1.1.1 “1-3;" 5(1) ". 11." I ' ‘ .' I "git. :- '1. V)- -,','r\ L; '3 a 1&1 _ . 1‘ . \ 3“ - _' ' , . 4". ‘ It”...qu..‘r “J1, ' “.' I , 1‘ _ r’, I a ‘1' . '3 "111 ‘ fl 1 4"” Vh‘fi‘i' mu: ’0. -' ‘ ' : ,11. g R: 5 ~21 . 3» .. .7-33'. .331? \' :1: rt.“ ’1'. n ' 3‘ ' ,. . . 311-1133111111? 511:1 a, '3: 47‘; I. V (J‘€:‘I‘f‘.‘v\‘$s‘y\i""’kl:9; If V13..'_3 ‘1’“ 3*;7321‘é'V-;“‘fi I 1' . “'7‘? “4: 4 I, :19 ’G : J (.1) ‘3” m . r7A.“'J'I“.".4,;‘.. I'FVPIQ'E 1‘” 1 “.74”! $1,,” ‘ ‘v'w 17x; {131212 9.5. 5‘1: ‘6“ 5.14”.» fanny)”- I 1'4 '1 1):)“ - «'1'» ;'I‘I (213:2; , - "21”" 1 1 “1‘13“; 3A,}; . “7‘" . . ,H 7“.” ". 4-. ‘1“ . ‘ V .‘w ‘1’. f1}'(‘:‘ If? ‘70... “5'. , . 11":7: ‘ Mi‘fi“; ' $5.13? . W ""' "11:, y ’ a1. “.1 , 11:” “2‘4' $w2$¥&.x‘ ”fix 3 1 . . » ' ' xiv-r 19%: ('1 5“? (if-'5 1 ' 1,1”! . 13‘V,‘13“'K\:‘u‘ ~31 1 . ,3 .131 . (111111. .1, 4 11 1 M1 1 1111.1: a" ' - -' " ‘ 1' -. 1.? " 1.; ‘ ““,‘(-1".§_.‘ 1,")! - ( ‘gfia . I 1t 3 . ‘ 3‘ I .- l- [I . ‘1‘ 7'“ . f“? 9"“ ‘7“ {'1‘ ”712-. ..f§zy‘éfp:t"“k’ I 2).. first}; ‘11:: 1 44¢“. . . b.4281; «9'3 1A fl:a5'.3,‘: ;§i§{ $1“ 1. 33; 1;... 1 .14. ~ "($131. 11! .14! £2." ' i wfifi 1‘7"”. '- _ , 111* 1473‘ TS? ' .\£"‘ ‘gyif'k'y‘nl ~ ‘1'“- ; A“??? ’wéqu‘ “if, .1 1:- 51,15“? 94‘, Jr“; - _ 5213‘“: 1 ‘ 95. H.- ...‘... 3. ,:.'7. 1. ‘7' ”Z“, ’ ,2‘ ' “1 '1‘f?‘ . "' ““1031 , ‘;' {'-‘};‘z;¢:érk' . .w E‘ “L”... ‘ Vr/cf'y‘fl ' ' 7'!” "v. S 1 r_ " ‘ . I‘ '4' ' " 1‘.‘:;C" J37.“ V". ' 1‘ r4... ,‘ 3, A )‘Inu... .’ 7 my '3.‘ .3 1.“ a .2 J In]? 1'3 " £17.15: .‘H'I‘if‘ 1‘- '1. . . -; ‘1‘: I “’2'; (3 if???" =';’-' ‘ 7' " . ’ .11“ " "“' "1 .1731.» "1C“ 1 1:41. ‘ 211- '.1rr“‘ . 12. 11. .. (.1 - .er 1111111? 1.1.1.11 ." ‘1’ ‘25. x x a“ h 5:32 ~ 3. i KG 154.“ f .‘l‘g .. ;.‘1‘,'-‘¢ ‘. i ‘ C‘LXN'.’ “to"! "1. '59,: : $11” I: Eiéwg I} 11' . x “$3‘2z‘g‘ 2E); 13 ; I. 1 JV- 1‘ 3g s3"§~r 1‘.-‘ A {i123 ‘. .9.x"' y. 4,. :Vfrl o 74‘ 'IW ,1 ”f .5 HA 7: ., .{c 4": ‘ V. . . W. 3" . ups». .- - o I"; z . . . . - 3,3, .‘.\t ' “3'3" x -9... . . 3: —v~.-‘ 13"! 2‘ ll 2 ~'.3.'.‘.'-f.r. v N“). A." 1111'“. 127x 753S(' Elam? Mm “3:33;" IIIIHHIIHHQWI;gluljjlljllll This is to certify that the dissertation entitled PHYSICIAN INTERACTICNS WITH PARENTS AND CHILDREN AS PSYCHOLOGICAL DETERMINANTS OF PARENT SATISFACI‘ICN presented by Anne B. Riley - .. has been accepted towards fulfillment of the requirements for Ph D degree in Ed 25%: n... Rama-J...) Major professor Date November 3, 1989 M5 U is an Affirmative Action/Equal Opportunity Institution 0-12771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or bdore due due. DATE DUE DATE DUE DATE DUE I'm 0 ~ 11 IL F—ll l MSU Ie An Affirmulve ActlorVEquel Opportunity Inemuion omens-o1 PHYSICIAN INTERACTIONS WITH PARENTS AND CHILDREN AS PSYCHOLOGICAL DETERMINANTS OF PARENT SATISFACTION BY Anne B. Riley A DISSERTATION submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education Department of Counseling, Educational Psychology, and Special Education 1989 C04; 57.x. ABSTRACT PHYSICIAN INTERACTIONS WITH PARENT AND CHILD AS PSYCHOLOGICAL DETERMINANTS OF PARENT SATISFACTION By Anne B. Riley This study examined the effects of specific physician behaviors on parent satisfaction in an acute care pediatric clinic. Seven physicians agreed to have five of their pediatric visits videotaped. After each visit, parents were asked to choose and explain their most and least satisfied moments and to complete the Parent Medical Interview Satisfaction Scale (P-MISS). Hypotheses and questions were developed relating parent satisfaction to physicians' verbal and nonverbal behavior, physician-child interaction, and length of visit. Qualitative analysis of the parent interviews found parents more influenced by physician verbal than nonverbal behavior when selecting their most and least satisfied moments. Diagnostic statements brought the most satisfaction. Relatedly, the majority of most satisfying moments occurred during the diagnosis and treatment of the visit; most of the unsatisfying moments occurred during the history-taking. Also, physician-parent interaction affected parents' satisfaction more than physician-child interaction. Videotape data analyzed using Hierarchial Linear Model (HLM) calculations allowed the variation in P-MISS scores to be partitioned into within- and between-doctor differences. Eighty-seven percent of the score variation was attributable to within-doctor differences. To explain within-doctor differences, variables, such as parent/child demographic data, visit length, and physician verbal and nonverbals behavior were used. Only one variable, visit length, correlated significantly with parent satisfaction; however, it explained only a small portion of the total observed variation. Shorter rather than longer physician visits increased parent satisfaction in this particular setting. The one between-doctor variable tested, years of experience, showed no correlation with parent satisfaction. A comparison was done of the verbal and nonverbal behaviors of the doctors rated high and low by parents in satisfaction. For high rated doctors, verbal and nonverbal behaviors were cited equally in parents' explanations of their most satisfying moment. For low rated doctors, verbal behavior was mentioned more frequently. When high and low rated doctors were compared in number of dissatisfied moments, half of the parents rating doctors high experienced no dissatisfaction; all but one of the parents rating doctors low had such a moment. A prototypic model of parent satisfaction was developed and implications were discussed. ACKNOWLEDGEMENTS I would like to express my appreciation to several people who have directly influenced the completion of this dissertation. Don Hamachek, chairperson and advisor, carefully edited the various drafts, providing both guidance and encouragement. The freedom to choose this topic and his belief in my ability to conduct this project were instrumental to its completion. Doug Campbell shared his theoretical knowledge and insights for the analysis of the qualitative data. His availability, constructive comments, and enthusiasm during the various stages of development were invaluable. Steve Raudenbush advised me on the quantitative analysis, providing basic insights in a number of areas. Martha Karson initially guided the project from a psychological perspective and was a consummate role model for interviewing. Richard Solomon gave administrative support that facilitated my site selection and data collection. Bill Rittenburg provided opportunities for my initial exposure to qualitative research and editorial comments. I am grateful to the physicians and nurses at the St. Lawrence Pediatric After-Hours Clinic for their participation in this project. As a result of observing them, I have a greater appreciation of the stress involved in caring for acutely ill children and their families. Finally, to my mother, my son, David, and my friends I owe thanks for their encouragement and interest. I know they share my sense of accomplishment and relief now that this task is done. iv Chapter II. III. TABLE OF CONTENTS THE PROBLEM. Introduction . . Statement of the Problem . . Justification for Medical Satisfaction Studies . Justification for the Proposed Research Purpose. . Theoretical Perspective. Parsons' Theory . . . Communication Theories. REVIEW OF THE LITERATURE . Verbal/Nonverbal Debate General Factors Influencing Adult Patient Satisfaction . . Physician Verbal Behaviors in Adult Patient Visits Physician Nonverbal Behaviors in Adult Patient Visits. . . . . . Parent Satisfaction Research . Physician Verbal Behaviors in Pediatric Visits. Physician Nonverbal Behaviors in Pediatric Visits. . Physician- -child Interaction . Extrinsic Factors . . Critique of the Literature. Summary and Implications for this Project DESIGN OF THE STUDY. Description of the Setting . . Description and Selection of the Samples . Physicians Parents . Children . . Procedures for Collecting Data . Instruments Page H mmO‘U‘PWNH . ll . ll . 14 . 15 . l6 . l8 . l9 . 21 . 22 . 25 . 25 . 26 . 29 . 29 . 29 . 29 . 30 . 3O . 31 . 32 Chapter Page Checklists for Physician Behaviors. . . . . . . . 33 Verbal Checklist . . . . . . . . . . . . . . 33 Nonverbal Checklist. . . . . 34 Parent Medical Interview and Satisfaction Scale . 34 Reliability. . . . . . . . . . . . . . . . . 35 Validity . . . . . . . . . . . . . . . . . . 35 Design . . . . . . . . . . . . . . . . . . . . 36 Research Hypotheses. . . . . . . . . . . . . . . . . . 37 Variables. . . . . . . . . . . . . . . . . . . . 38 Outcome Variables . . . . . . . . . . . . . . . . 38 NP- MISS. . . . . . . . . . . . . . . . . . . 39 Predictor Variables . . . . . . . . . . . . . . . 40 Visit Variables. . . . . . . . . . . . . . . 40 Doctor Variables . . . . . . . . . . . . . . 41 Procedures for Analyzing Data . . . . . . . 41 Data from the Post-visit Parent Interview . . . . 42 Data from Coders' Verbal and Nonverbal Checklists. . . . . . . . . . . . . . . . . . . . 43 IV. RESULTS. . . . . . . . . . . . . . . . . . . . . . . . 48 Overview . . . . . . . . 48 Hypotheses Relating to the Parent Interview . . . . 48 Hypotheses Relating to the Parent Satisfaction Scale . 51 Within- and Between- Doctor Variation . . . . . . 51 Physician Verbal and Nonverbal Behaviors. . . . . 54 Length of Physician Visit . . . . . . . . . . . . 51 Summary of Quantitative Analysis . . . . . . 59 Patterns, Quotes, and Interpretive Commentary from the Parent Interview . . . . 61 Parents' Responses to Physician Verbal Behavior . 61 Diagnosis . . . . . . . . . . . . . . 62 Diagnostic Confirmation . . . . . . . 62 Parents' Responses to Physician Nonverbal Behavior. . . . . 64 Occurrence of Most and Least Satisfied Moments. . 66 Responses Indicating Reluctance to Criticize Physicians. . . . . . . . . . . . . . . 67 High and Low Rated Doctors . . . . . 68 Reasons Cited for Most Satisfied Moments . . 69 Reasons Cited for Least Satisfied Moments . 69 Physician- parent vs. Physician-child Interaction. 71 Physical Exam . . . . . . . . . . . . . . 71 Responses Relating to Time. . . . . . . 72 Responses Considered Difficult to Interpret . . . 73 Summary of Qualitative Results . . . . . . . . . . . . 74 Final Summary of Hypotheses Testing. . . . . . . . . . 75 Vi Chapter Page V. DISCUSSION AND CONCLUSIONS. . . . . . . . . . . . . . . . . 77 Summary of Results . . . . . . . 77 Possible Reasons for Negative Quantitative Findings. . . . . . . . . . . . . . . . . . . . . 79 Setting. . . . . . . . . . . . . . . . . . . 79 Sample Size. . . . . . . . . . . . . . . . . 80 Low Reliability. . . . . . . . . . . . . . . 80 Technology . . . . . . . . . 80 Parent' 3 reluctance to criticize . . . . . . 80 Discussion of the Hypotheses . . . . . . 82 Physician Behavior and Parent Satisfaction. . . . 82 Impact of Verbal Behavior. . . . . . . . . 82 Impact of Receiving the Diagnosis. . . . . . 82 Occurrence of Most and Least Satisfied Moments. . . . . . . . . . . . . . . . . . . 83 Physician- parent vs. Physician—child Interaction and Parent Satisfaction . . . . . . . . 84 Length of Visit and Parent Satisfaction . . . . . 85 Between- and Within- doctor Variation . . . . . . 85 Models . . . . . . . . . . . . . . . . . . 86 Limitations of the Study . . . . . . . . . . . . . . . 88 Implications of the Study. . . . . . . . . . . . . . . 88 Theoretical Implications. . . . . . . . . . . . . 88 Parsons' Theory . . . . . . . 88 Communication Theory--Two Approaches, Structural and Intentional . . . . . . . . . 90 Practical Implications. . . . . . . . . . . . . . 92 Directions for Future Research . . . . . . . . . . . . 95 Conclusions . . . . . . . . . . . . . . . . . . . . . 97 vii Chapter I THE PROBLEM Introduction Parents' dissatisfaction with their children's medical care is a surprisingly common occurrence (Breslau & Mortimer, 1981; Korsch, Gozzi, & Francis, 1968). Although most parents leave the physician's office feeling satisfied, a sizable minority leave feeling dissatisfied, a puzzling phenomenon since medical treatment is far more effective today than even a few years ago (DiMatteo, 1979; Eisenberg, 1977). Because dissatisfaction frequently occurs in this "renaissance era" of medicine, the problem does not appear to be in technology but in the physician's relationship with the patient (Korsch & Aley, 1973; Pendleton & Hasler, 1983; Ware, Davies-Avery, & Stewart, 1978). The importance of physician-patient communication was recognized as far back as the fourth century B.C. when Hippocrates preached to fellow physicians that even a dying "patient may recover his health simply through his contentment with the goodness of the physician" (1923 English Translation). Until the late nineteenth century, this "bedside manner" and placebos were virtually the only "medicines" available to physicians (Houston, 1938). Yet, despite useless and even harmful medications and procedures, physicians were held in high regard (Jones, 1982). Although it cannot be empirically proven, the consensus then and now is that this high regard came from the physician-patient relationship (Shapiro, 1960). 2 Although a physician's ability to communicate with patients has been considered an integral part of the healing process throughout medical history (DiMatteo, 1979; Wexler, 1976), it was not until 20 years ago that the medical profession realized that patient dissatisfaction was so prevalent. A classic article poignantly illustrated the lack of effective communication between physicians and parents by noting that only 24 percent of the parents' main worries were mentioned to the doctor (Korsch et a1., 1968). Since then satisfaction with medical care has been a focus of scientific investigation; however, most of the research has been conducted on adult patients rather than parents. Statement of the Problem A host of problems currently plaguing the medical profession are linked to dissatisfaction with medical care (Friedman & DiMatteo, 1982; Pendleton, 1983). Problems, such as increasing malpractice litigation, high noncompliance rates, doctor-shopping, forgotten medical information and missed appointments are quite common, affecting not only the patients' health and medical costs, but also physicians' frustration. Good physician-patient communication is essential in determining patient satisfaction (Ley, 1979; Ware et a1., 1978), but it has been difficult to identify which communicative behaviors impress patients. Research has not gone beyond identifying a few predictable but vague interpersonal qualities, such as friendliness, empathy, or respect. The above problems affect not only adult patient care, but also pediatric care. The purpose of this study is to examine physician 3 communication in an initial pediatric visit to better understand the influence of physician verbal and nonverbal behavior on parent satisfaction. The impact of physician-child interaction on parent satisfaction will also be studied. Justification for Medical Satisfaction Studies Satisfaction studies predict a patient's future use of medical practitioners and facilities (DiMatteo & Hays, 1980; Roghmann, Hengst, & Zastowny, 1979). Relatedly, doctor-shopping (Cobb, 1954; Kasteler, Kane, Olsen, & Thetford, 1976), the use of nonmedical practitioners (Cobb, 1954), and malpractice suits (Blum, 1957; Vaccarino, 1977) all decrease when patient satisfaction increases. In addition, information learned from satisfaction studies has the potential to increase treatment effectiveness and reduce medical costs. Treatment effectiveness can be increased because of better compliance with treatment regimens. Compliance rates for adult patients are discouragingly low, 39 to 61 percent (Ley, 1982). Pediatric compliance rates are similarly low, about 54 percent (Jones, 1983). A direct link between compliance and patient satisfaction has been established (Francis, Korsch, & Morris, 1969; Hulka, Cassell, Kupper, & Burdette, 1976; Korsch et al., 1968). Compliance has also been linked to reduced medical cost, since noncompliance is estimated to be responsible for 20 percent of all hospital admissions (Ausburn, 1981). Satisfaction studies have also demonstrated that the doctor-patient relationship affects the length of hospital stay and patient recovery (Ley, 1977). 