous speech . However, as predicted a significant i ncrease in responsive speech with unfamiliar adults (p = .0 3 ; Tau - U: . 44 , p < . 01 ) was noted . Two of the five children experienced some acute occurrences of minor behavioral disinhib i tion , but did not experience any other side effects during the course of treatment. Behavioral disinhib i ti on did not correspond to changes in social anxiety symptoms , responsive speech, or spontaneous speech. P arents found the use of fluoxetine for the treatment of selective mutism highly acceptable . Copyright by JUSTIN A. BARTER IAN 2015 v ACKNOWLEDGEMENTS I am indebted to m any who supported me throughout my graduate studies and the completion of this dissertation project . First , I am gr a t e ful for the guidance of my dissertation chair , Dr. John S. Carlson , who continua lly pushed me to be better wit h high expectations , patience, and encouragement. I am appreci ative of the collaboration of Dr. Joel Sanchez and his willingness to be the psychiatrist on this project . This study could not have been completed without the collaboration of the professionals at the Michigan State University Psychiatry Clinic, especially Dr. Christopher Giuliano. Further, t his project substantially benefitted from the support of my dissertation committee members , Drs. Jed Magen, Sara Witmer, and Evelyn Oka, who helped shape the ideas proposed at the beginning of this process . Additionally , I am thankful for the time and effort provided by my projec t assistants Allison Siroky, Brittany Mash, Marianne Clinton, and Elizabeth Snyder. This study was funded by a generous grant from the MSU S chool P sychology P rogram and a D issertation C ompletion F ellowship provided by the Department of Counseling, Educatio nal Psychology, and Special Education . I could not have made it this far without the sacrifice and support of my parents, Anthony and Michelle, as well as the support of other family members such as my brother Jonathon, my grandmother Eleanor, my Aunt Debr a and those I have lost during my graduate studies: my grandparents Joseph and Rosina. I am forever grateful for the humor and camaraderie provided by the lifelong friends and colleagues I made during my time at M SU , who were excellent examples of working hard and having fun in the process . Finally, I would li ke to dedicate this paper to my fianc é , Alicia Spangler, whose unwavering support helped me persist and bring this paper to completion. I look forward to everything that comes next for us. vi TABLE OF C ONTENTS LIST OF TABLES ................................ ................................ ................................ .......................... x LIST OF FIGURES ................................ ................................ ................................ ..................... xii CHAPTER 1 ................................ ................................ ................................ ................................ .... 1 INTRODUCTION ................................ ................................ ................................ .......................... 1 CHAPTER 2 ................................ ................................ ................................ ................................ .... 6 LITERATURE REVIEW ................................ ................................ ................................ ................ 6 Selective M utism ................................ ................................ ................................ ........................ 6 Diagnostic criteria of selective mutism ................................ ................................ ............... 6 Associated features ................................ ................................ ................................ ............. 7 Differential diagnosis ................................ ................................ ................................ .......... 9 Prevalence rates ................................ ................................ ................................ ................ 11 Demograph ic characteristics ................................ ................................ ............................. 11 Prognosis ................................ ................................ ................................ ........................... 12 Etiology of selective mutism ................................ ................................ ............................ 12 A joint biopsychosocial framework of selective mutism and social anxiety disorder ................................ ................................ ................................ ..... 12 Biological factors of selective mutism and social anxiety ........................ 13 Psychosocial factors of selective mu tism and social anxiety .................... 15 Pediatric Psychopharmacology and Selective Mutism ................................ ............................. 1 6 Ethics in pediatric psychopharmacology research and practice ................................ ....... 17 Selective Serotonin Reuptake Inhibitors ................................ ................................ ................... 18 The intended effect of SSRIs ................................ ................................ ............................. 20 Side effects associated with SSRI s ................................ ................................ .................... 22 Intended effects of SSRIs vs. behavioral disinhibition related to selective mutism ................................ ................................ ................................ ................................ 24 Current Evidence - Base of Selective Mutism Treatments ................................ ........................ 25 Psychosocial treatments for selective mutism ................................ ................................ .. 25 Behavioral and cognitive - behavioral therapy ................................ ....................... 25 Selective s erotonin reuptake inhibitors for the treatment of selective mutism ................. 26 Fluoxetine ................................ ................................ ................................ ............. 26 Quasi - experimental studies ................................ ................................ ....... 27 Case studies ................................ ................................ ............................... 28 Critical summary of fluoxetine studies ................................ ...................... 30 Sertraline ................................ ................................ ................................ ............... 30 Critical summary of sertraline studies ................................ ....................... 32 Paroxetine and fluvoxa mine ................................ ................................ ................. 32 Critical summary of fluvoxamine and paroxetine studies ......................... 33 Single - Case Research Design ................................ ................................ ................................ .... 33 Rationale for single - case design within selective mutism research ................................ ... 33 vii Multiple - baseline designs ................................ ................................ ................................ .. 36 Concurrent versus nonconcurrent multiple baseline designs ................................ . 37 Visual analysis of single - case design data ................................ ............................. 39 Quantitative analysis of single - case design data ................................ .................... 41 Randomization tests for the analysis of single - case design data ............... 41 Effect size calculations for single - case design data ................................ ... 43 Single - case design in selecti ve mutism psychopharmacology studies .............................. 44 Current Study ................................ ................................ ................................ ............................ 45 Research questions and hypotheses ................................ ................................ .................. 46 CHAPTER 3 ................................ ................................ ................................ ................................ .. 54 METHODS ................................ ................................ ................................ ................................ .... 54 Participants ................................ ................................ ................................ ................................ 54 Measures ................................ ................................ ................................ ................................ ... 59 Social anxiety ................................ ................................ ................................ .................... 60 Direct behavior rating (DBR): Social engagement ................................ ............... 60 Multidimensional anxiety scale for children 2 nd edition (MASC - 2) - Social anxiety scale (SAS) ................................ ................................ ............................... 62 Clinical global impressions (CGI): Anxiety/Shyness ................................ ........... 63 Frequency of speech ................................ ................................ ................................ ......... 64 Direct behavior rating (DBR): Speech ................................ ................................ .. 64 Selective mutism questionnaire (SMQ) ................................ ................................ 64 Clinical global impressions (C GI): Mutism ................................ .......................... 65 Adverse events ................................ ................................ ................................ .................. 65 Clinical global impressions (CGI): Side effects ................................ ................... 65 Side effects form for children and adolescents (SEFCA) ................................ ...... 66 Behavioral disinhibition ................................ ................................ ................................ .... 67 Parent version of the young mania rating scale (P - YMRS) ................................ .. 67 Treatment acceptabilit y ................................ ................................ ................................ ..... 68 Treatment evaluation questionnaire acceptability scale (TEQ - P) ...................... 68 Medication compliance measure ................................ ................................ ........... 69 End of study form ................................ ................................ ................................ . 69 Procedures ................................ ................................ ................................ ................................ 69 Project personnel ................................ ................................ ................................ ............... 70 Project coordinator ................................ ................................ ................................ 70 Project assistants ................................ ................................ ................................ .... 71 Study psychiatrist ................................ ................................ ................................ ... 71 Training ................................ ................................ ................................ ............................. 71 Study ps ychiatrist ................................ ................................ ................................ .. 71 Project assistants ................................ ................................ ................................ .... 72 Teachers ................................ ................................ ................................ ................. 72 Parents ................................ ................................ ................................ ................... 73 Study Phases ................................ ................................ ................................ .............................. 73 Project overview ................................ ................................ ................................ ............... 73 Inclusion and exclusion criteria ................................ ................................ ......................... 74 Screening ................................ ................................ ................................ ........................... 75 viii Consent and assent procedures ................................ ................................ ......................... 76 Baseline ................................ ................................ ................................ ............................. 76 Treatment ................................ ................................ ................................ .......................... 78 End of treatment phase ................................ ................................ ................................ ....... 79 Experimental Design ................................ ................................ ................................ ................ 79 Data Analysis ................................ ................................ ................................ ............................ 80 Multi - gate analysis ................................ ................................ ................................ ............ 80 Question one ................................ ................................ ................................ ...................... 83 Question two ................................ ................................ ................................ ...................... 83 Question three ................................ ................................ ................................ .................... 83 Question four ................................ ................................ ................................ ..................... 84 Question five ................................ ................................ ................................ ...................... 84 CHAPTER 4 ................................ ................................ ................................ ................................ .. 88 RESULTS ................................ ................................ ................................ ................................ ...... 88 Question One ................................ ................................ ................................ ............................. 88 Question Two ................................ ................................ ................................ ............................. 90 Question Three ................................ ................................ ................................ ........................... 91 Question Four ................................ ................................ ................................ ............................. 93 Question Five ................................ ................................ ................................ ............................. 96 Individual Improve ments ................................ ................................ ................................ ........... 98 CHAPTER 5 ................................ ................................ ................................ ................................ 100 DISCUSSION ................................ ................................ ................................ .............................. 100 Fluoxetine Treatment for Associated Social Anxiety ................................ .............................. 100 Fluoxetine Treatment for Speaking Behaviors ................................ ................................ ....... 102 Behavioral Disinhibition and Other Adverse Events ................................ ............................... 105 Medication Compliance ................................ ................................ ................................ ........... 108 Treatment Acceptability ................................ ................................ ................................ ........... 109 Impli cations for Research ................................ ................................ ................................ ........ 110 Limitations ................................ ................................ ................................ ............................... 111 APPENDICES ................................ ................................ ................................ ............................. 115 Appendix A: Young Mania Rating Scale Parent Version (Adapted) ................................ .. 116 Appendix B: Letter to School Psychologists, Private Practitioners, and Early Childhood Mental Health W orkers ................................ ................................ .... 118 Appendix C: Recruitment Flyer ................................ ................................ ............................... 119 Appendix D: Letter to Principals ................................ ................................ ............................. 120 Appendix E: Inclusion and Exclusion Criteria ................................ ................................ ........ 121 Appendix F: Correlation Matrix ................................ ................................ .............................. 122 Appendix G: Medication Log ................................ ................................ ................................ .. 123 Appendix H: Early Exit Form ................................ ................................ ................................ .. 124 Appendix I: End of Study F orm ................................ ................................ ............................... 125 Appendix J: Direct Behavior Rating Forms Teacher and Parent ................................ .......... 126 Appendix K: Multidimensional Anxiety Scale for Children 2 nd Edition: Social Anxiety Scale ................................ ................................ ................................ ..... 130 ix Appendix L: Clinical Global Impressions (Severity and Change) Clinician, Parent, Teacher ................................ ................................ ................................ ... 131 Appendix M: Selective Mutism Questionnaire ................................ ................................ ........ 142 Appendix N: Adapted Side Effects for Children and Adolescents (SEFCA) ......................... 145 Appendix O: Answers on the Parent - Mania Rating Scale that May Indicate Behavioral Disinhibition as a Result of Fluoxetine Treatment ........................... 146 Appendix P: Treatment Evaluation Questionnaire, parent version ................................ ......... 148 Appendix Q: Consent Forms and Assent Procedures for Participation (Parent, Teache r, Child) ................................ ................................ ................................ . 149 Appendix R: Medical, Developmental, Psychosocial, Family, Substance Abuse, and Legal History Form MSU Psychiatry Clinic ................................ ............ 161 Appendix S: Consent for Release of Confidential Medical, Treatment, and School Records ................................ ................................ ................................ ............... 169 Appendix T: History of Treatment Form for Previous Treating Professional ......................... 170 Appendix U: Physical Exam ................................ ................................ ................................ .... 172 Appendix V: Visual Analysis Decision Tree ................................ ................................ ........... 173 REFERENCES ................................ ................................ ................................ ............................ 175 x LIST OF TABLES Table 1 DSM - 5 Criteria for Selective Mutism ................................ ............................... ...7 Table 2 FDA Approval of SSRIs for Anxiety Disorders in Children and Adolescents ................................ ................................ ................................ ......... 19 Table 3 Published Case Studies on Fluoxetine Treatment for Selective Mutism ............ 29 Table 4 Rationale for Single - Case Designs in Psychopharmacol ogy Studies with Children Diagnosed with SM ................................ ................................ ............. 34 Table 5 What Works Clearinghouse Requirements to B ................................ ................................ ........................... 36 Table 6 Research Questions, Hypotheses, and Planned Assessment Procedures ........... 48 Table 7 Demographic Cha racteristics and Baseline/Intervention Ratings for Six Participants ................................ ................................ ................................ ........ 56 Table 8 Pre - Treatment, During Treatment, and Post - Treatment Ratings by Treatment Schedule ................................ ................................ ........................... 58 Table 9 Assessments Planned for Each Phase ................................ ................................ 60 Table 10 Randomized Multiple Baseline Design Baseline and Treatment Schedule ................................ ................................ ................................ ........... 78 Table 11 Percent Completion Rate of Data Forms per Observer ................................ .... 82 Table 12 Means of Social Anxiety Symptom on Outcome Measure Across Baseline an d Treatment Phases ................................ ................................ ........ 85 Table 13 Direct Behavior Ratings Mean Speaking Behaviors Across Baseline and Treatment Phases ................................ ................................ ....................... 86 Table 14 Means of Supplementary Selective Mutism Outcome Measures Across Baseline and Treatment Ph ases ................................ ................................ ......... 87 Table 15 CGI Means of Global Side Effect Ratings ................................ ..................... 93 Table 16 Answers on the TEQ - P Across Participants ................................ ..................... 97 Table 17 Medication Compliance Rates per Phase ................................ .......................... 98 xi Table 18 Parent and Teacher Reports of Progress after the Onset of Fluoxetine Treatment ................................ ................................ ................................ .......... 99 Table 19 Correlation Matrix ................................ ................................ ......................... 122 xii LIST OF FIGURES Figure 1 Conceptual framework Adapted from Models of the Biopsychosocial Framework of Selective Mutism and Social Anxiety Developed by Carlson and Colleagues 1 4 Figure 2 Inconsistent Parent DBR Ratings of Social Engagement and Speaking Behaviors During Ba seline for Participant Removed from Data Analysis by Week ................................ ................................ ................................ ............................. 57 Figure 3 Parent DBR Social Engagement with Unfamiliar Adults by Week ........................... 89 Figure 4 Parent DBR Spontaneous and Responsive Speech with Unfamiliar Adults by Week ................................ ................................ ................................ ........................ 92 Figure 5 Child Two: Parent DBR Rating of Responsive Speech with Unfamiliar Adults and Onset of Behavioral Disinhibition ................................ ................................ ................ 94 Figure 6 Child Three: Parent DBR Ratings of Social Engagement, Responsive Speech, and Spontaneous Speech with Unfamiliar Adul ts and Onset of Behavioral Disinhibition ................................ ................................ ................................ .................. 95 Figure 7 Letter to Professionals ................................ ................................ ................................ .. 118 Figure 8 Recruitment Flyer ................................ ................................ ................................ .......... 119 Figure 9 Letter to Principals ................................ ................................ ................................ ....... 120 Figure 10 Early Exit Form ................................ ................................ ................................ .......... 124 Figure 11 End of Study Form ................................ ................................ ................................ ..... 125 1 CHAPTER 1 INTRODUCTION Selective mutism is a rare chi ldhood disorder , affecting less than 1% of children (APA, 2013). The disorder is t ypically diagnosed before age five and is characterized by the withholding of verbal communication in specific situations (e.g., school, playground, g rocery store) while disp laying developmentally appropriate speech abilities in other more comfortable settings , such as at home with family members ( APA, 2013 ). Selective mutism has a negative Bergman, Piacenti , & McCracken, 2 002 ). C hildren with selective mutism are especially at - risk for current and future difficulties with social anxiety ( Keeton & Budinger, 2012 ), and typically display a lack of age - appropriate social skills (Carbone et al. , 2010) . demi c performance may suffer, as group work , oral presentations, and contributing to class discussion may be impaired . Withholding speech may also prevent teachers from monitor ing academic progress (Bergman et al., 2002 ; Giddan, Ross, Sechler, & Becker, 1997 ) . C hildren with selective mutism are more likely to experience a psychiatric disorder as an adult and are at heightened risk for phobic disorder s (Steinhausen et al., 2006) . U ntreated, selective mutism may set a child on a negative developme ntal trajectory with educational and psychological ramifications throughout the lifespan . Given the importance of treating selective mutism, consideration of effective treatments is necessary. Selective mutism is particularly difficult to treat as mute be haviors are often reinforced in the environment and become entrenched (Kolvin & Fundudis, 1981). The American Academy of Child and Adolescent Psychiatry (2009) stressed that in cases where the child is not responding to behavioral treatment and the disorde r is having a significantly negative 2 pharmacological treatment . However, due to a lack of clinical trial data and the rarity of the selective mutism , it is not likely that the Food and Drug Administration (FDA) will ever approve a medication indicated for t his purpose. In the recently outdated Diagnostic and Statistical Manual of Mental Disorders 4 th Edition, Text Revision (DSM - IV - TR ; APA, 2000 ), select ive mutism wa s classified as a Disorder Usually First Diagnosed in Infancy, Childhood, or Adolescence. Despite this categorization, s e lective mutism has often been conceptualized as an anxiety disorder (Carlson, Mitchell, & Segool, 2008). In fact, the asso ciation between these two disorders has been so prominent throughout the literature that selective mutism was re - categorized as an anxiety disorder in the recently published Diagnostic and Statistical Manual of Mental Disorders Fifth Edition (DSM - 5; APA, 2013). S elective mutism and social phobia have many features in common , such as fear of speaking with others, fear of humiliation, fear of judgment from others, and fear of new situations and people (Westenberg, 1998). C oncep tualizing selective mutism as an anxiety disorder is empirically justified considering the shared etiological factors between selective mutism and social anxiety, more generally speaking. Chavira and colleagues (2007) pinpoint the familial (i.e., genetic) link between selective mutism and other anxiety - related conditions. Specifically, this evidence indicates parents of children with selective mutism demonstrat e high rates of comorbidity with social phobia or avoidant personality disorder. Serotonin has been connected to the development and etiology of anxiety, and more specifically, the etiology of social anxiety disorder (Nutt, Bell, & Malizia, 1998; Tancer, 1993). Since social anxiety has been associated with a dysfunction in the serotonergic system , S elective S erotonin R euptake 3 I nhib itor s (SSRIs) make sense as an appropriate approach for treatment resistant cases ( Akimova, Lanzenberger, & Kasper, 2009). A majority of research studies on psychopharmacological interventions with children diagnosed with selective mutism used unstructu red case studies, severely diminishing the potential of generalization and providing no definite conclusion about the mechanism of change (Carlson et al., 2008). Despite this, p sychopharmacological treatments show promise for inclusion within a comprehensi ve treatment approach to chronic and severe cases ( e.g., Bork & Snyder, 2013 ; Dummit et al., 1997 ) . The most common psychopharmacological interventions for selective mutism are SSRI medications , with f luoxetine having the most data supporting effectiveness (Carlson et al., 2008 ) . Although not approved by the FDA for the treatment of social anxiety disorder, previous research highlights the efficacy of fluoxetine for the treatment of social anxiety disorder in children. In a randomized control trial, Birmahe r and colleagues (2003) found fluoxetine at 20 mg/day for 12 weeks was more efficacious than a placebo in the reduction of anxiety symptoms for children diagnosed with general anxiety disorder, social anxiety disorder, and separation anxiety disorder. A fe w studies have been completed examining the use of fluoxetine for the treatment of selective mutism in children. One randomized control trial with a small sample (n=15 ; i.e., Black & Uhde, 1994) found fluoxetine at 12 to 27 mg/day for 12 weeks was superio r to placebo for improvements in teache r perceived anxiety and parent perceived symptoms of selective mutism. In an open label study (n =21) , Dummit , Lein , Tancer, Asche, & Martin ( 1997) found that fluoxetine at 20 to 60 mg/day for 9 weeks successfully redu ced anxiety symptoms while increasing speech in unfamiliar settings in 76% of participants. Several case studies examining the use of fluoxetine have also shown improvements in mutism symptomology and social anxiety symptoms (e.g., Bork & Snyder, 2013; Mil ne, 1998). 4 Previous research identifies the potential of fluoxetine ; however, these studies posse ssed several weaknesses, including: a) lack of standardization of data collection in case studies; b) lack of placebo conditions , and c) small samples for the employed statistical analyses . This study contributes by examining the effects of fluoxetine on symptoms of selective mutism and related behaviors (i.e., social anxiety symptoms) in five children through the use of a placebo controlled, single - blind , nonconcurrent multiple - baseline single - case design . This methodology provides experimental control while allowing the factors in each case to be examined closely through the collection of data at multiple time points (Riley - Tillman & Maggin, 2008) , allow ing for a direct comparison between the placebo and fluoxetine treatment . Data w ere analyzed through the use of visual analysis (Kratochwill et al., 2010) . If visual analysis indicated improvement , the Wampold and Worsham (1986) multiple - baseline design ra ndomization test was utilized to identify statistical significance. The Wampold and Worsham (1986) approach is superior to newer randomization approaches (i.e., Koehler & Levin, 1998) for data analysis with psychopharmacological interventions. Due to the d elayed onset of fluoxetine effects, the Koehler and Levin approach may provide false negative results, because it does not allow for the inclusion of an acquisition period in data analysis . If significant results were identified , the Mann - W hitney U (Tau - U; Parker et al. , 2011) was used to quantify effect size. These methods have not been used concurrently in prior studies to quantify improvements in core symptoms of selective mutism . B ecause children are randomly assigned to a treatment sche dule with varying start points, the use of a multiple - baseline approach provides external validity (Christ, 2007) . Furthermore, t his study sought to identify adverse events that commonly occur with the use of fluoxetine in children (e.g., behavioral disin hibition ) and medication compliance. This study provides preliminary yet needed information to assist practitioner s and 5 parents who are weighing the use of fluoxetine for cases of selective mutism. Additionally, study findings may result in identifying th e need for and focus of future group and single - case design studies examining the use of SSRIs for the treatment of selective mutism with comorbid social anxiety symptoms . 6 CHAPTER 2 LITERATURE REVIEW In order to inform the research questions and methodology of this study the following areas serve as a foundation and are discussed in detail : (a) the characteristics, associated features , developmental outcomes and prevalence of selective mutism, (b) issues pertaining to pediatric psychopharmacology, (c) an overview of selective serotonin reuptake inhibitors and their mechanism of action , (d) the current evidence - base for effective selective mutism treatments, and (e ) an overview of the importance and value of using single - case research design within pediatric psychopharmacology . Selective Mutism Diagnostic criteria of selective m utism . The Diagnostic and Statistical Manual of Mental Disorders 5 th Edition ( DSM - 5 ; APA, 2013 ) has provided a clear description of diagno stic criteria for selective mutism that can aid practitioners and researchers in the accurate assessment and linking of effective treatments to the disorder ( see Table 1) . Selective mutism is characterized by a con sistent lack of speech in certain social s ituations when speech is necessary and expected (e.g., school), while speech occurs freely in other situations, such as at home in the presence of parents and siblings. T he reluctance to speak must negatively impact the individual in the areas of education al achievement, occupational achievement, or social communication. For example, an academic impairment may result when a child fails to ask a teacher for clarification when he or she do e s not understand class content or course expectations ( Pionek Stone et al., 2002) . Mutism behaviors are typically first observed prior to age five. The presence of symptoms must persist longer t han a month. C hildren who do not speak during the first month of school should not be diagnosed with selective mutism , because this novel social situation may result in 7 heightened shyness that may spontaneously resolve as the school year progresses and children become more acclimated to their new social situation (Stein, Rapin, & Yapko, 2001) . P ractitioners need to distinguish typical shyness or avoidant behaviors associated with beginning a new school year from severe mutism lasting for a n extended period of time that significantly detracts from optimal development . Table 1 DSM - 5 Criteria for Selective Mutism A. Consistent failure to speak in specific social settings in which there is an expectation for speaking, (e.g., at school) despite speaking in other situations. B. The disturbance interferes with educational or occupational achievement or with social communication. C. The duration of the disturbance is at least 1 month (not limited to the first month of school). D. The failure to speak is not attributable to a lack of knowledge of, or comfort with, the spoken language required in the social situation. E. The dis turbance is not better accounted for by a Communication Disorder (e.g., stuttering) and does not occur exclusively during the course of a Pervasive Developmental Disorder, Schizophrenia, or other Psychotic Disorder. Adapted from APA (2013 p.127). Associa ted f eatures . Children with selective mutism experience disruptions in their lives in personal, social , and academic domai ns. In or der to develop effective interventions, it is imperative to be aware of these associated features . In children with selectiv e mutism, h igh levels of s ocial anxiety result in avoidant and shy behaviors such as hiding, fleeing, or attempting to minimize attention from others ( APA; 2013; Ford , Sladeczek, Carlson, & Kratochwill , 1998). N ew experiences and situations , such as g oing to a new preschool or daycare, getting a new babysitter, or play ing with a new friend , are difficult and distressing tasks (Ford et al., 1998). M ute behaviors typically occur in s ocial situations, such as being around unfamiliar peers 8 (Kolvin & Fundudis, 1981; Steinhausen & Juzzi, 1996). These behaviors become even more apparent in social situations with unfamiliar ad ults (Bergman et al., 2002). C hildren with this diagnosis display lower levels of social competence, exhibit internalizing symptoms in socia l situations, and appear to display lower levels of social assertion, self - control, and social skills (Carbone e t al., 2010). This lack of social skills may lead to problems forming friendships . When individuals with selective mutism do engage in speaking behaviors, speech occurs infrequently in the form of a whisper or a quiet voice, includes little spontaneity, and takes the form of brief responses to questions or prompts (Ford et al., 1998). Children with selective mutism frequently have secondary sp eec h and motor difficulties (APA, 2013) . B etween 19% (Ford et al., 1998) and 38% ( Steinhausen & Juzzi, 1996) of children with selective mutism experience speech and language problems . D evelopmental delays and toilet training problems are common in this popula tion (Ste inhausen & Juzzi, 1996). In regards to developmental history , parents of children with selective mutism report a noteworthy prevalence of pregnancy, labor, and neonatal diffi culties. As the children progress through infancy and the preschool years , they often experience relationship problems, separation anxiety, sleep disorders, and eating disorders (Steinhausen & Juzzi) . Children with selective mutism frequently experience difficulty with academic coursework (APA, 2013 ). However, there are mixed data regarding the degree of impact . Ford and colleagues (1998) found that individuals with selective mutism and parents of children with selective mutism reported very few academic problems associated with the mute behavior . However, Bergman and colleague s (2002) argue the impact of selective mutism on classroom performance is often underestimated . B ehavioral inhibition experienced by children with selective mutism may cause them to withdraw from others in the context of the classroom and 9 preclude them fro m obtaining key s ocial and academic experiences provi ded during the preschool and early elementary years (Pionek Stone et al. , 2002) . Historically, it has been suggested children with selective mutism engage in mute behaviors because they are opposition al , inattentive, or depressed (Elson, Pearson, Jones, & Schumacher, 1965 ; Ford et al., 1998 ) . R esearch has not substantiated these claims (Ford et al., 1998) . A n alternative explanation may be oppositional behaviors occur when children are forced to speak, a nd are a manifestation of anxiety surrounding the expectation to speak (Dummit et al., 1997). This is essential to consider when selecting and creating interventions for children with selective mutism, as treating the refusal to sp eak as an oppositional b ehavior will be in effective and possibly lead to worse behaviors . In addition, individuals with selective mutism report they do not have diffic ul ty maintaining attention and typically describe th eir mood as pleasant or neutral when not being forced to spea k. Differential d iagnosis . The main diagnostic criteria of selective mutism is having the capability to speak, but withholding speech in unfamiliar situations. Since several disorders, such as neurodevelopmental disorders, communication disorders , mental retardation, and psychotic disorders, lead to a lack of speech, it is essential to differentiate between selective mutism and these other conditions when considering a diagnosis and treatment approach (APA, 2013) . Selective m utism is currently listed as a n Anxiety Disorder in the DSM - 5 (APA, 2013) , and almost always presents comorbidly with social anxiety disorder, suggesting these disorders are etiologically related. Most studies examining comorbidity rates of selective mutism and social anxiety disorder found co - occurrence to be greater than 95% (Black & Uhde, 1995; Dummit et al., 1997). Additionally, in a study that examined the types of psychopathology commonly experienced in children with selective mutism, a significant level of social anxiety was the only 10 consistent abnormal behavioral feature identified throughout the sample (Black & Uhde, 1992). The notion that selective mutism and social anxiety disorder may be different points along the same biopsychosocial continuum is not surprising as many of t he diagnostic criteria of selective mutism and social anxiety disorder overlap. These criteria include: experiencing fear during social or performance situations, avoiding social situations to minimize the feelings of anxiety, and a lack of fear or anxiety symptoms when at home (Sharp, Sherman, & Gross, 2007). Although researchers mostly agree social anxiety is a component of selective mutism, there may be unique etiological factors that lead to the withholding of speech in children with selective mutism in contrast to those with social anxiety disorder who mainly avoid social situations, but still speak if necessary (Yeganeh et al., 2003; Yegeneh, Beidel, & Turner, 2006). Mannasis and colleagues (2003) found there was no significant difference in levels of anxiety between children diagnosed with social anxiety disorder and children diagnosed with selective mutism, suggesting selective mutism is not an extreme form of social anxiety. However, both groups presented similarly with high levels of anxiety on a st andardized measure. Therefore, Mannasis and colleagues (2003) suggest selective mutism and social phobia may have the same etiological factors, but children with selective mutism choose to cope with the anxiety by not speaking, while children with social anxiety choose to cope with the anxiety by avoiding social situations altogether. Interestingly, the researchers found children with selective mutism are more likely to present with communication and speech problems. Children with selective mutism were sig nificantly more likely to score lower on a standardized measure targeted at assessing suggest mild language impairment may make children with social anxiety refrain fr om speaking in order to avoid experiencing the anxiety. However, an alternative explanation offered by the 11 authors suggests children with selective mutism may refrain from speaking, preventing them from making speech gains commensurate with same - aged peers . Prevalence r ates . Due to the rarity of the disorder, potential differential diagnoses, varying diagnostic criteria (e.g., DSM - III) , and the varying level of expertise of those identifying children with selective mutism, examining the prevalence rate of the disorder has proved challenging as shown in the wide range of rates reported throughout the literature. Prevalence rates in the school setting have ranged from . 