4 Realizing the need for more effective communication, the medical profession has put much effort into "re-humanizing" medical care (Zastowny, Roghmann, & Hengst, 1983). Today a majority of medical schools require coursework pertaining to interpersonal skills training (Kahn, Cohen, & Jason, 1979). Yet, problems in relating to patients still exist, in part, because of the need for better skill specification (Carter, Inui, Kukull, & Haig, 1982; Roter, 1983; Strecher, 1982). Carroll and Monroe (1979), in their review of over 70 interpersonal skills courses for physicians, found little agreement about content and suggested research that examines the effectiveness of specific interpersonal skills. Finally, physicians benefit psychologically when patients leave feeling satisfied. Job satisfaction studies of primary care physicians show that effective patient communication and the resulting relationships are the most satisfying job aspects for these physicians (Ford, Liske, Ort, & Denton, 1967). Justification for the Proposed Research This study is important for three reasons: (1) Dissatisfaction with pediatric care is a frequent occurrence. Although there are no current rates of parent satisfaction available, the classic study done twenty years ago found that approximately 25 percent of parents were dissatisfied with their pediatric visit (Korsch et al., 1968). Adult patient dissatisfaction rates are even higher. Koo's survey (1955) of 1,000 families found that 64 percent were critical of 5 the physician-patient relationship. More recent estimates indicate that the median range of dissatisfied patients is between 35 and 40 percent (Ley, 1982). (2) Only a small percent of the medical satisfaction research deals with parent satisfaction, despite high utilization of medical facilities by children--89% of children under four see a physician at least once a year (National Center for Health Statistics, 1979). Parents are responsible for receiving the physicians' information and making key treatment decisions (Stewart, Pantell, Dias, Well, & Ross, 1981), so it is essential for them to have a comfortable relationship with their child's doctor. Pediatricians spend most of their time (60-80%) conversing with parents and the families of patients (Korsch & Aley, 1973), yet relative little emphasis is given to training medical students in interpersonal skills beyond taking a thorough history. (3) Because research on parents satisfaction is still in the exploratory stage, obtaining the insiders' (parents’) perspective is of value. This study will explore the parents' perceptions of the physicians' behavior and other aspects of a pediatric visit in order to develop a model of parent satisfaction. Purpose The purpose of this research is to examine the physician interaction in an initial pediatric visit to determine its effect on parent satisfaction. A specific focus will be on the impact of (a) the 6 physicians' verbal and nonverbal behavior, and (b) the physician-parent interaction and physician-child interaction. Theoretical Perspective This section discusses Talcott Parsons’ social interaction theory as it relates to a medical encounter and a description of the structural approach to the study of communication. The Parsons' theory provides a rationale for structure within a medical encounter and divides physician behavior into two components, instrumental (technical competence) and expressive (psycho-social). Parsons' theory is a systems (or sybsystem in the case of medical care) level interaction theory; the structural approach is an interaction theory of communication applied to individuals. Parsons' Theory Talcott Parsons saw health care as a subsystem of the larger structure of social action. He viewed the physician-patient interaction as a rigid, predictable exchange, an institutionalized role set revolving around the patient's medical needs (Parsons, 1951). Key to the Parsonian framework is the idea that illness is deviant behavior that disturbs the health functioning of the individual and, ultimately, the social system. Parsons believed in institutionalization of the physician-patient relationship to ensure the efficient functioning of society to which illness poses a major threat (West, 1984). 7 Parsons also believed in asymmetry in the physician-patient relationship, to promote the physician's control over patients in order to ensure the compliance needed for healing (Parsons, 1951). In essence, his concept of the physician-patient relationship "is predicated on institutionalized inequality between those who heal and those who must come to them for treatment" (West, 1984, p. 18). Although this description is not as directly applicable to a pediatric situation, the parent still is in the "sick-role" because the physician has sole possession of the knowledge with which to heal the child. Parsons (1951) divided physician behavior into "instrumental" (that dealing with technical aspects of medical care) and "expressive" (that dealing with social-psychological aspects of the doctor-patient relationship). He recognized the "art of medicine"—- "All good medical practice. . . has been and is to some degree psychotherapy" (1951, p. 478). Unfortunately, his recognition of this more personal aspect of physician-patient interaction is "reinstitutionalized" when he defines psychotherapy as a mechanism of social control which must be regarded as "automatic." Parsons' characterization of the physician-patient relationship as standardized behavioral expectations for both patients and practitioners was a powerful influence when first formulated, but his critics are now numerous. Critics, Bloom (1963) and Friedson (1970), attacked Parsons' neglect of significant others, such as family members, in the healing process. Szasz and Hollender (1956) have also taken issue with Parsons' idea that the patient must play a passive role (West, 1984). They 8 argued that patients both desire and need an active role in their treatment and point to patient involvement in controlling chronic diseases, such as diabetes. W The structural approach to communication can trace its roots to the beginning of this century when the idea of social interaction evolved. Anthropology and linguistics furthered the theory's development, so that structural linguistics (kinesis) came into being. Further refinements included incorporating the ideas of feedback and self-regulation. These concepts led to the realization that "communicating" is not an act in which one chooses to engage (Reusch & Bateson, 1951). Kendon, a leading proponent of this approach, further explains that " . . . so long as one is in the presence of another, all of one's behavior is a source of information for another, all of the time" (1979, p.70). Thus, one is always communicating or conversely; one cannot not communicate. Even when an individual chooses to remain silent, a "message" is being sent. Furthermore, communication flows in two directions because the behavior of each individual feeds back to and influences the other (Kendon, 1979). These concepts evolved into a methodology called ’context analysis' (McQuown, 1971), which states that behavior is described by the contexts in which it occurs (Kendon, 1982) and is both continuous and reciprocal (Kendon, 1979). Although context analysis has many adherents, critics contend that it is fraught with problems of knowing what rules to apply 9 in identifying structural units and hierarchical organization (Scherer & Ekman, 1982). Another approach to communication is based on the premise that communication is distinguished by intentionality (Ekman & Friesen, 1981). In other words, behavior is considered 'communicative’ only if the person providing it intends to convey some message, regardless of whether anyone else receives the message. Ekman and Freisen's idea of communicative behavior is more restrictive than Reusch and Bateson's viewpoint that 'all behavior is communication.’ This approach is not without its critics either. Kendon argues convincingly that "The question of intentionality is irrelevant because . . . to witness a behavioral event is to receive information. . ." (1981, p. 9). Furthermore, the intention of the speaker is often indeterminable. The research proposed here examines the relevance of Parsons' instrumental/expressive dichotomy when applied to an acute care pediatric setting. Relating to the structural approach, this project will examine what is communicated by the physician to the parent, regardless of the physician's intention. Stated another way, what "messages" did the parent perceive from the interaction with the physician. The parent (or the physician), according to this approach, could not have been a participant in the interaction without forming some impression. The impression of interest here is parent satisfaction and the parents' explanation of their most and least satisfied moments will be examined. There are four remaining chapters in this dissertation. Chapter Two contains the review of literature on both adult patient and parent 10 satisfaction. Chapter Three focuses on the qualitative and quantitative methodology. Chapter Four contains the results of the analyses and Chapter Five provides a discussion of the results with implications. Chapter II LITERATURE REVIEW This chapter includes 1) a brief historical perspective of the verbal-nonverbal debate, 2) a review of research on physicians' verbal and nonverbal behavior related to both adult patient and parent satisfaction, 3) a review of research on physicians’ interactions with children as it relates to parent satisfaction, and 4) a summary with implications for this study. Verbal/Nonverbal Debate Communication research has been a topic of interest only in the past 40 years with early emphasis on verbal behavior as it was "readily perceivable, codeable, and interpretable..." (Harrigan & Rosentha1,l986, p. 41). Then, in the 503 and 605 theoretical approaches to the study of communication promoted the concept that all behavior (spoken and unspoken) is potentially meaningful and could be considered communication (Reusch & Bateson, 1951). Nonverbal behaviors then became the focus of much scientific attention. Earliest interest in nonverbal language is traceable to Darwin's 1872 text, The Expressiong 9f the Emotione in Mag and Animals. Thirty years later Freud added credibility to the idea of a visual language by remarking about a patient, "if his lips are silent, he chatters with his fingertips; betrayal oozes out of him at every pore" (1905, p. 94). Interest in nonverbal behaviors waned during the next few decades, in 11 12 part, because they are difficult to define and quantify. However, during the past fifteen years nonverbal language has regained prominence and for some investigators surpassed the verbal communication in importance. The importance of nonverbal communication relates to the point Freud was illustrating, that people may "leak" their true feelings through cues other than verbal (Ekman and Friesen, 1969). Acting on this premise, research on deception proposed the concept of a leakage hierarchy, which states that all channels, verbal and nonverbal, can be placed on a continuum of controllability. Nonverbal behaviors are usually less controllable and thus are considered 'truer' reflections of feelings (Ekman & Friesen, 1969; Zuckerman, DePaulo, & Rosenthal, 1986). More recently, the primacy of nonverbal behavior has been questioned. [Primacy is defined as the "overwhelming importance" or "primary basis for communication" (Krauss, Apple, Morency, Wenzel, & Winton, 1981, p. 312)]. Even nonverbal advocates, Archer and Akert (1977) admit that nonverbal primacy "rests upon a highly specific experimental base, and the question (of primacy) has never been addressed more generally, using natural sequences of behavior" (p. 444), something this research attempts to do. In line with questioning the primacy of nonverbal behavior is the movement away from making specific estimates of its influence as Mehrabian's (1972) often quoted studied does. Research comparing verbal and nonverbal behaviors within the same study supported nonverbal primacy by a slight margin. However, as noted 13 by Archer and Akert (1977), the designs of many of these studies had contrived features, for example, the use of posed channel contradictions (positive content intentionally said in a negative voice), an uncommon occurrence outside a laboratory setting add other problems inherent in nonverbal research are the simultaneous nature of these behaviors and defining them operationally. To illustrate, a forward lean may vary from a 10 to 40 degree angle and it is the individual researcher who arbitrarily decides when such a behavior occurs (Harrigan & Rosenthal, 1986). When examining the research done in natural settings with spontaneous dialogue (most of which pertained to psychotherapy), the predominance of nonverbal behaviors over verbal content did not hold (Fretz, Corn, Tuemmler, & Bellet, 1979; Seay & Altekruse, 1979). Other support for the importance of verbal behavior comes from a meta-analysis of studies in the communication of deception (Zuckerman, DePaulo, & Rosenthal, 1986). This study of studies proved to be an exception to the rule of nonverbal primacy when results are obtained in a laboratory ,setting. In a comparison of three single channels--face, body and speech--speech was the most accurate, followed by the body, and the face. The authors found that, at least in the communication of deception, verbal cues (words, written or spoken) were more important that the nonverbal channels. Zuckerman and his associates conclude that the assumption that the nonverbal channels are more important that verbal cues in the communication of deception is not true. Krauss and colleagues (1981) reached the same conclusion as Zuckerman with regard to affective communication. 14 To summarize the current state of affairs, the dominance of nonverbal over verbal behaviors is being reconsidered, in large part, because of the artificiality of the research. Evidence is accumulating to suggest that in research with less contrived interactions, verbal behaviors may have a greater impact than nonverbal behaviors. The remainder of this review of literature examines the physicians' communication with adult patients and with parents. General Factors Influencing Adult Patient Satisfaction Factors identified in the adult patient satisfaction research include: 1) art of care, 2) technical competence, 3) availability, 4) continuity of care, and 5) treatment outcomes; the factor making the greatest impact is art of care or, in layman's terms, "bedside manner" (Bartlett, Grayson, Barker, Levine, Golden, & Libber, 1984; Young, Wasserman, McAuliffe, Long, Hagan, & Heath, 1985; Ware et al., 1978). Researchers have had difficulty specifying which physician behaviors best convey "art of care," and although interpersonal traits, such as empathy, understanding, and friendliness are frequently listed, they are rarely defined behaviorally (Dimatteo, Prince, & Taranta, 1979; Francis, Korsch, & Morris, 1969; Freeman, Negrete, Davis, & Korsch, 1971; Young et al., 1985). Technical competence is also influential (Ware et al., 1978) and has been closely linked to art of care by both physicians and patients (Ben-Sira, 1976; DiMatteo & Hays, 1980). Technical competence increased in importance for high SES patients and for those anxious about their 15 illness (Ben-Sira, 1980; DiMatteo & Hays, 1980). Availability and continuity of care were also influential factors in predicting satisfaction, but more so for chronically ill patients (Breslau & Mortimer, 1981; Shortell, 1976). As can be seen, there are many determinants of satisfaction and no theorist has fit them into a working model that predicts satisfaction. Moreover, most studies deal with only a few aspects at one time, so it is difficult to compare findings. It is probably correct to say that all the determinants of satisfaction have not been identified, although the review article by Ware and associates (1978) makes a creditable attempt. The next two sections review the literature on physician verbal and nonverbal behavior, integral parts of "art of care." Physieian Verbal Behaviors in Adult Patient Visits One physician verbal behavior appearing to relate to satisfaction is the amount of information patients receive about their illness (Carter et al., 1982; Roter, 1983; Waitzkin, 1984). Receiving such information not only increases compliance (Putnam et al., 1985) but