03 % (Krakaya et al., 2008) to 2% (Kumpulainen, Rasanen, Rasska, & Somnpi, 1998 ) ; however, 2% is likely an overestimate due to the use of DSM - III criteria, which does not take impairment of the symptomology into account. Rates of selective mutism are also rare in clinical settings. S elective mutism was reported to account for .11% of new patien ts for psychiatrists (Carlson, Kratochwill, & Johnston, 1994) and less than 1% of individuals treated in mental health settings (APA, 2013). The information found in clinical prevalence studies suggest that although selective mutism is rare in comparison t o other disorders such as depression or generalized anxiety disorder; selective mutism is still a disorder many practitioners will encounter throughout their career. Demographic characteristics. R esearch suggests selective mutism is somewhat more common in females than males (APA, 2000) , with ratios of females to males ranging from 2.3:1 (Black & Uhde, 1995) to 1.5:1 (Kopp & Gilberg, 1997). The finding that selective mutism is more prominent in females is congruent with the conceptu alization of selective mutism as an anxiety disorder , as anxiety disorders are typically more prevalent in females (APA, 2000 ). Previous research indicates the prevalence of selective mutism in immigrant populations may be higher than the prevalence in non - immigrant populations (Bradley & Sloman, 1975) . This higher prevalence rate may be the result of the increase d stress and social anxiety experienced by 12 immigrant children when interacting with native adults, because of the differences in language and cultural understanding (Eli zur and Perendik, 2003) . Prognosis . Children with selective mutism typically display clinically significa nt non - speaking behavior for three to five years (Ford et al., 1998) . Steinhausen and colleagues (2006) identified two predominant courses throughout childhood. The more prevalent course, which accounts for a little more than half of children, is defined by a slow decline in the intensity of symptoms across childhood. The less prevalent course is characterized by persistent symptomology across childhoo d. Regardless of the course, a majority of those in the Steinhausen and colleagues study report ed improvement with a 58% remission rate. Only 18% of those surveyed reported slight or mild improvements, while the remainder endorsed at least moderate improve ment. Although symptoms typically resolve by age 10, individuals who no longer meet criteria still experience higher levels of psychopathology than those not previously diagnosed with any type of mental health disorder , suggesting that untreated sel ective mutism may pla ce them on a trajectory for future psychological difficulties . More specifically, those diagnosed with selective mutism as children have higher prevalence rates of phob ic disorders and social phobia, and continue to feel uncomfortable during social gatherings (Ford et al.). Etiology of selective m utism . A joint b iopsychosocial framework of selective m utism and social anxiety disorder . Research has demonstrated both biological and psychosocial factors play a role in the development and displ ay of selective muti sm and social anxiety disorder. Therefore, it is essential to use a model that accounts for all of these factors when conceptualizing the disorder and selecting an effective treatment. According to Nutt and colleagues (1998), the innate anxiety circuit within the brain impacts several factors that may lead to the development of social 13 anxiety, including: autonomic symptoms, avoidance learning, cortical receptors, and negative cognitions. In addition, Shear and Beidel (1998) discuss ed psy chosocial factors impacted by various forms of psychotherapy leading to the reduction of social anxiety, which include: the physiological effects of social anxiety, negative evaluations and expectations regarding social situations, and the avoidance of soc ial situations. Carlson and colleagues (2008) combined the biological model for the development of social anxiety disorder developed by Nutt and colleagues (1998) and the psychosocial model deve loped by Shear and Beidel (see F igure 1; 1998) in order to cre ate one model of selective mutism an d social anxiety disorder . Biological f actors of s elective m utism and s ocial a nxiety. Research has demonstrated selective mutism and social anxiety are linked biologically, and there are several parallels between the e tiological underpinnings of anxiety disorders and selective mutism. First, biological and genetic factors, such as temperament characterized by behavioral inhibition, are similar between children with anxiety disorders and children with selective mutism ( Astendig, 1999; Beiderman et al., 2001). In families who have a child diagnosed with selective mutism, there is a higher prevalence of family members with selective mutism, a higher prevalence of family members with social anxiety, and higher prevalence l evels of other associated features of selective mutism, such as shy behavior and depression (Black & Uhde, 1995; Chavira et al., 2007; Steinhausen & Ademek, 1997). Black and Uhde (1995) found that 70% of families with a first degree relative with selective mutism displayed high levels of social anxiety and 37% of these families had another member with selective mutism. Chavira and colleagues (2007) found that high scores on scales of social phobia and avoidant personality disorder were 3 to 4 times 14 Figur e 1 Conceptual F ramework Adapted from Models of the Biopsychosocial Framework of Selective Mutism and Social Anxiety Developed by Carlson and Colleagues (2008), Nut, Bell, and Malizia (1998), and Shear and Beidel (1998). Actual or Perceived Social Thr eat Hearing own voice A lack of recognition of threat Being judged as sounding stupid Fear of social anxiety Negative Cognitions Social Skills Avoidance Learning Autonomic Symptoms/Cortical Receptors Psychosocial bases of social anxiety/selective mutism Biological bases of social anxiety/select ive mutism 15 more commo n in parents of children with selective mutism when compared to control parents who did not have a child with the disorder. This genetic link has also been displayed in more distant relatives. For example, Steinhausen and Ademek (1997) found that reserved behavior is more prevalent in the first degree, second degree, and third degree relatives of children with selective mutism when compared to the relatives of children without selective mutism. This finding suggests a genetic component related to social anx iety and selective mutism may play a significant role in the development of both disorders (Steinhausen & Adamek, 1997). The neurocorrelates of social anxiety and selective mutism have also been discussed in the literature . For example, Lorberbaum and co lleagues (2004) conducted an fMRI study examining the differences in brain activity between individuals with social phobia and those without social phobia when preparing to give a speech in front of others. All children in the social phobic group were diag nosed with social anxiety disorder, with one child diagnosed with social anxiety disorder and co - morbid selective mutism. Th e authors found those with social phobia demonstrated more acti vity in the subcortical system , the limbic system, specifically the a mygdala, and the lateral anterior paral imbic belt . The authors explain these findings are comme nsu rate with the current understanding of the functions of these structures , as these areas are more likely to be involved in automatic emotions. Psychosocial f actors of selective mutism and social anxiety. In addi tion to biological factors, selective mutism and social anxiety disorder have similar psychosocial etiological underpinnings . First, both those with social anxiety or selective mutism have difficulty wi th social skills (Astendig, 1999 ; Voncken & Bogels, 2008) . Second, fear of negative evaluation is a primary characteristic of both selective mutism and social anxiety disorder (APA, 2000). Finally , 16 both children with selective mutism and children with soci al anxiety disorder ar e likely to avoid social situations ( Dummit et al., 1997 ; Mesa, Beidel, & Bunnell, 2014 ). Pediatric Psychopharmacology and Selective Mutism Psychopharmacological treatments have been used for a significant amount of time to treat th e biological component s of behavioral disorders in children , including selective mutism (Carlson et al., 2008 ), and this practi ce ap pear s to be increasing (Debar, Lynch, Powell, & Gale , 2003; Olfson, Marcus, Weissman, & Jensen , 2002). I n order to provide i nformed treatment for children diagnosed with selective mutism, it is necessary to und erstand the context in which this practice began as well as how frequently these psychopharmacological agents are used with children currently diagnosed with mental healt h disorders. Research into psychopharmacological agents to change behavior began with Emil Kraepelin in the 1890s. Kraepelin was a student of Wundt and was curious about how drugs might impact psychological constructs in human beings. Kraepelin named the practice of using a psychopharmacological interventions with children began in the late 1930s with the work of Bradley (Popper, 2002). Bradley (1937) reported on the use of Benze drine to treat children with attention difficulties. Following the reports by Bradley (1937), the use of Benzedrine to treat behavioral and attention difficulties was further examined in the 1930s. Following this research, in 1939, Phenobarbital was studie d for the use of behavior disorders in children. After these discoveries, stimulant use and research in children remained relatively the same (Popper, 2002). In the 1950s, researchers began to explore o ther classe s of drugs. At this time, the biological revolution in psychiatry began, which emphasized the biological basis of disorders in addition or in contrast to the popular psychodynamic conceptualizations prominent throughout 17 the field (Popper, 2002). T he use of antidepressants first began in the mid to late 1950s with the accidental discovery that monoamine oxidase - inhibitors (MAO - Is), which were prescribed as a drug to treat tuberculosis, improved the mood of the patients. Tricyclic antidepressants (TCAs) also came into use during this time (Lieberma n, Golden, Stroup, and McEvoy, 2000) . Research with both of these drug families continued to demonstrate efficacy in treating depression. However, serious side effects and drug interactions occurred. In the 1980s, SSRIs were developed and had a significant ly lower side effect profile compared to the previous gold - standard MAO - Is and TCAs (Judd, 1998). The first research studies examining the use of SSRIs in the treatment of children occurred in the 1990s (Popper, 2002). As time progressed , research on psych opharmacological medication s in children and prescription rates continued to expand . Olfson and colleagues (2002) reported an increase in the number of children taking psychotropic medications from 1.4% to 3.9% between the yea rs of 1987 and 1996. Moreover, by 1996, the use of antidepressant medications in adolescents increased to 2.1% , which is 3.5 times higher than the prevalence rate in 1987 . Ethics in pediatric psychopharmacology research and practice. Despite advancements in t he field of psychopharmac ology and the increase in the use of psychopharmacological agents to treat mental health disorders in children, much still needs to be known about the costs and benefits and the short - and long - term c onsequences of their use ( American Academy of Child and Adolescent Psychiatry, 2009; Fanton & Gleason, 2009 ). In order to make informed treatment decisions, it is essential to examine the current viewpoints on the ethics of pediatric psychopharmacology . It is clear from the literature that e xperts have consiste ntly advised p sychosocial treatments be the first line approach given the decreased risk associated with these treatments . Despite this , psychopharmacological treatments hold promise 18 and are recommended in some cases for the treatment of children exper ienc ing severe difficulties, which do not respond to psychosocial treatment ( American Academy of Child and Adolescent Psychiatry, 2009; Gleason et al. , 2007; Vitiello, 2001). P sychopharmacological agents are being used to treat children with selective mutism i n practice despite a paucity of research supporting this practice (e.g., Harvey & Milne, 1998; Wright, Cuccaro, Leonhardt, Kendall, & Anderson, 1995) . In order for a study to be ethical, Vitiello (2001) argues that the possible harm that could occur from u sing a psychopharmacological agent in research must be smaller than the potential problems of leaving the disorder untreated. To improve this risk benefit relationship, the author suggests there be frequent monitoring of the child for intended outcomes and adverse effects, using standardized behavioral measures. Therefore, single - case design methodologies, especially those that use a multiple baseline approach, are well - suited for the examination of the effectiveness and safety of psychopharmacological ag ents for rare or unique cases like selective mutism as they include multiple and frequent assessments of the impact of an inte rvention (Horner et al., 2005). Selective Serotonin Reuptake Inhibitors Selective Serotonin Reuptake Inhibitors (SSRIs) are the most commonly used psychopharmacological approach to treat selective mutism in children (Carlson et al., 2008). This makes sense, as t he use of SSRIs to treat selective mutism is consistent with the conceptualization of selective mutism as an anxiety disor der (e.g., Black & Uhde, 1995; Dummit et al., 1997). Only a few SSRIs have been approved by the Food and Drug administration for the treatment of anxiety (i.e., o bsessive - c ompulsive d isorder) in children (s ee Table 2). Nonetheless, there is literature to s upport their efficacy with children and adolescents who experience other anxiety disorders. 19 Table 2 FDA Approval of SSRIs for Anxiety Disorders in Children and Adolescents Medication Anxiety Disorder Age Citalopram (Celexa) Escitalopram (Lexapro) Flu oxetine (Prozac) Fluvoxamine (Luvox) Paroxetine (Paxil) Sertraline (Zoloft) Not approved Not Approved Obsessive Compulsive Disorder Obsessive Compulsive Disorder Not Approved Obsessive Compulsive Disorder N/A N/A > 7 years 8 - 17 years N/A >6 Years Derived from ANI Pharmaceuticals (2011), Eli Lilly and Company (2011), and Pfizer (2011) The FDA has approved fluoxetine up to 20 mg/day to treat 8 to 18 year old children with major depressive disorder and fluoxetine up to 60 mg/day for ch ildren between the ages of 7 and 17 for the treatment of obsessive - compulsive disorder. Although not approved for other anxiety disorders, previous research indicates the efficacy of fluoxetine for this indication. Birma her and colleagues (2003) found flu oxetine at 20 mg/day for 12 weeks was successful in treating a variety of anxiety disorders (e.g., social, generalized) in 37 children , and that this impact was significantly greater than an equivalent placebo group. Moreover, other studies have demonstrat ed positive outcomes for children with anxiety disorders being treated with sertraline (Compton et al., 2001; Rynn, Siqueland, & Rickels, 2001), citalopram (Baumgartner, Emslie, & Crimson, 2002), fluvoxamine (Walkup, 2001), paroxetine (Stein et al., 1998; Wagner et al., 2004), and escitalopram (Isolan et al., 2007). In order to have an understanding of the mechanisms of action associated with SSRIs, it is important to discuss the purported role of serotonin in the etiology of anxiety disorders as well as u nderstand the impact that SSRIs have on the serotonergic system. Moreover, in order t o determine if SSRIs have their intended effect when treating selective mutism, it is essential to 20 discuss the mechanism of change and the timing of treatment effects . Fin ally, it is important to review the possible side - effects that are associated with SSRIs in order to provide patients with an informed understanding of the risks and benefits of treatment as well as to delineate between intended treatment outcomes and side effects that may be similar in appearance, such as an increase in the frequency of speech as a result of decreased anxiety and/or behavioral disinhibition . The intended effect of SSRIs. Serotonin (5 - HT) is a neurotransmitter in the brain that plays a rol e in the modulation of emotion and mood (Hensler, 2010). G enetic, pharmacological, and neuroimaging studies have demonstrated that the 5 - HT1A receptor in the serotonin system, which serotonin (5 - HT) binds to, plays a primary rol e in the neurobiology of anx iety (Akimova, Lanzenberger, & Kasper, 2009). However, the specific mechanism connecting serotonin to anxiety is still not well understood. Research has shown that by modifying levels of serotonin in the brain by implementing a 5 - HT1A agonist, such as an S SRI, anxiety can be reduced (Lowry & Hale, 2010). Selective serotonin reuptake inhibitors bind to the 5 - HT1A receptor in the brain, and block the re - uptake of serotonin from the synaptic cleft leaving a higher quantity of serotonin in the synaptic cleft, w hich is then available to stimulate the post - synaptic neuronal receptors which leads to a modification in neuronal functioning (Tschanz & Treiber, 2011). This receptor is located in high densities throughout the brain, including the following areas: the hi ppocampus, the hypothalamus, the amygdala, the cerebral cortex, the dorsal raphe, and the median raphe (Charney, Krysal, Delgado, & Heninger, 1990). Modifying the levels of serotonin through the use of an SSRI has been linked to altering the functioning o f these various components of the brain. For example, Stahl (1998) suggests that adding a serotonin agonist leads to an increase in serotonin in the parts of the brain responsible 21 for various psychological disorders, including anxiety disorders, and this i ncrease in serotonin moderates the underlying physiological processes associated with a specific psychological disorder. The mechanism of action in this change has been hypothesized to be the (Stahl, 1998 p. 215). Immediately after an SSRI is administered, the serotonin re - uptake pumps are inhibited leaving more serotonin in the synaptic cleft. However, there are several changes within the serotonergic system that take longer to occur. Serotoni n autoreceptors, when bound to by 5 - HT, function to lower neuronal firing rates when these autoreceptors are chronically stimulated, such as when there is an abundance of 5 - HT in the synaptic cleft. Over time, the somatodendritic serotonin 1A autoreceptors become desensitized changing neuronal functioning and leading to more 5 - HT within the serotonergic system. This change in neuronal functioning is what is hypothesized to lead to a delayed treatment impact (Stahl, 1998). Therefore, it is recommended by so me in the field that patients be informed that substantial treatment effects may take four to six weeks to occur (Garfield, Francis, & Smith, 2004). However, not all researchers and practitioners in the field of psychopharmacology believe that it takes fo ur to six weeks to see an onset of smaller treatment effects. Some argue there is a rapid onset of treatme nt effects (e.g., Mitchell, 2006 ), which does of S SRIs can occur within a week of the onset of the use of SSRIs (Taylor, 2007), and the reported delay in the effect of SSRIs has largely been due to how these effects have historically been measured (Gelenberg & Chelsen, 2000). For example, studies have req uired the use of a more subtle, clinically meaningful, changes may not be identified. In a meta - analysis, Taylor 22 (2007) concluded that the initial improvements in f unctioning that are observed after starting an SSRI treatment are not always placebo effects, but may be the sole impact of the medicati on. For example, Mitchell (2006 ) posits that participants can see changes within three days of the onset of treatment, and that 90% of individuals will experience a change in symptoms during the first two weeks. Neuroimaging also supports this notion. Through the use of fMRI technology, researchers demonstrated one dose of an SSRI can change the processing of th e n, & Harmer, 2009). Side effects associated with SSRIs. SSRIs are an improvement over their predecessors (e.g., tricyclic antidepressants and monoamine oxidase inhibitors), because they are more selective in their binding to neuronal receptors and do not interfere with the functioning of other receptors (e.g., histamine), leading to fewer side effects (Feighner, 1999). D espite these improvements, side effects have been noted. Zuckerman and colleagues (2007 ) examined the charts of children who had received treatment in a Boston medical facility with a selective serotonin reuptake inhibitor and reported on side effects that were mild, moderate , or severe. The authors defined mild side effects as events that d id not interfere with functioning, moderate side effects as events that interfered with functioning but not to an extreme degree, and severe side effects as events leading to difficulty with age - appropriate daily tasks . According to the chart review, no ch ildren had to be medically or psychiatrically hospitalized because of the adverse effects of the SSRI during treatment. In addition, the authors found that only 33% of the children who received an SSRI experienced any kind of adverse event and 28% of child ren who received an SSRI experienced a moderate adverse event (e.g., insomnia, headache ). These adverse events occurred within a range of 2 to 144 days since the onset of the medication and for 40% of the 23 children experiencing an adverse event resolved whi le still on the medication. Only about 18% of the children that experienced an adverse event discontinued the medication. The side effects associated with SSRIs include somatic side effects and behavioral side effects. Somatic side effects are commonly r eported throughout the literature for children (e.g., Dummit et al., 1996, Carlson et al., 2 008) , and , include : drowsiness, insomnia, decreased appetite, and nausea (Wilens et al., 2003). A behavioral side effect of SSRIs is behavioral disinhibition and is defined as increased activity that does not include a change in mood or impulse control. A rare behavioral side effect is a manic reaction, which is defined as increased activity accompanied by euphoric feelings and grandiosity (Walkup & Laballarte, 2001) . Behavioral side effects appear to be more prevalent in younger children when compared to adolescents. In one study, a pproximately 10.7% of pre - adolescent children experienced behavioral disinhibition during a trial of an SSRI, while only 2.1% of adolesce nts in SSRI studies experienced this side effect (Safer & Zito, 2006). Although significantly safer than their predecessors, SSRIs do have the potential to cause some serious side effects. However, the occurrences of these side effects are rare. First, SS RIs have been linked to an increased risk in suicide in adolescents and young adults causing the FDA to issue a black box warning for adolescents and young adults (FDA, 2007). However, there has been much debate about the link between suicidality and SSRIs and more research needs to be done to illuminate the rel ationship. For example, Olf son, Shaffer, Marcus, and Greenberg (2003) examined the change in prevalence of antidepressant prescriptions for adolescents and the number of completed suicides in 588 zip codes in the United States. The authors found an increase in antidepressants was negatively correlated to the overall levels of suicides. Although this is the case, no studies have been completed on the likelihood of pre - pubescent children 24 experiencing su icidal ideation while taking an antidepressant. In addition to the possibility of an increased likelihood of suicide, another serious side effect of SSRIs is serotonin syndrome. Serotonin syndrome occurs when there is a significant excess of serotonin. Sy mptoms may include: tremor, diarrhea, delirium, rigidness of the muscles, and ma y be life threatening (Boyer & Shannon, 2005). Although serotonin syndrome is potentially life threatening , the prevalence rate for those taking SSRIs as prescribed is extremel y minimal . For example, in an analysis of 200 patients taking fluvoxamine, none of the patients met the criteria for serotonin syndrome (Ebert et al, 1997). Serotonin syndrome usually only occurs when a patient is on multiple medications that impact the se rotonin system or when doses are very high . In addition to short - term adverse events , possible long - term side effects may occur ; however, little is known about this topic . T he potential for long - term negative side effects should not be an argument for a c omplete boycott of all psychotropic medications for children, because the impact and negative experiences imposed on t he children by a psychological disorder, such as selective mutism in its most serious entrenched form, may result in worse outcomes if not treated (Pine, 2002) . A cost - benefit analysis always needs to be conducted when deciding to prescribe psychotropic medications in children ( American Academy of Child and Adolescent Psychiatrists, 2009; Gleason et al., 2007). Intended effects of SSRIs vs. behavioral disinhibition related to selective mutism. Carlson and colleagues (2008) argue it is essential to examine whether the effects of SSRI treatment are the result of the intended mechanism of action (i.e., reduction in anxiety via modulation of se rotonin levels) or due to behavioral disinhibition . Behavioral disinhibition (e.g., increases in activity level , pressured speech ) and a decrease in anxiety leading to increased speech production may present similarly to the untrained eye . If i ncreased spe ech production is 25 observed without the presence of behavioral disinhibition such as increased risk taking , impulsive behavior, or oppositional behaviors , it is likely the intended effect of treatment is occurring . Since subtle changes in functioning can oc cur quickly after the onset of the intervention , frequent assessments of functioning during SSRI treatment studies using sensitive measures that examine small changes in functioning are necessary throughout treatment . S ingle - case design methodology works e xceptionally well when examining the treatment effects of SSRIs, because it can examine clinically meaningful aspects of behavior such as frequency in speech behaviors , while simultaneously monitoring symptoms of behavioral disinhibition . Current Evidenc e - Base of Selective Mutism Treatments Psychosocial t reatments for s elective m utism . R esearch ers and practitioners have implemented a variety of different intervention s based on various conceptual frameworks , including: psychodynamic therapy , family therap y, behavioral and cognitive - behavioral therapy (Cohan, Chavira, & Stein, 2006), psycho pharmacological treatments (Carlson et al., 2008) , and combinations of these approaches (Eke, 2001 ; e.g., psychopharmacological and psychosocial ) . Psychodynamic approache s were the first type s of treatments reported on for children with selective mutism (Krysanski, 2003) . C urrently, behavioral and psychopharmacological treatments appear to be the most common treatments f or selective mutism in children, and also appear to h ave the mos t evidence supporting their efficacy (Cohan et al., 2006 ; Pionek Stone et al. , 2002 ) . Behavioral and c ognitive - b ehavioral t herapy . Behavioral and cognitive - behavioral approaches operate under the notion that anxiety associated with selective m utism is re inforced to speak. To treat the anxiety associated with selective mutism using a behavioral approach , children must become desensitized by engaging in anxiety provoking social 26 tasks that increase in difficulty eventually l ead ing to speaking to unfamiliar individuals . A systematic review of the selective mutism treatment literature conducted by Pionek Stone and colleagues (2002) demonstrated that behaviorally orientated treatments (i.e., systematic desensitization, contingen cy management, shaping) were more effective than no treatment at all . Research has also demonstrated that cognitive behavioral techniques , which address negative cognitions surrounding speaking to unfamiliar individuals in addition to employing behavioral techniques , are efficacious for the treatment of anxiety disorders (Compton et al., 2004). Given cognitive - behavioral techniques often require verbalizations of negative cognitions to a therapist , this treatment may not always be appropriate for children w ith s elective mutism, who display a primary symptom of withho lding speech in front of others. Selecti ve serotonin reuptake i nhibitors for the treatment of selective m utism . SSRIs that have most commonly been used for the treatment of selective mutism inc lud e fluoxetine (Prozac) , sertraline (Zoloft) , citalopram (Celexa) , flu voxamine (Luvox) , and paroxetine (Paxil; Carlson et al., 2008). In order to understand the current evidence - base for these SSRI treatment s a review of the available literature is presen ted . Fluoxetine. Fluoxetine was the first SSRI to be studied as a treatment for selective mutism and , to date , is the most researched SSRI for this use (Carlson et al., 2008). The first publication on the successful use of fluoxetine at an unspecified d ose to treat selective mutism was c ompleted by Boon (1994), regarding the case of a six - year ol d female who had been experiencing symptoms for two years . Since then, s everal case studies and two small quasi - experimental studies have been published. However , the s e studies on the use of fluoxetine were conducted over a decade ago , aside from a case study disseminated by Bork and Snyder (2013) . 27 Quasi - experimental s tudies. T wo quasi - experimental studies have provided valuable information regarding the possible efficacy of fluoxetine for the treatment of selective mutism in children; however, the method o logical limitations of these studies leave several questions to be answered. Black and Uhde (19 94) were the first to conduct a structured study on the use of an SSRI (i.e., fluoxetine 12 to 27 mg/day for 12 weeks ) to treat selective mutism in children. The authors used a repeated meas ures analysis of variance to examine the difference in selective mutism behaviors between a control group provided with a placebo (n=9) and an experimental group provided with fluoxetine (n=6). Parent, teacher, and psychiatrist ratings were obtained during week four, eight, and 12 to determine what level of change in functioning was occurring. According to the results, there were no significant differences in teacher ratings and psychiatrist ratings from week four to week 12 between the control and ex perimental groups. However, there was a significant difference in parent ratings of behavior between the control group and the experimen tal group. Positive behavior changes included increased speaking behaviors with extended family members, neighbors, and store clerks. The results of the Black & Uhde (1994) study provide mixed results. For instance, only parents reported significant levels of changes. In addition , there were several limitations to the study, which inclu ded a small sample size (n=15) for the type of statistical analysis used and differences in baselin e scores between the two groups. The second experimental study to examine the use of fluoxetine in children was completed by Dummit and colleagues (1996). This study included 21 children and adolescents between the ages of five and 14 who were diagnosed with selective mutism. Participants were provided with a daily dose of fluox etine that was slowly increased as the weeks progressed in order to avoid the development of side effects. During the first week children received a dose of five mg/day, followed by 10 mg/day during the second week, 20 mg/day during week three, 40 28 mg/day a t week six, and 60 mg/day at week eight, if there were not substantial improvements at 40 mg/day. Behavioral changes and side effects were monitored during weekly visits with the study psychiatrist. The effectiveness of fluoxetine was determined by using a number of Liebowitz Social Anxiety scale, The Liebowitz Social Avoidance Scale, and the Social Behavior Scale (Self and Parent ratings) at pre - treatment (week zero) and post - treatment (week nine). There were significant improvements on all scale scores between week zero and week nine. Some side effects were noted throughout the duration of the study. The most serious side effect experienced was behavioral disinhibiti on (n=3). Two of these children were removed from the study by the investigators, and one child was removed from the stu dy at week eight by his parents although he was responding better to a reduced dose of fluoxetine. In addition to these two children, 43 % of the participants reported that they experienced a potential side effect, however, these side effects did not persist for more than a week in duration . Although the Dummit and colleagues (1996) study provided promising results for the use of fluoxetine to treat children diagnosed with selective mutism, there were several limitations to this study. First, the study did not include a control group. Therefore, it cannot be determined if the reduction in symptoms occurred because of the fluoxetine or becaus e of additional variables . Second, mean pre - test and post - test scores were assessed using t - tests; however, the small sample size makes it difficult to draw conclusions about the findings report ed when using this type of analysis. Case studies. S even case studies reporting on the use of fluoxetine to treat selective mutism in children have been published in scientific journals. All of these studies report positive outcomes (e.g ., speech at school; see Table 3 ) . For example, in a case study completed by Wri ght and colleagues (1995), a five - year old female prescribed fluoxetine ( i.e., four to 8 mg/day ) began 29 Table 3 Published Case Studies on Fluoxetine Treatment for Selective Mutism First Author (Year) Participant(s) (Sex; Age) Fluoxetine Dose Additional Treatments Outcomes Boon (1994) Bork (2013) Wright (1995) Guna - Dumitrescu (1996) Harvey (1998) RUPP (1999) Silveira (2004) Female; Six y/o Male; Nine y/o Female; Four y/o Male; Eight y/o Child One: Female; Five y/o Child Two: Fema le; Eight y/o Female; Five y/o Female; Six y/o Unspecified 1.5 mg/day to 12 mg/day Four mg/day to Eight mg/day 20 mg/day to 30 mg/day Child One: Two mg/day to four mg/day Child Two: unspecified to six mg/day Five mg/day to 30 mg/day U nspecified to 20 mg/day; reduced to 15 mg/day due to headaches None Behavioral Intervention Behavioral Intervention Behavioral Intervention Multimodal Psychotherapy Behavioral Intervention; Clonidine 0.025 mg; Haloperidol 0.5 mg Behavioral Int ervention; Psychoeducation Unspecified improvements Spontaneous speech in the school settting and community settings Speech in multiple settings including school; no side effects No treatment effect after six months; Behavior component was added and s peech occurred at school after three weeks Reduction in anxiety (Child One); Increased speech in unfamiliar settings (Child One); Increased speech at school (Child Two) Increase in speech in unfamiliar settings; possible aggression side effect Improved social skills; Increased speech in the school and community settings 30 speaking within 20 days after the onset of treatment without hesitation in several settings, including school . Despite the positive outcomes reported across case studies , it is diffic ult to determine what other variables may have been involved in influencing outcomes , because of the unstructured approach . For example, some of these studies included a behavioral intervention in addition to the fluoxetine treatment (e.g., Guna - Dumitrescu & Pelletier, 1996). Moreover, R upp (1999) reported on the use of fluoxetine (i.e., five to 30 mg/day) in combination with two other psychopharmacological agents (i.e., clonidine and h aloperidol). Therefore, these studies provide initial evidence of the po ssible benefits of fluoxetine treatment of selective mutism ; yet, more research needs to be conducted before conclusions from these case studies can be drawn. Critical summary of fluoxetine studies. The available quasi - experimental studies and case studi es suggest that fluoxetine from four to 60 mg/day is a promising treatment for selective mutism in children. However, as evidenced by the limitations associated with these two quasi - experimental studies and the lack of a specific measure of outcome variabl es and lack of scientific rigor in the case studies, it is evident that more work needs to be done to ensure that fluoxetine is indeed successful in treating selective mutism in children. In order to accomplish this, studies that adhere to rigorous experim ental designs examining the use of fluoxetine to treat selective mutism should be conducted. Sertraline. Two single - case design studies have examined the impact of an alternative SSRI (i.e., sertraline) treatment for children with selective mutism. First, Carlson and colleagues (1999) examined the use of sertraline to treat five children ages five to eleven diagnosed with selective mutism using a double - blind placebo procedure with a replicated multiple - baseline across participants design. Participants in the study had been experiencing selective mutism symptoms between two and seven years. All children were provided with a dose of 50 mg/day 31 for two weeks and 100 mg/day for the rest of the study. Several standardized methods were used to determine the effic acy of the sertraline treatment including: Goal Attainment Scaling, the Child Behavior Checklist, and Clinical Global Improvement Ratings conducted by the teachers, parents, and study psychiatrist. Moreover, parents determined how appropriate they believed the and acceptability. Final outcome data was analyzed using visual analysis and by calculating effect sizes. Evidence for the eff ectiveness of the sertraline w as demonstrated in that the parent GAS ratings were consistent for four of the five participants and changed with the onset of treatment in the hypothesized direction. However, there are some limitations to the Carlson and colleagues (1999) study. First , t he onset of treatment effects, which theoretically take multiple weeks to develop, emerged more quickly than hypothesized. Therefore, the authors cautioned that the cause of the increase in speech in this study might have been related to behavioral disinhi bition and not the intended mechanism of action of the medication. Second, s ince this study has been published, there has been much growth in the field of single - case design in regards to requirements for visual analysis and the number of available effect size measures that can be used to quant ify outcomes. Eke (2001) examined the use of sertraline at 50 to 100 mg/day paired with a behavioral consultation intervention for the treatment of selective mutism in four c hildren with ages ranging from six to 10 using a two group multiple - baseline single - case design methodology with a double - blind placebo control medication components and visual analysis to examine the treatment outcomes. Dependent measures in the study included: ratings of Clinical Global Improve ment provided by the teachers, parents, and study psychiatrist, the Child Behavior Checklist, Goal Attainment Ratings, and a standardized measure of side effects. Results 32 indicated the combined approach of the sertraline and psychosocial intervention w as e ffective in reduc ing symptoms. However, the author noted several limitations to this study. O ne of these limitations is missing behavioral observations and reliability data due to difficulties in recruiting school personnel and proj ect staff to complete ob servations. Moreover, this study also heavily relied on visual analysis . Unlike Carlson and colleagues (1999), this study did not calculate effect sizes for treatment outcomes. Critical s ummary of sertraline studies. Both the Carlson and colleagues (199 9) study and the Eke (2001) study provide initial evidence that sertraline may be beneficial for the treatment of s elective mutism ; however, because of the methodological limitations of these studies, more research is needed to draw stronger conclusions. S ince these studies were published, improved guidelines for visual analysis and effect size calculation s have been developed . The What Works Clearinghouse has accepted these guidelines as a means to identify evidence - based interventions (Kratochwill et al., 2013). Therefore, additional research with sertraline and other SSRI s in children with selective mutism is needed. Paroxetine and fluvoxamine. In addition to fluoxetine and sertraline, two additional case studies have been published on the use of SSRIs to treat selective mutism . First, Lafferty and Constantino (1998) examined the use of fluvoxamine to treat a six - year - old female, who had been displaying symptoms for approximately a year and a half. The child was put on a dosage of fluvoxamine of 50 mg/da y, which was increased to 100 mg/day after two weeks of no treatment response. According to the authors, two days after the dosage increase the child began to speak within the classroom and community settings, however, she continued to refrain from speakin g when behavioral disinhibition and was engaging in reckless behavior. This led to the tapering of the 33 kless behavior ceased , and her mutism behaviors did not reoccur. Second, Lehman (2002) reported on the case of an eight - year - old female who had been diagnosed with selective mutism at age five. The child was prescribed paroxetine at five mg/day. After two to three weeks, the child was no longer displaying symptoms of selective mutism, and her parents reported a dramatic increase in social skills, school attendance, and socializing with friends. Critical s ummary of fluvoxamine and paroxetine studies. The La fferty and Constantino (1998) study and the Lehman (2002) study suggest that other SSRIs such as paroxetine and fluvoxamine may be beneficial for the treatment of selective mutism. However, much more methodologically rigorous research needs to be completed before conclusions can be drawn about either of these SSRIs. T he findings in these case studies are consistent with the overall findings of other SSRI treatment quasi - experimental, single - case design, and case studies on the use of SSRIs to treat selectiv e mutism. Single - Case Research Design Rationale for single - case design within selective mutism r esearch . There are several unique characteristics to consider when examining psychopharmacological treatments for children with selective mutism , such as the rarity of the disorder and the need for frequent monitoring of improvements and potential side effects. Fortunately, single - case design methodologies have the capability to address these issues (see Table 4). The most salient of these characteristics is w orking with a small sample size due to the rarity of selective mutism. This reality makes it difficult to study treatments for selective mutism using randomized control trials, wh ich are often considered the gold standard for the measurement of the efficac y of a treatment 34 and require a large number of participants (Kratochwill & Levin, 2010 ). Single - case designs address this issue by requiring fewer participants to draw reliable and valid conclusions . Table 4 Rationale for Single - case Designs in Psychophar macology Studies with Children Diagnosed with SM Characteristics of SM Medication Studies Attributes of Single - Case Design Need for frequent monitoring of intervention effects and side effects Requires repeated measurement Rarity of Selective Mutism within the population Requires a small num ber of parti c i pan t s Lag of onset of treatment outcomes Ability to analyze when changes in functioning occur Need for standardized assessment with internal and external validity and quantitative analysis of outcomes Randomization of start point, within and between subject comparisons, calculations of effect sizes , non - parametric statistical calculations Need to determine if there is a socially valid change for the participants in the study Allows researche rs to examine clinical effectiveness and social validity of an intervention Derived from APA (2000), Carlson and colleagues (2008), Gleason and colleagues (2007) Koehler and Levin (1998), and Riley - Tillman and Burns (2009). Single - case designs are defi scientific methodology used to define basic principles of behavior and establish evidence - based Single - case design studies are intended to accomplish the following three things. First, single - case dependent variable. Second, single - case designs distinguish if the change caused in the dependent variable was a direct result of the application of an independent variable. Finally, single - case design methodologies are intended to determine if the change seen in the dependent - Tillman & Burns, 2009 p. 9). 35 Single - case designs were created as a way to provide experimental control and internal validity with a small sample, similar to the control provided in randomized control - group experimental designs that consist of two randomized groups with one group receiving the independent variable and a control group no t receiving the independent variable (Horner et al., 2005). In addition, single - case design methodology addresses some of the weaknesses associated with randomized control trials and traditional significance testing. For example, Carver (1993) believed th at results from randomized control trials were frequently being reported as significant because they were statistically significant at a p value of less than .05. However, even though the results were statistically significant, when examined more closely, the intervention did not have a clinically important impact on the outcome variable for individual participants . Kazdin (1977) states that small changes in the outcome variable that are not seen a s an acceptable level of change or a socially valid level of change by the individual or those around the individual do not a ttest to the effectiveness of an intervention . Therefore, behavior change must be observable and acceptable in order for an intervention to be determined eff ective. Specific criteria for the evaluation of si ngle - case design for school psychology interventions include: reliability of measures , multiple methods of assessment, multiple sources of data, validity of outcome measures, quality of the baseline data, educational and clinical significa nce, durability of the effects, identifiable intervention components, intervention implementation fidelity, replication, and the setting in which the intervention took place (Kratochwill & Stoiber, 2002) . More recently, Kratochwill and colleagues (2010, p. 14 ) in conjunction with the What Works Clearinghouse specified the necessary components of single - case e standards (See Table 5 ) , including (a) a methodical 36 manipulation of the independ ent variable, (b) a minimum of three data poi nts per series, (c) a minimum of three different phase repetitions for multiple baseline de sign studies, and (d) consistent assessment of the outcome variable across time . Table 5 What Works Clearinghouse R equirements to B e R # Requirement 1. 2. 3. 4. The independent variable (i.e., the intervention) must be systematically manipulated, with the researcher determining when and how the independent variable conditions change. Each outcome variable must be measured systematically over time by more than one assessor, and the study needs to collect inter - assessor agreement in each phase and on at least twenty percent of the data points in each condition (e.g., baseline, intervention) and the inter - assessor agreement must meet minimal thresholds. The study must include at least three attempts to demonstrate an intervention effect at three different points in time or with three different phase repetitions. For a phase to qualify as an attempt to demon strate an effect, the phase must have a minimu m of three data points . To meet standards a multiple baseline design must have a minimum of six phases with at least 5 data points per phase. Taken from Kratochwill and colleagues (2010 p. 14 - 16). Multiple - baseline designs. Single - case designs have improved over time in order to more accurately measure outcome variables . These methodologies have historically used a repeated baseline A - B - A - B design design . In this design, A stands for the baseline phase, which is the phase in which the independent variable is not provided to the subject. Next, stands for the intervention phase, where the independent variable (e.g., intervention) is implemented. In order to monitor th e impact of the independent variable, change is measured between phase A and phase B. In addition, to ensure that the independent variable is the reason for the change observed between phase A and phase B, a r eturn to baseline (i.e ., A) and then another im ple mentation of the intervention (i.e ., B) is conducted (Baer , Wolf & Risley, 1968; Riley - Tillman & Burns, 2009) . However, this design is 37 not well suited for psychopharmacology research for several reasons, including: a) the theoretical lag effect for some classes of psychotropic medications between the start of a treatment and the treatment effects (Stahl, 1998); b) the potential side effects caused by the abrupt removal of a medication (Haddad, 1998); and c) the potential for learned social behaviors to c arry on past the removal of the medication (Riley - Tillman & Burns, 2009). In some psychopharmacological studies , a multiple - baseline design may be more appropriate . In this design, multiple subjects each r eceive the independent variable (B) , but do so at different times throughout the experi ment. For example, participant one may receive a baseline phase ( A) for four weeks and then an interv ention phase (B) for four weeks , while participant two receives a baseline phase (A) for six weeks and the interventio n phase (B) for two weeks (Riley - Tillman & Burns, 2009) . The goal of multiple baseline research is to determine if changes in the data coincide with changes in condition across participants when the independent variable is implemented regardless of when it is added (Hayes, 1981). Concurrent versus nonconcurrent multiple baseline designs. There are two variations of the multiple - baseline design . Concurrent multiple - baseline designs include a staggered onset of the treatment across participants, but require the data for each participant to be collected at the same time. Concurrent data collection is argued to ward of f threats to internal validity by controlling for temporal affects (Christ, 2007). However, a weakness of the concurrent multiple baseline design is it requires all participants to proceed through a study simultaneously. Watson and Workman ( 1981) proposed th e nonconcurrent baseline design, to address the difficulties that the concurrent baseline posed for researchers working in applied settings. The authors argue d that it was not feasible for studies in applied settings to collect data on participants at the same time due to practical issues such as recruitment factors. This is a problem 38 for recruitment in selective mutism studies, as the rarity o f the disorder prevents researchers from finding several participants who can start a treatment at an identical time. In the nonconcurrent procedure, the baseline and intervention lengths are determined a priori . Through this a priori specification , rese arc hers can have confidence that a treatment effect is occurring , as the probability improvement would begin at the start of each intervention phase a cross participants is very low. As participants become available, they are randomly assigned to one of the pre - specified baseline/trea tment schedules . Here, r andom assignment insures participants do not get assigned to a baseline/treatment schedule based on any factor (i.e., time of enrollment in the study, age, severity of symptoms) other than chance . Overall , a priori determination of baseline and treatment lengths, repetitive assessments of outcome data, and the multiple replications of a potential treatment effect, provide substantial evidence that the nonconcurrent baseline design can be used to draw meani ngful conclusions about a treatment approach (Christ, 2007). One weaknesses of this approach is that it does not allow baseli ne level of functioning to stabi lize before beginning the treatment phase. In cases where a child shows high levels of variability s data may not be interpretable (Watson & Workman , 1981 ). This shortcoming is typically not a concern in selective mutism studies, where children have demonstrated a lack of speech for a significant amount of time, often s everal years. There has been much debate in regards to the scientific merits of concurrent baseline designs in comparison to nonconcurrent baseline designs. Specifically, criticism has been raised regarding nonconcurrent multiple baseline designs that do not select the baseline and treatment lengths before the beginning of the study (Christ, 2007). However, Christ examined the qualities of nonconcurrent single - case designs in light of several possible threats to internal validity, 39 which include: a) history ; b) maturation; c) testing; d) instrumentation; statistical regression ; and, d) interactions. The author concluded that the Watson and Workman (1981) version of th e nonconcurrent multiple baseline design address es these concerns comparably to conc urrent d esigns , except in the area of mortality (i.e., participant drop out/removal) , as participants who fail to establish a steady baseline should be removed from the study. This poses a potential problem as these participants who have a varying baseline level m ay have certain characteristics that will not be accounted for in the final analysis. Visual analysis of single - case design data. V isual analysis is the first step in identifying treatment effects , and is defined as the examination of graphed data betwee n the baseline phase and the intervention phase (Kratochwill & Stoiber, 2002) . For example, in a study examining selective mutism, data that demonstrates an effect for a treatment may display a consistent low level of speech behaviors in the baseline, and higher and increasing level of speech behaviors in the treatment phase. According to Kratochwill and colleagues (2010) visual analysis should include an examination of six variables. First, level , which refers to the frequency or intensity of a behavior, is considered . In the case of selective mutism studies, this would include the lack of speech when speech is expected. Second, t he trend of the targeted behavior should be observed in the baseline consistently before a treatment phase should be initiated. This is to ensure that the behavior that is targeted is consistent, which allows more confidence that a change in the treatment phase is indeed the result of the treatment and not just a natural variation of that behavior. In single - case design studies exa mining selective mutism treatment, the establishment of trend during the baseline phase should not pose a challenge given these children have likely refrained from speaking in front of unfamiliar people for a substantial period of time . Third, the variabil ity of the behavior in the baseline phase should be examined. If the behavior is 40 unpredictable and fluctuates greatly over time, it will be much more difficult to determine the effect of a treatment variable . Once these three variables are observed in the baseline phase, they should also be examined in the treatment phase to determine if there is a change in the level, trend, and variability in the treatment phase compared to the baseline phase. Depending on the hypothesis of a study, a change in level, tre nd, and/or variability can indicate a change that is attributed to the onset of the treatment variable. Fourth, the researcher should examine the immediacy of the effect of the treatment on the behavior of the participant. Fifth, researchers engaging in vi sual analysis of single - case design studies should determine how much the data from the baseline phase and the treatment phase overlaps in order to determine if a meaningful change has occurred from the baseline phase to the treatment phase. Finally, the r esearcher should take the data for each participant and combine it with the visual data from the rest of the participants to determine whether or not the data meets evidenc e standards (Kratochwill et al. ). Interestingly, the very observable changes in the target behavior of speaking makes selective mutism an optimal condition to examine treatment effects under single - case design methods. The What Works Clearinghouse (WWC) Standards proposed by Kratoch will and colleagues (2010) set several guidelines in ide ntifying single - case design stu dies that display strong evidence. First, the study must demonstrate an effect across three different phases as determined by two WWC raters that are trained in single - case analysis . In addition to this, there must be no cont radictory evidence when examining the following features of the data : a) level; b) trend; c) variability; d) immediacy of the effect; e) overlap of the data; f) an examination of the change from baseline to treatment phase; and g) anomalies in the data tha t occur within a phase (e.g., a sudden and dramatic increase in a behavior mid - phase with a known explanation ). If there are three demonstrations of the effect, but one of the previous variables has been 41 identified as a potential problem in the data, the s tudy will be determined to only Likewise, if there are three participants who demonstrat e the expected change , and one or more participants who do not demonstrate the expected change , the study will only have Quantitative analysis of single - case design data. S everal methods to quantify the impact of change in single - case design studies that go above and beyond visual analysis have been proposed . Best practice currently dictates these approaches only be utilize d after determining a n intervention has at least moderate evidence based upon visual analysis procedures (Kratochwill et al., 2010) . Despite the option to quantitatively examine single - case design outcomes, there is much debate in the field regarding the m ost appropriate way to complete this task , and, to date, it appears no consensus has been reached (Kratochwill et al., 2013). However, strong arguments have been made for the use of single - case design randomization tests ( Ferron & Levin, 2014) and measurem ent of intervention effect size (Kratochwill et al., 2013). Randomization tests for the analysis of single - case design data. Randomization tests in single - case de sign research work similarly to randomiz ation tests in group research. The goal is to determi ne i f the probability a child displayed a better outcome during a certain period of time , such as during an active treatment, compared to a control period, such as during a placebo treatment, is greater than chance (Edgington, 1980). By calculating this st atistic, one can determine if positive outcomes were likely due the active treatment (Ferron & Levin, 2014). Single - case design randomization tests were deve loped several decades ago ( Ed g ington, 1980, Wampold & Furlong, 1981 ; Wampold & Worsham, 1986 ), and are capable of providing meaningful conclusions about treatment effectiveness. 42 The Wampold & Worsham (1986) technique appears to be the foundat ional randomization test for the quantitative analysis of multiple - baseline design data . A majority of the cur rent tests appe ar to be variants on this procedure , tweaked to solve various types of research questions . The Wampold and Worsham test requires all treatment schedules stagger the onset o f the treatment (e.g., active medication ), allowing the researcher to better rule out for "history" effects. Next, a test statistic is calculated. If there are K subjects, then there are K! number of potential permutations (i.e., potential orders of intervention onset). Then , a randomization distribution is created by compu ting W for K! permutations. These permutations are rank ordered, and the significance of the findings is calculated by the following formula: P(Type I Error)= =(number of Ws as large or larger than the obtained value) /k!. The test statistic can be compa red to an alpha of .05. As noted, v ariations on the Wampold & Worsham (1986) test appear throughout the literature. Marascuilo and Busk (198 8) used a similar concept to Wampold and Worsham; however, instead of using fixed staggered start points, each part randomly selected fro m all available time points in a particular treatment schedule to which the participant was randomly assigned . This increased the number of potential randomized start points, thus increasing the statistical po wer of the test. While there is a beneficial increase in statistical power, one drawback to this technique is individual s may not have enough of a baseline period to establish a consistent trend since any potential time point can be randomly selected as th e intervention start - point. Additionally, participants may not receive an intervention until the last few available intervention start points. This approach is problematic for psychopharmacological research (e.g., SSRIs) , as extended periods of time are of ten needed to show an effect. Koehler & Levin (1998) developed another variation of the Wampold and 43 Worsham approach. In this design, participants are first rand omized to a treatment schedule. In each treatment schedule, the researcher designates a block o f time points as potential start points. These potential start points are stagg ered for each participant so participants do not start the intervention at the same time. Next, a start point is randomly selected from a block of potential start points for eac h treatment schedule, thus improving the power of the statistical analysis . Again, while this increased power is desirable, it has drawbacks for research with psychopharmacological medications. Specifically, the issue of latency of effect negates the u tili ty of this design for psychopharmacology research. Because each child may take a slightly different amount of time to respond to the medication, the Koehler and Levin approach could make an effective medication look ineffective , given the focus on individu al time points . This design is most effective when there is an immediate expected outcome like when studying treatment response to psychostimulants ( M. Koehler, personal communication, 2012). Effect size calculations for single - case design data. As with g roup research, effect sizes can be used to quantify single - case design research after visual analysis indicates a treatment effect (Kratochwill et al., 2013). Some frequently utilized effect size approaches include the n o a ssumptions e ffect s ize , p ercentag e of n on - o verlapping d ata, p ercentage of a ll n on - o verlapping d ata , and R 2 (Riley - Tillman & Burns, 2009) . However, all four of these methods have been criticized in the literature for weaknesses. Parker and colleagues (2011) developed the Tau - U effect size to address the problems with the previously mentioned effect sizes . The Tau - U was created thr ough the combination of K endal l r ank c orrelation and the M ann - W hitney - U test. Tau - U improves on previous approaches by controlling phase A trend and combining no noverlap between phases with trend from within the intervention phase. The Tau - U approach consists of multip le components. The first optional step is to create a time series data difference matrix of 44 pairwise data comparisons between all of the data poin ts for a particip ant in a time forward direction, as a visual aid. Here, the left margin of the matrix contains the data series, while the top of the matrix contains the data series in reverse. In the matrix it is noted whether each comparison indicates a n increase in score (+), a decrease in score ( - ) or a tie (T). The first objective consists of determining the improvement trend in Phase B by inputting the available data into a formula determined to establish the Kendal rank correlation (KRC) coefficient : - L - U S ) / ( U L + U S ) = S / #Pairs = (#pos - #neg) / pairs = Tau. Second, t he same formula can be used to determine the effect size and significance o f improvement trend in Phase A. Third, overall improvement is calcu lated via the following formula by contrasting phase A against phase B, while including Phase B trend, and inputting the variables into the same formula: Tau=S/#pairs. Finally, overall improvement is calculated while controlling for Phase A improvement. To accomplish this calculation, the sign of Phase A trend is reversed followed by the recalculation of full trend. After the reversal of the sign, this procedure is completed via the following previously used formula: Tau=S/#pairs. Significance values, stand ard deviation, and z - scores for each task are obtained by inputting the previous information into a KRC m odel via a statistical program, such as SPSS. Single - case design in selective mutism psychopharmacology s tud i es. Although single - case design is a ve ry suitable approach for psychopharmacology research with children who have selective mutism, only two studies have used single - case design methodology for this purpose. First, Carlson and colleagues (1999) used a non - concurrent multiple baseline single - c ase design to examine the impact that sertraline had in the treatment of selective mutism. T he authors were able to identify change across phases with the addition of a medication treatment. This nuanced level of individual change would not have been detec ted with a randomized control 45 trial, as much more participants would be needed to run a t - t est or ANOVA to examine the average change across participants of children who received sertraline and those that did not. Second, Eke (2001) use d a two - group single - case design to examine a combined psychopharmacosocial treatment for children with selective mutism. In this study, the use of a single - case design allowed for an examination of behavioral a nd sertraline treatments for selective mutism. T o complete this t ype of study with a randomized control approach, many participants would have been needed, which would be difficult to recruit given the rarity of selective mutism. Moreover, the use of a single - case design allowed for the examination of the social validit y of the outcomes for each individual as opposed to a randomized control study, means across groups. The methods utilized in the current study are similar to t hose used by Carlson and colleagues (1999) examination of sertraline. However, the data analytic method s used in this project are more sophisticated given the use of visual analysis, the Wampold and Worsham (1986) randomization test, and the quantification of treatment effect using the newly developed Tau - U effect size (Parker et al., 2007) . Current Study When psychosocial approaches fail to yield meaningful outcomes, a psychopharmacological approach may be appropriate to improve functionality and increase well - being ( American Academy of Child and Adolescent Psychiatrists, 2009 ) . T herefore, the identification of psychopharmac ological interventions that can improve the quality of life of children affected with selective mutism children is necessary. Due to the current state of the literature, there is limited information available for practitioners and parents regarding the eff ectiveness of these treatments. This is troubling given that selective mutism has been linked to 46 difficulties with academics (Carbone et al., 2010) and later psychopathology (Steinhausen et al., 2006). Previous case studies (e.g., Bork & Snyder, 2013), a small open label study (Dummit et al., 19997), and a small - randomized control trial (Black & Uhde, 1994) suggest fluoxetine may be eff ective in the treatment of selective mutism with elevated social anxiety symptomology . The current study contributes to the greater understanding of the utility of fluoxetine for this indication through the use of a non - concurrent, multiple baseline, sing le - case design procedure, wi th a multi - gate data analy tic approach including visual analysis (Kratochwill et al., 2010), the Wampold and Worsham (1986) randomization test, and the Tau - U effect size approach (Parker et al., 2011) for quantifying the impact of fluoxetine treatment in five children between the ages of 5 and 14. Side effects were documented, and the relationship between speaking behaviors and the possible development of behavioral disinhibition associated with the fluoxetine was examined to ens ure that improvements were due to the hypothesized intended effect of the medication (i.e., social anxiety reduction). Treatment acceptability was examined in order to determine if this was perceived to be a reasonable intervention for families seeking tre atment for selective mutism. Research q uestions and h ypotheses . Question 1: W ill fluoxetine treatment lead to a significant reduction in social anxiety symptoms between the baseline/placebo phases and the treatment phase in five children di agnosed with selective mutism involving elevated levels of social anxiety symptoms ? According to the literature, selective mutism almost always occurs with high levels of social anxiety (APA, 2013; Black and Uhde, 1995; Dummit et al., 1997). Selective sero tonin reuptake inhibitors are hypothesized to address the biological component of selective mutism disorder by increasing serotonin levels. Fluoxetine at 20 mg/day has demonstrated efficacy with 47 anxiety disorders in children and adolescents, and is approve d by the FDA for o bsessive - c ompulsive d isorder in children (Birmaher et al., 2003; Eli Lilly and Company, 2009). Black & Uhde (1994) found that teachers rated children with comorbid social anxiety and selective mutism receiving fluoxetine 12 to 27 mg/day a s significantly more improved compared to a placebo - anxiety. In an open - label study, Dummit and colleagues (1996) found that there was a significant decrease in social anxiet y symptoms on the Liebowitz Social Anxiety Scale and the Liebowitz Social Avoidance scale for children taking fluoxetine 20 to 60 mg/day . Finally, there have been several case studies conducted that have reported a reduction in anxiety in association with fluoxetine treatment for selective mutism (e.g., Boon, 1994; Harvey & Milne, 1998; Silveira et al., 2004; Wright, 1995). Therefore, it was hypothesized there would be a significant effect of the fluoxetine treatment on social anxiety symptoms. The primary outcome measure of social anxiety in this study was Direct Behavior Ratings (DBRs) of social engagement with unfamiliar adults completed by parents. Data on s ocial anxiety symptoms wer e also collected using the following measures: 1) DBRs of social engage ment completed by teachers or an additional parent if a teacher was unavailable (e.g., during summer break); 2 ) the MASC - 2 Social Anxiety Scale (MASC - 2: SAS) completed by parents; and 3 ) the Clinical Global Impression (CGI) improvement and severity rating scales, which were completed by teachers, parents, the study psychiatrist, or both parents if one of the previous raters was unavailable (e.g., teacher during summer break ; see Table 6). 48 Table 6 Research Questions, Hypotheses, and Planned Assessment P rocedures Question Hypotheses Measures Will fluoxetine treatment lead to a significant reduction in social anxiety symptoms ? Fluoxetine treatment will result in a significant decrease in symptoms of social anxiety. Primary 1) Direct Behavior Ratings of Social Engagement with Unfamiliar Adults Parent Supplemental 1) Multidimensional Anxiety Scale for Children (2 nd Edition) Social Anxiety Scale Parent 2) Clinical Global Impression - Improvement Ratings (Global Shyness and Global Anxiety) - Parent, Teac her, Project Psychiatrist 3) Clinical Global Impression Severity Ratings (Shyness Severity and Anxiety Severity) Parent, Teacher, Project Psychiatrist 4) Direct Behavior Ratings of Social Enagagement - Teacher Will fluoxetine treatment increase the frequency of spontaneous speech and responsive speech? Fluoxetine treatment will lead to a significant increase in spontaneous speech and responsive speech. Primary 1) Direct Behavior Rating s of Responsive and Spontaneous Speech with Unfamiliar Adults Parent Supplemental 1) Selective Mutism Questionnaire Parent 2) Direct Behavior Ratings - Teacher 3) Clinical Global Impression Severity Ratings (Mutism Severity) Parent, Teacher, Project Psychiatrist 4) Clinical Global Impression - Improvement Ratings (Glo bal Mutism) - Parent, Teacher, Project Psychiatrist 5) Diagnostic Interview Project Psychiatrist 49 What adverse side effects will children experience during fluoxetine treatment? Is there a positive risk to benefit profile ? Child ren will not experience any serious adverse events, but will experience mild adverse events. The benefits of the fluoxetine treatment will outweigh negative side effects. Primary 1) Adapted Side Effect Form for Children and Adolescents (SEFCA) Paren t 2) Clinical Global Impression Ratings (Global Side Effect Severity) Parent and Project Psychiatrist Supplemental 1) Medication Management Meetings - Project Psychiatrist Does onset of active fluoxetine medication correspond with an increase in beh avioral disinhibition? The onset of active fluoxetine medication will not correspond with an onset of behavioral disinhibition. 1) Young Mania Rating Scale Parent Will parents find the use of fluoxetine for the treatment of selective mutism accepta ble? Parents will find the use of the fluoxetine treatment as acceptable for the treatment of selective mutism. 