has also been shown to improve post-operative healing (Waitzkin, Stoeckle, Beller, & Mons, 1978). Yet in more than half of their office visits, physicians underestimate patients' interest in receiving information (Roter, 1983; Waitzkin, 1984), and this misattribution may be one of the most common errors in clinical practice (Waitzkin, 1984). In one study, doctors were averaging only a little more than one minute out of a twenty minute encounter to impart knowledge, yet estimated that they were spending 16 considerably more time than that (Waitzkin, 1984). The situation is exacerbated because patients routinely fail to communicate their need for more information (Roter, 1983). Using medical jargon is another verbal behavior negatively related to patient satisfaction. Although patients are more familiar with medical terminology than physicians think, patients still do not know very much (McKinlay, 1975). Mutual understanding is not achieved when doctors fail to communicate in terms understandable to their patients (Ley & Spellman, 1967). Patients respond best when communication about their illness is in simple language (Bartless, Johnston, & Meyer, 1973; Schwartz, 1970), and, not surprisingly, simplifying information for patients also enhances patient recall (Bradshaw, Ley, Kincey, & Bradshaw, 1975; Ley, Goldman, Bradshaw, Kincey, & Walker, 1972). a o verbal ehavio s ' du t Pat'ent Vis'ts Nonverbal communication is divided into decoding and encoding skills. Decoding skills involve "reading" an individual; encoding skills relate to the transmission of nonverbal cues. According to DiMatteo (1979), physician decoding skills are crucial because patients may attempt to monitor and control their own verbal messages of dissatisfaction to a high status professional. (Nonverbal behavior being more difficult to control relates to the "leakage hierarchy" discussed earlier.) If the physician decodes or recognizes such cues, problems can be dealt with immediately and minimized (DiMatteo, 1979; Vaccarino, 1977). A patient's nonverbal cues can also indicate the level of l7 understanding and willingness to comply with the recommended treatment (DiMatteo, 1979). Not surprisingly, physicians who were more sensitive to body movement and posture cues received higher ratings from their patients in "art of care" than physicians who were not as adept at decoding nonverbal behavior (DiMatteo & Hays, 1980). The nonverbal cues projecting empathy or caring and concern have been studied most frequently (Harrigan & Rosenthal, 1986), with correlations noted between patient satisfaction and distance from physician (Byrne & Heath, 1980; Larsen & Smith, 1981; Weinberger, Greene, & Mamlin, 1981). Physician gaze and body orientation were also important (Byrne & Heath, 1980; Larsen & Smith, 1981). Higher satisfaction was particularly associated with a forward lean, directly facing the patient, open arm positions, and nodding head movements; a physician's backward lean was associated with lower satisfaction (Larsen & Smith, 1981). In another study of high and low rapport doctors, distance, body orientation, and arm and leg and gaze position continued to be significant discriminators (Harrigan & Rosenthal, 1986). However, contrary to previous findings, high rapport doctors maintained less mutual gaze because they were looking at the patient's chart, which was felt to convey interest in the patient's condition. Another physician behavior appearing to relate to satisfaction is listening, although it has rarely been a focus in the research. [Listening, though not classified as either verbal or nonverbal behavior, appears to be nonverbal because it "excludes the specifically linguistic" (Reber, 1985)]. Patients want the physician to listen to 18 their comments and questions, and when this occurs both patient satisfaction and compliance increase (Stiles, Putnam, Wolf, & James, 1979). Physicians who take the time to listen to their patients give them a sense that adequate time has been spent. Interestingly, DiMatteo and colleagues (1979) found that perceived time with the physician was more important than actual time, that is, the patients’ perception that enough time had been taken to talk with the doctor was more important than the actual time spent. Confounding the issue of how much time should be spent listening to the patient is the need for a thorough medical history. Traditionally, the history is taken at the beginning of the visit with the doctor asking specific questions. In West’s (1984) analysis of questions and answers during a medical visit, she found 91 percent of the questions were doctor-initiated. In four out of the five visits involving adolescents, all of the questions were asked by the physician. Concerning talking in general, one estimate is that 75 percent of the statements are doctor-initiated (Byrne & Long, 1976). A delicate balance needs to be struck between the patient's need for medical information and the need for the doctor to listen, competing needs both of which enhance satisfaction. Parent Satisfaction Research The classic study on parent satisfaction was done 20 years ago in an acute care setting, the Emergency Clinic at Children's Hospital of Los Angeles, on 800 parents who brought their children for treatment of 19 common pediatric problems (Korsch, Gozzi, & Francis, 1968). Since there were no long-standing relationships between parents and physicians, the study focused on the initial medical interview. The participating physicians were residents or fellows. Data was collected by tape recording the physicians' visits, reviewing the charts, and conducting follow-up interviews with the parents. Only the physicians' verbal communication was studied. The results of the study are succinctly expressed in the article title, "Gaps in Doctor-Patient Communication." Using Parsons' divisions of physician behavior--"instrumental" (technical aspects of medical care) and "expressive" (psychological and social aspects of care), Korsch and her colleagues (1968) found that parent satisfaction was almost entirely influenced by the expressive component. This finding has also been supported by others (Deisher, Engel, Spielholz, & Standfast, 1965; Liptak, Hulka, & Cassel, 1977). In general, mothers specifically expressed a wish for more warmth and friendliness, and a greater show of concern, (Francis et al., 1969; Korsch et al., 1968). A link between parents' expectations, compliance, and satisfaction was noted (Francis et al., 1969). Physieien Verbal Behavior in a Pediatric Visit Looking first at verbal behavior, one "gap" uncovered in the doctor-patient relationship was the physicians' frequent use of technical terms (Korsch et al., 1968; Young et al., 1985). In more than half the cases, pediatricians used difficult technical language (Korsch et al., 1968). Oddly enough, although the use of medical jargon blocks effective 20 communication, Korsch and her associates found it did not always lead to dissatisfaction, a finding contradicting that in the adult patient satisfaction literature. Korsch explained that some mothers, even though they could not understand the physician, were impressed with the quantity and technical nature of the explanation and were flattered. Another "gap" in the doctor-patient relationship was the mothers' intense need for an explanation of their child's illness (Korsch et al., 1968). Speculating that parents have a universal tendency to blame themselves for their child's illness, Korsch found that almost half of the 800 mothers did blame themselves. Others expressed this self-blame indirectly through defensive statements or feelings of guilt. Yet the physicians rarely explored these feelings or made any attempt to relieve the mothers' fears, and in a few isolated cases they specifically blamed the mother even though she had not blamed herself. Finding out the cause of the illness was central to meeting parental expectations. In adult patient visits, the expectation for a medical encounter is usually to obtain symptom relief. In pediatric visits, however, parents have a need to learn the cause of their child's illness in order to feel satisfied and relieved of guilt (Korsch et al., 1968). A third "gap" in physician-parent communication was found in the physicians' limited use of civilities and social commentary. Korsch and Negrete (1972) found "less than 6 percent of the doctor’s communication to the mother carried positive affect in the form of friendly remarks, joking, agreement or support" (pp.73-74), yet the authors found that social commentary produced favorable impressions in the parents. 21 Although it is helpful to know why parents feel dissatisfied in an initial pediatric visit, it is also important to know when parents "reach their breaking point" and decide to change pediatricians. A study of dissatisfied parents found they left their regular pediatricians for reasons very similar to those causing dissatisfaction in an initial visit, that is, unhappiness over personal components of care. In these situations, dissatisfaction with the personal component spilled over into questioning the physician's technical competence (Young et al., 1985). P ia 0 ve ba Behavior n Pediat V'sit No research was found that examined physicians' nonverbal behavior in a pediatric visit. However, in Korsch's discussion of parents' comments, she stressed the need for the physician to listen to the parent. Listening, as previously mentioned, will be considered a nonverbal behavior for purposes of this study. Listening to the parent was not directly addressed in the Korsch study, but the limited amount of time given to parents by the doctors for asking questions was a complaint (Korsch et al., 1968). Yet, most pediatricians felt they offered adequate opportunity for questions. In actuality, 10 percent of the mothers asked no questions and an additional 27 percent asked only one or two questions that did not relate to the presenting problem, for example, "Where is the hospital pharmacy?" In postvisit interviews, one out of four mothers stated they would have liked more time to question the doctor. For those who did ask questions, 22 Korsch found the doctor frequently ignored them, gave a vague answer, or changed the subject (Korsch et al., 1968). Related to parents' not having time to ask questions was the finding that 76 percent of the parents' main worries were not mentioned to the doctor during the medical visit. Better educated parents had a significantly larger proportion of their concerns and expectations met than did parents who had not finished high school (Korsch et al., 1968). Limited time is often mentioned by pediatricians as one cause for ineffective communications with parents during the medical visit; however, Korsch and her colleagues (1968) found no correlation between diagnosis and interaction time or between diagnosis, interaction time, and satisfaction. Korsch concluded that the physician-parent communication can be satisfactory in as short a time as five minutes. Since the average time spent with a pediatric patient ranges from a low of 8.4 minutes with adolescents to 11.7 minutes with infants younger than six months, satisfaction should easily be achievable (Reisinger & Bires, 1980). P a -C d te action In a pediatric visit most of the physician’s communication is with the parent, regardless of the age of child. Traditionally, the feeling was that relatively little could or should be directly communicated to the child after limited involvement in the history-taking and physical exam. Therefore, discussion of the treatment was usually aimed at the parents (Stewart et al., 1981). Logically, children's participation 23 during the visit, especially in the interview and treatment segments, is limited not only by their cognitive level and parental controls, but also by accepted medical norms (Pantell et al., 1982). The child's participation is further limited by the physicians' knowledge of children's cognitive levels (Campbell, 1978; Stewart et al., 1981). Physicians' communication with children is not well researched, but even pediatricians sometimes do not have adequate knowledge of children's cognitive development. Perrin and Cerrity (1981) found that physicians, when asked to evaluate children's statements, were poor judges of children's ages. Also, physicians' knowledge of a child's understanding of basic health concepts is limited (Pantell et al., 1982). There is a growing body of literature on a child's conception of health, illness, and bodily functions, concepts which develop in a fashion similar to other concepts (Pantell et al., 1982) and are well developed by adolescence (Campbell, 1975). Recently more interest has been shown in increasing the physicians’ involvement with children. In 1968, as Korsch and her colleagues noted, only 12.5% of all interactions were between the physician and child and most of the statements to the child were social rather than medical. In 1982, Pantell and his coworkers found physician-child interaction had increased to 45.5% of the total interaction in a pediatric visit. But surprisingly, the number of previous medical visits or the seriousness of the illness had no bearing on the extent of communication. The noted increase in physician-child communication suggests that children can assume more responsibility than either physicians or parents give them 24 (Campbell, 1978; Stewart et al., 1981), and have an inherent interest in doing so, especially for certain medical problems, such as enuresis and encopresis (Vaughn, 1957). Concerning the content of physician-child communication, Stewart and his associates (1981) found that 35 percent of the physicians' statements related to the physical exam, 20 percent to the presenting problem, 15 percent to social exchange, and 12 percent to encouragement and reassurance. Substantive communication increased with the child's age, and boys received more information than girls (Pantell et al., 1982). The American Academy of Pediatrics (1977) has stated that preventive pediatrics is one of the main goals of quality medical care for children, yet little time is devoted to illness prevention and good health promotion even in well-child visits (Reisinger & Bires, 1980; Stewart et al., 1981). One encouraging finding by Reisinger and Bires (1980), however, is that recently trained physicians were more likely to spend time talking to children about general health issues. Many questions are raised when it comes to assessing the impact of physician-child interaction on parent satisfaction. Freemon, Negrete, Davis, and Korsch (1971) noted that physicians who talked more extensively with children were rated higher in parent satisfaction and compliance with treatment regimens. Yet, a recent study (Scott, 1985) that compared the impact of physician-parent to physician-child interaction on parent satisfaction found a stronger correlation to the physician's ability to communicate with the parent. 25 When it comes to the impact of extrinsic factors related to the visit--1ength of waiting time, courteousness of office staff, proximity to doctor's office, problems with parking and billing--the research suggests that some factors have a more direct impact on parent satisfaction than others. Although length of waiting time was mentioned by 25 percent of the parents in the Korsch study (1968), a long waiting time apparently did not negatively influence their impressions about other aspects of the visit. In another study, communication skills of the office staff were judged to be important in determining parent satisfaction. (Charney, 1972; Young et al., 1985). MW Some basic limitations of the research just reviewed should be noted. Satisfaction studies lack definitions, models, and theories, which makes comparisons of the results confusing and difficult (Locker & Dunt, 1978). Only one researcher attempted a definition of patient satisfaction--"the individual's positive evaluation of the health care he or she experienced" (Linder-Peltz, 1982) which provides little conceptual enlightenment. Researchers frequently mention the multi-dimensional nature of patient satisfaction (Linder-Peltz, 1982; Pendleton & Hasler, 1983; Ware et al., 1978), which is explained as the evaluation of different aspects of the care as well as an overall evaluation of the health care event. It is not clear whether all aspects or dimensions of satisfaction have been clearly identified. In 26 addition, many of the methods used to measure satisfaction were single-item measures and in many instances, reliability data was not reported (Ware et al., 1978). This project attempts to overcome some of the above limitations. A definition of satisfaction is offered. Parent satisfaction refers to the fulfillment of expectations about the delivery and outcome of pediatric 'care. A model of parent satisfaction is one purpose of this project and will be described in the last chapter. It will be multi-dimensional based on the findings of this research, yet at this developmental point in the parent satisfaction literature, can only be considered prototypic. Finally, the methods of data collection will go well beyond the single- item measures frequently used in the past. Both qualitative and quantitative methods will be used, and the one quantitative measure used will have reliability data reported. Segmery and Implications for this Project The following is a summary of the main points of this review of literature: 1. Controversy exists as to the relative impact of verbal and nonverbal behavior in communication, in part, because of the methodological limitations in the nonverbal research. 