1) Treatment Evaluation Questionnaire, Acceptability Scale - Parent Note: Adapted from Carlson and colleagues (1997) and Eke (2001) 50 Quest ion 2 : Will the fluoxetine tre atment increase the frequency of spontaneous speech and responsive speech across the baseline/placebo phases and the treatment phase in five children diagnosed with selective mutism involving elevated levels of social anxiety symptoms? According to the model of selective mutism proposed by Carlson and colleagues (2008), underlying biological aspects of anxiety may play a role in the reluctance to speak in children with selective mutism . By treating the anxiety, it can be expe cted that speaking behaviors will increase . Several case studies suggest fluoxetine may be effective in increasing the frequency of speech in children with selective mutism ( e.g., Bork & Snyder, 2013; Silveira et al., 2004 ). Additionally, a small randomize d control trial ( Black & Uhde 1994) and an open label study (Dummit et al., 1997) found parents perceived improvement in speaking behaviors after treatment with fluoxetine at 12 to 60 mg/day . However, in these studies , ratings of speaking behavior were ma de infrequently , and raters were not asked to differentiate between spontaneous speech (e.g., initiating speech with others) and responsive speech (e.g., answering a question posed directly to the child). Delineating between spontaneous speech and responsi ve speech is essential, as some have argued that spontaneous speech is a better criteria to examine whether a child is engaging in speech for communication, which is the goal of treatment (Pionek Stone et al., 2002). Carlson and colleagues (1999) used sert raline, another SSRI, to treat selective mutism and found that there were increases in both spontaneous and responsive speech. Therefore, it was hypothesized that children would display a significant increase in all types of speech behaviors in the school and community setting. This study adds to the literature by requiring parents and teachers to provide Direct Behavior Ratings (Chafouleas, Riley - Tillman, & Christ, 2009 ) multiple times per week in the school and community settings. Direct Behavior Ratings completed by parents are the primary outcome variable; however, several other assessments will 51 be used to collect supplementary data on speaking behaviors , including: 1 ) the parent completed Selective Mutism Questionnaire (SMQ: Bergman et al., 2008) ; 2 ) CG I ratings provided by parent s teachers, and the psychiatrist ; and 3 ) a psychiatrist administered diagnostic interview for selective mutism (see Table 6) . Question 3: What adverse side effects, if any, will children experience during the fluoxetine treatm ent? Is there a positive risk to benefit profile for children taking fluoxetine for selective mutism ? SSRIs have a relatively low risk of moderate to severe adverse effects (Vaswani, Linda, & Ramesh, 2003). However, previous research indicates there are side effects associated with the use of SSRIs in children, including behavioral problems (i.e., behavioral disinhibition) , somatic complaints (e.g., headache, upset stomach), and mood difficulties (e.g., increased anxiety). Safer (2011) argue d that it is e ssential to examine the side effects associated with psychopharmacological agents in children as children may experience different side effects or more intense side effects than adults. A lthough the previous research on the use of fluoxetine for selective mutism provides evidence for the overall safety of this approach , it was hypothesized that children would experience mild to moderate adverse events such as upset stomach and a reduction in appetite at the onset of treatment. However, it was hypothesized t hat children would not experience any serious adverse events from the fluoxetine treatment. The frequent assessment of side effects in this study illuminate s when and if side effects occur ed as well as when side effects dissipate d during the course of tre atment. It was hypothesized that if adverse effects were experienced the negative effects would not outweigh the benefits of the fluoxetine treatment such as improved anxiety ratings and an increase in speech production. In order to examine the adverse eff ects that were experienced by the children taking fluoxetine, a modified 52 version of the Side Effects Form for Children and Adolescents (SEFCA) that examines the most common side effects specific to the use of SSRIs with children was used. In addition, info rmation provided to the psychiatrist regarding possible side effects during the bi - weekly medication management meeting was reported. Finally, the CGI - Global Side Effects Severity rating scale completed by parents on a weekly basis and the study psychiatri st on a bi - weekly basis was used ( s ee Table 6). Question 4 : Does onset of active fluoxetine medication correspond with an increase in behavioral disinhibition? Children treated with an SSRI may sometimes experience behavioral disinhibition (Walkup & Label larte, 2001). When determining whether an SSRI has successfully treated selective mutism, it can be difficult to identify what behaviors can be attributed to the intended effects of the SSRI (e.g., anxiety reduction) and what behaviors can be attributed to the unintended effects of the SSRI ( i.e., behavioral disinhibition ) , as the outcomes may appear similar (e.g., increased speech) . Given this issue, Carlson and colleagues (2008) state that it is essential to examine the possibility that behavioral disinhi bition may be occurring when examining the impact of SSRI treatment for selective mutism. Despite the possibility that behavioral disinhibition may arise as a result of the fluoxetine treatment, it was hypothesized that behavioral disinhibition w ould not o ccur after the onset of fluoxetine treatment . The literature indicates that the occurrence of behavioral disinhibition for children treated with SSRIs is rare. For example, Wilens and colleagues (2003) found that the development of behavioral disinhibition as a result of SSRI treatment was only noted in 6% of children who received treatment with an SSRI at a pediatric clinic. Additionally, in a review article of studies examining the use of SSRIs in children, Safer and Zito (2006) reported that only 10.7% o f pre - 53 adolescent children and only 2.1% of adolesc ents in reviewed studies experienced this side effect. If behavioral disinhibition occurred , it was hypothesized it would not solely coincide with improvements in social engagement and speaking behaviors . B ehavioral disinhibition was identified via the parent completed adapted version of the Parent - Young Mania Rating Scale ( Appendix A ; Gracious, Youngstrom, Findling, & Calabrese, 2002 ; see Table 6 ). Question 5 : Will parent s find the use of fluoxetine for th e treatment of selective mutism with elevated social anxiety symptoms acceptable? Prescription rates of psychopharmacological agents for the treatment of internalizing disorders in children are on the rise (Debar et al., 2003). Therefore, understanding pa rent satisfaction of this intervention approach is essential. According to Kazdin (1977), an intervention is effective only if stakeholders can identify a socially valid level of change ; therefore, a behavioral change needs to be both observable and accept able . Carlson and colleagues (1999) and Eke (2001) examined the use of sertraline, another SSRI, for the treatment of selective mutism . Results from both studies indicated that parents found the intervention highly acceptable . Given these previous findings and the hypothesized improvements in social anxiety and speech in this study , it was hypothesized that parents would find fluoxetine as an acceptable approach for the treatment of selective mutism. Treatment acceptability was assessed by the Treatment Eva luation Questionnaire Parent (TEQ - P; Kratochwill, Elliot, Loitz, Sladeczek, & Carlson, 2003 ; see Table 6 ). 54 CHAPTER 3 METHODS Participants Participants (N= 6 ) were recruited by sending out letters via email and the postal service to mental he alth professionals (N= 874 ) within a 90 - mile radius of East L ansing, Michigan (see Appendix B ). Each letter contained an attached flyer for the mental health practitioner to provide to parents of children between the ages of five and 18 who appeared to mee t the diagnostic criteria for the disorder (see Appendix C ). First, the letter and attached flyer were sent to members of the Michigan Association of School Psychologists (MASP: N= 130 ), National Association of School Psychologist s (NASP: N= 133 ), and Americ an Psychological Association (APA: N= 201 ) in the specified radius. Second, local mental health practitioners in private, group, and community mental health practices (N= 410) , including psychologists, psychiatrists, and social workers, in this radius were identified through an on - line search and sent the letter and attached flyer. Finally, p rincipals at school districts within a 90 - mile radius were also sent a letter (N= 2,129 ; see Appendix D ). This mailing included the letter for mental health practitione rs and the letter for parents. In the letter, principals were asked to pass the mental health practitioner letter and parent flyer to a mental health professional within their school. Twenty - one potential participants were identified via these methods. Of the 2 1 , four children were not accepted into the study because they were previously or currently taking medication for selective mutism symptoms (n=2 fluoxetine; n=2 sertraline), four did not want to participate in the study because of travel, two could no t participate because they were from out of state (Arizona and Colorado), two were averse to medication treatments and preferred a psychosocial ap proach, one spok e English as a second language , and two families did not respond to multiple attempts to sched ule 55 a screening phone call. Of the six remaining eligible children , three became aware of the study via letters sent to school principals, while the remaining three were informed of the study through a psychologist in private practice who specializes in se lective mutism treatment. Five females and one male between the ages of 5 and 14 were enrolled after qualifying based on the inclusion and exclusion criteria (Appendix E ; see Table 7 ). Despite meeting all screening requirements, ratings of social engageme nt and speaking behaviors indicated th at one of the children, who was initially assigned to treatment schedule three, was consistently engaging in the desired behavior prior to the onset of the fluoxetine medication. A stable baseline trend indicating sele ctive mutism symptomology could not be established for her across measures, necessitating her removal from the analysis of treatment outcomes per the guidelines provided by Watson and Workman (1981 ) and Christ (2007 ; see Figure 2). For the remainder of thi s paper, This treatment schedule was re - opened and a nother participant was invited to participate in the study. The remaining five children who completed the screening process and qualified per the inclusion and exclusion criteria decided to participate and completed treatment (see Table 8 ). All five children began displaying symptoms at 4 or 5 years of age, and received some form of behavioral or cognitive behavioral therapy before enrolling in the study provided by private practitioners and school mental health professionals. All met the age - related duration criteria for prior psychosocial treatment (i.e., 10 weeks for children seven - years - old and older; 12 weeks for children six - years - old and young er) . Previous practioners for all participants reported resistance to psychosocial treatment . Child Four continued to receive cognitive - behavioral therapy on a weekly basis throughout the study. Given he was not experiencing improvement with cognitive - be havioral therapy alone according to his therapist , he was allowed to enroll in the study while continuing to receive psychosocial 56 Table 7 Demographic Characteristics and Baseline/Intervention Ratings for Six Participants Child Age/ Sex Age of Onset Rel ative w/ Anxiety MASC - 2: SAS Performance Fears - Screen Previous Treatment Concomitant Treatment One Two Three Four Five R* 7/F 5/F 7/F 12/M 14/F 10/F 5 4 5 5 5 4 Sister Undiagnosed Social Anxiety Fath er Undiagnosed Generalized Anxiety Sister - Undiagnosed Social Anxiety Father Generalized Anxiety Disorder Father Generalized Anxiety Disorder Mother Generalized Anxiety Disorder T = 80** T = 76** T = 76** T = 80 T = 83 T = 69 Behavior Therapy; Social Skills Behavior Therapy Behavior Therapy; Social Skills Cognitive - Behavioral Therapy Behavior Therapy Behavior Therapy None None None Cognitive - Behavioral Therapy None None *Participant removed from final an alyses due to inconsistent baseline ratings across measures **Based on 8 - year - old normative sample. 57 58 Table 8 Pre - Treatment, During Treatment, and Post - Treatment Ratings by Treatment Schedule MASC - 2 SMQ Behavioral Disinhibition End of Study SM Diagnosis TEQ - P Rating Screen Placebo Mean (SD) End of Treatment Screen Placebo Mean (SD) End of Treatment Child One 20 21.25 (0.5) 27 13 12 (0) 24 No Yes 60 Child Two 25 25 (1.58) 26 5 8.8 (3.03) 11 Yes Week s 6, 7, and 10 Yes 56 Child Three 20 23.25 (2.85) 25 11 12 (0) 23 Yes Weeks 13 and 14 Yes 60 Child Four 18 23 (0.63) 19 19 N/A* N/A* No Yes 62 Child Five 27 26.27 (0.65) 26 17 10 (0) N/A* No Yes 65 *School ratings not available for the SMQ due to summer vacation 59 treatment. All but one of the participants had never taken any kind of psychopharmacological medication prior to enrollment in the study. Child Four was previously treated under the care of his pediatrician with a low dose (10 mg/day) of fluoxetine for approximately four weeks over three years prior to enrolling in the research study. Given the medication treatment was discontinued quickly without the opportunity to have a lasting impact on functioning, this child was allowed to participa te. Measures The dependent variables of this study were examined through the use of several measures completed once per school day (DBR - teacher), three times per week (DBR - Parent), twice per week (i.e., Clinical Global Impression Ratings Parents/Teach ers, MASC - 2: SAS, SMQ, P - YMRS), bi - weekly throughout the project (Clinical Global Impression Ratings Study Psychiatrist, SEFCA), and at the end of treatment (DSM - IV - TR Diagnostic Interview for Selective Mutism; Treatment Evaluation Questionnaire - Parent; See Table 9 ). See Appendix F for a correlation matrix of all parent - completed measures of anxiety and mutism symptoms. In addition, other forms were on hand to collect supplemental information regarding treatment integrity as well as examine the perceptio ns of parents and participants if a child was withdrawn in the middle of the study or when a participant completes the study. These forms included the medication compliance form (See Appendix G), which was completed everyday by the parent, the early withdr awal form (see Appendix H), which did not need to be used as no children withdrew from the study early, and the End of Study Form (see Appendix I), which was completed at the end of the study. The measures used to identify the utility of fluoxetine for th e treatment of selective mutism with elevated social anxiety symptoms are discussed below. 60 Table 9 Assessments Planned for Each P hase Phase Assessment Plan Screening Baseline/ Treatment End of Treatment 1) Phone Interview - Project C oordinator 2) Diagnostic Interview Project Coordinator and Project Psychiatrist 3) Medical/Psychosocial History Form - Parent 4) Multidimensional Anxiety Scale for Children (2 nd Edition) Social Anxiety Scale - Parent 5) Parent - Young Mania Rating Scale Parent 6) Physical Exam Project Psychiatrist 1) Direct Behavior Rating Parent (3x per week) 2) Parent - Young Mania Rating Scale Parent (2x per week) 3) Adapted Side Effect Form for Children and Adolescents Psychiatrist (bi - weekly) 4) Medication Compliance Form Pare nt (daily) 5) Multidimensional Anxiety Scale for Children (2 nd Edition) Social Anxiety Scale Parent (2x per week) 6) Selective Mutism Questionnaire Parent (2x per week) 7) Direct Behavior Rating Teachers (5x per week; once per school day) 8) Clinical Global Impression Ratings Parent/Teacher (2x per week) 9) Clinical Global Impression Ratings - Project Psychiatrist (bi - weekly) 1) Treatment Evaluation Questionnaire Parent 2) Diagnostic Interview Project Psychiatrist 3) End of Study Form Parent Note: A dapted from Carlson and colleagues (1997) and Eke (2001). Social anxiety . Direct behavior rating (DBR) : Social engagement . Direct Behavior Rating (DBR; Chafouleas, Riley - Til lman, & Christ, 2009; Appendix J ) is an assessment technique that can be complete d by observers, such as teachers, parents, and school psychologists. The procedure requires observers (e.g., teachers, parents) to provide a rating of an operationally defined behavior (e.g., social interaction) on a scale (e.g., 1 to 10; Never to Always; 0% to 100%), during a pre - specified period of time (e.g. hour, day, class period). DBRs are designed to supplement or be an alternative to systematic direct observations (SDO) , which often require extensive 61 resources to complete over multiple periods of t ime (Chafouleas et al.). DBRs are a less resource intensive solution when SDOs cannot be completed , as SDO s and DBRs have been shown to be significantly correlated (r p < .01) , with regression analysis reveal ing DBR ratings accounted for 76% of the variance of SDO ratings (Riley - Tillman, Chafouleas, Sassu, Chanese & Glazer, 2008). Research has demonstrated that DBRs have the capacity to accurately quantify targ et behaviors. In a study examining the concurrent validity between the Social Skills Rating System (SSRS) and DBRs in identifying problem behaviors of kindergarten students, the authors found significant correlations ranging from 0.28 to 0.88, with most co rrelations significant at the p < .01 level (Chafouleas, Kilgus, & Hernandez, 2009). Repeated SDOs in the school and community context were not feasible due to several barriers. For example, training parents to document social engagement and speaking beha viors in the community using a SDO method would be challenging and possibly inaccurate given their inexperience with this complex assessment procedure. Further , funding and distance prevented study personnel from conducting SDOs at schools multiple times p er week for each child . The alternative to have teachers videotape classroom activities was considered but deemed impractical within this dissertation study. Parents completed two ratings on a ten - point scale three times per week ranging from zero (Never) to 10 (Always) to address social anxiety ( e.g., my child appeared comfortably and socially engaged with other unfamiliar adults , with one rating focusing on social engagement with unfamiliar adults and the other addressing social eng agement with friends . Parents were instructed not to base the ratings targeting social engagement on spe aking behaviors, as the purpose of these items were to determine frequency and ease of engagement not frequency of speech . Children in this study were consistently socially engaged with their friends during the baseline phase , which makes 62 sense given the familiarity participants had with these children. Therefore, parent ratings of social engagement with unfamiliar adults were the primary outcome measure for social anxiety symptomology. Teachers also completed supplementary DBR ratings of social engagement on a daily basis using the same 10 - point scale and similar questions tailored for the classroom ( e.g. , nd socially engaged with the classroom Given DBRs are designed to track behaviors in individual children, no standardized scoring system has been developed. In this study, higher ratings indicate a higher rate of social engagement . Multidimensi onal anxiety scale for children 2 nd e dition (MASC - 2) Social anxiety scale (SAS). The M ultidimensional Anxiety Scale for Children, 2 nd Edition: Social Anxiety Scale (MASC - 2: SAS ; March, 2013; Appendix K ) is a nine - item parent - report measure used to asse ss social anxiety symptoms in children between the ages of eight and 19 . Given the lack of norm - referenced assessments for social anxiety designed for children younger than eight years of age, t he MASC - 2 was used as a supplementary measure of social anxie ty for all children in this study given face validity . The MASC - 2: SAS has excellent test - retest reliability (r=. 90, p < .001), and Parent (CBRS - P; r= .55, p < . 01). Pa rents completed the MASC - 2: SAS twice per week, and provide d wit - 2: SAS raw scores can be converted into T - s cores for comparison to same - aged peers. When T - scores were required (i.e., inclusion criteria) for children younger than eight, eight - year - old norms were used as an approximation of the sever ity of symptomology. In 63 this study, raw scores will be used to track progress over time, with higher scores indicating higher levels of social anxiety symptomology. Clinical global impressions (CGI) : Anxiety/Shyness . An adapted Clinical Global Impressions scale ( CGI; National Institute of Men tal Health, 1985; See Appendix L ), which was used by Carlson and colleagues (1999) to examine the effectiveness of sertraline with children diagnosed with selective mutism, was used as a supplementary measure to examin e the impact of fluoxetine on anxiety and shyness symptoms. Ratings were provided by parents and teachers twice per week and by the study psychiatrist bi - weekly. Research has demonstrated that the CGI has adequate reliability and validity. For example, in an adapted version of the CGI to examine the symptoms of social anxiety disorder, correlations between the CGI and the Social Interactions Anxiety Scale were all significant at the .01 level, with correlations ranging from .44 to .74 (Zaider, Heimberg, Fre sco, Schneir, & Liebowitz, 2003). In this study, p arents provided shyness and anxiety severity ratings on five - point L ikert scale s from one (absent) to five (severe), across the home, school, and community settings , while teachers provided the same ratings only for the classroom setting. The study psychiatrist also completed anxiety and shyness severi ty r atings based on parent report and observations during medication management meetings. Parents and psychiatrists provided global change ratings for the cons tructs of anxiety and shyness by using seven - point L ikert scale s , which range d from one ( much improved) to seven ( much worse) . Teachers complete d five - point l ikert - scale s on global shyness and anxiety change that range d from one ( much improved) to five ( mu ch worse) . Given CGIs are global ratings of improvement and severity, there is no standardized way of scoring CGIs aside from the ratings themselves. Higher ratings across severity and improvement scales indicate worse symptomology. 64 Frequency of speech. Direct behavior rating (DBR) : Speech . The same DBR technique ( Chafouleas et al., 2009 ; Appendix J ) used to identify levels of social engagement was also used to evaluate speaking behaviors . Parents provided ratings for spontaneous and responsive speech with unfamiliar adults and friends three times per week ( spontaneously spoke ) on a scale of zero (never) to 10 (always) during situations where speaking behaviors would be expected . Given participants spoke freely with friends on a regular basis at the beginning of the study , parent ratings of responsive and spontaneous speech with unfamiliar adults were used as the primary outcome measure. For children who received treatment during the school year, te achers provided supplementary ratings of responsive and spontaneous speaking behaviors in the classroom on a daily basis When appropriate, the student spontaneously spoke ) on the same scale (i.e., zero to 10). Selective mutism que stionnaire (SMQ). The Selective Mutism Questionnaire (SMQ; Berg man et al., 2008; See Appendix M ) is a 17 - item parent - report assessment used to determine the degree of mutism symptoms a child is experiencing, and was completed by the parent twice per week a s a supplementary measure of speaking behaviors . Research has demonstrated that the selective mutism questionnaire has adequate reliability and validity. Bergman and colleagues found that internal consistency ratings for the SMQ were excellent and ranged f rom .88 on the Home/Family scale to .97 on the School scale, with an overall internal consistency coefficient of .97. In addition, Bergman and colleagues found that the SMQ was sensitive enough to capture the impact of treatment. Moreover, Letamendi and co lleagues (2008) found that the SMQ correlated with the Anxiety Disorders Interview Schedule for Children for DSM - IV parent Version (ADIS/CP) with a correlation coefficient of 0.48 suggesting that the measure has 65 adequate convergent validity. Parents provi de ratings of each speaking behavior in each setting as occurring never (0) , seldom (1) , frequently (2) , or often (3 ). For example, one statement is Lower scores on the Sc hool, Home, and Community and Total scales indicate more significant symptomology, with average scores for children with selective mutism on the Total scale ranging from 13.18 for three to five year olds to 15.73 for nine to eleven year olds . Clinical g lo bal i mpressions (CGI) : Mutism . The same adapted CGI technique (National Institute of Mental Health, 1985; s ee Appendix L ) used to gather information on shyness and anxiety symptoms was used to collect supplementary information on mutism symptomology at ho me, in the community, at the clinic, at school , and overall . Ratings wer e provided by parents and teachers twice per week and the study psychiatrist bi - weekly. Parents and the study psychiatrist provided selective mutism severity ratings on a scale of one ( absent ) to five ( severe ), while providing global mutism change ratings on a scale of one (much improved) to seven (much worse). For children who received treatment during the school year, t eachers also provided mutism severity ratings on a scale of one ( a bsent) to five (severe) and global change ratings of mutism behaviors on a scale of one ( much improved ) to five (much worse) . Adverse events. Clinic al global impressions (CGI): Side effects . Parents, two times per week, and clinicians, bi - weekly, provid ed side effect severity ratings using the same CGI (see Appendix L ) approach discussed for social anxiety and mutism behaviors in order to examine the perceived risk to benefit profile of fluoxetine for each of the participants. Ratings ranged from one ositive changes greatly outweigh negative changes. Medication effects are overall extremely 66 Side effects form for children and adolescents (SEFCA). A 12 - item adapted version of t he Side Effects Form for Children and Adolescents (SEFCA; Klein , Abikoff, & Barkley, 1994 ; See Appendix N ) was used to examine the side effects participants experience d while taking fluoxetine and was administered bi - weekly by the study psychiatrist . This original SEFCA consists of 54 - items that inquire about the frequency and severity of side effects of several different classes of psychopharmacological medications . If an item is endo rsed, the adminis t rator of the rating scale ha s the parent rate the severity from one to three, with one being defined as mild and three being defined as severe. Side effects examined on the SEFCA include the broad categories of cardiovascular , gastrointestinal, central ne rvous system, ocular, mouth and nose, genito - urinary, dermatology, and musculo - skeletal side effects. The measure does not have psychometric properties. However, it has been used frequently in published studies examining psychotropic medication use in chil dren (e.g., Birmaher et al., 2003). Scoring is completed on an item level basis, allowing for the identification of types and severity of side effects each child experienced. In order to reduce the data collection burden on parents, an adapted version of the SEFCA that only includes adverse events associated with SSRIs that occur in 5% or more of children and adolescent s (Wilens et al., 2003), such as drowsiness, difficulty falling asleep, and irritability , w as used. In addition, an item capturing behavior al disinhibition w as also included as this kind of adverse event is important to note as a possible confounding variable (Carlson et al., 2008). 67 Behavioral disinhibition . Parent v ersion of the y oung m ania r ating s cale (P - YMRS). In addition to the item on the adapted SEFCA completed by the study psychiatrist, parent ratings on the P - YMRS were also examined to identify the potential occurrence of behavioral disinhibition. Gracious and colleagues (2002) developed t he Parent Version of the Young Mania Rat ing Scale (P - YMRS; Appendix A ) as an adaptation of the Young Mania Rating Scale ( Young et al. , 1978 ). The P - YMRS , which w a s completed twice per week by parents, consists of 11 multiple - choice questions and was designed to examine manic symptoms in children ages five to 17. For example, the P - definite elevation - more optimistic , self - confident; cheerful, appropriate to t Gracious and colleagues (2002) found that the P - YMRS has an internal consistency of .75, and has a dequate discriminative ability when identifying children who have bipolar disorder compared to other diagnoses such as depression. Scores on the P - YMRS range from 0 to 60, with higher scores indicating greater psychopathology and a score of 21 or higher in dicating a likely episode of mania (Gracious et al.). Since the purpose of the P - YMRS in this study was to identify possible behavioral disinhibition associated with the onset of fluoxetine treatment, this measure was not scored conventionally . Instead, r esponses that may indicate the onset of manic symptom have been determined, a priori, and hig hlighted in yellow in Appendix O . Three items have been removed for this study, as they have been deemed inappropriate for the age group. First, an item about appe arance was determined to be inappropriate due to the young age of some of the children . Second, an item regarding insight 68 into manic symptoms was removed , as these children are not expected to have manic symptoms at the beginning of the study. Finally, an item about sexual interest was removed due to the anticipated age of the majority of participants . Treatment acceptability . Treatment e valuat ion questionnaire - acceptability scale (TEQ - P ) . The acceptability scale on the Treatment Evaluation Q uestionnair e , parent version (TEQ - P; Kratochwill et al . , 2003 ; See Appendix P ) was used to examine treatment acceptability as rated by the parents of the participating children at the end of the study . The TEQ - P was developed from the Treatment Evaluation Inventory ( TEI; Kazdin, 1980) , which was developed by conducting a factor analysis on data gathered piloting the measure with college students rating the appropriateness of various treatment options for externalizing behaviors . Kazdin (1980) found that the TEI was ab le to delineate between treatments that individuals considered acceptable and not acceptable. T he TEQ - P consists of 21 statements that parents are asked to rate on a six point Likert scale . However, only the 11 questions associated with the acceptability s cale w ere provided to parents, as these are the items most closely aligned to the research questions in this study. Each statement is rated on a Likert scale with a rating of one being strongly disagree and a rating of six being strongly agreed . Exampl es of questions on the modified TEQ - P acceptability scale include d : use of this intervention to other par Possible scores on the acceptability scale range from 11 to 66. A score of 55 or higher has historically been used to indicate high treatment acceptability (Kratochwill et al., 2003). Carlson and colleagues (1999) and Eke (2001) used the TEQ - P to examine the acceptability and effectiveness of SSRI treat ments for children with selective mutism. Carlson and colleagues (1999) and Eke (2001) found high treatment acceptability in 69 studies examining the use of sertraline, with average scores of 58.6 and 56.8, respectively. Therefore, using this measure to exam ine treatment acceptability created continuity with previous studies that have examined parental treatment acceptance for children with selective mutism who are being treated with a psychopharmacological approach. Medication compliance measure. Parent s re port ed medication compliance on a form requiring them to document the time they provided the medication each day (e.g., Monday at 8:00AM ; see A ppendix G ) . This allowed for the examination of missed doses, and provided insight into treatment acceptability , as higher rates of compliance indicate d parents were successful at meeting the requirements of the treatment procedure. This data was compiled in the form of the percentage children were provided their medication within 6 hours of the recommended time. In addition, this information was helpful to examine in conjunction the timing of the onset of treatment effects. For example, if a child misse d a significant amount of doses, the treatment outcomes may have become difficult to interpret without compliance da ta. There is no reliability and validity data available on this measure, as it was designed specifically for this study. End of study f orm . The End of Study Form (Carlson, 1999; Appendix I ) inquired about follow up care and parent perceptions of the po sitive and negative aspects of participation . This brief interview provides additional information on treatment satisfaction , while also ensuring children received appropriate follow up care. No reliability or validity data exists for this interview. Proc edures The Michigan State University Institutional Review Board (MSU - IRB) approved the procedures used in this study. Of note, the MSU - IRB required a number of conditions be met 70 before granting approval. First, children could be no younger than five ye ars of age. At the start of the study, the lower age limit was seven - years - old, which was decreased to five after selective mutism experts wrote letters to the IRB describing the common practice of prescribing SSRIs to young children with selective mutism and the need for systematic data collection on this approach. Second, for children younger than seven, the project psychiatrist was required to write a treatment summary following every medication management meeting. These treatment summaries along with th e most recent participant data forms were provided to a safety committee, which was composed of a pediatrician and a child and adolescent psychiatrist who were not affiliated with the project in any other way. The two members of the safety committee indivi dually reviewed the information after each medication management meeting, and approved all of the medication decisions made by the psychiatrist. Finally, the IRB stipulated a maximum dose of 20 mg/day. In addition to these requirements, all project personn el completed the IRB training regarding responsible practices in research. Project personnel . Project c oordinator . The project coordinator was a Michigan State University graduate stud ent in school psychology who is completing this project in partial fulf illment of the requirements to obtain a Doctorate of Philosophy in School Psychology. The project coordinator was responsible for the following: (a) organizing and completing participant recruitment and screening efforts; (b) contacting schools, parents, and mental health professionals; (c) explaining the treatment procedure to parents and teachers ; (d) training the project assistants , psychiatrists, teachers, and parents ; (e) organizing the responsibilities o f the project assistants ; (g) meeting with fami lies before their medication management meetings (h) obtaining consent and assent; (i) 71 distribution and collection of data from parents, teachers, and project psychiatrists ; (j ) visual analysis of graphs; and k) organizing and compiling the data . Project a ssistants . The project assi stants were four Michigan St ate University graduate student s who completed the requirements to receive a Masters of Arts degree in school psychology. The primary responsibi lity of the project assistants was visual analysis of g raphs. The project coordinator also delegate d some of his responsibilities to project assistants in the scenario of scheduling conflicts or other events that prohibit ed h im from carrying out his duties. Thes e responsibilities include d : a) meeting with fami lies before their medication management m eetings; b ) distribution and collection of data from parents, teachers, and project psychiatrists; and c ) organizing and inputting data obtained from the various outcome measures. The project assistant s w ere reimbu rsed for hours spent working on the project (i.e. , $10 per hour) . Study psychiatrist. The study psychiatrist was a resident at the Michigan Sta te University Psychiatry clinic, working under the supervision of a board certified child and adolescent psychia trist. The study psychiatrist was responsible for the following: (a) confirming a diagnosis of selective mutism; (b) a physical evaluation; (c) prescription of fluoxetine and dosage decisions; (d) biweekly medication management meetings; (e) monitoring of adverse treatment effects; and (d) admin istering assessments (e.g ., CGI ; SEFCA ). The study psychiatrist received incentives for collaboration in this project including authorship on any publications that arise from this st udy as well as $250 for books for a professional library. T raining . Study psychiatrist. T he project coordinator met with the study psychiatrist prior to the study to discuss the protocol in detail . During this ti me, the scope and goals of the study, the multiple baseline design procedu re , the responsibilities of the psychiatrist , the Clinical Global 72 Improvement rating system, the SEFCA, and the goals of the bi - weekly medication management meeting were discussed . Project a ssistants. T he project coordinator met with the project assistan ts prior to their participation in the project . The following topics were discussed: ( a) the responsibilities of the project assistant, ( b) meetings with participants, and (c ) visual analysis of single - case design data . P roject assistants received addition al training on visual analysis of data. They were asked to read the Kratochwill and colleagues (2010) guidelines for the analysis of single - case design data . Next, there was a n additional meeting to en sure an accurate understanding of visual analysis prin ciples. During this meeting, the following variables were discussed: a) level; b) trend; c) within phase variability; d ) proportion of overlap; e ) comparison of baseline and intervention phase data; e) immediacy of effect ; and f ) anomalies in the data. Pro ject assistants were each asked to code simulated data for 10 participants. The project coordinator reviewed the rati ngs and discuss ed discrepancies between ratings provided by the project coordinator and the project assistants . Project assistants were giv en more data to code until a minimum of . 