2. Verbal and nonverbal behavior have not been examined in the same study in a medical setting. In other naturalistic settings, verbal behavior appears to have more impact than nonverbal behavior. 3. Art of care or "bedside manner" is the most influential aspect of the physician-patient interaction in predicting adult patient satisfaction, but it is not well defined. 27 4. The following types of statements by physicians have been singled out as impactful in adult patient satisfaction: a. medical information (positive) b. medical jargon (negative) 5. In the adult patient satisfaction literature, certain physician nonverbal behaviors are beginning to emerge as most impactful: physical distance between patient and doctor, trunk position and orientation, arm and leg positions, gaze, and listening. 6. An adult patient's perceived quality time with the physician is more important in determining satisfaction than actual time with the physician. 7. In a pediatric setting, physician verbal behaviors linked to parent satisfaction are: a. medical information (positive) b. medical jargon (negative) c. social commentary and humor (positive) 8. No research could be located on physician nonverbal behavior in a pediatric visit, but listening is speculated to have a positive effect. 9. The impact of physician-child interaction has only recently been a focus of interest and its impact on parent satisfaction is unclear. 10. Related to length of the medical visit, parents wanted more time to question the doctor, although little correlation between length of visit and satisfaction was noted. Most parents (76%) did not express their main concerns to the physician during them medical visit. Little doubt exists that effective physician communication increases adult patient satisfaction. Defining effective communication is an onging process and this study attempts to make a contribution by using qualitative and quantitative methods to judge the effectiveness of specific verbal and nonverbal physician behavior. The findings related to adult patient satisfaction will be examined as they relate to parent satisfaction. The impact of physician-child interaction and length of the visit on parent satisfaction will also be examined. Finally, the relative impact of both physician verbal and 28 nonverbal behavior will be studied, something that has not been done either in adult patient or parent satisfaction investigations. Implications of this research pertain to utilization of health care services, compliance with treatment regimens, and ultimately children's health risks and costs. Other implications relate to teaching future physicians interpersonal skills. Chapter III DESIGN OF THE STUDY This chapter presents the plan of operation for this study, beginning with descriptions of the setting and the samples--physicians, parents, and children. Subsequent sections include procedures for collecting data, selection and description of the instruments, explanations of the research hypotheses and question, the design, and finally, the procedures for analyzing data. Description of the Setting The setting was a local hospital's acute care pediatric clinic called the After-Hours Clinic, which is open only after regular office hours. The clinic is staffed on a rotating basis by 18 Board Certified pediatricians, with each physician acting in the capacity of attending physician approximately twice a month. These staff physicians also work in private or group pediatric practices. Medical students and residents are part of the staff, but neither were included in the physician sample. During winter when data for this project was being collected, approximately 30 pediatric patients were seen per day. Description and Selection of the Samples P ns 0f the attending physicians, seven (41%) agreed to participate--six males and one female. Their ages ranged from late twenties to early 29 30 sixties with an average of fifteen years experience in private practice. Five were Caucasian; two were Asian. According to their medical supervisor, the physicians who volunteered for this project were representative of the interpersonal skill level of the entire staff. listens: The 35 parents participating in this study were a volunteer sample of those who brought their children to the After-Hours clinic between the months of January and April. Although no attempt was made to control for demographic data, it was collected so an accurate description of the sample could be written. Eighty—eight percent were mothers. Eighty-three percent of the parents were married; three percent were divorced; fourteen percent were single. The mean age of the sample was 32. Parents' education ranged from a high school diploma to a Ph.D. with the average education level being two years of college. C ' d en The children ranged in age from six months to 16 years with a slight majority (51%) being younger than four. Fifty-four percent were male. Due to the purpose of the After-Hours clinic, all pediatric visits were for acute medical problems. Acute illnesses, those with symptoms severe enough to limit activity or require a doctor's attention, account for nearly half of the patient volume of pediatricians in any setting. Children under 5 have about 3.5 acute illnesses per year with 8.8 days of restricted activity (Haggerty & Green, 1984). As in most pediatric 31 settings, ear infections were the most common presenting problem (60%) in this research. Procedures for Collecting Data Pediatricians employed at the After-Hours Clinic were sent a cover letter and a consent form (see Appendix A), which explained the procedures and nature of the project. Parents were first approached by a nurse, who briefly explained the procedures and asked if the parents would consider participating. If they agreed, the nurse put them in one examination room designed for videotaping behind a one-way mirror. (This meant that the physicians were always aware of the possibility they were being taped.) After the nurse took a brief history, I entered and explained the procedures in detail. If the parent(s) was still interested (and the child four or older gave his or her verbal agreement) a written explanation of the project and consent form were given to the parent to read and sign (see Appendix B). The parent was also asked to fill out a brief demographic data sheet (adapted from Scott, 1985) prior to seeing the physician (see Appendix C). To insure confidentiality, all subjects were assigned a code number which was then the only identification used on the research instruments. The letter of consent, the only document containing the name of the parent, was then separated from other documents to prevent identification of subjects. The visit with the physician was then videotaped from behind a one-way mirror. The permanently—placed microphone hung from the ceiling 32 so that the only equipment added during the taping was an extra light in one corner . At the conclusion of the physician visit, the parents were escorted to another room where a VCR, black and white TV, and audiotape recorder were set up. They were asked to recall two moments during their visit-~the most and least satisfying. These two moments were then viewed on the TV and the parents were asked for comments with a focus on the physicians' behavior. This idea for viewing and having the parents comment was adapted from Erickson's work (1982) on social interactions. The insiders' viewpoint was essential for a more complete understanding of what helped parents form impressions of the doctor. Having them view and comment on the videotapes was felt to the be most accurate way of helping them recall what they were seeing and feeling at a particular moment. Viewing the tapes was also thought to be helpful in remembering nonverbal behaviors. Parents' comments were audiotaped for later transcription. Parents then completed a satisfaction rating scale, the Parent Medical Interview Satisfaction Scale (P-MISS). Instruments Three instruments were used in this project--two checklists, one each for coding verbal and nonverbal behaviors and a parent satisfaction rating scale, (P-MISS). 33 Checklists for Physician Behaviors Verbal Checklisg. Although a few categories on the verbal coding checklist were taken from available checklists (Bales, 1950; Roter, 1977), the one used for this research is much more extensive (see Appendix D). It includes six major categories--l) Phatic, 2) Requests for Information, 3) Statements Conveying Information, 4) Evaluating Statements, 5) Requests for Action, and 6) Requests for Questions. With the exception of the Phatic category, each of the other major categories has five subcategories--current symptoms/problems, sociopersonal considerations, generalized health care, treatment, and follow-up. The three subcategories under Phatic are general, humor, and empathic. Phatic language contributes to the establishment and maintenance of communicative contact, e.g., "Great day, isn't it." (Lyons, 1988, p. 53). The major categories, again with the exception of the Phatic category, were established on the principle of function as opposed to form. Function refers to how verbal expressions are used and does not rely on a specific syntactical order or form. An example of this is when the physician says, "Give this medicine to her every four hours." This sentence has two functions--convey information about the medicine and request an action from the parent. The syntactical form of the sentence classifies it not as a request or question, but as a command-- "(You) Give this medicine to her every four hours." To correctly classify the function of a sentence, it is necessary to examine it in the context in which it occurs. This project was more interested in the function of the physicians’ statements than their syntactical form. 34 Nogverbal Checklisg. The nonverbal coding checklist was adapted from a checklist by Harrigan and Rosenthal (1986) for doctor-patient interactions (see Appendix E). It was one of two nonverbal checklists located and the only one specifically designed for analyzing physician behavior. The checklist adapted for use in this project includes the nonverbal categories of body orientation, trunk angle, arm position, leg position, head movement, facial expression, listening, and writing. Most of these categories had subcategories of behaviors. The adaptations of the checklist included omitting the categories of proxemics and gaze. The placement of the camera and limited lighting made it difficult to judge distance between physician and parent, i.e., proxemics, and direction of gaze. e edical terview atis action Sca e -MISS Although numerous scales exist for measuring adult patient satisfaction, the Parent Medical Interview Satisfaction Scale (P-MISS) is the only known scale for measuring parent satisfaction (see Appendix F). It assesses one aspect of satisfaction: parents' impression of physician conduct during the medical visit (Lewis, Scott, Pantell, & Wolf, 1986). The 26 statements on the scale were generated from parent and physician interviews and from revised items on the Medical Interview Satisfaction Scale (MISS) (Wolf, Putnam, James, & Stiles, 1978), a measure of adult patient satisfaction. All statements are rated on a seven-point Likert scale, with a 1 signifying Very Strong Disagreement and a 7 Very Strong Agreement. Nine of the 26 statements are negatively worded to reduce the potential of an acquiescence response set (Lewis et al., 1986). The 35 P-MISS statements are divided into four subscales--1) Physician Communication with the Parent, 2) Physician Communication with the Child, 3) Distress Relief, and 4) Adherence Intent, that is, intent to adhere or comply with the recommended treatment regimen. Reliability. The P-MISS is considered a highly reliable, internally consistent scale and has demonstrated a total score alpha reliability of 0.95 (with or without Adherence Intent items); for the four subscales, the alpha reliabilities have ranged from 0.81 to 0.93 (Lewis et al., 1986). Lewis states that these reliability coefficients compare favorably with the 0.90 suggested by Helmstadter (1964) for scales designed to compare individuals. The distribution of scores was moderately negatively skewed, with 60% falling in the top point interval--the mean satisfaction score was 6.0 with a range from 3.5 to 7 and a standard deviation of 0.82 (Scott, 1985). This score distribution compares favorably with other patient satisfaction measures (Korsch et al., 1968; Ware, Snyder, Wright, 1976; Wolf et al., 1978), which have found that patients are hesitant to give strongly negative evaluations of physicians (Wolf et al., 1978). Finally, demographic characteristics did not affect either the total satisfaction score or any of the subscale scores (Lewis et al., 1986). Validity. Both concurrent and construct validity of the P-MISS have been assessed (Lewis et a1. 1986). Concurrent validity is the relationship between test scores and indicies of criterion status obtained at approximately the same time (Anastasi, 1982), and was gathered by videotaping the actual pediatric visits for which parents 36 later completed the P-MISS (Scott, 1985). These videotapes were then coded for physician behavior reflecting empathy and respect. Scales developed by Carkhuff (1969) were used to check validity. Parent satisfaction showed an acceptable correlation (0.56) between the objective ratings of the physician interpersonal skills and the pediatric visit (Scott, 1985). Construct validity refers to the extent to which the test measures a "theoretical construct" or trait (Anastasi, 1982). Preliminary evidence of the P-MISS's construct validity comes from significant correlations of the P-MISS and all subscales with objective ratings of physicians' interpersonal skills during the medical interviews (Lewis et al., 1986; Scott, 1985). Design This study is basically descriptive, the purpose being to explore parent satisfaction in an acute care setting. The natural setting for this project limits possibilities for manipulating variables, yet possibilities for ecological validity are increased. The design of this study can best be understood by picturing a chart in which the seven doctors are listed across the top and the five visits (by individual parents) to each doctor are listed down the side. This is a repeated measures design-~the five visits to each doctor being the repeated measures. The visits (observations) each involving a different patient are nested within doctors. 37 Research Hypotheses This study investigated several factors in an initial pediatric visit felt to be associated with parent satisfaction. As mentioned earlier, parent satisfaction is defined as the fulfillment of expectations about the delivery and outcome of pediatric care. Most of the hypotheses focus on the physicians' behavior, but the impact of physician-child interaction and length of the physician visit were also studied. The first hypothesis deals with the parents' impression of whether the physicians' verbal or nonverbal behavior was more important in their selection of most and least satisfied moments. Hypothesis 1: Parent w choose mo e h sician verba b hav 5 th no ve behav ors as 'n ue n the' cot on of mos t' 'ed nd lees; eetlsfied moments. The next hypothesis compares the impact of physician-parent interaction to physician-child interaction on the parents' choice of a most and least satisfied moment. Hypothesis 2: P nts wi choos mor h s - are inte actio s a er the h 51 ian-child interactions when selectin thei most and leest gatlefled moments. The next question examines the impact length of visit had on parent satisfaction. Research Question 3: Will length of physlcian visit have an effect on pegent satlefactiog (P-MlSS)? 38 The next set of hypotheses relates to specific physician verbal behaviors. Hypothesis 4a--Certa h sician verba statements e a car n an ern ° udin oc o e sona consideratio s med ca t‘ n e ts fo uestio s will have a as t ve e fee 0 eegiefeetlon (E-MlSS). Hypothesis 4b: Certa 5 ve a1 ta ' a e e t o 1 0.1m; 0 -1('l b ta 1_ a an: as 0 I". a,_ 8-,103 W 1-V- a v 0 rent a tie P- The final set of hypotheses examines specific physician nonverbal behaviors. Hypothesis 5a: e ta h s v a vio e d e t n w a e b orwa d lea towa d are c o e rm peelelene, (d) opeg leg poeltiege, age (e) lieteeigg, will have a v e 0 en 5 s t o -M . Hypothesis 5b: Ce tai h a onve a v o e a t’ e wit oriented awa from the arent b closed arm ositions a c 05 ed e as tions wi have a e at ve effect 0 arent t’ fact (P-MISS). Variables Outcome V ab es This study has two outcome variables. For Hypotheses l and 2, the outcome variable is the parent-selected most and least satisfying moments explained in the procedures for collecting data section. For Question 3 and Hypotheses 4 and 5, the outcome variable is the parent satisfaction rating scale (P-MISS). Further discussion of the P-MISS is needed. 39 N£;M1§§. The P-MISS yields both a total scale score and four subscales scores. I decided not to figure any of the subscales scores for two reasons--l) obtaining a more global impression of satisfaction was the purpose of this research; and 2) all subscales could not be used with all parents as will be explained in the next paragraph. In this project, the total parent satisfaction score was based on 22 items--the four adherence intent items were not included. According to Lewis et al. (1986), the Adherence Intent Subscale may be omitted and this was done because it measures compliance, an outcome factor of satisfaction rather than a contributing factor. The P-MISS was designed for use with parents who had children four or older because it contains physician-child communication items, such as "The doctor talked to my child about what (s)he can do to become more healthy" and "The doctor listened closely to my child talk." For those parents of children under four, the 10 items in the Physician Communication to Child subscale were omitted in figuring the total score. To make an appr0priate composite variable, given this situation, I used the mean score of the answered items as the representative composite variable and called this variable, NP-MISS. Cronbach's alpha, a measure of reliability referring to generalizability, was high (.94). Finally, the P-MISS version used in this project varies slightly from the one published by Lewis and associates (1986). The wording of one item in the most recent revision has been changed slightly from "The doctor listened closely to my child" to "The doctor listened closely to my child talk." The word "talk" was added so the parent would focus on 40 the physician communicating with the child rather than listening to the child through the stethoscope. WM There were several predictor variables, some relating to visit or within doctor differences, others relating to between doctor differences. Vlslt variables. Preliminary analysis, correlations, indicated that Length of Physician Visit should be used in an attempt to explain within-doctor variance. In order to have more reliable measures of the physicians' verbal and nonverbal behavior, scaled variables, both verbal and nonverbal, were created. After examining the correlation matrices, intercorrelated items were chosen to create scale variables. Since individual behaviors were measured in frequencies and the behaviors are believed to differ from one another in their chance of occurrence, those selected items (behaviors) were standardized first and added together. The behaviors in the new verbal variable differ slightly from the ones listed for Hypothesis 4a. Verbal behaviors in Hypothesis 4b could not be tested due to low counts. The new verbal variable is comprised of physician verbal behaviors of requests for questions, caring and concern, and medical information. Reliability of this variable was moderate (.55). Two nonverbal variables differing slightly from behaviors listed in Hypotheses 5a and 5b were formed due to conceptual similarities and intercorrelations--nonverbal 1 ("positive" behaviors) and nonverbal 2 41 ("negative" behaviors). Nonverbal l is the behaviors of forward lean, arms open, listening, head nodding, and smiling. Legs crossed was omitted because of insufficient data; three others were added, listening, head nodding, and smiling. These five behaviors were standardized and added together resulting in Nonverbal 1, which had a reliability of .46. Nonverbal 2 consists of sitting upright, trunk oriented away from parent, arms closed, head shaking, and writing. Again, behaviors comprising Nonverbal 2 vary from those listed in the hypothesis for similar reasons. The same computation used to form Nonverbal 1 was used for Nonverbal 2. Reliability was low (.24). Qeeto; veglebles. There was only one doctor variable, years of experience. Procedures for Analyzing Data The data was analyzed both qualitatively and quantitatively. For Hypotheses 1 and 2, a combination of quantitative and qualitative techniques were applied. For the three remaining hypotheses, only quantitative techniques were used. Dat om e P st-vis t Parent Inte iew The audiotaped data from the post-visit interview was subjected to both quantitative and qualitative analysis. The parents' comments explaining their choice of a most and least satisfied moment were transcribed from the audiotapes. This data was used to test Hypothesis 1 42 which compares the impact of physicians' verbal behavior to nonverbal behavior in the parents' explanation of a most and least satisfied moment. It was also used for Hypothesis 2, which compares the impact of physician-parent to physician-child interaction in the parents' choice of a most and least satisfied moment. The following steps were used for this analysis, with steps 6 and 7 adapted from ua ’tative Me ds in Reeeagch en Ieaehing (Erickson, 1985). 1. Parents' responses were categorized under two headings-~most and least satisfied moments. 2. The responses were then divided into physician interactions with the parent or child. 3. Under the two interaction categories, each response was then further classified as either verbal, nonverbal, or both. 4. The number of responses in each of these categories was tabulated. 5. The responses were then examined for typical patterns. A pattern was said to exist if 20% of the responses in a category expressed similar ideas. 6. Direct quotes from the parents' responses were used to illustrate the various patterns. 7. Finally, an interpretive commentary explains each pattern of a parent response. (Chapter V will contain further interpretive information, but of a more general nature.) Of the above steps, only the last two pertain to qualitative analysis. The quantitative analysis for Hypothesis 1 and 2 was limited to calculating percentages made more meaningful when complemented by qualitative illustrations and interpretation. I felt that the essential information to be learned from the first two hypotheses could be obtained best by examining direct quotes. Although the frequency with which 43 verbal and nonverbal behaviors is mentioned is important, the parents' description of the physicians’ behavior could be even more informative. P55. 01'. 0" ' ”1‘ ,, ‘t‘ 0110 -' d g3 5. .8d 1.11:1,8 5 d The verbal and nonverbal behaviors analyzed were taken from the most and least satisfied moments and from samplings of the entire visit. Three one-minute samplings of physician verbal behavior were taken, one each from the history-taking (first third), physical exam (middle third), and diagnosis and treatment (last third). The three one-minute samplings of each visit were combined to obtain summary indicators. In some visits, a "pure" sampling from each of the three segments was not possible because some of the segments were too short. The verbal behavior was transcribed and coded by two trained coders. The coders received several hours of instruction, including practice sessions. All the verbal coding was done before the nonverbal coding was introduced. The coders had a written transcript of the physician's statements and worked from the videotape so the statements were heard in context. The coders were not aware of the parents' comments or ratings made during the post-visit interview. The coders coded independently, but were told they could discuss ambiguous statements and change their coding, if they chose. The prOportion of agreement between the two coders was .90. For nonverbal behavior, one ninety-second sampling was taken during the last third of the visit, the diagnosis and treatment segment. This 44 segment was chosen because the longest physician-parent interaction is likely to occur here. Although two coders were used for the nonverbal coding, they were coding different behaviors as they simultaneously viewed the tape, so no proportion of agreement could be figured. The nonverbal coding was more difficult to learn than the verbal coding. As mentioned earlier, quantitative analyses was done on all five hypotheses. For Hypotheses 1 and 2, percentages were calculated to identify various response patterns; Pearson correlation coefficients were figured for Question 3; the Hierachical Linear Model (HLM) (Bryk, Raudenbush, Seltzer, & Congdon, 1976) was used for Hypotheses 4 and 5. HLM is built around the idea that "parameters estimated at one level in the model become the outcome variables at the next level" (Raudenbush & Bryk, 1986, p. 3). This method enables separate specification of within- and between-doctors models and allows intercepts and within-doctor regression coefficients to be either fixed or random. In general, there are two steps in HLM analysis. First, I used the model in which no predictor variables are included in order to get the information of variance decomposition. This tells how much variation lies either within or between doctors, given the total observed variatiOn. After identifying the proportion of variation at each level, the next step is to specify the within-doctor and between-doctor models so the related hypotheses can be tested. The within-doctor model as applied to this research specifies the relationships among physician characteristics, such as physician verbal and nonverbal behavior and length of visit, and the outcome variable, 45 the parent satisfaction scale (P-MISS). The within-doctor model is as follows: Yij - 801 + Bllxl + BZLXZ + . . . + Bpixp + elj for i - 1,. . .,1 doctors, j - 1,. . .,J visits and p - 0, 1, . . .P, where Yij is the NP-MISS score of doctor 1 in visit j, Bpi is the effect of the predictor variable, Xp, on the NP-MISS across the five visits, and eij is the random error term that is normally distributed with mean 0 and variance 0'2. This is a standard linear model except the within-doctor regression coefficients are allowed to vary across doctors. If we set the variables to have fixed effects on NP-MISS, then HLM becomes the traditional regression model. For the between-doctor model, I used each slope of the within-doctor model as an outcome variable whose variation can be explained by doctor level variables, such as years of experience. Thus, the between-doctor model allows a direct representation of the effects of doctor variables on parent satisfaction and a simultaneous investigation of effects on parent satisfaction (NP-MISS) and within-doctor regression coefficients. The between-doctor model is: 801 - 60° + 60121 + 60222 + . . . + Oquq + 301 for i - 1,. . .,1 doctors, 301 is the slope of the within-doctor model; eoq are the coefficients of between-doctor variables, Zq; and aoi is the 46 error term that is assumed normally distributed with a mean 0 and a variance To . In other words, the slope of the within-doctor model is the function of doctor level variables, such as years of experience, plus the error (aoi)- Here the variance, To, can be seen as a parameter variance, because the parameter value of 301 varies around some fixed unknown, 600. If there are no predictor variables in both within and between doctor models, we just have the models containing only base slopes. Since we use two-level models we need to decompose the variation of outcome variable. In other words we need information on how much variation of NP-MISS lies within doctors and how much lies between doctors. We can get the information from the base only models where no predictors are used. To figure the proportion of between-doctor variance, this formula was used: where the values of T and 0'2 come from the models using only base slopes for both within and between doctor models. The percent of between-doctor variation equals the variation between-groups divided by the total observed variance, that is, the sum of between doctor variance, T, and within doctor variance, 6 2. The advantage of the HLM model allows the use of both within-doctor information and between-doctor information. That is, we can specify 47 within-doctor model with information of each visit and specify between-doctor model with information of doctors, such as years of experience. The next chapter presents results based on the analysis explained in this chapter. Chapter IV RESULTS Overview The primary purpose of this research was to analyze the interaction between physicians and parents in an initial pediatric visit in order to better understand parent satisfaction. This chapter presents the results of those analyses. The results reported here are organized around the presentation of each of the hypotheses, and the research question with the results from the quantitative analysis presented first. For Hypotheses 1 and 2, the quantitative analysis includes only percentages with data coming from the parent interview. For Research Question 3, data comes from the timing of each of the visits; Pearson correlation coefficients are calculated. Data for Hypotheses 4 and 5 comes from coded information taken from the videotapes of physicians' verbal and nonverbal behavior and HLM analysis is used. The second part of this chapter is the qualitative analysis of the parents' post-visit interview. It includes descriptions, quotes, and interpretations of patterns in the parents' explanations of their most and least satisfied moments. Hypotheses Relating to the Parent Interview The data analyzed for Hypothesis 1 and 2 comes from parents’ responses to two specific questions in the post-visit interview 48 49 concerning their choice of most and least satisfied moments. The parents were asked the following two questions: Most Satisfied Moment--What was the high point of the visit, that is, when were you feeling most satisfied and comfortable in your interaction with Dr. and why? Least Satisfied Moment--Again, remembering everything that happened in your visit with Dr. , and in comparison to the high point, there must be a low point. When did the low point or least satisfied moment occur and why? Although 35 parents participated in this project, only 30 post-visit parent interviews were obtained. Five interviews were not completed. In these 30 interviews, one parent was unable to choose a most satisfying moment, and three parents made comments related solely to the physical exam, which could not be construed as either verbal or nonverbal behavior of the physician. So a total of 26 interviews were analyzed for reasons why a certain moment was chosen as most satisfying. The data on least satisfied moments for Hypothesis 1 is based on eleven interviews. Of the 30 interviews completed, eight parents could or would not select a least satisfied moment, and nine interviews contained parent responses that did not relate specifically to the physician visit just completed. Of the remaining 13 interviews, two contained comments relating again to the child's physical exam and could net be categorized as either verbal or nonverbal. Hypothesis 1: Parents will chooee moge physiciae vegeel beh v'o an ve bal behavi s as ' luenc n heir selectlog of both most and least satisfied moments. 