9 0 reliability was achieved. Teachers . Teache r s were required to me e t with the project coordinator in person or by phone to receive training on DBRs that included the following components based on research conduc ted by Chafouleas, Kilgus, Riley - Tillman, Jaffery, & Harrison (2012): a) modeling of the rating procedure on pre - recorded video tapes displaying children engaging in simulated selective mutism behaviors; b) practice and feedback rating six 1 - minute videota ped recordings; c) frame - of reference training (e.g., viewing a performance and discussing the important aspect of a performance); and d) a discussion of common rater errors (e.g., halo effect, central tendency). Teachers were provided with a $50 Target gi ft card for school supplies for their time. 73 Parents. Parents also met with the project coordinator during their first meeting to the MSU psychiatry clinic to review the assessment measures. D uring this time, parents received the same training on DBR ra tings as teachers. Additionally, they were instructed on the completion of all of the other required forms. In order to compensate parents for their data collection efforts, medication management appointments at the MSU psychiatry clinic and the fluoxetine medication was provided without charge . Additionally, parents were reimbursed for mileage accrued while driving to the clinic , with those driving less than or equal to 25 miles receiving $100, those traveling 26 to 50 miles receiving $150, and those drivi ng 50 miles or more receiving $270. Study Phases Project o verview . Participants were required to attend medication management meetings at the MSU psychiatry clinic to meet with the project coordinator/ ass istants and psychiatry resident at the beginning o f weeks 1, 2, 4, 6, 8, 10, 12 , 14, and after week 15 . During these meetings, t he projec t coordinator /assistant collected parent completed measures (e.g., SMQ, P - YMRS, CGI, CPRS - R , DBR ) from the previous two weeks and provided families with measures for the following two weeks . Starting at the second meeting with the psychiatrist, families w ere provided with container s of medication for the two - week period before the next medication management meeting . Alternatively, at times, the medication bottles were mai led to families directly from the pharmacy . The container s w ere marked with the week paren t s w ere to give children the elixir from that container, and p arents were provided with a medication schedule. The fluoxetine and placebo were indistinguishable elixi rs, made by BioMed Pharmacy in Lansing, Michigan. 74 I nclusion and exclusion criteria. Several rule out criteria were implemented ( s ee Appendix E) to ensure symptoms were solely the result of selective mutism, and, secondly, to align with the ethical requir ements of conducting a risk - benefit analysis when using psychopharmacological medications with children. First, according to the differential diagnosis criteria in the DSM - 5 (APA, 2013), children were not diagnosed with a speech condition, mental retardati on, a pervasive developmental disorder or schizophrenia, and spoke English as a first language. Since fluoxetine is hypothesized to address a biological component in selective mutism symptomology (see Carlson et al., 2008; Nutt et al., 1998), only children who had an immediate family member (i.e., biological mother, father, and siblings) diagnosed with an anxiety disorder or an immediate family member who displayed symptoms of an anxiety disorder were included. Only children who failed to respond to at leas t 10 weeks of an evidence - based psychosocial treatment for selective mutism according to their practitioners were included to ensure psychosocial approaches were attempted first . Ten weeks is the typical amount of time that children can receive psychologic al services through managed care insurance programs. For children younger than seven, 12 weeks of failed psychosocial treatment were required to meet the guidelines provided by Gleason and colleagues (2007) for the use of psychopharmacological medications in young children. In order to control external validity and to prevent possible complications associated with multiple psychopharmacological medications, children taking any psychopharmacological medication were excluded. Moreover, in order to ensure chil d safety and promote an adequate risk benefit profile for each child, children who had a negative experience with a psychopharmacological drug were not included in this study. C hildren with a medical illness that could become worse during psychopharmacolog ical treatment as determined by the project psychiatrist were not considered. Finally , children had to demonstrate significant social 75 anxiety symptoms as indicated by a T - score greater than 65 (moderately atypical) on the MASC - 2: SAS/Performance Fear scale . T - Scores are only provide d for children eight - years and older on the MASC - 2. Therefore, participants younger than eight - years - old were evaluated based on the T - scores for eight - year old children , g iven there are no normed assessments for young children t hat reliably and validly assess s ocial anxiety symptoms. Screening . The flyer parents received from the mental health practitioners instructed parents to call the project coordinator. During that phone call, the project coordinator conducted a diagnostic interview to determine if the child met DSM - 5 criteria . Parents were asked if anyone in the immediate family displayed symptoms of anxiety. Finally, the project coordinator requested contact information for teacher s . If criteria were met, parents were mai led the screening packet. The packet included: a) the consent form (Appendix Q ) ; b) the MSU psychiatry clinic medical/psychosocial history form (Appendix R ) ; c) the MASC - 2 : SAS ; d) the P - YMRS ; and c) the Release of Confidenti al Information form (Appendix S ; s ee Table 9 ) . This information was returned to Michigan State University using a pre - paid envelope. To ensure children received the required psychosocial treatment, a form detailing treatments and interventions was sent to previous mental health care pro vider s (s ee Appendix T ). Parents signed a release of confidential information form (Se e Appendix S ) to allow the project coordinator to collect this information. I f children met inclusion criteria, they were invited to the MSU Psychiatry clinic. During thi s meeting, the project psychiatrist : a) confirmed the diagnosis of selective mutism ; b) clarif ied outstanding questions regarding medical and treatment history ; and c) conducted a physical exam to ensure the child did not have any medical problems that may be exacerbated by fluoxetine (Appendix U ; s ee Table 9 ) . 76 Consent and assent p rocedures . Parents were provided with a consent form that discussed the purpose and scope of the study, possible risks and benefits, costs of participation, and alternative trea tment options ( s ee Appendix Q ) . Informed assent from the children was also obtained. In adherence with the MSU IRB guidelines, participants under 13 - years of age were required to sign an assent document. Participants who were 13 years and older signed the consent form to signify their assent to the project (see Appendix Q ). Since some of this study overlapped with the summer break, only three teachers participated in data collection. Teachers were required to consent to participation, as they were frequentl y asked to complete ratings of the ( s ee Appendix Q ). Baseline. Previous studies show children with selective mutism demonstrate very little response to placebo (Carlson, 1999; Eke, 2001) . Despite this finding , it is possible that a pla cebo response may occur. To account for this possibility , a one - week no - medication component was included in the baseline . This allowed for visual inspection of a possible placebo response after the introduction of the placebo. Therefore, children did not receive any elixir during their first week in the study . Families were provided with rating scales for the week and returned to the clinic the following week. At the beginning of week two, children were provided with the placebo treatment. The placebo was administered for four to six weeks based on the randomized treatment schedule (see Table 10 ). The no - medication component and the placebo component were grouped together in the overall baseline phase, which was compared to the outcomes of the treatment pha se. The firs t child (i.e., Child R) to enroll in this study demonstrated consistent engagement in the desired outcome behaviors prior to the onset of fluoxetine across the domains of social engagement, responsive speech, and spontaneous speech. Therefore, she was removed 77 from the final analyses per the guidelines of Watson and Workman (1981) and Christ (2007). Visual inspection indicated there were placebo responses for the remaining five children , but these occurrences quickly returned to the established trend for each child . Child F our and Child Five experienced a brief improvement in social engagement during the placebo phase ; however, both of these responses returned to normal within one week . Therefore, these children were included in the analyses for social engagement. Occurrences of responsive speech were evident during the placebo phase for Children One, Two, and Five. Again, these behaviors quickly returned to baseline. Child Four displayed responsive speech during both the no medication phase and the placebo phase; however, th e speaking behavior did not establish a consistent trend of response as it frequently returned to low levels. Therefore, this child as well as Children One, Two, and Five were included in the analyses for responsive speech. F inally, n one of the children consistently spoke spontaneously with unfamiliar adults prior to the onset of the fluoxetine medication. However, Child Two and Child Five had brief responses during the placebo phase, which quickly returned to the original tre nd . Therefore, the se children were included in the analyses for spontaneous speech. During the baseline phase, several data collection procedures occurred. One parent for each participant completed the medication compliance measure daily, the DBR ratings three times per week and, the CGI scales, the MASC - 2: SAS, the SMQ, and the P - YMRS twice per week. Available teachers (n=4) provided DBR ratings five times per week (i.e., once per school day) and CGI ratings twice per week. For children whose teacher coul d not provide ratings, a second parent completed CGI ratings twice per week and DBRs three times per week. Finally, the study psychiatrist completed CGI ratings and adapted - SEFCA at bi - weekly medication management meetings. 78 Table 10 Randomized Multiple Ba seline Design Baseline and Treatment Schedule Child /Week 1* 2* 3 4* 5 6* 7 8* 9 10* 11 12* 13 14* 15 Child One A B B C C D D D D D D D D D D Child Two A B B B C C D D D D D D D D D Child Three A B B B B C C D D D D D D D D Child Four A B B B B B C C D D D D D D D Child Five A B B B B B B C C D D D D D D *Meetings with psychiatrists occur at the beginning of these weeks (Participants meet with psychiatrists after week 15 for debriefing) A= No Medication B= Placebo C= Fluoxetine - In troductory Dose D= Fluoxetine - Therapeutic Dose Treatment . The treatment phase encompassed the treatment with active medication. Children were randomly assigned to a schedule. The onset of the active medication varied in accordance with the multiple - baseline design ( see T able 10 ) . The same assessment schedule used during the baseline phase was implemented during the treatment phase (see Table 9 ). The pharmacological treatment consisted of two doses of fluoxetine: ( a ) an introductory dose and (b) a t herapeutic dose . The elixirs were identical in flavor, color , consistency, and quantity (i.e., ml) to the placebo elixir , and were obtained from BioMed Pharmacy in Lansing, Michigan . A child and adolescent psychiatry resident, under the supervision of a bo ard certified child and adolescent psychiatrist, at the Michigan State University Psychiatry Clinic treated all of the participants. Since each child is unique, varying doses are often required (American Academy of Child and Adolescent Psychiatry, 2009). W hile the psychiatry resident decided medication doses, the Michigan State University - Investigational Review Board (MSU - IRB) 79 required a dose range be specified apriori to ensure optimum patient safety . Through consultation with experts in the psychiatry d epartment, the MSU IRB approved a dose range of 2.5 mg/day to 20 mg/day. Based on his expertise, the study psychiatrist prescribed 10 mg/day as the introductory dose and 20 mg/day as the therapeutic dose for all participants. The initial dose (10 mg/day) was prov ided for two weeks , starting when children began the treatment phase of their randomized condition . After two weeks on the introductory dose, the psychiatrist increased to a therapeutic dose (20 mg/day) for all children. During week 10 of the stu dy, the psychiatrist and the project coordinator/assistant bega n to discuss end of treatment options with the families . Parents were encouraged to identify a provider that was a b le to continue treatment at the end of the study . To assist in this search, fa milies were informed about various options for continued treatment, and provide d with phone numbers to experienced practitioners . End of t reatment p hase . Several data collection tasks were completed during the end of treatment phase (see Table 9 ). This p hase began for each participant at the last medication m anagement meeting, after week 15 of treatment and after the r egular data collection tasks were completed for the bi - weekl y medication management meeting . During th is meeting , the TEQ - P and the End of Study form were completed. Experimental Design A randomized non - concurrent multiple baseline single - case design was used to answer the research questions . This study was single blind, as the prescribing psychiatrist was aware of the treatment schedule an d prescribed the medication. This specification was necessary, as the psychiatrist determined the dose of fluoxetine for each child based on unique individual characteristics. Five treatment schedules were determined apriori , and varied based on when 80 child ren receive d the placebo or f l uoxetine treatment . The five participants were randomly assigned to one of these treatment schedules once enrolled in the study (see Table 10 ). While the first child to enroll in this study (i.e., Child R) was assigned to trea tment schedule three at the start of the study , the family did not return the rating forms for the last several weeks of the study despite multiple attempts to collect them. Therefore, when a parent of a set of twins with selective mutism symptoms became i nterested after four participants had already been recruited, this treatment schedule was re - opened for random assignment (i.e., Child Three). It was intended that b oth children who received this treatment schedule be included in the data analysis . However , Child R did not establish consistent baseline trends and already was engaging in the desired behaviors at a frequent rate . Therefore, she was subsequently removed from the analyses per the guidel ine s of Watson and Workman (1981) and Christ (2007). Data A nalysis Multi - gate analysis. In order to determine if fluoxetine was effective for the treatment of social anxiety and speaking behaviors, a multi - gate analysis approach consisting of the Kratochwill and colleagues (2010) visual analysis procedure, the Wampold and Worsham (1986) randomization test, and the Tau - U effect size was used to analyze parent DBRs of social engagement and speaking behaviors with unfamiliar adults . Of note, selective serotonin reuptake inhibitors have a delayed effect on behaviora l outcomes; however, research suggests 90% of individuals experience positive changes in symptomology within the first two weeks of treatment (Mitchell, 2006). Therefore, an acquisition period of two weeks of the therapeutic dose of medication was factored into the analys i s, meaning data collected during the introductory dose and first two weeks of the therapeutic dose of fluoxetine treatment were not included in visual or statistical analys i s. 81 The analysis procedure include d three gates . If one step did n ot indicate sign i ficant results, the remainder of the procedure was discontinued . First, two project assistants using the Kratochwill and colleagues (2010) guidelines for visual analysis reviewed each graph to determine if there was a n improvement in sympt omology using the Visual Analysis Decision Tree (see Appendix V ). If raters disagreed regarding a response to treatment for a child on an outcome measure (i.e., social engagement), the project coordinator made the final decision. Inter - rater reliability be tween the project assistants was calculated using Kappa and indicated moderate agreement ( , p = .04). Ultimately, the project coordinator had to make the final decision for three cases (i.e., social enga ge ment: n=1; responsive speech: n = 2 ). If there was evidence of at least a moderate effect based on the Kratochwill and colleagues guidelines (i.e., three replications with one or more incident of contradicting evidence), significance was assessed at the p .05 level using the Wampold and Worsham (1986) randomization test. The Wampold and Worsham technique requires that there be no missing data at time points when children switch from the baseline to treatment phase. However, raters failed to complete some scheduled assessments. For exampl e, o n Direct Behavior Ratings of social engagement and speaking behaviors, which were the primary outcome measures, completion rates varied. Adherence for DBR ratings ranged f rom 62 % to 98 % for primary raters (i.e., mothers) . Overall, completion rates for DBRs completed by the primary raters across the baseline phase ( i.e., 93% ) and treatment phase (i.e., 94%) were consistent . See Table 1 1 for a summary of completion rates by measure and rater. To account for missing data at critical time points during the Wampold and Worsham (1986) analysis , the conservative approach of averaging the previous and following 82 Table 1 1 Percent Completion Rate of Data Forms per Observer. Baseline Treatment DBR MASC - 2 SEFCA P - YMRS SMQ CGI TIF Child One Mom Teache r Psychiatrist 98% 28% 66% 99% 39% 50% 98% 45% 97% 56% 100% 100% 100% 35% 56% 100% Child Two Mom Teacher Psychiatrist 94% 39% 66% 96% 71% 83% 98% 59% 93% 78% 100% 97% 97% 63% 78% 100% Child Three Mom Teacher Psychiatrist 100% 43% 66% 97% 57% 50% 96% 53% 97% 56% 100% 100% 100% 37% 56% 100% Child Four Mom Dad Psychiatrist 74% 67% 100% 89% 92% 100% 62% 73% 93% 100% 90% 93% 93% 90% 100% 100% Child Five Mom Dad Psychiatrist 99% 6% 100% 89% 16% 100% 91% 20% 97% 100% 97% 100% 97% 27% 100% 100% 83 data point s of the missing data point was employed. Data points only needed to be estimated for parent DBRs of responsive speech with u nfamiliar adults (n = 4) . If the randomization test indicated signi fi cant results, the Tau - U effect size measure was calculated to quantify the impact of treatment , with a large effect (i.e., Tau - U 0.50) considered as evidence for improvement. Question one . T he multi - gate analysis procedure was used to determine if changes occurred across the baseline and intervention phase s on parent completed DBRs of social engagement with unfamiliar adults . A n examination of the standard deviation across parent ratings during the baseline phase demonstrated minimal variability (SD =1.72) , indicating the parent ratings of social engagement were a reliable measure of social anxiety. In addition to the multi - gat e procedure, the means of s everal supplementary assessments ( i.e. , DBRs, CGI, MASC - 2: SAS) completed by parents, teachers, and/or the psychiatrist were calculated to gain additional insight into the effect of fluoxetine on social anxiety symptoms in child ren diagnosed with selective mutism (see Table 1 2 ) . Question two . In order to determine if fluoxetine treatment increased s peaking behaviors, the multi - gate analysis procedure was used to identify changes in parent DBRs of responsive and spontaneous speec h with unfamiliar adults . An examination of the standard deviation for DBRs across participants during the baseline phase for responsive speech (SD = 1.72 ) and spontaneous speech (SD = .66 ) indicate d these ratings were reliable. To provide additional infor mation of the impact of fluoxetine on speaking behaviors, means of supplementary assessments (i.e., DBRs, SMQ, CGI, Diagnostic Interview) completed by p arents, teachers, an d/or the psychiatrist were calculated (see Table 1 3 and 1 4 ). Question three . In ord er to examine adverse events , a descriptive analysis of parental reports on the adapted SEFCA was conducted. This analysis focused on the types (e.g., nausea, 84 irritability, behavioral disinhibition ) and intensity of side effects reported . Next, descriptive statistics regarding CGI Severity of Side Effect ratings provided throughout the study by the parents and study psychiatrists were examined. In addition, a mean parent rating and a mean psychiatrist rating were calculated. For the purposes of calculating the mean outcome, a rating of - p ositive and n egative c hanges a re a pproximately e Effect Scale to prevent skewing th e ratio in a positive direction if no changes were noted. A Question four . In order to ensure that possible improvements in selective mutism symptoms were not the result of behavioral disinhibition, which is a common side effect of SSRI treatment in children, responses on the P - YMRS were analyzed individually. Item responses tha t reflect the possible development of behavioral disinhibition have been highlighted in Appendix O. Positive reports of behavioral disinhibition were examined to determine if they occured primarily when a child is showing improvement in social anxiety, res ponsive speech, or spontaneous speech. Question five . In order to assess parental acceptability of the fluoxetine treatment, parent answers on the Treatment Evaluation Questionnaire were examined. Previous research has determined a score of 55 on this sca le to be interpreted as high acceptability (Kratochwill et al., 2003). Therefore, parent scores of 55 or higher were used as the criterion to determine if parents found the fluoxetine treatment acceptable for the child with selective mutism. Additionally, descriptive statistics of medication compliance were calculated to determine the feasibility of the intervention for families. 85 Table 12 Means of Social Anxiety Symptom on Outcome Measures Across Baseline and Treatment Phases Measure Possible Ratings Baseline (SD, Range) Intervention (SD; Range) DBR 0 (Never) 10 (All of the Time) Mother (Adults) 1.13 (1.72, 0 - 10) 2.96 (2.84, 0 - 9) Mother (Friends) 5.7 (3.8, 0 - 10) 5.7 (3.02, 1 - 10) Teacher (Teacher) 4.04 (3.46, 1 - 9) 8.01 (5 - 9) Teacher (Students) 5.77 (2.34, 3 - 9) 8.42 (6 - 10) MASC - 2: SAS 0 (Absent) 27 (Severe) 23.82 (2.31, 21 - 27) 23.95 (2.76, 19 - 27) CGI Anxiety Severity Mother (Home) Mother (Community) 1 (Absent) 5 (Severe) 1.71 (1.01, 1 - 2) 3.92 (1.07, 3 - 5) 1.38 (.46, 1 - 2) 2.46 (.85, 2 - 4) Teacher 3.92 (.35, 3 - 4) 2.28 (.44, 2 - 3) Psychiatrist 4 (0, 4) 3.72 (.97, 3 - 4) CGI Shyness Severity 1 (Absent) 5 (Severe) Mother (Home) Mother (Community) 1.06 (.28, 1 - 2) 4.12 (.40, 4 - 5) 1 (0, 1) 3.2 (.89, 3 - 5) Teacher 3.92 (.35, 3 - 4) 2.68 (.44, 2 - 3) Psychiatrist 4.3 (.26, 4 - 5) 3.73 (.97, 3 - 4) CGI Anxiety Change Mother Teacher Psychiatrist 1 (Much Improved) 7 (Much Worse) 1 (Much Improved) 5 (Muc h Worse) 1 (Much Improved) 7 (Much Worse) 3.88 (.31, 3 - 4) 2.58 (.53, 2 - 3) 4 (0, 4) 2.49 (.83, 1 - 4) 1.58 (.50, 1 - 2) 2.83(.74, 2 - 4) CGI Shyness Change Mother 1 (Much Improved) 7 (Much Worse) 3.88 (.31; 3 - 4) 2.73 (.69, 2 - 4) Teacher 1 (Mu ch Improved) 5 (Much Worse) 2.58 (.53, 2 - 3) 1.5 (.51, 1 - 2) Psychiatrist 1 (Much Improved) 7 (Much Worse) 4 (0, 4) 3.33 (.53, 3 - 4) Note: Adapted from Carlson (1997) 86 Table 13 Direct Behavior Ratings Mean Speaking Behaviors Across Baseline and Treatment Phases Condition Baseline (SD; Range) Intervention (SD; Range) Responsive Speech - Friends/Students Mother (Primary) Teacher Spontaneous Speech - Friends/Students Mother (Primary) Teacher Responsive Speech Unfamiliar Adults/Teacher Mot her (Primary) Teacher Spontaneous Speech Unfamiliar Adults/Teacher Mother (Primary) Teacher 5.58 (3.83, 0 - 10) 4.15 (3.78, 0 - 9) 5.06 (4.02, 0 - 10) 2.11 (1.74, 0 - 6) 0.71 (1.72, 0 - 10) 3.15 (4.09, 0 - 9) 0.08 (.66, 0 - 5) 0.96 (1.15, 0 - 3) 7.18 (2. 97, 1 - 10) 6.63 (3.26, 0 - 9) 6.22 (3.2, 0 - 10) 5.05 (2.9, 0 - 10) 2.61 (2.75, 0 - 10) 5.19 (3.02, 0 - 9) 0.81 (1.59, 0 - 8) 3.58 (3.27, 0 - 9) 87 Table 14 Means of Supplementary Selective Mutism Outcome Measures Across Baseline and Treatment Phases Measure Possible Ratings Baseline (SD; Range) Intervention (SD; Range) SMQ Home/Family 0 (Severe) 16 (Absent) 9.81 (2.49, 4 - 12) 12.07 (1.81, 7 - 14) Social Situations 0 (Severe) 14 (Absent) 0.82 (1.03, 0 - 3) 1.59 (1.27, 0 - 8) Interference/Distress 0 (Extreme Interference) 18 (No Interference) 2.29 (2.19, 0 - 5) 4.36 (3.9, 0 - 11) CGI - SM Severity 1 (Absent) 5 (Severe) Mother 4.52 (.49, 4 - 5) 3.76 (.88, 2 - 5) Teacher 4.5 (.71, 4 - 5) 2.8 (.65, 2 - 4) Psychiatrist School Home Clinic 4.35 (.49, 4 - 5) 1.4 (.42, 1 - 4) 4.5 (.51, 4 - 5) 3.85 (.32, 3 - 4) 1.6 (.67, 1 - 3) 4 (0, 4) Peers 4 (0, 4) 3.83 (.42, 3 - 4) Overall 4 (0, 4) 3.8 (.32, 3 - 4) CGI SM Cha nge Mother 1 (Much Improved) 7 (Much Worse) 3.89 (.37, 3 - 4) 2.75 (1.06, 1 - 4) Teacher 1 (Much Improved) 5 (Much Worse) 2.83 (.42, 2 - 3) 1.83 (.31, 1 - 2) Psychiatrist School Home Clinic 1 (Much Improved) 7 (Much Worse) 4 (0, 4) 3.9 (.29, 3 - 4) 4 (0, 4) 3.2 (.71, 2 - 4) 3 (1.05, 1 - 4) 3. 4 (1.07, 1 - 4) Overall 4 (0, 4) 3.56 (.53, 3 - 4) Note: Adapted from Carlson (1997). 88 CHAPTER 4 RESULTS Question One Will fluoxetine treatment lead to a significant reduction in social anxiety symptoms between the baseline/placebo phases and the treatment phase in five children diagnosed with selective mutism involving elevated levels of social anxiety symptoms? As a group, f luoxetine did not lead to a significant reduction in soc ial anxiety symptoms for the five children in this study. Visual analysis of parent DBR ratings of comfortable social engagement with unfamiliar adults indicated there were changes in level, trend, variability, overlap, immediacy of effect and consistency after the onset of fluoxetine treatment for only one of the five participants (i.e., Child Three ; see Figure 3) . Raters indicated the remaining four participants did not show improvement because of contradictory evidence in the domains of slope ( i.e., Chil d One, Child Two, Child Four ), mean ( Child Five ), a nd immediacy of effect ( Child One ). Although children were rated as not improving via visual analysis, the Wampold and Worsham (1986) randomization test was still performed due to the less than optimal rel iability between the raters performing visual analysis. The randomization test confirmed there was no improvement in social engagement with unfamiliar adults across participants after the onset of the fluoxetine treatment (p=.38). Supplementary ratings on the MASC - 2 indicated similar results , with baseline (M= 23.82 , SD= 2.31 ) and treatment (M= 23.95 , SD= 2.76 ) means across participants remaining stable. Overall psychiatrist ratings of improvement indicate similar findings with a mean rating of 4 (SD=0) during the baseline phase and 3.73 (SD=.97) during the intervention phase on the CGI Anxiety Severity scale. For an overview of DBR ratings and supplementary measures provided by parents, teachers, an d the psychiatrist, see Table 1 2 . 89 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Figure 3 Parent DBR - Social Engagement with Unfamiliar Adults by Week Baseline = No Medication (Week One) and Placebo Acquisition =Introductory Dose (10 mg/day, Two Weeks); Therapeutic Dose (20 mg/da y; Two Weeks) Treatment= Therapeutic Dose (20 mg/day) Note: Baseline/Intervention administered non - concurrently across participants. Baseline Acquisition Treatment DBR R Child One Child Two Child Three Child Four Child Five Week 90 Question Two Will the fluoxetine treatment increase the frequency of spontaneous speech and responsive speech across the baseline/placebo phases and the treatment phase in five children diagnosed with selective mutism involving elevated levels of social anxiety symptoms? Fluoxetine treatment resulted in a significant improvement in responsive speaking behaviors; however, th e effect of the treatment was not as large as hypothesized. Visual analysis revealed three of the participants experienced improvements in responsive speech consistent with the onset of fluoxetine in level, trend, variability, overlap, immediacy of effect and consistency (see Figure 4 ). Two of the chi ldren were deemed non - responders because of a lack of immediacy of effect ( i.e., Child One ) and no change in mean ( i.e., Child Five ). Given three replications of the treatment effect were identified, there is moderate visual evidence for improvement as a g roup . The Wampold and Worsham (1986) randomization technique indicated the improvement in responsive speech was significant (p=.0 3 ) . However, calculation of the Tau - U effect size indicated only a moderate improvement (Tau - U = . 44 ; p < .001 ), with a value be low the apriori value (i.e., Tau - U=.50) set to identify a substantial improvement. Fluoxetine treatment did not lead to a significant improvement in spontaneous speaking behaviors. Visual analysis of parent DBR ratings of spontaneous speech with unfamilia r adults indicated changes in level, trend, variability, immediacy of effect, overlap and consistency after the onset of fluoxetine treatment for only two participants ( Child Three and Child Four ; s ee Figure 4 ) , indicating there were not enough replication s to identify a treatment effect . Three of the children were determined to be non - responders because raters perceived unexpected outcomes in slope ( i.e., Child One, Child Two ) , mean ( i.e., Child Five ), and immediacy of effect ( i.e., Child One, Child Two, C hild Five ) from the baseline to treatment phases. Although there 91 was not enough evidence to indicate a treatment effect using the guidelines provided by Kratochwill and colleagues (2010), the Wampold and Worsham (1986) test was used to identify the possibi lity of a signif i cant change in spontaneous speaking behaviors due to the less than optimal reliability between visual analysis raters. The randomization test confirmed that there was not a significant change in spontaneous speaking behaviors for all five children across the baseline and treatment phases (p=.10). Despite improvements in responsive speech, all children continued to meet DSM - 5 criteria for selective mutism at the end of the study. While there w as no significant improvement in spontaneous spe aking behaviors , s upplementary ratings indicated improvements in unspecified (i.e., not responsive or spontaneous) speaking behaviors from the placebo phase to the treatment phase. On the SMQ parents perceived an overall improvements from baseline to treat ment phase for speaking behaviors at home with family (Baseline: M=9.81, SD= 2.49 ; Treatment: M=12.07, SD= 1.81 ) and in s ocial s ituations (Baseline: M=0.82, SD= 1.03 ; Treatment: M=1.59, SD= 1.27 ). The study p sychiatrist saw a similar improvement in overall mu tism symptoms as rated on the CGI - Mutism when comparing baseline ratings (M=4, SD=0) to intervention ratings (M=3.56, SD= .53 ). For an overview of scores on Direct Behav i or Ratings as well as additional supplementary ratings for speaking behaviors provided by teachers, parents, and the child psychiatrist, see Table s 1 3 and 1 4 . Question Three What adverse side effects, if any, will children experience during the fluoxetine treatment? Is there a positive risk to benefit profile for children taking fluoxet ine for selective mutism? 92 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Figure 4 Parent DBR Spontaneous and Responsiv e Speech with Unfamiliar Adults by Week Baseline = No Medication (Week One) and Placebo Acquisition =Introductory Dose (10 mg/day, Two Weeks); Therapeutic Dose (20 mg/day; Two Weeks) Treatment= Therapeutic Dose (20 mg/day) Child One Note: Baseline/Intervention administered non - concurrently across participants. DBR R Baseline Acq uisition Treatment Child Two Child Three Child Four Child Five Week 93 According to the psychiatrist completed SEFCA, none of the participants experienced adverse events during the fluoxetine treatment. As a group, parents perceived more positive th an negati v e cha nges during the fluoxetine phase ( = 2.12 , SD = .83 ). Further, the study psychiatrist also perceived a positive benefit to risk ratio ( M see Table 1 5 ). Table 15 CGI Means of Global Side Effect Ratings Baseline (SD; Ran ge) Intervention (SD; Range) Child One Mother Psychiatrist 4 (0, 4) 4 (0, 4) 2.13 (.81, 1 - 3) 1 (0, 1) Child Two Mother Psychiatrist 2.5 (1.6,1 - 4) 4 (0, 4) 1.38 (.50, 1 - 2) 2.5 (.71, 2 - 3) Child Three Mother Psyc hiatrist 4 (0,4) 4 (0, 4) 2 (.85, 1 - 3) 2 (0, 2) Child Four Mother Psychiatrist 3.9 (.33, 3 - 4) 3.75 (3 - 4) 1.6 (.73, 1 - 3) 2.67 (.58, 2 - 3) Child Five Mother Psychiatrist 4 (0, 4) 4 (0, 4) 3.5 (.53, 3 - 4) 4(0, 4) Question Fo ur Does onset of active fluoxetine medication correspond with an increase in behavioral disinhibition? The onset of active fluoxetine medication did correspond with incidents of behavioral disinhibition for two participants as rated by parents on the P - YMRS (Child Two and Child Three ) . These two children were also noted to have improved in at lea s t one domain of selective mutism symptomology. Despite this finding , visual inspection indicate d improvements in 94 selective mutism symptomology did not solely co rrespond to the onset of behavioral disinhibition (see Figures 5 and 6). Figure 5 Child Two: Parent DBR Rating of Responsive Speech w ith Unfamiliar Adults and Onset of Behavioral Disinhibition Baseline = No Medication (Week One) and Placebo Acquisition =Introductory Dose (10 mg/day, Two Weeks); Therapeutic Dose (20 mg/day; Two Weeks) Treatment = Therapeutic Dose (20 mg/day) 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Baseline Acquisition Treatment DBR R Week 95 Figure 6 Child Three: Parent DBR Ratings of Social Engagement, Responsive Speech, and Spontaneous Speech with Unfamiliar Adults and Onset of Behavioral Disinhibition Bas eline = No Medication (Week One) and Placebo Acquisition =Introductory Dose (10 mg/day, Two Weeks); Therapeutic Dose (20 mg/day; Two Weeks) Treatment= Therapeutic Dose (20 mg/day) 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 5 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Baseline Baseline Baseline Acquisition Acquisition Acquisition Treatment Treatment Treatment DBR R DBR R DBR R Week Week Wee k 96 Question Five Will parents find the use of fluoxetine for the treatment of selective mutism and social anxiety symptoms acceptable? Parents found the use of fluoxetine for the treatment of selective mutism and social anxiety symptoms highly acceptable. A score of 55 or higher on the TEQ - P indic ates an intervention is highly acceptable (Kratochwill et al., 2003), with the average across raters in this acceptable intervention for the problem behavior, with t he parent of Child Four indicating she negative side effects, with th See Table 16 for all ratings across parents. Parent reported medication compliance data was collected for each child via a daily parent - completed form indicating the time the medication was provided. The study psychiatrist instructed parents to either give the medication in the morning (n= 3 ) or in the evening (n =2 ) each day if they provided the medication within six hours of the instructed time. Parents provided the medication 82% to 100% of the tim e during the baseline phase. The parents of Child Two who indicated 82% compliance noted they had forgotten to give the medicine some days. They were encouraged to be consistent with providing the medication, and were informed that SSRIs work over an extended period of tim e, which requires doses to be provide d consistently. During the treatment phase, parents were compliant 98 % to 100% of the time (see Table 17). 97 Table 16 Answers o n the TEQ - P Across Participants Question/Participant Child One Child Two Child Three Child Four Child Five Average 1. This was an acceptable intervention 6 6 6 5 6 5.80 2. Most parents would fi nd this intervention appropriate for behavior problems in addition to the one described. 6 5 6 5 6 5.60 3. This intervention was effective in changing the problem behavior. 5 5 5 5 5 5 4. I would suggest the use of this intervention to other parents. 6 5 6 6 6 5.80 5. severe enough to warrant use of this intervention. 6 5 6 6 6 5.80 6. Most parents would find this intervention suitable for the behavior problem described. 5 5 5 6 6 5.4 7. The intervention did not result in negative side effects for my child. 6 5 6 6 6 5.80 8. The intervention would be appropriate for a variety of children. 5 5 5 5 6 5.20 9. The intervention was a fair way to 5 5 5 6 6 5.4 10. I liked the procedures used in this intervention. 5 5 5 6 6 5.4 11. The intervention was a good way to 5 5 5 6 6 5.4 Sum of Ratings 60 56 60 62 65 60.6 Adapted from Carlson (1997) & Eke (2001) Ratings: 1 (strongly disagree) to 6 (strongly agree); Sum of 55 or higher considered acceptable intervention 98 Table 17 Medication Compliance Rates per Phase Baseline Fluoxetine Child One 95 % 99% Child Two 82% 100% Child Three 100% 100% Child Four 94% 100% Child Five 100% 98% Individual Improvements While improvements were only identified as a group in the domain of responsive speech, parents found the intervention highly acceptable. Given the lack of expected results in the other domains, it was deemed important to examine i ndividual improvements in order to account for the high level of acceptability. All childre n were noted as improving in at least one area. For example , Child One was rated as improving in the domain of responsive speech; however, she Visual analysis revealed that Child Two was a responder in the domain of responsive speech. Visual analysis and the Tau - U effect size measure revealed that Child Three was a responder in the domains of socia l engagement (Tau - U= .44, p=.04) and responsive speech (Tau - U=.75, p=.001), and visual analysis alone indicated Child Three was a responder in the domain of spontaneous speech. Child Four was deemed a responder via visual analysis and the Tau - U effect size in the domains of responsive speech (Tau - U=.99, p=.001) and spontaneous speech (Tau - U=.71, p=.03). Finally, although Child Five was not rated as improving via visual analysis, her mother did report - Shyness and Anxiety ratings at the end of the study. In addition to these indicators of change for individual children, parents and teachers provided feedback and comments on forms and during medication management meetings indicating progress after the onset of fluoxetine treatment ( see Table 1 8 ). 