50 Of the 26 interviews analyzed for most satisfied moments, sixteen (62%) stated that their choice was due to reasons attributed to verbal behavior; five (19%) said it was due to reasons relating to physicians' nonverbal behavior; and five (19%) mentioned reasons that included both verbal and nonverbal behavior. There was a statistical significance between the two proportions (z - 4.43; p ®:——>© > Ix D = Doctor P a Parent C - Child 87 The second model illustrates the components of parent satisfaction in an acute care setting based on the findings of this study. It is presented here with the understanding that parent satisfaction is a complex concept with many determinants, only a few of which were examined in this study. 82121.2. Phxsicjan Verbal . Shorter Physician- Be avuor - diagnostic Visits Parent statements Interaction Parent Satisfaction < Inaybe Physician. Interaction Longer . during Visuts History-taking phase It is hoped that future research will provide a model incorporating reciprocal interaction of the determinants of parent satisfaction. However, before that can be accomplished, all determinants must be identified and weighed according to the influence they exert. This research has examined specific physician behaviors and identified some of the factors influencing parent satisfaction. 88 Limitations of the Study Generalizability is limited because of the specific medical setting, acute care, and by the fact that only initial physician encounters were studied. A majority of children are cared for by physicians in private practice, but acute care settings, such as after-hours clinics, hospital emergency rooms, and docs-in-the-box, are becoming more prevalent. The findings in this study may have limited relevance for longstanding physician-parent relationships, such as those in private pediatric practices, but the results do have relevance for all initial pediatric visits regardless of the setting, private practice or acute care. In addition to the above limitations, this study, as with all investigations in natural settings, had little control over many possible confounding variables. One of the drawbacks of medical satisfaction studies is the impossible task of trying to control all the people, setting, and treatment variables. This study attempted to focus on physician behavior and its impact on parent satisfaction, but even so the number of factors that could have influenced the physician-parent interaction was difficult to contain. Implications of this Study Ihggrgrigal Implications Parsons' Theory. Two implications relate to Parsons' view of structure of the medical system. The first has to do with Parsons' instrumental versus expressive acts in a medical encounter. 89 "Instrumental" acts are the technical aspects of medical care; "expressive" acts are the social-psychological aspects of medical care. In this study, the parents' choice of a most and least satisfying moment was influenced by both instrumental and expressive functions of the physician. Parents were most concerned about receiving a diagnosis (instrumental), yet many also mentioned the manner in which the doctor related to them (expressive). The Korsch study (1968), perhaps because satisfaction was measured differently, strongly supported the expressive function as being most influential in parent satisfaction. Relating the difference in between- and within-doctor variations to Parsons' view that the physician-patient (in this case parent) interaction is a rigid, predictable exchange, appears, at first glance, to hold up. That is, the differences in mean satisfaction ratings between doctors were small. These ratings were based on the parents' judgment of the doctors' behavior, so it is assumed that, on the whole, doctors were viewed similarly by parents in terms of satisfaction. Does this mean Parsons was correct in his assumption of an institutionalized role for physicians? Not really, because the concept of standardized physician behavior breaks down when visits to the same doctor were compared to one another. The large variation in satisfaction ratings of the same doctor was a surprising result, since it seems reasonable to assume that individuals will differ more when compared to other individuals than when compared to themselves. Knowing the large within-doctor variation encourages speculation about visit variables. To recall, specific clusters of verbal and 90 nonverbal behavior of the physician did not explain the differences. Visit length showed a negative correlation with parent satisfaction and accounted for some of the within-doctor variation. Other visit variables which may be influential are the seriousness of the illness, especially the presence of a high fever, chronicity of the ailment, child's general health, length of time spent waiting to see the doctor, parents' diagnostic and treatment expectations, and interaction with other medical personnel, for example, nurses, receptionists, and laboratory technicians. Commu t o Theo --Two A oach : t ct a a nte l. The structural approach to the study of communication posits that an individual is always communicating and that all behavior is communication, even silence. In contrast, another approach states that communication is distinguished by intentionality; that is, only behavior intended to convey some message can be considered communicative. Intentionality is the key concept distinguishing these two views. Relating to the structural approach, physicians would probably be surprised at some of the behaviors parents mentioned in explaining their most and least satisfying moments. What sometimes appeared to be inconsequential physician behavior was capable of causing parents to be either most or least satisfied. Here is an example: A doctor was writing his findings and treatment recommendations on the chart for several minutes and did not converse with the parent. A couple of parents stated that this was the low point (least satisfying moment) of 91 the visit. As one parent explained, "...that was the time I was sort of the only one there." The physician's behavior was having a significant negative impact on the parents' satisfaction at a time when the doctor was not consciously communicating. The approach based on the idea of intentionality did not hold up as well as the structural approach. Taking again the example of the "silent" doctor who was writing, it is quite likely that the doctor did not consciously intend to exclude the parent. Yet as shown by the parent's comments, that is exactly how she felt. Another physician behavior, teaching the parent, obviously intended to help, was viewed by two parents quite differently. One doctor’s effort to inform the parent came across as intended and was chosen as the most satisfying moment--"She explained it so well." Another doctor's attempt to inform was chosen as the least satisfying moment--". . . too much teaching." The intention of the two doctors was presumably the same, to inform the parents. Yet because the second doctor was oblivious to signals given by the parent or chose to ignore them; his detailed teaching, as the parent commented later, actually increased her stress. Although some of the ideas of the intentional approach have appeal, the idea that behavior is not communicative unless that is the intention requires reexamination. Kendon (1981) called the idea of intentionality irrelevant. That judgment may be a bit harsh, because in some situations the study of what is intended as opposed to what is communicated is a proficient way to learn more about why communication mixups occur. In conclusion, it appears that most, if not all, physician behavior has 92 communicative potential and the 'message' intended is not always the one received. P t'ca Im ca ons The results of this research have implications for physicians, parents, and children who are involved in medical care delivered at acute care settings. The major qualitative finding of this study was that when parents chose a most and least satisfying moment, physician verbal behavior was cited most frequently as the reason for their choice. Physician verbal behavior created the greatest satisfaction during the treatment and diagnosis segment of the visit and the lowest satisfaction during the history-taking segment. Physician interpersonal skills training should emphasize the parents' needs xary from the beginning to the end of the medical visit. More research on the various phases of a medical visit is needed. Although most of the burden for effective communication in a medical interview falls on the physician, educating parents (and adult patients) to communicate their needs to medical professionals is important. The uniqueness of this study in directly asking the parents to explain their satisfaction should be replicated in future research and become an integral part of the medical students' interpersonal skills training. Attempts to identify specific physician verbal and nonverbal behaviors influencing the degree of parent satisfaction were unsuccessful. Although the findings in this study were probably influenced by methodological problems inherent in any study of 93 communication, other methods to study communication might have proved more effective. For instance, rather than focusing on either individual in the dyadic exchange, in this case, the doctor, perhaps the emphasis should have been on the interaction between the two for a clearer understanding of the context in which the exchange took place. Warnings have been issued not to become too microscopic in the investigation of medical exchanges (Carter et al., 1982); although this project did not conduct the microanalysis characteristic of some researchers (Frankel, 1984), perhaps it was still too fragmented. The impression gained from this research is that more ethnographic research is needed to provide a clearer understanding of what is essential, then quantitative methods can be applied. Using qualitative results to assist in forming hypotheses for quantitative analysis which are then further refined using qualitative measures is a method that deserves further consideration. Analysis using both methods has the best chance to offer the reciprocal complementation necessary to understand communication complexities. Finally, interviewing the doctors would have been helpful in understanding whether they felt they met the parents' needs, but in a setting designed for rapid treatment delivery, doctors are rarely available immediately after seeing the patient. Although methodological limitations may, in part, explain the lack of significant results for verbal and nonverbal behavior, it was surprising to find that physician variables, such as years of experience, did not predict satisfaction either. Interestingly, Scott (1985) found a correlation between younger physicians and ability to convey empathy and respect to both parents and children. He further noted that previous 94 research has found decreases in empathy as medical education increases, although this was not supported by his own research because he used a Sample of physicians with a limited range of experience. If it is true that some younger physicians may have more effective communication skills than older physicians, this may be due to medical school curricula that now require interpersonal skills training. Implications are that experienced physicians may also benefit from courses on patient relationships in their continuing education. There are several implications of the finding that physician-parent interaction had a greater influence than physician-child interaction in choosing both a most and least satisfied moment. According to two pediatricians, Korsch and Aley (1973), much of a child's medical workup must be done through the parent, making the relationship with the parent extremely important. Put more specifically, they state that a fraction of an hour is spent with the child but it is the parent who will carry out the advice . . . all day long, every day of the week. (p. 5). Evidently, the parents in this study agreed with the conclusion; that is, exchanges with the parents were more important than with their children. Yet the importance of physician-child interaction should not be overlooked. In the findings here, parents focused on the physical exam for both their most satisfied and least satisfied moments. One implication is that pediatric training should include specific child management techniques that cover both invasive and noninvasive exams. In this study, for example, some doctors routinely completed a physical exam while the infant was in the security of its mother's lap; other 95 doctors had the infant lie on the examination table and then requested the parent restrain the child's head or arms sometimes causing both the parent and child distress. Learning techniques to reassure both the child and parent during this time is important. Although communication courses in medical school stress thorough, efficient history-taking, techniques that facilitate good communication with parents and children are neglected (Hornsby, 1983). Finally, physicians may find it reassuring to learn that at least in acute care settings parents can be satisfied and, in fact, sometimes are more satisfied with a short rather than a long visit. Korsch and her colleagues (1968) also found g9 relationship between interaction time and satisfaction and concluded that communication can be satisfactory in as short a time as five minutes. Since limited time is often mentioned by pediatricians as one cause for ineffective parent communication, it is encouraging to find that parents can have their expectations met in a short visit while in acute care settings. Directions for Future Research The present study examines only a few aspects of the vast complexities in a medical exchange. The results indicate a need for further investigation of the physician-parent dialogue, physician-child interaction, and parent expectations for these visits. Such research may help identify the variables responsible for the within-doctor differences. This study also poses questions for researchers concerning the appropriate level of communication analysis. 96 More information is needed regarding the physicians' perception of satisfaction with the parent interaction. For example, what are the physicians' expectations for a satisfactory pediatric visit, how do these change depending on the medical setting, the age of the child, and the diagnosis and treatment? How are these expectations communicated to the parent and the child? Do pediatricians and family practitioners differ from other specialists in the importance placed on patient communication? Interpersonal skills training for physicians is effective in the short run (Engler, Saltzman, Walker, & Wolf, 1981; Quirk & Babineau, 1982), but is it effective in the long run or does the emphasis shift back to efficient treatment delivery? A second line of inquiry is needed in the physicians' interaction with children. Although in this study physician-parent communication had a greater impact on parent satisfaction, the literature suggests a growing need for more effective communication between physician and child. Pantell and his colleagues (1982) are leading advocates of increased involvement of children in all phases of the pediatric visit. As they state, communication with children is not just an amenity but an actual therapeutic lever. Benefits include reduced surgical complications, increased parent satisfaction, and adherence to treatment regimens. The question does not appear to be whether physicians' involvement with children should be increased, but rather, what kind of involvement would be most appropriate at various ages? Stewart and his associates (1981) found the physicians communication with the child often lacked 97 medical substance. The task now is to define the goals for more effective communication and to learn what and how medical information can be conveyed at various stages of child development. Research on parent satisfaction and its contributing factors is still exploratory in nature. Although much can be learned from the adult patient literature, the pediatric visit presents a challenge to both practitioners and researchers because of the unique situation of the patient (the child) not being the main communicator. Further research on physician communication is necessary to determine the content and balance between parent and child that is most effective. Conclusions The American Academy of Pediatrics (1977) has stated that "discussion and counseling is the most important element of child health care." The results of qualitative analysis agree, highlighting the physicians' verbal behavior, especially diagnostic statements, and its impact on parent satisfaction. Attempts to identify other specific verbal and nonverbal behaviors using quantitative methods were largely unsuccessful. In future research, capitalizing further on the interplay between qualitative and quantitative methods will be useful. Each method has the capability of further refining the other until the desired level of analysis is achieved. As in this study, most analyses of physician-patient (and parent) interactions have been confined to the short-term effects of a single interaction. Such knowledge now needs to be supplemented by longitudinal 98 studies that examine changes occurring over time in a physician-parent relationship. Knowing what evolves out of these long term interactions will provide further insight into the composition of effective physician communication. APPENDIX A John C. Doe, M.D. 125 Medical Arts Building Lansing, MI 48921 Dear Dr. Doe: Dr Solomon and I are requesting your participation in a research project. I am a doctoral student, who is examining parent satisfaction in a pediatric visit. Participants will be chosen from the attending staff physicians at the After-Hours Clinic, who volunteer for this project. The only procedure necessary for your participation is to allow five to six of your clinic visits to be videotaped. It is anticipated that most of the videotaping could be completed by mid December. While there will probably not be any immediate benefits to you, i will be most willing to discuss with you the final project results, which should be available by next April or May. If you have other questions about the project before you decide whether or not to participate, please feel free to call me at 355-7821 or Dr. Solomon at 332-2310. Your participation would be greatly appreciated and if you should decide to be a part of this project, a physician consent form is enclosed along with a stamped envelope addressed to me at my MSU address. Sincerely, Anne Riley 99 100 PHYSICIAN CONSENT FORM Ms. Anne Riley, MSU doctoral student, under the direction of Dr. Richard Solomon and MSU Professor Don Hamachek, is conducting a research project on patient satisfaction in a pediatric visit. Ms. Riley and Dr. Solomon would like to ask your permission to videotape some of your After-Hours Clinic visits for this study. 1. Your participation in making the videotapes is voluntary and can be withdrawn at any time. You may have the tapes erased at any time. 2. The tapes will be analyzed by Ms. Riley, and 3-4 trained research assistants from MSU. To aid in analysis the tapes may also be viewed privately by appropriate members of Ms. Riley's dissertation committee. The tapes will not be viewed by anyone else in the Lansing area without your written consent. 3. The results of the research will be treated with discretion and confidence. The tapes will not be loaned to others without your written consent and will only be shown to scientific audiences outside the Lansing area where it is unlikely you would be recognized. If it is shown, all names will be erased first. Your name and others will be disguised in all written materials based on the tapes, such as notes, transcripts, papers, or publications. 4. Ms. Riley and Dr. Solomon would be pleased to talk further with you about this project as well as share the results with you which should be available in April or May, 1988. They can be reached at (517) 355-7821 or (517) 332-2310. I give my consent to be videotaped as part of this project relating to parent satisfaction. I have read this consent form and understand the tapes will only be used as described above. Signed: Date: Clinic: Witness: APPENDIX B RESEARCH PROJECT ON PARENT SATISFACTION IN A PEDIATRIC VISIT Dear Parent: Researchers from Michigan State University are conducting a project at our After-Hours Clinic. The project is about communication between the doctors and patients and its purpose is to better understand what factors in the visit with the doctor contribute to parent satisfaction. If there is time tonight, the researchers will come to talk to you to invite you to participate in the project. If you are willing to participate, you will be asked to fill out a brief questionnaire before your visit with the doctor, to have your visit with the doctor videotaped, and then to take a few minutes after the visit to be interviewed. This project is completely voluntary. You do not have to participate if you do not want to. If you agree, however, the researchers will be most grateful. They will keep information about your visit strictly confidential and you will receive a small gift for your time. After-Hours Clinic Staff 101 102 PARENT/CARETAKER CONSENT FORM Ms. Anne Riley, MSU doctoral student, under the direction of Dr. Richard Solomon and MSU Professor Don Hamachek, is conducting a research project on patient satisfaction in a pediatric visit. Ms. Riley and Dr. Solomon would like to ask your permission to videotape your child's visit with the doctor. They are also asking your permission to record a brief post- visit interview and to complete a 5-minute rating scale on parent satisfaction. 1. Your child's care will not be affected if you decline to participate in this study. 2. Your participation is voluntary and can be stopped at any time without and repercussions. You can request that the tape be erased at any time. 3. The tape, interview, and rating scale will be analyzed by Ms. Riley, trained research assistants from MSU, and appropriate members of Ms. Riley's dissertation committee. It will only be shown to scientific audiences outside the Lansing area where it is unlikely you would be recognized. If it is shown, all names will be identifiable by code only. In all written materials based on the use of these tapes, such as notes, transcripts, papers, or publications, no names will be used. In summary, the results of this study will be treated with strict confidence and you will remain anonymous in all written or oral reports based on this study. 4. Ms. Riley and Dr. Solomon would be pleased to talk further with you about this project at any time. They can be reached at (517) 355- 7821 or (517) 332-2310. The results from this project should be available in April or May, 1988. I give my consent to have a videotape made of my child’s clinic visit, to participate in a brief post-visit interview with Ms. Riley, and to complete a rating scale on parent satisfaction. I have read this consent form and understand that the videotape, interview and rating scale will only be used as described above. Signed: Date: Clinic: Witness: APPENDIX C Parent Code: Doctor Code: Date: 10. THE FOLLOWING INFORMATION WILL BE USED FOR STATISTICAL PURPOSES AND NO INDIVIDUALS WILL BE IDENTIFIED. Parent Information Surrey How old is your child? What grade is your child in? What is your relationship to this child? Mother _____ Father _____ Grandmother _____ Other Have you ever seen today's doctor before? Yes _____ No _____ # of times What is your age? What is your sex? M _____ F _____ What is your child's sex? M F Race: Caucasian;___. Black;___ Asian____ Hispanic____ Other____ Marital Status: Single___ Married___ Divorced___ Separated;__ Does this child live in a single parent family? Yes ____ No ____ *ANSWER ITEMS 12-16 ONLY IF YOU ARE THE PARENT OR GUARDIAN OF THE CHILD* ll. 12. 13. 14. Parents Education (circle the number of years completed): Yourself: 7 8 9 10 ll 12 l3 14 15 16 17 18 19 20 Other Parent: 7 8 9 10 ll 12 13 14 15 16 17 18 19 20 What is your occupation? Kind of business/organization What is other parent's occupation? Kind of business/organization 103 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 104 What is your family's approximate gross yearly income? $0"10»000._____ $10,000--20,000 _____ $20,000--30,000 _____ $30,000--40,000 _____ $40,000--50,000 _____ $50,000--60,000 _____ How many people are dependent on this income? What is the main reason for today's doctor visit? Illness Type Accident Type Other Has your child been seen anywhere for this problem in the past 4 weeks? Yes No How many times? How many times has you child seen any doctor in the last 6 months? What is your child's birth order? (circle the answer) oldest 2nd 3rd 4th 5th 6th or more Please describe your child's general health in the last 3 months. Very Good ______ Good _____ Fair _____ Poor _____ Does your child have an illness or medical condition that has contained for more than 3 months? Yes Illness No Is this visit related to the above illness? Yes No Briefly describe your expectations for today's visit with the doctor. APPENDIX D VERBAL CATEGORIES I. Phatic A. General---utterances that keep an interaction going, i.e., the "glue", e.g., How are you, Uh-huh, M-m-m-m, I see, Well..., Interesting, Okay. B. Humor, including laughter. C. Empathy---statements indicating an awareness and understanding of the emotions and feelings of another person, e.g., I understand (know) what you’re going through right now, I have felt that way, too. II. "Requests" for information A. Symptoms/medical---requests related specifically to the presenting problem, e.g., Tell me about your symptoms; How long have you been feverish?; So you've been feeling sick? B. Sociopersonal---requests related to nonmedical factors that may have an influence on the presenting problem, e.g., How are you doing in school now that you're receiving speech therapy; I don't understand why you aren't happy with medicaid; What else are you doing besides holding down one and a half jobs?; You look depressed so what's going on? C. Generalized health care---requests relating to general health habits and health maintenance, e.g., How often do you take your child for a general checkup; Are his immunizations up to date?; You are drinking plenty of fluids and getting eight hours of sleep, right? D. Treatment a. Treatment preferences---requests for patient's preference of various treatments, e.g., Would you prefer oral or intravenous form?; For your treatment, would your prefer chemotherapy, surgery, or both, which do prefer?; Have you had better luck with amoxicylin or meladrine? How effective was the medication? Do you understand why I want to see her in two weeks? b. Treatment possibilities---requests for patient factors that may limit treatment possibilities, e.g., Are you allergic to penicillin as that is usually what is prescribed; We recommend physical therapy three times a week and this can been done at home with an OT or you may come to the hospital. III. "Statements" conveying information/explanations A. Symptoms/Medical---statements related specifically to the presenting problem, e.g., articulation problems are common in children who have a history of ear infections; Using a cane or walker for several hours a day will lessen the chances of the 105 IV. 106 problem reoccurring; Studies have shown that for this particular condition that . Sociopersonal---statements related to nonmedical factors that may have an impact on the treatment, e.g., You'll need to consider working only halftime; A private school should be considered until he’s better; Marital therapy may relieve some of your anxiety; Because of his small size, he should not be enrolled in kindergarten until he's six Generalized health care---statements related to longterm or preventive health maintenance, e.g., Are you aware of the well-baby clinics on every fourth Saturday of the month; Studies have shown that... ; In my experience, overweight infants grow into overweight children and adolescents; Treatmento--statements conveying information regarding treatment, e.g., These are our options for treatment at this time, continue with the medication or minor outpatient surgery; We have this medication in oral or intravenous form. For her weight and age, two teaspoons of amoxycillin is appropriate, I'm going to need to do a throat culture since I suspect strep. Follow-up-—-statements related to medical instructions after the completion of this particular treatment regimen, e.g., I'm going to refer you to a specialist; Dr. Johns and Dr. Will are the pediatric surgeons to consider; A prosthetic will need to be considered depending on the results of the surgery. Evaluating Statements A. Medical---general statements of opinions/judgements specifically relating to the presenting problem or illness. 1. Approval---That's great, ; He looks really healthy now; Yes, that's a good idea; Of course; That's right. 2. Disapprova1---I told you not to . . .; How could you do that; it makes no sense. Sociopersonal---evaluative statements concerning nonmedical factors that may affect the presenting problems, e.g., I believe a private school would provide the best setting for Johnny right now; I really think the best thing for you to do is take a short medical leave; You really have not attended to the stressors in your life very well. Generalized health care---opinions/judgements regarding general health care, including preventive measures, e.g., I'm disappointed that you've not been attending the well-baby clinics on a regular basis; I'm so pleased you’ve been consistently keeping his inoculations current; How could you have let his general health deteriorate to this point without getting concerned? You must take care of your body—-it’s the only one you'll have. Treatment---opinions/judgements concerning compliance with the recommended course of treatment, e.g., You must not have VI. 107 followed the recommended course of treatment; You've done a fine job of following the recommended course of treatment. "Requests" for action A. Medical---requests for action regarding the presenting problem, including the related physical exam, e.g., Help me hold her arms so I can examine her ears; Lie still while I'm examining you; Try to get a urine sample from her; Sociopersonal---requests for action related to nonmedical considerations that relate indirectly to the present problem, e.g., Check with your school district's special education director to see what services are available while she's recuperating; Why don't you use the health department's free clinic for immunizations; Please consider marital counseling to reduce your stress level; Can you keep her home from school tomorrow. Generalized health care---requests for action concerning general health care and preventive measures, e.g., Bring him back in six months for his regular checkup; You will need to watch her development more closely and bring her in as soon as you suspect a delay in any of her skills; Make sure he eats a balance diet most of the time. Treatment---requests for action relating to the specific treatment regimen, e.g., call the lab tomorrow for the results of his tests; Try to get her to take capsules, if possible; Take two teaspoons three times a day; Stop at the pharmacy downstairs, if you can, to have this prescription filled; Start this medication immediately; If his fever goes above 101 in the next 24 hours, please bring him in again. Follow-up---requests for action related directly to medical care after the treatment regimen has been concluded, e.g., Can you bring her back in 10 days so I can recheck her ears; You know what the usual symptoms are now for this condition, call me anytime they reappear. "Requests" for questions A. Medical---requests for questions concerning the presenting problem and related symptoms, e.g., Do you have any questions about the diagnosis or the procedures used to make a diagnosis; Ask me about anything that we've just discussed that you don't understand; Would you like to ask anything? Sociopersonal---requests for questions relating to nonmedical factors, which may or may not impact the presenting problem, e.g., Don't be afraid to ask me about other stressors in your life that may be of concern to you; I know being unemployed is difficult for you and the family, so ask any questions that are on your mind, It looks like something is bothering you, do you want to ask me something. Generalized health care---requests for questions about general health care, including preventative measures, e.g., Do you have 108 any questions concerning the general care of the baby; Feel free to ask me or the nurse any questions regarding the necessity of regular checkups; You look puzzled, is there something about the immunizations you don't understand. D. Treatment---requests for questions regarding the recommended course of treatment, including possible alternatives and preferences, e.g., Do you have any questions about why I chose ampecylin over mylantin; Feel free to ask me anything related to the treatment regimen, surgery, etc.; I'm sure you must have somequestions because the directions are rather complicated. VII. Medical Jargon---any word, phrase, or statement containing specialized language not readily understood by nonmedical individuals. VIII. Miscellaneous/Unintelligible 109 VERBA CA EGOR S Video tape #: .