99 Table 18 Parent and Teacher Reports of Progress after the Onset of Fluoxetine Treatment Sample of Positive Reports Child One Child Tw o Child Three Child Four Child Five 100 CHAPTER 5 DISCUSSION The findings and impli cations of this multiple - baseline single - case design study are summarized below and discussed in the context of previous research. The results of this study shed light on the effectiven e ss of fluoxetine for the treatment of selective mutism in children and adolescents . While generalizabili ty is limited, the outcomes can significantly inform future research and may provide some insight for practicing mental health clinicians. There are many questi ons regarding psychopharmacological treatment s for children wi th selective mutism that are unanswered. The findings from this study are small step s to wards answering these questions by demonstrating how some symptomology changes occurred or did not occur in correspondence with the active medication treatment . Fluox etine Treatment for Associated Social Anxiety Visual analysis revealed there was not an improvement in social anxiety with unfamiliar adults across the baseline and fluoxetine phases. Given findings from previous research , it was surprising there were not significant improvements in social anxiety symptomology. Previous research has documented positive improvements in social anxiety symptoms in children with and without selective mutism in response to treatment with fluoxetine and other SSRIs. However, in t hese studies, there were also children who did not benefit from treatment. For example, in an open label study with 21 children, Dummit and colleagues (1997) found that 76% of children with selective mutism experienced a reduction in social anxiety symptom s when treated with fluoxetine. In a study examining the use of fluoxetine for anxiety disorders non - specific to selective mutism, Birmaher and colleagues (2003) found that 61% of children improved while 101 taking fluoxetine, which was statistically significa nt when compared to the response of a placebo group . It is possible that children in this study may have needed higher doses to obtain a more optimal response. In children with obsessive - compulsive disorder, the FDA has approved doses up to 60 mg/day (Eli Lilly and Company, 2011). In the Dummit and colleagues study (1997) doses as high as 60 mg/day were prescribed to some participants with selective mutism, with higher doses prescribed to older children. Potentially, the dose of fluoxetine used in this stu dy may have been too conservative for each participant to achieve a substantial gain in social anxiety symptomology. Additionally, all children in this study were resistant to psychosocial treatments according to their pracitioners and have exhibited sympt oms of social anxiety for a long period of time. Children with severe selective mutism symptomology that has persisted for years may need a more aggressive psychopharmacological approach. While not fully examined in the anxiety literature for children, the re are examples of those with severe depression symptoms needing higher doses of SSRIs (Montgomery, Rasmussen, Lyby, Conner, & Tanghosj, 1992). Contradictive to the findings in this study, previous SSRI studies conducted with children with selective mutis m and elevated social anxiety symptoms and those with social anxiety disorder alone have noted improvements for children on measures of social anxiety. However, in these studies children received treat ment for longer periods of time. In this study, childre n received a therapeutic dose of the medication for six to ten weeks . In prior fluoxetine studies using a double - blind design to examine the effecti ve ness of fluoxetine on social anxiety symptoms , children received treatment for 12 weeks (Birmaher et al., 2003; Black & Uhde, 1994) . However, prior research suggests that some response s to SSRIs should be seen by at least 102 4 to 6 weeks (Garfield et al., 2004) , with some studies suggesting that 90% of individuals given an SSRI show a response within two weeks (M itchell, 2006) . Given the children in this study have been experiencing social anxiety symptomology for many years, more time may have been needed to identify more substantial improvements in functioning. Timing of the onset of fluoxetine treatment based on the severity of symp t omology has not been studied in children with any anxiety diagnosis. However, research suggests that regardless of the severity of symptoms , 75% of adults treated for depression with fluoxetine exhibited a response by four weeks of treatment ( Nierenberg et al., 2000) . Children in this study were taking fluoxetine for four weeks, including the introductory and therapeutic dose phases, suggesting that a response would have been expected at this point. More research into the onset of fl uoxetine treatment effects on social anxiety symptoms for children with selective mutism is imperative to help practitioners judge if fluoxetine should be used to treat selective mutism and how long children should be provided with the medication before ch anging the intervention approach . Fluoxetine Treatment for Speaking Behaviors Selective mutism is currently listed as an anxiety disorder in the recently published DSM - 5 (APA, 2013). Previous research has demonstrated that selective mutism almost always occurs comorbidly with symptoms of social anxiety (Black & Uhde, 1994; Dummit et al., 1997). Furthermore, i n an examination of comorbid psychopathology experienced by children with selective mutism, social anxiety was the only abnormal characteristic cons istent within the sample (Black & Uhde, 1992). The joint biopsychosocial framework of social anxiety and selective mutism put forth by Carlson and colleagues (2008) theorized that that the etiological underpinnings of social anxiety and selective mutism in clude both biological ( e.g., autonomic symptoms) and psychosocial factors ( i.e., social skills; see Figure 1). Therefore, in this study, an 103 intervention targeting the hypothesized underlying biological cause of social anxiety (i.e., serotonin system) was i ntroduced to lower social anxiety symptoms . I t was hypothesized that speaking behaviors would increase in conjunction with a decrease in social anxiety . Since non - significant outcomes were noted for social anxiety symptoms , improvements in speaking behavio rs were not expected. However , results indicated that fluoxetine led to a significant improvement in responsive speech. Previous research has also identified improvements in speaking behaviors with fluoxetine treatment ; however, these studies posse ssed we aknesses . In an open label study that did not include a control group , Dummit and colleagues (1997) found that 76% of their sample (n = 21) exhibited an increase in speech production after treatment with fluoxetine , as indicated by CGI ratings . Black and Uhde (1994) conducted a small - randomized control trial with six children receiving active fluoxetine medication and nine children receiving placebo. Analogous to the findings of Dummit and colleagues (1997), Black and Uhde (1994) found parents of children who received fluoxetine rated mutism behaviors as improved on the CGI ; however, results were limited due to the small sample size for the data analysis procedure. In addition to these two studies, a handful of case studies have also noted increased speech in children treated with fluoxetine (Boon, 1994; Guna - Dumestrica & Pelletier, 1996; Harvey & Milne, 1998; Rupp, 1999; Silveira et al., 2004; Wright, 1995 ). The current study contributes additional evidence to the phenomenon of fluoxetine increasing speaki ng behaviors, and adds to these previous findings , as the type of speech (i.e., responsive), and with whom the speech was occurring was specified (i.e., unfamiliar adults) . The finding that responsive speaking behaviors improved without changes in social anxiety symptomology highlights a potentially complicated relationship between social anxiety 104 and selective mutism. Two schools of thought regarding th is relationship are predominant in the literature. First, Black and Uhde (1995) suggested that selective mutism might be an extreme form of social anxiety , noting that the anxiety is so severe in these children that it prevents a child from speaking during a social situation. In co ntrast , children only experiencing social anxiety disorder have anxiety in soci al situations, but not enough to prevent speech. Second, Mannasis and colleagues (2003) postulated that selective mutism is similar to social anxiety disorder, but unique etiological factors (e.g., speech problems, social skills) have resulted in the refu sal of speech as a way to cope with the anxiet y. This theory is supported by their research indicating similar levels of social phobia in children diagnosed with social anxiety disorder and children diagnosed with selective mutism . A recent review complete d by Muris and Ollendick (2015) highlight ed the lack of longitudinal research investigating the etiological origins of selective mutism. The authors note d that it is important that selective mutism be classified as an anxiety disorder given the consistent findings that these children experience heightened levels of anxiety , but stresse d th e lack of understanding of the relationship between social anxiety and selective mutism. One possibility is that selective mutism is a specific phobia of expressive speech in certain situations and with unfamiliar people (Omdal & Galloway, 2008). Therefore, children with selective mutism are not afraid of being around or engaging non - verbally with others , but are afraid of speaking or being required to speak in these social interactions. Here, an improvement in speaking might be observed if an SSRI treatment were to lower anxiety associated with expressive speech . Delineating between responsive and spontaneous speech is essential in selective mutism treatment studies . S pont aneous speech has been deemed a better indicator of progress, as it is considered speech that is intended to communicate (Pionek Stone et al., 2002). Conceptually, 105 the notion children may first begin engaging in responsive speech after the onset of treatm ent before engaging in spontaneous speech makes sense . For example, effective behavioral treatment for selective mutism focuses on systematic desensitization of anxiety provoking speaking situations (Pionek Stone et al., 2008), where less anxiety provoking behaviors are engaged in first followed by more difficult behaviors. Here, it is possible that the fluoxetine treatment impacted anxiety surrounding speaking behaviors just enough to allow children to provide a brief response, but did not lower the anxiet y enough for children to engage in conversation using spontaneous speech. Given the relatively low dose of medication and shortened treatment time compared to previous studies (i.e., Black & Uhde, 1994; Dummit et al., 1997) it is possible that children may have displayed spontaneous speaking behaviors if they received a higher dose of the medication or were treated for a longer period of time. Second, it is possible that the fluoxetine treatment was impacting the serotonin system in another unknown way othe r than reducing anxiety symptomology . Given that serotonin receptors are widespread throughout the brain (Stahl, 1998) , it may be difficult to untangle the unintended ways fluoxetine may be impacting speaking behavior s . Future research should continue to e xplore this issue. Behavioral Disinhibition and Other Adverse Events Previous studies have noted rapid improvement s in speaking behaviors for children with selective mutism after the onset of treatment with an SSRI. To explain this phenemenon, researcher s have suggested that the unintended side effect of behavioral disinhibition may be playing a role in the improvement of speech instead of the intended anxiety reduction mechanism (i.e. , Carlson , 1999 ) . Carlson and colleagues (2008) noted that researchers should examine the potential that behaviors associated with behavioral disinhibition are mimicking an improvement in symptomology of selective mutism symptoms. For example, children who are experiencing 106 behavioral disinhibition may take more risks, such as yelling out in class or interrupting the games of other children. To the untrained eye, these behaviors may look like a reduction in social anxiety, while they are a display of increased inappropriate risk taking behaviors. It was hypothesized that childr en in this study would not experience behavioral disinhibition , as it is a relatively rare side effect of treatment with selective serotonin reuptake inhibitors (Wilens et al., 2003). However, one child who experienced improvement in responsive speech and one child who was rated as improving in social engagement, responsive speech, and spontaneous speech experienced symptoms indicative of behavioral disinhibition for brief periods of time . Although the occurrences of behavioral disinhibition in the current study were brief , they were interestingly timed right before a rapid improvement in in responsive speech for one child. Additionally, they occurred during the highest positive ratings in social engagement, responsive speech, and spontaneous speech for the other child. There was not enough evidence in this study to indicate that behavioral disinhibition was the cause of improvements in individual children; however, future studies should continue to examine the potential relationship between behavioral disinh ibition and improvements in the frequency of speech in children with selective mutism. Aside from the behavioral disinhibition experienced by two participants, p arents of children in this study reported no adverse events during treatment with the active m edication . This lends evidence to the notion that fluoxetine is a safe treatment for selective mutism, especially at the doses (i.e., 10 mg/day and 20 mg/day) utilized in this study. Overall, this finding is consistent with previous research examining the use of fluoxetine for selective mutism and other internalizing disorders. For example, Black and Uhde (1994) examined the use of fluoxetine for the treatment of selective mutism in children between the ages of six and 11, finding no statistical difference in adverse events between the placebo group and the treatment 107 group. In a study examining fluoxetine for internalizing disorders at a dosage of 20 mg/day, Birmaher and colleagues (2003) found a higher prevalence of adverse events, with approximately half of the seven to 17 year old children treated with fluoxetine experiencing gastrointestinal side effects. This was a statistically significant difference from the placebo group. The difference between the low frequency of side effects in this study with a s imilar dose of medication in the Birmaher and colleagues study may be due to the small sample size in this project. Given all of the side effect data collected in previous clinical trials indicating a small but reliable level of side effects with fluoxetin e (FDA, 2011), it is likely there would have been a greater percentage of mild side effects with a larger pool of participants. Interestingly, Dummit and colleagues (199 7 ) provided evidence indicating fluoxetine is a safe and effective treatment for selec tive mutism; however, participants in their study experienced more side effects than participants in this project with 19% of children experiencing behavioral disinhibition and another 10% experiencing other side effects such as headaches, insomnia, and or jitteriness. One possible reason for the higher rates of adverse events in the Dummit and colleagues study may be the use of higher doses of fluoxetine (e.g., up to 60 mg/day). However, it should be noted Dummit and colleagues found more robust results wi th these higher doses. This highlights the delicate balance between finding a dose that is effective for a child with selective mutism while also finding a dose that does not lead to unpleasant side effects (American Academy of Child and Adolescent Psychia trists, 2009). It also calls into question the possibility that these children were improving based upon symptoms of behavioral disinhibition , instead of the intended anxiety reducing effect of the medication. Overall, parents and the study psychiatrist re ported that there were more positive gains than negative side effects on the CGI Global Side Effect Scale. Therefore, in this study, the risk was low while the 108 benefit was perceived to be high given parent acceptability ratings ; however, it is possible t hat children may have experienced a more substan t ial improvement in symptomology from higher doses. Future research should continue to look at the optimal doses for children with selective mutism and the kinds of side effects that are associated with thos e doses. Medication Compliance Medication compliance was high for this study, with compliance equal to or higher than medication compliance reported in previous research studies in child and adolescent psychopharmacology (Hack & Chow, 2001). According to Hack and Chow (2001), psychotropic medication compliance in children and adolescents can be encouraged by several practical methods including: a) establishing a treatment alliance with the family/child; b) providing education on the rationale for the u se of the medication; c) minimizing the frequency of providing a dose (e.g., one time per day); d) ensuring the timing and simplicity of the dosing regimen is acceptable to the family; e) ensuring the medication is palatable to the child; and f) minimizing cost. This study utilized all of these methods as a strong rapport was built with families via the project coordinator/assistants and the treating psychiatrist during the screening phase and weekly medication management meetings. Additionally, parents an d families were informed about the hypothesized mechanism of action of fluoxetine in the treatment of selective mutism. The fluoxetine only needed to be administered once daily, and was provided to the family in pre - measured weekly bottles. Most children f ound the mint flavoring of the medication acceptable; however, one did not. This child was provided with suggestions (e.g., take with a spoonful of peanut butter immediately after) to help improve palatability. Additionally, the treatment was provided fre e of charge to families, further eliminating barriers to compliance. 109 Other factors promoting treatment feasibility in this study included frequent monitoring of medication compliance at bi - weekly medication management meetings and the desperation of pare nts to find a treatment that works. Parents were required to bring in forms indicating the time the provided the medication each day. Making parents document the time they provided the medication, and providing performance feedback to parents on a regular basis likely boosted treatment compliance (Hagermoser Sanetti & Kratochwill, 2008). All of the children in this project had failed a previous course of psychosocial treatment. Given selective mutism is a debilitating disorder that affects several areas of life (Bergman et al., 2002; Carbone et al., 2010; Steinhausen & Juzzi, 1996), par ents appeared more than willing to comply with treatment recommendations. Treatment Acceptability All children experienced some positive changes in symptomology a lbeit these changes varied across participants. Given each family experienced failed treatments in the past, it makes sense that all parents rated the intervention as highly acceptable. While consistent improvement across all five cases was not as robust a s hoped for, all parents reported the requirements of the intervention were highly acceptable. They further reported they believed this intervention was highly acceptable for other children and families . Overall acceptability ratings were higher than previ ous studies examining the use of sertraline (i.e., Carlson et al., 1999; Eke, 2001) for the treatment of selective mutism symptoms . In this study, parents provided an overall acceptability rating of 60.6, while parents in the Carlson and colleagues (1999) provided an average rating of 58.6 and parents in the Eke (2001) study reporting an average rating of 56.8. This high acceptability rating may be useful for practitioners to share with families of children diagnosed with selective mutism who may benefit f rom treatment with fluoxetine. Overall, this study 110 suggest s fluoxetine treatment for selective mutism with co - occurring social anxiety symptoms was an acceptable treatment approach according to parents . Implications for Research Selective m utism is a severely debilitating disorder with social and academic implications (Bergman et al., 2002) . Future research needs to continue to search for effective treatments in order to improve the quality of life for these children, and subsequently improv e long - term outcomes. First, research should continue to focus on the relationship between social anxiety and speaking behaviors in children with selective mutism. As noted by Muris and Ollendick (2015), much of the research examining the relationship betw een social anxiety and selective mutism is cross sectional leaving questions about the relationship between social anxiety and selective mutism unanswered. One possible explanation is that selective mutism is better described as a specific phobia of expre ssive speech (Omdal & Galloway, 2008). Given the lack of data regarding this relationship as well as the findings from this study that indicate there was an improvement in speaking behaviors without an overall improvement in social anxiety symptoms, more r esearch is needed. Second, more single case design studies need to be conducted to examine the individual changes children experience in response to fluoxetine treatment for selective mutism symptoms. Single case designs have the capability to determine if an intervention effect is observable and important, and can determine if the change in behavior was due to the implementation of an independent variable. Kazdin (1977) stated that the impact of an intervention needs to be both observable and acceptable i n order for an intervention to be determined effective. In this study, every child showed improvement in at least one domain, and every parent rated the intervention as highly acceptable. Future single case design studies may better illuminate treatment ou tcomes by recruiting children who are similar in their current 111 functioning (i.e., engaging in social interactions but not speaking) and demographic characteristics (e.g., age, gender, socio - economic status) , and identifying the effect fluoxetine has across these similar participants . Third, future research needs to focus on the types of speech (i.e., responsive, spontaneous) children engage in during SSRI treatment. T his study only noted improvements in responsive speech, which is not considered conversati onal speech . Finally, future research should focus on the combination of psychopharmacology and behavior therapy to determine how psychosocial treatments can augment psychopharmacological treatments for selective mutism. Limitations This study used a sin gle blind , placebo - controlled , non - concurrent multiple baseline design to examine the impact of fluoxetine treatment on symptoms of social anxiety and selective mutism . This method ology has several benefits in regards to providing internal and external validity within the study; however, the sole reliance on pre - specified parent observations and ratings for the primary outcome measures across all children is a limitation of this stu dy. F uture studies should work to identify an evaluation approach that specifically targets the individual symptom profile s of each of the participant s . To be included in this study, children needed to be resistant to psychosocial treatment according to their prior practitioner for at least 10 to 12 weeks depending on age. Relying on practitioner reports of resistance to psychosocial treatment was a limitation of this study. Data regarding intervention integrity was not provided to the study coordinator o r the study psychiatrist . Consequently , it is possible that the interventions provided by these practitioners were not completed with fidelity making it seem like children were non - responders . Future 112 research on psychopharmacological interventions with chi ldren with selective mutism should strive to ensure that children are truly treatment resistant by monitoring the integrity of an evidence - based manualized treatment approach for selective mutism. The reliance on DBRs as a primary outcome measure is a li mitation of this study. Research has demonstrated that DBRs can perform adequately in comparison to SDOs, which are considered the gold standard in observations of problem behaviors. For instance, research has demonstrated that SDOs and DBRs of the same ob servation periods are highly correlated, and DBRs account for 75% of the variance of SDOs (Riley - Tillman et al., 2008). Despite this, DBR is a relatively new observation approach and more research needs to be conducted regarding their sensitivity to change s in behavior due to treatment. The sole reliance on parents to complete DBRs for the primary outcome variables of social anxiety, responsive speech, and spontaneous speech was a lso a limitation of this study. P arents are not trained mental health practiti oners and, therefore, may not have always reliably report ed the observed data. Analyzing teacher data in the same manner as parent data would have added more information regarding selective mutism symptoms across contexts; however, some teachers were not available due to the time of year of the intervention ( i.e. , summer vacation) , thus rendering the sample size too small for any kind of statistical analysis . Systemic direct observations (SDOs) completed by an observer trained in mental health and assessme nt are the gold standard in documenting treatment outcomes. Despite receiving training in DBRs, it is possible that parents committed observation errors (e.g., observer drift, observer bias) given the lack of mental health and assessment training. While th e reliance on parent - completed ratings is a limitation, Chafouleas and colleagues (2012) reported improvements in accuracy on DBRs completed by non - mental - health practitioners after training was implemented. The training parents received in the current stu dy was modeled after the 113 Chafouleas and colleagues approach. Future studies should strive to involve SDOs conducted by trained mental health practitioners. The removal of Child R from this study is a limitation and is one of the weaknesses of a non - concu rrent multiple - baseline design, as additional time cannot be provided to wait for ratings to stabilize (Christ, 2007 ; Watson & Workman, 1981 ). The consistent engagement in the target behaviors according to parent report was unexpected, given the intensive screening process. Although it was not ideal to have to remove this child from the analysis, there were no specific variables that were apparent to the researchers that make this child significantly different from the remainder of the sample. The use of a single blind and t he constraints of the dosing schedule were also limitations. In order for the psychiatrist to be able to adjust doses as necessary in the best interest of the child, a double - blind procedure was not feasible. Alt hough this is the case, there were multiple pieces of data collected from multiple stakeholders. The data from the psychiatrist was viewed in lieu of the data collected fr om parents and teachers who were not aware of the treatment schedule. T he constraints of the dosing schedul e (i.e., two weeks of an introductory dose followed by the therapeutic dose) left little flexibility for the study psychiatrist to increase doses to an optimal level. Future single - case design studies examining SSRI treatment for selective mutism may benef it from having an introductory dose, a first therapeutic dose, and then a second therapeutic dose if there is still potential for improvement. A reason for the lack of a more robust response in social anxiety symptoms may be that more time was need for th e fluoxetine treatment to be effective . While some studies report tangible improvements can be obtained within the first week of treatment (Mitchell, 2006), it may take six weeks or longer for some individuals to begin to show an improvement (Garfield et 114 a l., 2004). In a larger scale study of childhood anxiety disorders, approximately two thirds of children responded to 20 mg/day of fluoxetine after four to six weeks of treatment (Birmaher et al, 2003). It is possible that children in this study may have ex hibited a more robust response given more time on the therapeutic dose of the medication. Future studies should extend the treatment phase to allow a demonstration of a treatment effect after six to eight weeks of the therapeutic dose. 115 APPENDICES 116 A ppendix A Y oung M ania R ating S cale Parent Version (Adapted) Directions: Please read each question below and circle the answer number that best describes your child. 1. Mood ? A. No B. Mildly or possibly increased C. Definite elevation more optimistic, self - confident, cheerful, appropriate to their conversation. D. Elevated but inappropriate to content; joking, mildly silly E. Euphoric, inappropriate laughter; singing/making noises, very silly 2. Motor Activity/Energy greater than usual? A. No B. Mildly or possibly increased C. More animated; increased gesturing D. Energy is excessive; hyperactive at times; restless but can be ca lmed E. Very excited; continuous hyperactivity; cannot be calmed 3. Sleep A. No B. Sleeping less than normal amount by up to one hour C. Sleeping less than normal amount by more than one hour D. Need for sleep appears decreas ed; less than four hours E. Denies need for sleep; has stayed up one night or more 4. Irritability Has your child appeared irritable? A. No more than usual B. More grouchy or crabby C. Irritable openly several times throughout the day; recent episodes of anger wit h family, at school, or with friends D. Frequently irritable to point of being rude or withdrawn E. Hostile and uncooperative about all the time 117 5. Speech (rate and amount) Is your child talking more quickly or more than usual? A. No Change B. Seems more ta lkative C. Talking faster or more to say at times D. Talking more or faster to point he/she is difficult to interrupt E. Continuous speech; unable to interrupt 6. Thoughts Has your child shown changes in his/her thought patterns? A. No B. Thinking faster; some decreas C. Distractible; loses track of the point; changes topics frequently; thoughts racing D. Difficult to follow; goes from one idea to the next; topics do not relate; makes rhymes or repeats words E. Not understandable; 7. Content Is your child talking about different things than usual? A. No B. He/she has new interests and is making more plans C. Making special projects; more religious or interested in God D. Thinks more of him/herself; belie ves he/she has special powers; believes he/she is receiving special messages E. Is hearing unreal noises/voices; detects odors no one else smells; feels unusual sensations; has unreal beliefs 8. Disruptive - Aggressive Behavior Has your child been more disrup tive or aggressive? A. No; he/she is cooperative B. Sarcastic; loud; defensive C. More demanding; making threats D. Has threatened a family member or teacher; shouting; knocking over possessions/furniture or hitting a wall E. Has attacked a family member, teacher, or peer; destroyed property; cannot be spoken to without violence 118 Appendix B Letter to School Psychologists, Private Practitioners, and Early Childhood Mental Health Workers Figure 7 Letter to Professionals 119 Appendix C Recr uitment Flyer Figure 8 Recruitment Flyer 120 Appendix D Letter to Principals Figure 9 Letter to Principals 121 Appendix E Inclusion and Exclusion Criteria linic visit. A. Inclusions: Please check all those that apply. If all criteria are not met then child is not eligible for participation in this research project. _____ Male or female from five to 17 years of age at their last birthday _____ Met DSM - 5 criteria for Selective Mutism _____ No history of medication treatment for Selective Mutism _____ Child has an immediate biological family member who is diagnosed with an anxiety disorder or has experienced symptoms of an anxiety disorder at s ome point in time _____ Child has received 10 weeks of an evidence - based psychosocial treatment if seven years or older; Child has received 12 weeks of an evidence based psychosocial treatment if six years or younger _____ Child has never had a negati ve reaction to a psychopharmacological medication _____ T - Scores of 65 or higher on MASC - 2:SAS scale B. Exclusions: do not apply. If any of the criteria are not negated, the child is not elig ible for participation in this research project . _____ Diagnosed with a speech condition, mental retardation, pervasive developmental disorder, or has a diagnosis of schizophrenia _____ Child is an English language learner or from a different cultur e than the culture predominately represented within his or her school _____ Child is taking or has taken any kind of a psychopharmacological medication (e.g., SSRI, MAO - I, stimulant, etc.) _____ Child has a medical illness that may be complicated through the use of a psychopharmacological treatment as determined by the project psychiatrist Adapted from Carlson (1997). 122 Appendix F Correlation Matrix Table 19 Correlation Matrix Appendix G Medication Compliance Reporting Form 123 A ppendix G Medication Log Please fill out this form for the following week: _____________________________ Please write the time when you provided your child with the medication for each day of the week. Sunday Monday Tuesday Wednesd ay Thursday Friday Saturday Time 124 Appendix H Early Exit Form Figure 10 Early Exit Form 125 APPENDIX I End of Study Form Figure 11 End of Study Form 126 A ppendix J Direct Behavior Rating Forms Teacher and Parent Te acher - Daily Behavior Rating Form Date:__________________ Day (Please Circle) : M T W TH F Directions: Place a mark along the line that best reflects the percentage of total time the student exhibited each target behavior each day he/she is at school. Note that the percentages do not need to total 100% across behaviors because some behaviors may co - vary. 1. Behavior: When appropriate, the student appeared comfortably and socially engaged with other stude nts (e.g., completing group work, playing). *Note : This rating does not include whether the child spoke to other students. 2. Behavior: When appropriate, the student was comfortably and socially engaged with the classroom teacher. *Note: This rating does not include whether the child spoke to the teacher. 3. Behavior: When appropriate, the student spontaneously spoke to other students. 127 4. Behavior: When appropriate, the student spontaneously spoke to the teacher. 5. Behavior: When appropriat e, the student responded (e.g., saying yes or no) when spoken to by other students. 6. Behavior: When appropriate, the student responded (for example, saying yes or no) when spoken to by the teacher. 128 Parent - Daily Behavior Rating Form Date: __________________ Day (Please Circle) : M T W TH F Directions: Place a mark along the line that best reflects the percentage of total time the student exhibited each target behavior. Note that the perce ntages do not need to total 100% across behaviors. Please cross out behaviors you did not get a chance to observe during a day. It is important that you attempt to rate each behavior (1 - 4) at least 3 times per week. 1. Behavior: My child appeared comfortabl y and socially engaged with friends (e.g., playing with a neighborhood friend). *Note : This rating does not include whether the child spoke to the children . 2. Behavior: My child appeared comfortably and socially engaged with the unfamiliar adults. *Note : This rating does not include whether the child spoke to the unfamiliar adult. 3. Behavior: When appropriate, my child spontaneously spoke to friends. 4. Behavior: When appropriate, my child spontaneously spoke to unfamiliar adults. 129 5. Behavior: W hen appropriate, my child responded (for example, saying yes or no) when spoken to by friends. 6. Behavior: When appropriate, my child responded (for example, saying yes or no) when spoken to by unfamiliar adults. 130 Ap pendix K Multidimensional Anxiety Scale for Children 2 nd Edition: Social Anxiety Scale Instructions: These sentences ask you how your child might have been thinking, feeling, or acting recently. For each item, please circle the number that describes how often the statement is true about your child. Circle 0 if a sentence is Never true about your child. Circle 1 if a sentence is Rarely true about your child. Circle 2 if a sentence is Sometimes true about your child. Circle 3 if a sentence is Often tru e about your child. Remember, there are no right or wrong answers, just answer how your child has been feeling since the last time you completed this form. Here is an example to show you how to complete the questionnaire. In the example, if your child about your child. Never Rarely Sometimes Often Example My child is scared of dogs. 0 2 3 Never Rarely Sometimes Often 1. My child worr ies about other people 0 1 2 3 2. My child is afraid that other kids will 0 1 2 3 3. My child worries about getting called 0 1 2 3 4. My child is afraid other people wi ll 0 1 2 3 5. My child worries about what other 0 1 2 3 6. My child worries about doing 0 1 2 3 7. My child gets nervous if he/she has to perform in 0 1 2 3 8. My child has trouble asking other 0 1 2 3 9. 0 1 2 3 1 131 A ppendix L Clinical Global Impressions (Severity and Change) Clinician, Parent, Teacher Severity of Beha vior Form Clinician Participant _____________ Date______________ Clinician Name __________________ Phase of study: ____________________________________ I. Selective Mutism Severity Provide a rating for each of the following settings: A. School ____ _ B. Home ______ C. Clinic______ D. Peers ______ E. Overall______ 1. Absent No apparent difficulty or reluctance to speak or communicate in any situation. 2. Minimal Some reluctance to speak or communicate in one or more situations, in speak or communicate in any situation. For example, child speaks up or raises his or her hand to speak less often than most children, may be slow to respond when spoken to or may speak softly. Does not appear particularly distressed when expected to speak. 3. Moderate Modest reluctance or unwillingness to speak or communicate in one or more situations. The reluctance or unwillingness to speak or communicate is significant eno to speak or communicate in one or more situations. For example, the child rarely speaks spontaneously to teacher, and is moderately reluctant to respond even when spoken to, responding slowly or very quietly. May appear somewhat distressed when expected to speak. 4. Marked Reluctance or unwillingness to speak or communicate in one or more situations is markedly abnormal, and is significant enough that there is marked interference with t one or more situations. For example, the child never speaks spontaneously to teacher, and is notably reluctant to respond even when spoken to, responding very slowly, or in a mere whisper, and sometimes not at all. May appear quite distressed when expected to speak. 5. Severe Child is completely unwilling or unable to speak or communicate in one or more situations, and that is significant enough that there is severe y to speak or communicate in one or more situations. For example, the child does not speak at all to the teacher even in a whisper and even in response to direct questions. May appear extremely distressed when expected to speak, for example, by crying. X Unable to assess. 132 II. Shyness Severity (Circle one of the items below) 1 Absent No problem, comfortable, conversant. 2 Minimal Somewhat shy, some avoidance of eye contact, delayed verbal responses to questions, short answers. 3 Moderate Moder ately shy, very brief responses (yes or no) 4 Marked Very shy, no verbal responses, but nodded or provided written responses, some eye contact 5 Severe Virtually no interaction with examiner, no responses at all to any questions, virtually no e ye contact. X Unable to assess. III. Anxiety Severity 1 Absent No anxiety, fear, or nervousness, and normal school, family, and social activities. 2 Minimal Some anxiety, fear, or nervousness some days; but they are not interfering with normal s chool, family, and social activities. 3 Moderate Anxiety, fear, or nervousness most days, interfering with normal school, family, and social activities in minor ways. 4 Marked Considerable anxiety, fear, or nervousness most days or everyday; interferin g with normal school, family, or social activities. 5 Severe Severe anxiety, fear or nervousness every day or nearly every day; normal school, family, and social activities are very disrupted. X Unable to assess. 133 Severity of Behavio r Form - Teacher Participant_________ Date___________ Teacher__________________ I. Selective Mutism Severity: unwillingness, or inability to speak since the last rating. Circle one number. 1 A bsent No apparent difficulty or reluctance to speak or communicate in any situation. 2. Minimal Some reluctance to speak or communicate in one or more situations, in speak or c ommunicate in any situation. For example, child speaks up or raises his or her hand to speak less often than most children, may be slow to respond when spoken to or may speak softly. Does not appear particularly distressed when expected to speak. 3. Moderate Modest reluctance or unwillingness to speak or communicate in one or more situations. The reluctance or unwillingness to speak or communicate to speak or communicate i n one or more situations. For example, the child rarely speaks spontaneously to teacher, and is moderately reluctant to respond even when spoken to, responding slowly or very quietly. May appear somewhat distressed when expected to speak. 4. Marked Relu ctance or unwillingness to speak or communicate in one or more situations in markedly abnormal, and is significant enough that there is one or more situations. For example, th e child never speaks spontaneously to teacher, and is notably reluctant to respond even when spoken to, responding very slowly, or in a mere whisper, and sometimes not at all. May appear quite distressed when expected to speak. 5. Severe Child is completely unwilling or unable to speak or communicate in one or more situations, and that is significant enough that there is severe more situations. For example, the child does not speak at all to the teacher even in a whisper and even in response to direct questions. May appear extremely distressed when expected to speak, for example, by crying. X Unable to assess. 134 I. SELECTIVE MUTISM SEVERITY (contd.) To your knowledge, in which of the following situations has the child been reluctant, unwilling, or unable to speak since the last rating? Please use the following ratings: 0=no difficulty in speaking 1=somewhat difficulty in speaking 2=very reluctant to sp eak 3=never speaks in this situation Situations: In the classroom, to other children? _______ In the classroom, to teacher? _______ In the classroom, to other adults? _______ In other situations (e.g., in hallways, on playground, in gym), to other children? _______ In other school situations, to teacher? _______ In other school situations, to other adults? _______ II. SHYNESS SEVERITY (Circle one of the items below): 1 Absent No problem, comfortab le, conversant 2 Minimal Somewhat shy, some avoidance of eye contact, delayed verbal responses to questions, short answers 3 Moderate Moderately shy, very brief responses (yes or no). 4 Marked Very shy, no verbal responses, but nodded or provided wri tten response, some eye contact. 5 Severe Virtually no interaction with examiner, no response at all to any questions, virtually no eye contact X Unable to assess. III. Anxiety Severity (Circle one of the items below): 1 Absent No anxiety, fea r, or nervousness, and normal school, family, and social activities. 2 Minimal Some anxiety, fear, or nervousness some days; but they are not interfering with normal school, family, and social activities. 3 Moderate Anxiety, fear, or nervousness most days, interfering with normal school, family, and social activities in minor ways. 4 Marked Considerable anxiety, fear, or nervousness most days or everyday; interfering with normal school, family, or social activities. 5 Severe Severe anxiety, fear o r nervousness every day or nearly every day; normal school, family, and social activities are very disrupted. X Unable to assess 135 Severity of Behavior Form Parent Participant _____________ Parent__________________ Date: I. Selective Mutism Severity : unwillingness, or inability to speak since the last rating. Circle one number 1 Absent No apparent difficulty or reluctance to speak or communicate in any situation. 2. Minimal Some reluc tance to speak or communicate in one or more situations, in speak or communicate in any situation. For example, child speaks up or raises his or her hand to speak less often than m ost children, may be slow to respond when spoken to or may speak softly. Does not appear particularly distressed when expected to speak. 3. Moderate Modest reluctance or unwillingness to speak or communicate in one or more situations. The reluctan ce or unwillingness to speak or communicate is significant enough that there is some interference with example, the child rarely speaks spontaneously to teacher, and is moderately reluctant to respond even when spoken to, responding slowly or very quietly. May appear somewhat distressed when expected to speak. 4. Marked Reluctance or unwillingness to speak or communicate in one or more situations in markedly abnormal, and is significant enough that there is one or more situations. For example, the child never speaks spontaneously to teacher, and is notably reluctant to respond even when spoken to, re sponding very slowly, or in a mere whisper, and sometimes not at all. May appear quite distressed when expected to speak. 5. Severe Child is completely unwilling or unable to speak or communicate in one or more situations, and that is significant enou gh that there is severe more situations. For example, the child does not speak at all to the teacher even in a whisper and even in response to direct questions. May appear extreme ly distressed when expected to speak, for example, by crying. X Unable to assess. 136 To your knowledge, in which of the following situations has your child been reluctant, unwilling, or unable to speak since the last rating? Please use the followi ng ratings: 0=no difficulty in speaking 1= somewhat difficulty in speaking 2=very reluctant to speak 3= never speaks in this situation Situations with mother in your home? _________ with mother away from home? _______ __ with father in your home? _________ with father away from home? _________ with siblings in your home? _________ with siblings away from home? _________ with extended family in your home? _________ with extended family away from home? ______ ___ with familiar adults in your home? _________ with familiar adults away from home? _________ with unfamiliar adults in your home? _________ with unfamiliar adults away from home? _________ with familiar children in your home? _________ with famil iar children away from home? _________ with unfamiliar children in your home? _________ with unfamiliar children away from home? _________ other _______________? II. Shyness Severity Please provide a rating for each of the following settings. Home_ _______ School________ Community________ 1 Absent No problem, comfortable, conversant 2 Minimal Somewhat shy, some avoidance of eye contact, delayed verbal responses to questions, short answers. 3 Moderate Moderately shy, very brief responses (yes or no). 4 Marked Very shy, no verbal responses, but nodded or provided written responses, some eye contact. 5 Severe Virtually no interaction with examiner, no responses at all to any questions, virtually no eye contact. X Unable to assess. 137 I II. Anxiety Severity Please proved a rating for each of the following settings: Home___________ School___________ Community___________ 1 Absent No anxiety, fear, or nervousness, and normal school, family, and social activities 2 Minimal Some anx iety, fear, or nervousness some days; but they are not interfering with normal school, family, and social activities. 3 Moderate Anxiety, fear, or nervousness most days; interfering with normal school, family, and social activities in minor ways. 4 Marked Considerable anxiety, fear, or nervousness most days or everyday; interfering with normal school, family, and social activities. 5 Severe Severe anxiety, fear or nervousness every day or nearly every day; normal school, family, and social act ivities are very disrupted. X Unable to assess. 138 Global Change Form Clinician Participant ______________ Date_______________ Clinician Name__________________ I. GLOBAL MUTISM CHANGE A. School ________ B. Home_______ C. Clinic_________ D. Overall__________ 1. Markedly improved 5. Minimally worse 2. Moderately improved 6. Moderately worse 3. Minimally improved 7. Markedly worse 4. No change II. GLOBAL SHYNESS CHAN GE A. Clinic setting ________ 1. Markedly improved 5. Minimally worse 2. Moderately improved 6. Moderately worse 3. Minimally improved 7. Markedly worse 4. No change III. GLOBAL ANXIETY CHANGE A. Generalized Anxiety_______ B. Social Anxiety________ 1. Markedly improved 5. Minimally worse 2. Moderately improved 6. Moderately worse 3. Minimally improved 7. Markedly worse 4. No change IV. GLO BAL SIDE EFFECT SEVERITY RATING SCALE How do medication side effects compare with beneficial medication effects? 0 Not applicable (there have been no positive changes or negative changes) 1 Positive changes greatly outweigh negative changes. Medication effe cts are overall extremely positive. 2 Positive changes significantly outweigh negative changes. Medication effects are overall somewhat positive. 3 Positive changes outweigh negative changes. Medication effects are overall somewhat positive. 4 Positive and negative changes are approximately equal. 5 Negative changes outweigh positive changes. Medication effects are overall somewhat negative 6 Negative changes significantly outweigh positive changes. Medication effects are overall very negative 7 Negative chang es greatly outweigh positive changes Medication effects are overall extremely negative 139 Global Change Form Teacher Participant ______________ Date_______________ Teacher__________________ I. SELECTIVE MUTISM CHANGE : Do you think it has become an y easier for the child to talk in school (or in other situations or places where it has been hard for him or her to talk) than it was before he or she began participating in this treatment study, or is it getting more difficult, or has there not been any c hange? Circle one number. 1 Much easier 2 A little easier 3 No change 4 A little more difficult 5 Much more difficult II. S HYNESS CHANGE: Do you think the child has been any less shy recently compared to before he or she began participating in this treatment s tudy, or has he or she been more shy, or has there not been any change? Circle one number. 1 Much less shy 2 A little less shy 3 No change 4 A little more shy 5 Much more shy III. ANXIETY OR NERVOUSNESS CHANGE: Do you think the child has been any less anxious or nervous recently compared to before he or she began participating in this treatment study, or has he or she been more anxious or nervous, or has there not been any change? Circle one number. 1 Much less nervous 2 A little less nervous 3 No change 4 A little mo re nervous 5 Much more nervous IV. COMMENTS: mentioned above? 140 Global Change Form - Parent Participant______________ Date____________ Parent_____________________ Person comple ting this form: Mother__________ Father________ Both parents_______ I. SELECTIVE MUTISM CHANGE: Do you think it has become any easier for your child to talk in school (or in other situations or places where it has been hard for him or her to talk) than i t was before he or she began participating in this treatment study, or is it getting more difficult, or has there not been any change? Circle one number. 1 Much easier 5 A little more difficult 2 Somewhat easier 6 Somewhat more difficult 3 A little easier 7 Much more difficult 4 No change II. S HYNESS CHANGE: Do you think your child has been any less shy recently compared to before he or she began participating in this treatment study, or has he or she been more shy, or has there not been any change? Ci rcle one number. 1 Much less shy 5 A little more shy 2 Somewhat less shy 6 Somewhat more shy 3 A little less shy 7 Much more shy 4 No change III. ANXIETY OR NERVOUSNESS CHANGE: Do you think your child has been any less anxious or nervous recently co mpared to before he or she began participating in this treatment study, or has he or she been more anxious or nervous, or has there not been any change? Circle one number. 1 Much less nervous 5 A little more nervous 2 Somewhat less nervous 6 Somewhat m ore nervous 3 A little less nervous 7 Much more nervous 4 No change 141 IV. GLOBAL SIDE EFFECT SCALE How do medication side effects compare with beneficial mediation effects? 0 Not applicable (there have been no positive changes or negative chang es) 1 Positive changes greatly outweigh negative changes Medication effects are overall extremely positive. 2 Positive changes significantly outweigh negative changes. Medication effects are overall very positive. 3 Positive changes outweigh negative ch anges Medication effects are overall somewhat positive. 4 Positive and negative changes are approximately equal. 5 Negative changes outweigh positive changes. Medication effects are overall somewhat negative. 6 Negative changes significantly outweigh positive changes. Medication effects are overall extremely negative. V. COMMENTS: mentioned above? 142 A p pendix M Selective Mutism Questionnaire Selective Mutism Q uestionnaire (SMQ) © statement is true for your child. AT SCHOOL 1. When appropriate, my child talks to most peers at school. Always Often Seldom Never 2. When appropriate, my child talks to selected peers (his/her friends) at school. Always Often Seldom Never 3. When my child is asked a question by his/her teacher, s/he answers. Always Often Seldom Never 4. When appropriate, my child asks his or her teacher questions. Always Often Seldom Never 5. When appropriate, my child speaks to most teachers or staff at school. Always Often Seldom Never 6. When appropriate, my child speaks in groups or in front of the class. Always Often Seldom Never HOME/ FAMILY 7. When appropriate, my child talks to family members living at home when other people are present. Always Often Seldom Never 8. When appropriate, my child talks to family members while in unfamiliar places. Alway s Often Seldom Never grandparent, cousin). Always Often Seldom Never 143 10. When appropriate, my child talks on the phone to his/her parents and siblings. Always Often Seldom Never 11. When appropriate, my child speaks with family friends who are well - known to him/her. Always Often Seldom Never 12. My child speaks to at least one babysitter. Always Often Seldom Never N/A IN SOCIAL SITUATIONS (OUTSIDE OF SCHOOL) Always Often Seldom Never Always Often Seldom Ne ver 15. When appropriate, my child speaks with his or her doctor and/or dentist. Always Often Seldom Never 16. When appropriate, my child speaks to store clerks and/or waiters. Always Often Seldom Never 17. When appropriate, my child talk s when in clubs, teams or organized activities outside of school. Always Often Seldom Never Interference/Distress* 18. How much does not talking interfere with school for your child? Not at all Slightly Moderately Extremely 19. How much doe s not talking interfere with family relationships? Not at all Slightly Moderately Extremely 144 20. How much does not talking interfere in social situations for your child? Not at all Slightly Moderately Extremely 21. Overall, how much does not talking interfere with life for your child? Not at all Slightly Moderately Extremely 22. Overall, how much does not talking bother your child? Not at all Slightly Moderately Extremely r you? Not at all Slightly Moderately Extremely Scoring: Always = 3; Often = 2; Seldom = 1; Never = 0 *These items are not included in total score and are for clinical purposes only. 145 A p pendix N Adapted Side Effects for Children and Adolescents (SEFCA) To be obtained from patient and caretaker: Are complaints or signs present? 1= No _________ 2 = Yes If YES, complete appropriate items below rating frequency and severity as follows: Frequency (days per week) 1= 1 - 2 days 2= 3 - 4 days 3= 5 - 7 days Severity : 1 = mild, does not interfere with functioning 2= moderate, some interference with functioning 3= severe, functioning is significantly impaired because of side effects Frequency Severity_ 1. Dro _____ ________ 6. Depression (reduction of ability to enjoy 7. Appetite D _ ________ 12. Behavioral Activation/Manic Behaviors/ 146 A ppendix O Answers on the Parent - Mania Rating Scale that May Indicate Behavioral Disinhibition as a Result of Fluoxetine Treatment 1. Mood A. No B. Mildly or possibly increased C. Definite elevation more optimis tic, self - confident, cheerful, appropriate to their conversation. D. Elevated but inappropriate to content; joking, mildly silly E. Euphoric, inappropriate laughter; singing/making noises, very silly 2. Motor Activity/Energy motor activity appear to be greater than usual? A. No B. Mildly or possibly increased C. More animated; increased gesturing D. Energy is excessive; hyperactive at times; restless but can be calmed E. Very excited; continuous hyperactivity; cannot be calmed 3. Sleep A. No B. Sleeping less than normal amount by up to one hour C. Sleeping less than normal amount by more than one hour D. Need for sleep appears decreased; less than four hours E. Denies need for sleep; has stayed up one night or more 4. Irritability Has your child appeared irritable? A. No more than usual B. More grouchy or crabby C. Irritable openly several times throughout the day; recent episodes of anger with family, at school, or with friends D. Frequently irritable to point of being rude or withdrawn E. Hostile and uncooperative about all the time 147 5. Speech (rate and amount) Is your child talking more quickly or more than usual? A. No Change B. Seems more talkative C. Talking faster or more to say at times D. Talking more or faster t o point he/she is difficult to interrupt E. Continuous speech; unable to interrupt 6. Thoughts Has your child shown changes in his/her thought patterns? A. No B. C. Distractible; loses tra ck of the point; changes topics frequently; thoughts racing D. Difficult to follow; goes from one idea to the next; topics do not relate; makes rhymes or repeats words E. 7. Content Is your child talking about different things than usual? A. No B. He/she has new interests and is making more plans C. Making special projects; more religious or interested in God D. Thinks more of him/herself; believes he/she has special powers; believes he/she is receiving special mes sages E. Is hearing unreal noises/voices; detects odors no one else smells; feels unusual sensations; has unreal beliefs 8. Disruptive - Aggressive Behavior Has your child been more disruptive or aggressive? A. No; he/she is cooperative B. Sarcastic; loud; defens ive C. More demanding; making threats D. Has threatened a family member or teacher; shouting; knocking over possessions/furniture or hitting a wall E. Has attacked a family member, teacher, or peer; destroyed property; cannot be spoken to without violence 148 A p pendix P Treatment Evaluation Questionnaire, parent version 149 A p pendix Q Consent Forms and Assent Procedures for Participation (Parent, Teacher, Child) EVALUATING THE RESULTS OF PHYSICIAN AND PARENT DECISIONS TO TREAT SELECTIVE MUTISM W ITH FLUOXETINE PARENT INFORMATION AND CONSENT FORM YOU ARE INVITED TO GIVE PERMISSION FOR YOUR CHILD TO PARTICIPATE IN A RESEARCH STUDY ON THE TREATMENT OF SELECTIVE MUTISM. Researchers and Titles: 1) Justin Barterian M.A., Doctoral Candidate in Schoo l Psychology 2) John Carlson Ph.D., Professor and Psychologist 3) Jed Magen D.O., Professor and Psychiatrist Institution: Michigan State University Departments: 1) Department of Counseling, Educational Psychology, an d Special Education 2) Department of Psychiatry Contact Information: Mr. Justin Barterian Michigan State University 620 Farm Lane 401 C Erickson Hall East Lansing, MI 48823 Phone: 586 - 899 - 3715 Email: barteria@msu.edu PURPOSE OF RESEARCH The purpose of this research study is to see if a medication called fluoxetine sometimes called Prozac can help your child talk with oth ers. Fluoxetine is a medication approved by the U.S. Food and Drug Administration for anxiety (Obsessive - Compulsive Disorder) in children ages 7 to 18 years old. You and your child have been asked to be in this research study because your child has had tro uble talking to others. Many families will be interviewed to see if their child is a good candidate for the treatment. Only five families, who have children ages 5 to 17 years old, will be selected to come to the MSU Psychiatry Clinic to receive treatment. ALTERNATIVE OPTIONS If you decide not to take part in this research study, you should know that there are other treatments that may be helpful for selective mutism. Other medications, like Zoloft, and behavior therapy have been helpful for some child ren. Behavior therapy often is used to increase how these other treatments. 150 If you leave the study, there are two different choices the project coordinator wil l offer you to get school. Second, you may be sent to a mental health worker in the community. The payment for this treatment will be your responsibility. W HAT WILL YOU DO Overview This research study occurs in two phases. First, we will conduct a phone interview to see if your symptoms, the mental health histor for us. After the interview, you may be sent a screening packet in the mail that will contain forms for you to complete, including release of information permission forms, questionnaires about If your child is invited to the clinic, you and your child will be enrolled in the treatment part of the study for approximately 15 weeks. You and your child will come to the clinic about once every two weeks (you can come sooner if needed). You will come to the clinic 9 times. These visits will take about one hour. The researchers will the study to see if the symptoms are getting better and to make sure your child is not feeling worse. You should feel free to contact the psychiatrist if you have questions about the medication. Randomization Dur behaviors. During this week, your child will not receive any treatment. After the first week, your child will receive the medication (fluoxetine) or syrup that looks an d tastes like the medication, but has no medication in it. This is done to see if the real medication is helping your child. This means your child will not get the real medication during some weeks of the study. However, your child will get the real medic ation at some point in the study. While you, your child, and will know. The time your child gets the real medication is decided by chance and is not determin ed by any information you give us. Since you will not know if your child is getting the real medication you will not know how the medication worked for your child until the end of the research study. School and Home Involvement Information about your times during this research study. This information will help the researchers know if the medicat ion is helping your child. 151 Clinic Visits You will have to come to the Michigan State University Psychiatry Clinic in East Lansing, MI nine times during this study. During these visits, you and your child will meet with the project coordinator or one of his assistants. The project coordinator and assistants have Masters degrees in psychology. You and your child will also met with the psychiatrist who will prescribe the medication. During these meetings you may be asked complete forms and talk about how y our child is doing with project coordinator, project assistants, and psychiatrist. This will help us know how the medication may be helping your child. of th e study, the staff will help you chose other treatments or help you secure treatment for your child with a professional in your community. Fluoxetine treatment may be continued after the study; however, you would have to pay for the treatment. More info rmation about your visits to the MSU psychiatry clinic is give below. Visit 1 You will meet with a researcher who will get you ready for your meeting with the psychiatrist. He or she will answer any questions you have about the research study. The resea rcher will also talk to your child to make sure they agree to be in the study. After this, you family, and psychiatric history. The psychiatrist will physicall y examine your child to make sure they are healthy and meet the requirements for treatment. The psychiatrist may ask other questions to make sure your child is a good fit for the study. Next, if your child is still thought to be a good fit for the study, t he researcher will meet with you again for about one hour to give you training on the forms you will be asked to complete every week. You will be asked to complete several forms during the week and asked to come back next week. Visits 2 - 8 During these visits, your child will be prescribed the medication or syrup that tastes and looks like the medication. When your child is given the medication, s/he will be prescribed a lower dose for two weeks, which will be raised to a higher dose after those two wee ks. The psychiatrist will decide the dose for your child, because each child is different and may need different doses than other children. When your child receives the syrup that does not contain the medication, the dose might also look to change so you c annot tell if your child is on the medication or the syrup. During these visits you and your child will meet with the psychiatrist to discuss how your child psyc Visit 9 During the last visit, you and your child will do the same things as visits two through eight. However, the study psychiatrist will also complete a physical examinati on with your child and will discuss ending the research study. 152 POTENTIAL BENEFITS It is possible the medication may help our child talk. It may also help your child feel more comfortable in social situations. However, it is also possible that your ch ild may not improve behavior, which may help you decide how to best help your child. other children who also have trouble speaking in public. For example, the information from this study may add to help doctors and psychologists better treat selective mutism. The results of the study may also benefit other children who might later be treat ed with the same medication. POTENTIAL RISKS Medical Risks The side effects of fluoxetine have not been studied fully in children, especially in children younger than 7 years of age. Because of this, the Food and Drug Administration (FDA) has not ap proved the use of fluoxetine for children younger than age seven for any disorder. In addition, the FDA has not approved the use of fluoxetine for the treatment of selective mutism in children. It is possible that unexpected immediate and long - term side ef fects may occur if your child is below the age of 7. Some of these unexpected side effects may include headaches, stomachaches, and behavior changes. In children who are older than 7 who have taken the medication for other anxiety disorders and depression , most children have not had bad side effects. However, sometimes people do experience side effects when taking fluoxetine. These side effects are listed below: Asthenia (loss of strength) Flu Syndrome Vasodilation (Widening of Blood Vessels) Nausea Di arrhea Anorexia Dry Mouth Dyspepsia (Upset Stomach/Indigestion) Insomnia Constipation Flatulence Vomiting Weight Loss Thinking Abnormal Mania/Hypomania Thirst Hyperkinesia (Abnormal Movements) 153 Agitation Personality Disorder Epistaxis (Nose bleeds) Anxie ty Nervousness Somnolence (Drowsiness) Tremor Libido Decreased (decreased sex drive) Urinary Frequency Mennorrhagia (Vaginal Bleeding) Abnormal Dreams Pharyngitis (Inflammation of the Throat) Sinusitis (Inflammation of the Sinuses) Yawn Sweating Rash Pru ritus (Itching) Abnormal Vision Impotence Abormal Ejaculation Some uncommon but serious side effects have been linked to the use of fluoxetine. For example, other medications like fluoxetine have been connected to an increase in suicide in children, ado lescents, and young adult patients. However, this is rare. Also, in some patients, serotonin syndrome, which is an uncommon condition that happens when there is too much serotonin in the brain can develop and be life threatening. However, serotonin syndrom e is very uncommon in children receiving normal doses of this type of medication. As with any medication, some unexpected side effects may come about. If the researchers learn of any information that might change your mind about allowing your child to st ay in this study, the research staff will tell you. We encourage you and your child to tell us about any side effects he or she is experiencing or if he or she is feeling worse. Although research has shown no effects of fluoxetine on mental or physical p erformance, medications like fluoxetine have been shown to weaken the mental and physical skills needed to do dangerous activities like driving a car or using heavy machinery. Since this may happen with fluoxetine, it is important to be careful. Any othe r medications used during the study should be discussed with the study psychiatrist, who will inform you if this additional medication is safe to use during the research study. PRIVACY AND CONFIDENTIALITY As with any psychiatric/medical treatment, t here is a small risk that confidentiality will be broken. However, your confidentiality will be protected to the maximum extent allowable by 154 law. Only the investigators, the study coordinator, the MSU Human Research Protection Program (HRPP), and under ce rtain circumstances, the U.S. Food and Drug Administration will confidential. Confidentiality is also insured by the use of patient initials on case records, instea d of patient names. Identification numbers will be used when entering information into a database and the codebook will remain in a locked file. The locked file will be located on Michigan State rs after the study is completed. If may be shared with a safet y committee of medical professionals. No identifying information (for example, names, birthdates, addresses) will be provided to the committee. A description of this clinical trial will be available on http://www.ClinicalTrials.gov, as required by U.S. L aw. This Web site will not include information that can identify you. At most, the Web site will include a summary of the results. You can search this Web site at any time. YOUR RIGHTS TO PARTICIPATE, SAY NO, OR WITHDRAW Participation in this study is vo luntary, and you can refuse to participate or withdraw at any time without penalty or loss of benefits to which you or your child are otherwise entitled . You have the right to say no. You may choose not to answer specific questions. Under certain circums tances, patients may be withdrawn from the study without their consent if the psychiatrist decides that it is not in their best interest, or if they fail follow the procedures. COSTS AND COMPENSATION FOR BEING IN THE STUDY All assessment and treatment procedures (including medication) will be provided at no cost to you or your child and your health insurance (if applicable) will not be charged. In addition, you will receive financial assistance at the end of the study for your trips to the clinic based on the distance from your residence to Michigan State University, using the following criteria: Distance Reimbursement 25 miles $100 26 miles to 50 miles $150 51 miles to 90 miles $270 155 THE RIGHT TO GET HELP IF INJURED If you are i njured as a result of your participation in this research project, Michigan State University will assist you in obtaining emergency care, if necessary, for your research related injuries. If you have insurance for medical care, your insurance carrier will be billed in the ordinary manner. As with any medical insurance, any costs that are not covered or are in excess of what are paid by your insurance, including deductibles, will be your responsibility. The pensation for lost wages, disability, pain or discomfort, unless required by law to do so. This does not mean that you are giving up any legal rights you may have. You may contact Dr. John Carlson at 517 - 432 - 0843 with any questions or to report an injury. CONTACT INFORMATION If you have any questions about this study, such as scientific issues, how to do any part of it, or to report an injury, please feel free to contact Justin Barterian, MA at (586) 899 - 3715 or barteria@msu.edu, John Carlson, PhD at (51 7) 432 - 4856 or carlsoj@msu.edu, or Jed Magen, DO at (517) 353 - 4363 or jed.magen@hc.msu.edu . Additionally, you can contact Mr. Barterian by mail at Michigan State University, 620 Farm Lane, 401c Erickson Hall, Eas t Lansing, MI 48823. If you have questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this study, you may contact, anonymously if you wish Protection Program at 517 - 355 - 2180, Fax 517 - 432 - 4503, or email irb@msu.edu or regular mail at Olds Hall, 408 West Circle Drive #207, MSU, East Lansing, MI, 48824. DOCUMEN TATION OF INFORMED CONSENT Your signature below means you voluntarily agree to participate in this research study. Signature of Parent or Guardian: ______________________________ Date: ____________ Signature of Child/Adolescent: ________________ _______________ Date: ____________ (Ages 13 - 18) Signature of Investigator: ____________________________________ Date: ____________ 156 EVALUATING THE RESULTS OF PHYSICIAN AND PARENT DECISIONS TO TREAT SELECTIVE MUTISM WITH FLUOXETINE Proje ct Overview and Teacher Consent Form You are being asked to participate in a research study. Researchers are required to provide a consent form to inform you about the research study, to convey that participation is voluntary, to explain risks and benefi ts of participation, and to empower you to make an informed decision. You should feel free to ask researchers any questions you may have. Study Title: Evaluating the results of physician and parent decisions to treat selective mutism with fluoxetine. R esearchers and Titles: 1) Justin Barterian M.A., Doctoral Candidate in School Psychology 2) John Carlson Ph.D., Professor and Psychologist 3) Jed Magen D.O., Professor and Psychiatrist Institution: Michigan State Unive rsity Departments: 1) Department of Counseling, Educational Psychology, and Special Education 2) Department of Psychiatry Contact Information: Mr. Justin Barterian Michigan State University 620 Farm Lane 401 C Erickson Hall East Lansing, MI 48823 Phone: 586 - 899 - 3715 Email: barteria@msu.edu Purpose of the Research A research study concerning the treatment of children who have selective mutism is being conducted, and we would appreciate your participation. Many teachers of children who exhibit characteristics of selective mutism often wonder what can be done to help the child feel more comfortable speaking in certain si tuations. In order to help address these concerns, this research study will provide evaluation and treatment to children , ages five to eighteen, who exhibit the characteristics of selective mutism. The treatment will consist of a medication (fluoxetine). The families and teachers of children who meet the study criteria will be invited to participate in this study. Since selective mutism is a disorder that results in the withholding of speech within the school context, it is important to understand teacher viewpoints regarding an improvement in functioning that has occurred for the child. Children will be assessed on various measures over 157 the course of the study, in which teachers, if willing, will play an important role. The length of this research study i s approximately 15 weeks. Your participation in this study will include a 1 - hour training on data collection. In addition, you will be asked to complete rating scales that will take approximately 20 to 30 minutes per week. In addition to examining a treat ment for selective mutism, this study will, in part, fulfill requirements for a graduate degree at Michigan State University for the project coordinator, Mr. Justin Barterian. What You Will Do You have been selected to participate in this study, as a child in your classroom meets criteria for enrollment. The role of the teachers who decide to participate will be to complete regarding the child confidential. Tea functioning within the classroom on a measure twice a week. It is estimated that the completion of this measure will take no longer than 10 to 20 minutes per week. In addition, teachers will be completion of the daily rating will take no longer than 2 minutes per day. Potential Risks and Benefits This study will examine a medication treatment for ch ildren with selective mutism. Therefore, there are potential medical risks involved for the child. However, there are no known risks to teachers who are involved in this project. While there is no anticipated direct benefit for teachers in this study, they may find they are better able to communicate with the child in the classroom setting if the medication treatment is successful. Privacy and Confidentiality All information will be held strictly confidential, and no identifying information will be use d in any discussions or reports. The data you provide for this study will be kept confidential ratings in an attempt to give them a better understanding of how th e fluoxetine treatment is the data you provide include researchers and research staff and the MSU Institutional Review Board. In addition, there is a possibil ity that the U.S. Food and Drug Administration may inspect the records. All data provided by you will be kept in a locked file on the MSU campus. Identification numbers will be used when entering information into a computer database and the codebook wil l remain in a locked file. All records collected during this study will be destroyed five years after the completion of the study. 158 A description of this clinical trial will be available on http://www.ClinicalTrials.gov, as required by U.S. Law. This Web site will not include information that can identify you. At most, the Web site will include a summary of the results. You can search this Web site at any time. Your Rights to Participate, Say No, or Withdraw Participation is voluntary. Refusal to partic ipate will involve no penalty or loss of benefits to which you are otherwise entitled. You may discontinue participation at any time without penalty or loss of benefits to which you are otherwise entitled. Costs and Compensation for Participation There is no cost for you to participate in this study and you will receive a $50 gift card for school supplies at the end of the study for your participation. Contact Information If you have any questions about this study, such as scientific issues, how to d o any part of it, or to report an injury, please feel free to contact Justin Barterian, MA at (586) 899 - 3715 or barteria@msu.edu, John Carlson, PhD at (517) 432 - 4856 or carlsoj@msu.edu, or Jed Magen, DO at (517) 353 - 4363 or jed.magen@hc.msu.edu . Additionally, you can contact Mr. Barterian by mail at Michigan State University, 620 Farm Lane, 401c Erickson Hall, East Lansing, MI 48823. If you have questions or concerns about your role and rights as a research part icipant, would like to obtain information or offer input, or would like to register a complaint about this study, you Protection Program at 517 - 355 - 2180, Fax 517 - 432 - 4503, or email irb@msu.edu or regular mail at Olds Hall, 408 West Circle Drive #207, MSU, East Lansing, MI, 48824. Documentation of Informed Consent Your signature below means that you voluntarily agree to participate in th is research study. _________________________________ ________________________________ Signature Date You will be given a copy of this form to keep. 159 Informed Assent Script for Minors (Ages 5 to 7) We wanted you to come here today because you sometimes have a hard time talking to h might help children talk more. - up from the doctor. He will do things like listen to your heart. Also, your mom/dad and you will be asked about how you have been feeling and what you have been thinking an d doing. To help us know if the medicine we are going to give you works, we will sometimes give any of the stuff that makes the medicine. This helps us know if the real medicine helps kids like you. You will be given real medicine or pretend medicine for 15 weeks , and you will be given the real or pretend medicine mom and dad, and teacher will be asked how you are feeling. Sometimes, but not all of the time, the medic ine might make you feel funny and it may make your stomach or head hurt, or might make you feel weird in other ways. If this happens, you can tell your mom and dad who can tell us. We will fix it right away so you do not have to feel bad for long. If the m medicine. Do you have any questions? Do you want to help us see if this medicine will help you and other kids talk more by taking the 160 Informed Consent for Minors (Ages 8 +) We asked you to be in a research study because you have a hard time talking sometimes. A office, we would tell you more about these if you want him/her to. Before we start, you will get a check - up from our doctor. You and your mom/dad will be asked questions about how you have been feeling and what you have been thinking and doing. If you want to take the medicine, you will take the medicine for about 15 weeks. You will be asked to take medicine every day and to come here for visits sometimes. We might give you real medicine or a pretend medicine that tastes and looks like the real medicine, but does not ha ve any of the real medicine that may help you talk. This is to help us know if the real medicine is helping you. with the doctor to find out how you have been doi ng. Your mother/father and teacher will also fill out forms to help us know how you are doing. At the end of the study, you will get another check - up from the doctor. Sometimes, but not always, this medicine may make kids feel a little uncomfortable. You might get an upset stomach or get diarrhea; your mouth might feel dry; you might have problems change it, you do not have to keep taking the medicine. If the doctor thinks the medicine is not helping you then the doctor may stop giving it to you. If you want to be in the study and try the medicine, write your name on the line below. Writing your name on the line means that I told you about the study and the medicine. It also means what I told you makes sense and that I told you that you could ask questions. Writing your name on want to. We now want you to write your name to let us know that what we said about the study makes sense. When I write my name it means that I agree to do the study. My parents will also have to agree to have me in the study. Signature of the patient: _____________________________________ Date: ____________ Signature of Investigator: ______________________________________ Date: ____________ Investigator: Justin Barterian, MA 161 A p pendix R Medical, Developmental, Psychosocial, Family, Substance Abus e, and Legal History Form MSU Psychiatry Clinic 162 163 164 165 166 167 168 169 A p pendix S Consent for R elease of C onfidential M edical, T reatment, and S c hool R ecords 170 A p pendix T History of Treatment F orm for Previous Treating Professional Dear Professional, In the recent past, you treated ____________________ for symptoms associated with a diagnosis of selective mutism. Currently, this child is enro lled in a selective mutism treatment study at Michigan State University to examine the effectiveness of fluoxetine in the treatment of selective mutism. In order to determine if ___________________ is eligible for participation in the selective mutism stud y at Michigan State University, it is essential to obtain information regarding the previous types of treatment that child has received under your care. Please check the following boxes that apply to the types of treatments that were provided to the child: A. Behavioral Treatments ______ Contingency Management ______ Shaping ______ Positive reinforcement program ______ Stimulus fading ______ Response initiating ______ Response Cost ______ Systematic desensitization ______ Behavioral Play therapy B. Psychodynamic Treatments ______ Psychoanalysis ______ Affect Sharing ______ Play therapy ______ Art therapy C. Cognitive - Behavioral ______ Cognitive restructuring ______ Anxiety Management ______ Social sk ills training ______ Self - Modeling ______ Relaxation training D. Family Therapy ______ General Family Therapy ______ Structural family therapy E. Psychoeducation ______ Parent Training ______ Teacher Training 171 Please provide the number of sessio ns and the length of each session in which you worked with the participant: ______________________________________________________________________________ ____________________________________________________________________________ _______________________ ______________________________________________________ ______________________________________________________________________________ ____________________________________________________________________________ Please provide a brief paragraph regarding t he scope and goals of treatment: _____________________________________________________________________________ _____________________________________________________________________________ _________________________________________________________________ ____________ _____________________________________________________________________________ _____________________________________________________________________________ _____________________________________________________________________________ _________ ____________________________________________________________________ _____________________________________________________________________________ _____________________________________________________________________________ Please list any treatment manu als that were used in whole or in part during treatment: _____________________________________________________________________________ _____________________________________________________________________________ _________________________________________ ____________________________________ Please provide any additional information about the treatment that was not accounted for in the previous questions: _____________________________________________________________________________ _______________________ ______________________________________________________ _____________________________________________________________________________ _____________________________________________________________________________ _____________________________________________ ________________________________ _____________________________________________________________________________ 172 A p pendix U Physical Exam To be completed by the study psychiatrist on the participants first/last clinic visit DATE: _________________ C OMPLETED BY: ________________ PROCEDURES TO BE COMPLETED HEIGHT: __________________ WEIGHT: _________________________ BLOOD PRESSURE: ____________________ PULSE: __________________ OTHER: Taken from Carlson (1997) 173 A p pendix V Visual Analysis Decision Tree Complete the following decision tree for each graph for each participant: 1) Is there evidence of the problem behavior (e.g., lack of speech)? Is there a stable baseline (i.e. no medication and placebo) pat tern of data? : ________ Yes to all (Go to question 2) ________ No to any (Discontinue Participant) 2) Compare the level (mean) of the data form the baseline phase to the level (mean) of the treatment phase. Is the level of the treatment phase visually hi gher than the level for the baseline phase? ________ Yes (Go to question 3) 3) Compare the baseline phase trend (slope) to the treatment phase trend (slope). Is the treatment phas e slope significantly more pronounced in the hypothesized direction than the baseline slope? ________ Yes (Go to question 4) 4) Does the change in level and trend occur within the e xpected time frame of treatment effects (i.e. during the acquisition period or during the first three time points after the acquisition period). ________ Yes (there is an observable treatment effect) __________ Observable Treatment Effect __________ No Observable Treatment Effect 174 Overall Change When examining all of the baseline and treatme nt pha se replications together, does the data consistently indicate that there is a significant change in the treatment phase? ________ Replicated Effect ________No Replicated Effect ______ Mixed Data* *If mixed data, please explain how many replications were observed and how many participants did not show a predicted treatment effect: 175 REFERENCES 176 REFERENCES Akimova, E., Lanzenberger, L., & Kasper, S. (2009). The serot onin - 1a receptor in anxiety disorders. Biological Psychiatry, 66, 627 - 635. doi: 10.1016/j.biopsych.2009.03.012 American Academy of Child and Adolescent Psychiatrists (2009). Practice parameter on the use of psychotropic medication in children and adolesce nts. Journal of the American Academy of Child and Adolescent Psychiatry, 48, 961 - 973. doi: 10.1097/CHI.0b013e3181ae0a08 American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (4 th ed., text revision). Washington, DC : Author. ANI Pharmaceuticals (2011). Fluvoxamine medication guide. Retrieved from http://www.accessdata.fda.gov/drugsatfda_docs/label/2011/021519s002lbl.pdf Astendig, K. (1999). Is selective mutism an anxiety disorder? Rethinking its DSM - IV classificat ion. Journal of Anxiety Disorders, 13, 417 - 434. doi: 10.1016/S0887 - 6185(99)00012 - 2 Baer, D., Wolf, M., & Risley, T. (1968). Some current dimensions of applied behavior analysis. Journal of Applied Behavior Analysis, 1, 91 - 97. doi : 10.1901/jaba.1968.1 - 91 Baumgartner, J., Emslie, G., Crimson, M. (2002). Citalopram in children and adolescents with depression or anxiety. Annals of Pharmacotherapy, 36, 1692 - 1697. doi: 10.1345/aph.1C078 Beiderman, J., Hirshfeld - Becker, D., Rosenbaum, Faraone, S. (2001). Further evidence of association between behavioral inhibition and social anxiety in children. American Journal of Psychiatry, 158, 1673 - 1679. doi: 10.1176/appi.ajp.158.10.1673 Bergman, R., Keller , M., Piacentini, J., & Bergman, A. (2008). The development and psychometric properties of the selective mutism questionnaire. Journal of Clinical Child & Adolescent Psychology, 37, 456 - 464. doi: 10.1080/15374410801955805 Bergman, R., Piacentini, J., & Mc Cracken, J. (2002). Prevalence and description of selective mutism in a school - based sample. Journal of the American Academy of Child and Adolescent Psychiatry, 41, 938 - 946. doi: 10.1097/00004583 - 200208000 - 00012. Birmaher, B., Axelson, D., Monk, K., Kala Fluoxetine for the treatment of childhood anxiety disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 415 - 423. doi: 10.1097/01.CHI.0000037049.04952.9F 177 Black, B., & Uhde, T. (1 992). Case study: Elective mutism as a variant of social phobia. Journal of the American Academy of Child & Adolescent Psychiatry, 31, 1090 - 1094. doi: 10.1097/00004583 - 199211000 - 00015 Black, B., & Uhde, T. (1994). Treatment of elective mutism with fluoxet ine: A double - blind, placebo - controlled study. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 1000 - 1006. doi: 10.1097/00004583 - 199409000 - 00010 Black, B., & Uhde, T. (1995). Psychiatric characteristics of children with selective mu tism: A pilot study. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 847 - 856 . doi: 10.1097/00004583 - 199507000 - 00007 Boon, F. (1994). The selective mutism controversy (continued). Journal of the American Academy of Child & Adolescen t Psychiatry, 33, 282. doi: 10.1097/00004583 - 199402000 - 00023 Bork, P. & Snyder, D. (2013). When behavioral intervention failed: Report of a successful treatment of selective mutism using fluoxetine. Clinical Medicine and Diagnost ics, 3, 6 - 10. doi: 10.5923/j.cmd.20120301.02 Boyer, E., & Shannon, M. (2005). The serotonin syndrome. The New England Journal of Medicine, 352, 1112 - 1120. doi: 10.1056/NEJMra041867 Bradley, C. (1937). The behavior of children receiving Benzedrine. Americ an Journal of Psychiatry, 94, 577 - 585. Retrieved from http://psycnet.apa.org/psycinfo/1938 - 03205 - 001 Bradley, S., & Sloman, L. (1975). Elective mutism in immigrant families. Journal of the American Academy of Child and Adolescent Psychiatry, 14, 510 - 514. doi: 10.1016/S0002 - 7138(09)61450 - 3 Briesch, A., Chafouleas, S., & Riley - Tillman, T. (2010). Generalizability and dependability of behavior assessment methods to estimate academic engagement: A comparison of systematic direct observation and direct behavio r rating. School Psychology Review, 39, 408 - 421. Retrieved from http://go.galegroup.com.proxy2.cl.msu.edu/ps/i.do?id=GALE%7CA239530646&v=2.1& u=msu_main&it=r&p=AONE&sw=w Carbone, D., Schmidt, L., Cunningham, C., McHolm, A., Edison, S., St. Pierre, J., & Bo yle, M. (2010). Behavioral and socio - emotional functioning in children with selective mutism: A comparison with anxious and typically developing children across multiple informants. Journal of Abnormal Child Psychology, 38, 1057 - 1067. doi: 10.1007/s10802 - 0 10 - 9425 - y Carlson, J. (1997). Sertraline treatment of selective mutism: A single case research trial . (Doctoral dissertation). Retrieved from ProQuest Dissertations & Theses. (9724309) 178 Carlson, J., Kratochwill, T., & Johnston, H. (1994). Prevalence and tr eatment of selective mutism in clinical practice: A survey of child and adolescent psychiatrists. Journal of Child and Adolescent Psychopharmacology, 281 - 291. doi: 10.1089/cap.1994.4.281 Carlson, J., Kratochwill, T., & Johnston, H. (1999). Sertraline trea tment of 5 children diagnosed with selective mutism: A single - case research trial. Journal of Child and Adolescent Psychopharmacology, 9, 293 - 306. doi: 10.1089/cap.1999.9.293 Carlson, J., Mitchell, A., & Segool, N. (2008). The current state of empirical s upport for the pharmacological treatment of selective mutism. School Psychology Quarterly, 23, 354 - 372. doi: 10.1037/1045 - 3830.23.3.354 Carver, R. (1993). The case against statistical significance testing, revisited. The Journal of Experimental Educationa l, 61, 287 - 292. doi: 10.2307/20152382 Chafouleas, S., Kilgus, S., & Hernandez, P. (2009). Using direct behavior rating (DBR) to screen for school social risk: A preliminary comparison of methods in a kindergarten sample. Assessment for Effective Intervent ion, 34, 214 - 223. doi: 10.1177/1534508409333547 Chafouleas, S., Kilgus, S., Riley - Tillman, T., Jaffery, R., & Harrison, S. (2012). Preliminary evaluation of various training components on accuracy of direct behavior ratings. Journal of School Psychology, 50, 317 - 334. doi: 10.1016/j.jsp.2011.11.007 Chafouleas, S., Riley - Tillman, T., & Christ, T. (2009). Direct behavior rating (DBR): An emerging method for assessing social behavior within a tiered intervention system. Assessment for Effective Instruction, 3 4, 195 - 200. doi: 10.1177/1535508409340391 Chafouleas, S., Riley - Tillman, T., & Christ, T. (2010). DBR Standard Form. Retrieved from: http://www.directbehaviorratings.org/cms/files//pdf/V%201.3%20DBR%20Standard%20 Form%20 - %20Fill - in%20Behaviors.pdf Charney , D., Krystal, J., Delgado, P., & Heninger, G. (1990). Serotonin - specific drugs for anxiety and depressive disorders. Annual Review of Medicine, 41, 437 - 446. doi: 10.1146/annurev.med.41.1.437 Chavira, D., Shipon - Blum, E., Hitchcock, C., Cohan, S., & Stein , M. (2007). Selective mutism and social anxiety disorder: All in the family? Journal of the American Academy of Child and Adolescent Psychiatry, 46, 1464 - 1472. doi: 10.1097/chi.0b013e318149366a Christ, T. (2007). Experimental control and threats to inter nal validity of concurrent and nonconcurrent multiple baseline designs. Psychology in the Schools, 44, 451 - 459. doi: 10.1002/pits20237 Christ, T., Riley - Tillman, C., & Chafouleas, S. (2009). Foundation for the development and use of direct behavior rating (DBR) to assess and evaluate student behavior. Assessment for Effective Intervention, 34, 201 - 213. doi: 10.1177/1534508409340390 179 Cohan, S., Chavira, D., & Stein, M. (2006). Practitioner review: Psychosocial interventions for children with selective mutis m: A critical evaluation of the literature from 1990 - 2005. Journal of Child Psychology and Psychiatry, 47, 1085 - 1097 . doi: 10.1111/j.1469 - 7610.2006.01662.x Compton, S., Grant, P., Chrisman, A., Gammon, P., Brown, V., & March, J. (2001). Sertraline in chil dren and adolescents with social anxiety disorder: An open trial. Journal of the american academy of child and adolescent psychiatry, 40, 564 - 571. doi: 10.1097/00004583 - 200105000 - 00016 Compton, S., March, J., Brent, D., Albano, A., Weersing, R., & Curry, J. (2004). Cognitive - behavioral psychotherapy for anxiety and depressive disorders in children and adolescents: An evidence - based medicine review. Journal of the American Academy of Child & Adolescent Psychiatry, 43, 930 - 959. doi: 10.1097/01.chi.0000127589 .57468.bf DeBar, L., Lynch, G., Powell, J., & Gale, J. (2003). Use of psychotropic agents in preschool children. Archives of Pediatric and Adolescent Medicine, 157, 150 - 157. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/12580684 Dummit, E., Klein, R ., Tancer, N., Asche, B., & Martin, J. (1996). Fluoxetine treatment of children with selective mutism: An open trial. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 615 - 621. doi: 10.1097/00004583 - 199605000 - 00016 Dummit, E., Klein, R., Tancer, N., Asche, B., Martin, J., & Fairbanks, J. (1997). Systematic assessment of 50 children with selective mutism. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 653 - 660. doi: 10.1097/00004583 - 199705000 - 00016 Ebert, D., A lbert, R., May, A., Merz, A., Murata, H., Stosiek, I., & Zahner, B. (1997). The serotonin syndrome and psychosis - like side - effects of fluvoxamine in clinical use an estimation of incidence. European Neuropsychopharmacology, 7, 71 - 74. doi: 10.1016/S0924 - 9 77X(96)00043 - 0 Edgington, E. (1980). Validity of randomization tests for one - subject experiments. Journal of educational and behavioral statistics, 5, 235 - 251. doi: 10.3102/10769986005003235 Eke, A. (2001). Pharmacopsychosocial treatment of selective mut ism: A research investigation . (Doctoral dissertation). Retrieved from ProQuest Dissertations & Theses. (3020624) Eli Lilly and Company (2011). Prozac: Highlights of prescribing information. Retrieved from http://pi.lilly.com/us/prozac.pdf Elizur, Y., & Perednik, R. (2003). Prevalence and description of selective mutism in immigrant and native families: A controlled study. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 1451 - 1459. doi: 10.1097/01.chi.0000091944.28938.c6 180 Elson, A., Pearson, C., Jones, D., & Schumacher, E. (1965). Follow - up study of childhood elective mutism. Archives of General Psychiatry, 13, 182 - 187. doi: 10.1001/archpsyc.1965.01730020084010 Fanton, J., & Gleason, M. (2009). Psychopharmacology and preschoolers: A critical review of current conditions. Child and Adolescent Psychiatric Clinics of North America, 18, 753 - 771. doi: 10.1016/j.chc.2009.02.005 Feighner, J. (1999). Mechanism of action of antidepressant medications. Journal of Clinical Psychiatry, 60(supp4 ), 4 - 11. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10086478 Ferron, J. & Levin, J. (2014). Single - case permutation and randomization statistical tests: Present status, promising new developments. In T. R. Kratochwill, & J. R. Levin (Eds.). Single - Case intervention research: Methodological and statistical advances. Washington, DC: American Psychological Association. Food and Drug Administration (2007). FDA proposes new warnings about suicidal thinking, behavior in young adults who take antidepress ant medications. Retrieved from http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/2007/ucm108905.htm Ford, M., Sladeczek, I., Carlson, J., & Kratochwill, T. (1998). Selective mutism: Phenomenological characteristics. School Psychology Quarterly, 1 3, 192 - 227. doi: 10.1037/h0088982 Garfield, S., Francis, S., & Smith, F. (2004). Building concordant relationships with patients starting antidepressant medication. Patient Education and Counseling, 55, 241 - 246. doi: 10.10116/j.pec.2003.09.011 Gelenberg , A., & Chesen, C. (2000). How fast are antidepressants? Journal of Clinical Psychiatry, 61, 712 - 721. doi: 10.4088/JCP.v61n1002 Giddan, J., Ross, G., Sechler, L., & Becker, B. (1997). Selective mutism in elementary school: Multidisciplinary interventions. Language, Speech, and Hearing Services in Schools, 28, 127 - 133. Retrieved from http://www.selectivemutism.org/resources/library/School Issues/SM in Elementary School.pdf C. (2007). Psychopharmacological treatment for very young children: Contexts and guidelines. Journal of the American Academy of Child and Adolescent Psychiatry, 46, 1532 - 1572. doi: 10.1097/chi.0b013e3181570d9e Gracious, B., Youngstrom, E., Findling, R., & Calabrese, J. (2002). Discriminative validity of a parent version of the young mania rating scale. Journal of the American Academy of Child and Adolescent Psychiatry, 41, 1350 - 1359. doi: 10.1097/01.CHI.0000024841.60748.9A 181 Guna - Dumitrescu, L., & Pelletier , G. (1996). Successful multimodal treatment of a child with selective mutism: A case report. The Canadian Journal of Psychiatry, 41, 417. doi: 10.1002/clc.21952 Hack, S., & Chow, B. (2001). Pediatric psychotropic medication compliance: A literature revie w and research - based suggestions for improving treatment compliance. Journal of Child and Adolescent Psychopharmacology, 11, 59 - 67. doi: 10.1089/104454601750143465 Haddad, P. (1998). The SSRI discontinuation syndrome. Journal of Psychopharmacology, 12, 30 5 - 313. doi: 10.1177/026988119801200311 Hagermoser Sanetti, L., & Kratochwill, T. (2008). Treatment integrity in behavioral consultation: Measurement, promotion, and outcomes. International Journal of Behavioral Consultation and Therapy, 4, 95 - 104. Retriev ed from: http://files.eric.ed.gov/fulltext/EJ861339.pdf Harvey, B., & Milne, M. (1998). Pharmacotherapy of selective mutism: Two case studies of severe entrenched mutism responsive to adjunctive treatment with fluoxetine. Southern African Journal of Child and Adolescent Mental health, 10, 59 - 66. doi: 10.1080/16826108.1998.9632346 Hayes, S. (1981). Single - case experimental design and empirical clinical practice. Journal of Consulting and Clinical Psychology, 49, 193 - 211. doi: 10.1037//0022 - 006X.49.2.193 H ealey, (2002). The creation of psychopharmacology. Cambridge, MA: Harvard University Press. Hensler, J. (2010). Serotonin in mood and emotion . In C. Muller and B. Jacobs. Handbook of the Behavioral Neurobiology of Serotonin (pp. 367 - 378). San Diego, CA: Elsevier. Horner, R., Carr, E., Halle, J., McGee, G., Odom, S., & Wolery, M. (2005). The use of single - subject research to identify evidence - based practice in special education. Exceptional Children, 71, 165 - 179. Retrieved from http://cec.metapress.com/i ndex/G7H873H63R42367J.pdf Isolan, L., Pheula, G., Salum, G., Oswald, S., Rohde, L., & Manfro, G. (2007). An open - label trial of escitalopram in children and adolescents with social anxiety disorder. Journal of Child and Adolescent Psychopharmacology, 17, 751 - 759. doi: 10.1089/cap.2007.0007 Judd, L. (1998). A decade of antidepressant development: The SSRIs and beyond. Journal of Affective Disorders, 51, 211 - 213. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/10333978 Kazdin, A. (1977). Assessing the c linical or applied importance of behavior change through social validation. Behavior Modification, 1, 427 - 452. doi: 10.1177/014544557714001 182 Kazdin, A. (1980). Acceptability of alternative treatments for deviant child behavior. Journal of Applied Behavior A nalysis, 2, 259 - 273. doi: 10.1901/jaba.1980.13 - 259 Kazdin, A., & Kendall, P. (1998). Current progress and future plans for developing effective treatments: Comments and perspectives. Journal of Clinical Child Psychology, 27, 217 - 226. doi: 10.1207/s1537442 4jccp2702_8 Kazdin, A. , & Ko e pel, S. (1975). On resolving ambiguities of the multiple baseline design: Problems and recommendations. Behavior Therapy, 6, 601 - 608. doi: 10.1016/S0005 - 7894(75)80181 - X Keeton, C., & Budinger, M. (2012). Social phobia and se lective mutism. Child and Adolescent Psychiatric Clinics of North America, 21, 621 - 641. doi: 10.1016/j.chc.2012.05.009 Klein, R., Abikoff, H., & Barkley, L. (1994). Clinical trials in children and adolescents. In R. Prien and D. Robinson (Eds.). Clinical evaluation of psychotropic drugs: Principles and guidelines. New York: Raven Press. Koehler, M., & Levin, J. (1998). Regulated randomization: A potentially sharper analytical tool for the multiple - baseline design. Psychological Methods, 3, 206 - 217. doi : 10.1037//1082 - 989X.3.2.206 Kolvin, I., & Fundudis, T. (1981). Elective mute children: Psychological development and background factors. Journal of Child Psychology and Psychiatry, 22, 219 - 232. doi: 10.1111/j.1469 - 7610.1981.tb00548.x Kopp, S., & Gillber g, C. (1997). Selective mutism: A population - based study: A research note. Journal of Child Psychology and Psychiatry, 38, 257 - 262. doi: 10.1111/j.1469 - 7610.1997.tb01859.x Kratochwill, T., Elliott, S., Loitz, P., Sladeczek, I., & Carlson, J. (2003). Conjo int consultation using self - administered manual and videotape parent - teacher training: Effects on School Psychology Quarterly, 18, 269 - 302. doi: 10.1521/scpq.18.3.269.22574 Kratochwill, T., Hitchcock, J., Horner, R., Le vin, J., Odom, S., Rindskopf, D. & Shadish, W. (2010). Single - case designs technical documentation. Retrieved from http://ies.ed.gov/ncee/wwc/pdf/wwc_scd.pdf . Kratochwill, T., Hitchcock, J., Horner, R., Levin, J., Odom, S., Rindsckopf, D., & Shadish, W. ( 2013). Single - case intervention research design standards. Remedial and Special Education, 34, 26 - 38. doi: 10.1177/0741932512452794 Kratochwill, T., & Levin, J. (2010). Enhancing the scientific credibility of single - case intervention research: Randomizati on to the rescue. Psychological Methods, 15, 124 - 144. doi: 10.1037/a0017736 183 Kratochwill, T., & Stoiber, K. (2002). Evidence - based interventions in school psychology: Conceptual foundations of the procedural and coding manual of division 16 and society for the study of school psychology task force. School Psychology Quarterly, 17, 341 - 389. doi: 10.1521/scpq.17.4.341.20872 Krysanski, V. (2003). A brief review of selective mutism literature. The Journal of Psychology, 137, 29 - 40. doi: 10.1080/00223980309600597 Kumpulainen, K., Rasanen, E., Rasska, H., & Somppi, V. (1998). Selective mutism among second - graders in elementary school. European Child & Adolescent Psychiatry, 7, 24 - 29. doi: 10.1007/s007870050041 Lafferty, J., & Constantino, J. (1998). Fl uvoxamine in selective mutism. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 12 - 13. doi: 10.1097/00004583 - 199801000 - 00008 La Greca, A., & Lopez, N. (1998). Social anxiety among adolescents: Linkages with peer relations and friend ships. Journal of Abnormal Child Psychology, 26, 83 - 94. doi: 10.1023/A:1022684520514 Lehman, R. (2002). Rapid resolution of social anxiety disorder, selective mutism, and separation anxiety with paroxetine in an 8 - year - old girl. Journal of Psychiatry & Ne uroscience, 27, 124 - 125. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/pmc161642/ Letamendi, A., Chavira, D., Hitchcock, C., Roesch, S., Shipon - Blum, E., & Stein, M. (2008). Selective mutism questionnaire: measurement structure and validity. Jou rnal of the American Academy of Child and Adolescent Psychiatry, 47, 1197 - 1204. doi: 10.1097/CHI.0b013e3181825a7b Lieberman, J., Golden, R., Stroup, S., & McEvoy, J. (2000). Drugs of the psychopharmacological revolution in clinical psychiatry. Psychiatric Services, 51, 1254 - 1258. doi: 10.1176/appi.ps.51.10.1254 Neural correlates of speech anticipatory anxiety in generalized social phobia. NeuroReport, 15, 2701 - 270 5 . Retrieved from http://journals.lww.com/neuroreport/Abstract/2004/12220/Neural_correlates_of_speech_ anticipatory_anxiety.3.aspx Lowry, C., & Hale, M. (2010). Serotonin and the neurobiology of anxious states. In C. Muller and B. Jacobs. Handbook of the B ehavioral Neurobiology of Serotonin (pp. 379 - 397). San Diego, CA: Elsevier. Mannasis, K., Fung, D., Tannock, R., Sloman, L., Fiksenbaum, L., & McInnes, A. (2003). Characterizing selective mutism: Is it more than social anxiety disorder ? Depression and A nxiety, 18, 153 - 161. doi: 10.1002/da.10125 184 Marascuilo, L., & Busk, P. (1988). Combining statistics for multiple - baseline AB and replicated ABAB designs across subjects. Behavioral Assessment, 10, 1 - 28. Retrieved from: http://psycnet.apa.org/psycinfo/1989 - 10682 - 001 Mesa, F., Beidel, D., & Bunnell, B. (2014). An examination of psychopathology and daily impairment in adolescents with social anxiety disorder. PLoS ONE, 9, e93668. doi: 10.1371/journal.pone.0083668 Mitchell, A. (2006). Two - week delay in onset of action of antidepressants: New evidence. The British Journal of Psychiatry, 188, 105 - 106. doi: 10.1192/bjp.bp.105.011692 Muris, P. & Ollendick, T. (2015). Children who are anxious in silence: A review on selective mutism, the new anxiety disorder in DS M - 5. Clinical Child and Family Psychology Review, 1096, 1 - 19. doi: 10.1007/s10567 - 015 - 0181 - y does of citalopram on amygdala response to emotional faces. The Britis h Journal of Psychiatry, 194, 535 - 540. doi: 10.1192/bjp.bp.108.056093 National Institute of Mental Health. (1985). Ratings scales and assessment instruments for use in pediatric psychopharmacology research. Psychopharmacology Bulletin, 21, 839 - 843. Nier enberg, A., Farbaugh, A., Alpert, J., Gordon, J., Worthington, J., Rosenbaum, J, & Fava, M. (2000). Timing of onset of antidepressant response with fluoxetine treatment. The American Journal of Psychiatry, 157, 1423 - 1428. Nutt, D., Bell, C., & Malizia, A . (1998). Brain mechanisms of social anxiety disorder. Journal of Clinical Psychiatry, 59 (S17), 4 - 9. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/9811424 Olfson, M., Marcus, S., Weissman, M., & Jensen, P. (2002). National trends in the use of psych otropic medications by children. Journal of the American Academy of Child and Adolescent Psychiatry, 41, 514 - 521. doi: 10.1097/00004583 - 200205000 - 00008 Olfson, M., Shaffer, D., Marcus, S., & Greenberg, T. (2003). Relationship between antidepressant medica tion treatment and suicide in adolescents. Archives of General Psychiatry, 60, 978 - 982. doi: 10.1001/archpsyc.60.9.978 Omdal, H. & Galloway, D. (2008). Could selective mutism be re - conceptualised as a specific phobia of express speech? An exploratory post - hoc study, 13, 74 - 81. Doi: 10.1111/j.1475 - 3588.2007.00454.x Parker, R., Hagan - Burke, H., & Vannest, K. (2007). Percentage of all non - overlapping data (PAND): An alternative to PND. The Journal of Special Education, 40, 194 - 204. doi: 10.1177/0022466907040 0040101 185 Parker, R., Vannest, K., Davis, J., & Sauber (2011). Combining nonoverlap and trend for single - case research: Tau - U. Behavior Therapy, 42, 284 - 299. doi: 10.1016/j.beth.2010.08.006 Pfizer (2011). Zoloft. Retrieved from http://labeling.pfizer.com/S howLabeling.aspx?id=517 Pine, D. (2002). Treating children and adolescents with selective serotonin reuptake inhibitors: How long is appropriate? Journal of Child and Adolescent Psychopharmacology, 12, 189 - 203. doi: 10.1089/104454602760386888 Pionek Ston e, B., Kratochwill, T., Sladezcek, I., & Serlin, R. (2002). Treatment of selective mutism: A best - evidence synthesis. School Psychology Quarterly, 17, 168 - 190. doi: 10.1521/scpq.17.2.168.20857 Popper, (2002). Child and adolescent psychopharmacology at the turn of the millennium. In S. Kutcher (Ed.), Practical Child and Adolescent Psychopharmacology. (pp. 1 - 38). New York, NY: Cambridge University Press. Riley - Tillman, T., & Burns, M. (2009). Evaluating educational interventions: Single - case design for mea suring response to intervention. New York, NY: The Guilford Press. Riley - Tillman, T., Chafouleas, S., Sassu, K., Chanese, J., & Glazer, A. (2008). Examining the agreement of direct behavior ratings and systematic direct observation data for on - task and d isruptive behavior. Journal of Positive Behavior Interventions, 10, 136 - 143. doi: 10.1177/1098300707312542 Journal of the American Academy of Child and Adolescent Psychiatry, 38, 7. doi: 10.1097/00004583 - 199901000 - 00011 Rynn, M., Sisqueland, L., Rickels, K. (2008). Placebo - controlled trial of sertraline in the treatment of children with generalized anxiety disorder. The American Journal of Psychiatry, 158, 2008 - 2014. doi: 10.1176/ appi.ajp.158.12.2008 Safer, D. (2011). Age - grouped differences in adverse drug events from psychotropic medication. Journal of Child and Adolescent Psychopharmacology, 21, 299 - 309. doi: 10.1089/cap.2010.0152 Safer, D., & Zito, J. (2006). Treatment - emerge nt adverse events from selective serotonin reuptake inhibitors by age group: Children versus adolescents. Journal of Child and Adolescent Psychopharmacology, 16, 159 - 169. doi: 10.1089/cap.2006.16.159 Sharp, W., Sherman, C., & Gross, A. (2007). Selective m utism and anxiety: A review of the current conceptualization of the disorder. Journal of Anxiety Disorders, 21, 568 - 579. doi: 10.1016/j.janxdis.2006.07.002 186 Shear, M., & Beidel, D. (1998). Psychotherapy in the overall management strategy for social anxiety disorder. Journal of Clinical Psychiatry, 59 (S17), 39 - 44. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9811429 Silveira, R., Jainer, A., & Bates, G. (2004). Fluoxetine treatment of selective mutism in pervasive developmental disorder. Internationa l Journal of Psychiatry in Clinical Practice, 8, 179 - 180. doi: 10.1080/13651500410006143 Stahl, S. (1998). Mechanism of action of serotonin selective reuptake inhibitors: Serotonin receptors and pathways mediate therapeutic effects and side effects. Journ al of Affective Disorders, 51, 215 - 235. doi: 10.1016/S0165 - 0327(98)00221 - 3 Stein, M., Liebowitz, M., Lydiard, B., Pitts, C., Bushnell, W., & Gergel, I. (1998). Paroxetine treatment of generalized social phobia (social anxiety disorder). Journal of the Ame rican Medical Association, 280, 708 - 713. doi: 10.1001/jama.280.8.708 Stein, M., Rapin, I., & Yapko, D. (2001). Selective Mutism. Pediatrics, 107, 926 - 929. doi: 10.1016/j.biopsych.2010.11.008 Steinhausen, H., & Adamek, R. (1997). The family history of chi ldren with elective mutism: A research report. European Child & Adolescent Psychiatry, 6, 107 - 111. doi: 10.1007/BF00566673 Steinhausen, H., Wachter, M., Laimbock, K., & Metzke, W. (2006). A long - term outcome study of selective mutism in childhood. Journal of Child Psychology and Psychiatry, 47, 751 - 756. doi: 10.1111/j.1469 - 7610.2005.01560.x Tancer, M. (1993). Neurobiology of social phobia. Journal of Clinical Psychiatry, 54 (S12), 26 - 30. Retrieved from: http://psycnet.apa.org/psycinfo/1994 - 29974 - 001 Tayl or, M. (2007). Rapid onset of true antidepressant action. Current Psychiatry Reports, 9, 475 - 479. doi: 10.1007/s11920 - 007 - 0064 - 0 Tschanz, J., & Treiber, K. (2011). Serotonin. In Encyclopedia of clinical neuropsychology (pp. 2269 - 2270). New York, NY: Sprin ger. Vaswani, M., Linda, F., & Ramesh, S. (2003). Role of selective serotonin reuptake inhibitors in psychiatric disorders: A comprehensive review. Progress in Neuro - Psychopharmacology & Biological Psychiatry, 27, 85 - 102. doi: 10.1016/S0278 - 5846(02)00338 - X Vitiello, B. (2001). Psychopharmacology for young children: Clinical needs and research opportunities. Pediatrics, 108, 983 - 989. doi: 10.1542/peds.108.4.983 187 Voncken, M., & Bogels, S. (2008). Social performance deficits in social anxiety disorder: Real ity during conversation and biased perception during speech. Journal of Anxiety Disorders, 22, 1348 - 1392. doi: 10.1016/j.janxdis.2008.02.001 (2004). A multicenter, r andomized, double - blind, placebo - controlled trial of paroxetine in children and adolescents with social anxiety disorder. Archives of General Psychiatry, 61, 1153 - 1162. doi: 10.1001/archpsyc.61.11.1153 Walkup, J. (2001). Fluvoxamine for the treatment of a nxiety disorders in children and adolescents. New England Journal of Medicine, 344, 1279 - 1285. doi: 10.1056.NEJM200104263441703 Walkup, J., & Lebellarte, M. (2001). Complications of SSRI treatment. Journal of Child and Adolescent Psychopharmacology, 11, 1 - 4. doi: 10.1089/104454601750143320 Wampold, B. & Furlong, M. (1981). Randomization tests in single - subject designs: Illustrative examples. Journal of Behavioral Assessment, 3, 329 - 341. doi: 10.1007/BF01350836 Wampold, B. & Worsham, N. (1986). Randomizati on tests for multiple - baseline designs. Behavioral Assessment, 8, 135 - 143. Retrieved from: http://psycnet.apa.org/psycinfo/1987 - 09267 - 001 Watson, P., & Workman, E. (1981). The non - concurrent multiple baseline across - individuals design: An extension of the tradi tional multiple baseline design, Journal of Behavior Therapy and Experimental Psychiatry, 12, 257 - 259. doi: 10.1016/0005 - 7916(81)90055 - 0 Westenberg, H. (1998). The nature of social anxiety disorder. Journal of Clinical Psychiatry, 59 (S17), 20 - 24. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/9811426 Wilens, T., Biederman, J., Kwon, A., Chase, R., Greenberg, L., Mick, E., & Spencer, T. (2003). A systematic chart review of the nature of psychiatric adverse events in children and adolescents tr eated with a selective serotonin reuptake inhibitor. Journal of Child and Adolescent Psychopharmacology, 13, 143 - 152. doi: 10.1089/10445603322163862 Wright, H., Cuccaro, M., Leonhardt, T., Kendall, D., & Anderson, J. (1995). Case study: Fluoxetine in the multimodal treatment of a preschool child with selective mutism. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 857 - 862. doi: 10.1097/00004583 - 199507000 - 00008 Yeganeh, R., Beidel, D., & Turner, S. (2006). Selective mutism: More th an social anxiety? Depression and Anxiety, 23, 117 - 123. doi: 10.1002/da.20139 188 Yeganeh, R., Beidel, D., Turner, S., Pina, A., & Silverman, W. (2003). Clinical distinctions between selective mutism and social phobia: An investigation of childhood psychopa thology. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 1069 - 1075. doi: 10.1097/01.CHI.000007026 Zaider, T., Heimberg, R., Fresco, D., Schneier, F., & Liebowitz, M. (2003). Evaluation of the clinical global impression scale among individuals with social anxiety disorder, Psychological Medicine, 33, 611 - 622. doi: 10.1017/S0033291703007414 Zuckerman, M., Vaughan, B., Whitney, J., Dodds, A., Yakhkind, A., MacMillan, C., - Heydrich, G. (2007). Tolerability of selective serotonin reuptake inhibitors in thirty - nine children under age seven: A retrospective chart review. Journal of Child and Adolescent Psychopharmacology, 17, 165 - 174. doi: 10.1089/cap2007.008