____ Code #: Segment: 1 2 3 Sat or Dissat Counter #: to Instructions: Put the number of the physician's statement in the box corresponding to the category chosen. More than one category may be used to categorize the function of the statement. L___P_b.atis A. General B. Humor C. Empathic II " e uests" for ormat'o Current symptoms/problem Sociopersonal Generalized health care Treatment 1. Possibilities 2. Preferences tzcuut> III "Statements" conve in nformation A. Current symptoms/problem B. Sociopersonal C. Generalized health care D. Treatment E. Follow-up IVI Evaluating statements A. Current symptoms/problem 1. Approval 2. Disapproval B. Sociopersonal C. Generalized health care D. Treatment V "Re ue ts" or act 0 A. Current symptoms/problem B. Sociopersonal C. Generalized health care D. Treatment E. Follow-up VI "Re uests" o uestio s A. Current symptoms/problem B. Sociopersonal C. Generalized health care D. Treatment E. Follow-up VII, Medical Jargon VII M 5 el aneous n'ntelli ible APPENDIX E NONVERBAL BEHAVIOR CATEGORIES Total Time Tape # Code: Segment: 1 2 3 Sat Dissat Counter: to SGCS l. Bgdy Qrigntntign Inward Parenr forward 20 toward parent upright backward 20 away from parent could not tell 2. Trunk Anglg directly facing--0 turned away from parent--4S perpendicular to parent--90 could not tell 3. A a 0 io symmetrical open arms w/o hands touching symmetrical closed arms w/ hands touching could not tell 4. Leg EogItIon crossed uncrossed could not tell 5. Hea Movement nodding shaking could not tell 6. EaciaI Expression smiling could not tell listening writing 110 APPENDIX F PARENT REACTION QUESTIONNAIRE Instructions: He want to know how you feel about today's visit with your child‘s doctor. Please answer carefully and honestly. answers to these items. There are no “right“ or “wrong" Each of the items contains a statement describing your visit with the doctor. Example: If you very strongly disagree, mark '1'. Your personal answers will not be shown to the doctor. Very Strongly Disagree g—a The doctor was neatly dressed ...... Strongly Disagree N Disagree U Unsure .5 Agree Strongly Agree 05 ten Strongly Agree \1 If you very strongly agree, mark "7“. If your opinion is somewhere in between, circle the number under the words which reflect your opinion. Please try to answer all of the items. The doctor listened very carefully to What I said OOOOOOOOOOOOOOOOOOOOOO00...... The doctor seemed to think about my child's problem carefully ................ The doctor seemed to feel it was important for my child to understand the ViSit 0.00000000000...OOOOOOOOOOOOO... The doctor did not really give me a chance to say what was on my mind ........ The doctor used words and ideas too difficult for my child to understand ..... After talking with the doctor, I feel that I am handling my child's illness well] .OCCCOOOOOO...OOOOOOOOOOOOOOOOOCOOOOO 111 Very Strongly Disagree Strongly Disagree Disagree Unsure Agree Strongly Agree Very Strongly Agree 10. ll. 12. 13. 14. IS. 16. 17. 18. 19. 20. 21. 112 The doctor seemed to care about my child's feelings ......................... The doctor talked to my child about what (s)he can do to become more healthy ...00.0.0000.........OOOOOOOOOOOOO The doctor encouraged my child to talk ... The doctor failed to understand my main reason for cming ......OOOOOOOOOOOOO The doctor listened closely to my Chi‘d 0.00.0000.........OOOOOOOOOOCCOO.... The doctor made me feel I've done a good job of caring for my child .......... The doctor excluded my child from most of the discussion ................... The doctor knows how to talk to Children ....O0.00.00.00.00.......OOOOOOOO My child could not understand most of what the doctor said ....3............. After talking with the doctor, I feel much better about my child's prOD'ems ....000.000.00.000...00.0.0000... The doctor seemed to have other things on his m‘nd .....OOOOOOOOOOOOOOO... The doctor really understood how my Child feels O................OOOOOOOOOOOOO The doctor seemed to know just what to do for my child‘s problem ............. The doctor gave me a poor explanation Of my Child's il‘ness ......OOOOCOOOOOOOOO The doctor explained things very well to my Chi‘d ......OOOOOOOOOOOOOOOCOO...... Very Strongly Disagree Strongly Disagree N Disagree u Unsure 0 Agree Strongly Agree Oi rm Strongly Agree 22. 123. 24. 25. 26. 113 I really felt understood by my Child's dOCtor ......COCOOCOOCCOOOOO ..... 0 It may be too difficult for us to do exactly what the doctor told us to do .... I intend to follow the doctor's instructions O.......OOOOOOOOOOOOOOOOO.... I expect that it will be easy for me to follow the doctor's advice ............ It will be too much trouble to follow the doctor's adv‘ce .....OOOOOOOOOOOOOOOOO very Strongly Disagree 1 Strongly Disagree N Disagree U Unsure Agree Strongly 397‘9 Very Strongly Agree APPENDIX C Demographic Data e d i d NP-Miss Parent's Age .1177 p - .250 Parent's Income .0767 p - .331 Parent's Education -.l394 p - .212 Family Size .0924 p - .299 Child's Sex .0608 p - .364 Child's Age .1466 p - .200 Ailment .1027 p - .279 Length of Visit -.3581 p - .017 2993.: Years of Experience -.0108 P - .475 114 LIST OF REFERENCES (1977). American Academy of Pediatrics Committee on Standards of Child Health: Preventive Child Health Care. Sta dards 0 Ch' d Healtn Care, (3rd ed.), Evanston, IL: American Academy of Pediatrics. Anastasi, A. (1982). Ps cholo ca sti . New York: Macmillan Co. Archer, D., & Akert, R. M. (1977). Words and everything else: Verbal and nonverbal cues in social interpretation. JournaI of Personnlity and Sngini PsychoIogy, g2, 443-449. Ausburn, L. (1981). Patient compliance with medical regimens. In J. Sheppard (Ed.), Behaviorni medigine. Lidcome, NSW: Cumberland College of Health Sciences. Bales, R. F. (1950). nte ct o c s a a 5' A me hod for the firnny_nf_§ngli_grgnn§. Cambridge, MA: Addison-Wesley. Bartless, M. H. Johnston, A., & Meyer, T. C. (1973). Dial access library -- patient information service: an experiment in health education. New EngIand Journal of Medicine, 288, 994-998. Bartlett, E. E., Grayson, M., Barker, R., Levine, D. M., Golden, A., & Libber, S. (1984). The effects of physician communication skills on patient satisfaction, recall, and adherence. Journal of Chronic Qiggasgs, 31, 755-764. Ben-Sira, Z. (1980). Affective and instrumental components in the physician-patient relationship: An additional dimension of interaction theory. J u a 0 He 1t d oc 1 ea v 0 170-180. Bloom, 8. W. (1963). The docto and s a ent' A soc'olo ical inrernrgrarign. New York: The Free Press. Blum, R. H. (1957). Inn ngygnoiogy 9f naInractice suits. San Francisco: California Medical Association. Bradshaw, P. W., Ley, P., Kincey, J. A., & Bradshaw, J. (1975). Recall of medical advice: Comprehensibility and specificity. 'tis Jgnrnni of §ogiaI QIinicai Psygboiggy, IQ, 55-62. 115 116 Breslau, N. (1982). Continuity reexamined: Differential impact on satisfaction with medical care for disabled and normal children. Medical Care, 22(4), 347-360. Breslau, N., & Mortimer, E. (1981). Seeing the same doctor: Determinants of satisfaction with specialty care for disabled children. flggipnl_gnrg, 12, 741-758. Bryk, A., Raudenbush, S., Seltzer, & Congdon, Z. (1976). An ntroductio to ' Com ute ro r user's uide. University of Chicago. Byrne, P. S., & Heath, C. C. (1980). Practitioners' use of non-verbal behaviours in real consultations. Jour a o 0 al C e e General Practioners 30 327-331. Byrne, P. S., & Long, B. E. (1976). Doctors talking to patients. London: H. M.S.0. Campbell, J. (1978). The child in the sick role: Contributions of age, sex, parental status, and parental values. Jou na of ea th and SociaI fiehavior, I2, 35-41. Campbell, J. D. (1975). Illness is a point of view: The development of children's concepts of illness. gnild QeveIppmenr 46 92-97. Carkhuff, R. R. (1969). Helping and human relations (Vols. I & II). New York: Holt, Rinehart, & Winston. Carroll, J. G., & Monroe, J. (1979). Teaching medical interviewing: a critique of educational research and practice. Journa ed'cal Education, 24, 498-500. Carter, W. B., Inui, T. S., Kukull, W. A., & Haigh V. H. (1982). Outcome-based doctor-patient interaction analysis. MedicaI Care, 29 (6), 550-566. Charney, E. (1972). Patient-doctor communication: Implications for the clinician. Pediatric Clinics in North America, I2, 263-279. Cobb, B. (1954). Why do people detour to quacks? Ine Psychiatric Bulletin, 3, 66-69. Darwin, C. (1872). The expression of the enotiong in man and animals. London: John Murray. [More currently published, Chicago: University of Chicago Press, (1965)]. Deisher, R. W., Engel, W. L., Spielholz, R., & Standfast, S. J. (1965). Mothers' opinions on their pediatric care. Pediatrics, 22, 82-90. 117 DiMatteo, M. R. (1979). A social-psychological analysis of physician-patient rapport: Toward a science of the art of medicine. Journai of SociaI Issues, 22, 12-33. DiMatteo, M. R. & Hays, R. (1980). The significance of patients' perceptions on physician conduct. Journni pf Communiry Healrh, lé. 18-34. DiMatteo, M. R., Prince, L. M., & Taranta, A. (1979). Patients' perception of physicians' behavior: Determinants of patient commitment to the therapeutic relationship. lpprnnl_pfi_gpnnnniry fleaitn, 5(4), 280-290. Eisenberg, L. (1977). The search for care. angninp, 122, 325-246. Ekman, P., & Friesen, W. V. (1981). The repertoire of nonverbal behavior: Categories, origins, usage, and coding. In T.A. Sebeok J. Umiker-Sebeok, & A. Kendon (Eds)., Nonverbai communication ang gesture (pp. 57-106). The Hague: Mouton Publishers. Ekman, P., & Friesen, W. V. (1974). Detecting deception from the face or body. Journal pf Perspnality and §ppini Psychology, 22, 288-98. Ekman, P., & Friesen, W. V. (1969). Nonverbal leakage and cues to deception. Psychiatry, 22, 88-106. Engler, C. M., Saltzman, G. A., Walker, M. L., & Wolf, F. M. (1981). Medical student acquisition and retention of communication and interviewing skills. Journal of Medical Education, 22, 572-79. Erickson, F. (1985). Qualitative methods in research on teaching. In M. C. Wittrock's (Ed.) finndbook of research on teaching, (pp. 119-161) (3rd ed.), New York: Macmillan. Erickson, F., & Shultz, J. (1982). e counse as atekee er' Social interaction in intgrviews. New York: Academic Press. Ford, A. B., Liske, R. E., Ort, R. S., & Denton, J. C. (1967). In; doc 's ers ective' cia v ew t e tie t a d ctice. Cleveland, OH: Case Western Reserve University Press. Francis, V., Korsch, B. M., & Morris, M. J. (1969). Gaps in doctor-patient communication: Patients' response to medical advice. New England JournaI pf Medigine, 229, 535-540. Frankel, R. M. (1984). From sentence to 'sequence': Understanding the medical encounter through microinteractional analysis. Qiscourse Processes, 1, 135-170. Freemon, B., Negrete, V. F., Davis, M., & Korsch, B. M. (1971). Gaps in doctor-patient communications: Doctor-patient interaction analysis. Eediatric Research, 2, 298-311. 118 Freidson, E. (1970). o e s on o med cine' tud t e soc 010 of nppiign_knpglgggg. New York: Harper & Row. Fretz, B.R., Corn, R., Tuemmler, J.M., & Bellet, W. (1979). Counselor nonverbal behaviors and client evaluations. ourna 0 use n Melon. ZQ. 304-311. Freud, S. (1905). Fragments of an analysis of a case of hysteria (1905). QoIIgcted papers, 2, New York: Basic Books. Friedman, H. S., & DiMatteo, M. R. (Eds.) (1982). Inter on Issues in Henirn Care. London: Academic Press. Harrigan, J. A., & Rosenthal, R. (1986). Nonverbal aspects of empathy and rapport in physician-patient interaction. In P. D. Blanck, R. Buck, & R. Rosenthal (Eds.), No verba communicat'on the clinicaI context (pp. 36-73). University Park: The Pennsylvania State University Press. Helmstadter, G. C. (1964). Principles of psychologipal measurement. New York: Appleton-Century-Crofts. Hippocrates, (1923). o e ' dec a d t e h si an. (W. H. S. Jones, Trans.). London: William Heinemann, Ltd. Hornsby, J. L. (1983). Strategies for training physicians in effective interpersonal communication skills. In S. M. Thornton & W. K. Frankenburg (Eds.), gnild heaitn pare communications; enhancing interaction Among Professionals, parents, and children (pp. 131-140). USA: Johnson & Johnson Baby Products Company. Houston, W. R. (1938). The doctor himself as therapuetic agent. Annals o nte a ed'cine, ii, 1416-25. Hulka, B. S., Cassel, J. C., Kupper, L. L., & Burdette, J. A. (1976). Communication, compliance, and concordance between physicians and patients with prescribed medications. e on ub ic Heairn, 22, 847-853. Jones, J. G. (1983). Compliance with pediatric therapy. Clinical Endinrrips, 22(4), 262-265. Jones, R. A. (1982). Expectations and Illness. In H. S. Friedman & M. R. DiMatteo (Eds.), Inrerpersonal Issues in Hgnlth Care (pp. 145-167). New York: Academic Press. Kahn, G. S., Cohen, B., & Jason, H. (1979). The teaching of interpersonal skills in U.S. medical schools. Journal of Medical Edncation, 22,29, 29-35. 119 Kasteler, J., Kane, R. L., Olsen, D. M., & Thetford, C. (1976). Issues underlying prevalence of "doctor-shopping" behavior. Journal of fisslrh sng §ocial Behavior, I1, 328-339. Kendon, A. (1982). The organization of behavior in face-to-face interaction: Observations on the development of a methodology. In K. R. Scherer & P. Ekman (Eds.), flangpppk pf methogs in nonverbal bsnsyipr_rsssnrpn, (pp. 440-505), Cambridge: Cambridge University Press. Kendon, A. (1981). Introduction: Current issues in the study of "nonverbal communication". In T. A. Sebeok, J. Umiker-Sebeok, & A. Kendon (Eds.). MW (pp- 1-53). The Hague: Mouton Publishers. Kendon, A. (1979). Some methodological and theoretical aspects of the use of film in the study of social interaction. In G. P. Ginsburg's (Ed.), Emerging strategies in spsial psypnplpgical ressarch. New York: Wiley. Koos, E. (1955). "Metropolis" - what city people think of their medical service. Anerican Journal pf Enplis flsalrn, gs, 1551-1557. Korsch, B. M., & Aley, E. F. (1973). Pediatric interviewing technics. In Morris Green (Ed.) WW2 nsg_psgisrriss. (pp. 3-43) Lexington, MA: Lexington Books. Korsch, B. M., Gozzi, E. K., & Francis, V. (1968). Gaps in doctor-patient communication. Esgisrrips,,g2, 885-871. Korsch, B. M., & Negrete, V. F. (1972). Doctor-patient communication. W. 2.21. 66-74- Krauss, R. M., Apple, W., Morency, N., Wenzel, C., & Winton, W. (1981). Verbal, vocal and visible factors in judgments of another's affects. Jpnrnsl pf Eersonality ang §psial Esyshplpgy, 32, 312-319. Larsen, K. M., & Smith, C. K. (1981). Assessment of nonverbal communication in the patient-physician interview. Ine Journal of Eamily Eractice, I2, 481-488. Lewis, G. S., Scott, D. E., Pantell, R. H., & Wolf, M. H. (1986). Parent satisfaction with children's medical care: Development, field test, and validation of a questionnaire, Msnipnl_§nrs, 22(3), 209-215. Ley, P. (1982). Satisfaction, compliance, and communication. British MW. 21. 241-254. Ley, P. (1979). The psychology of compliance. In D. J. Oborne, M. M. Gruneberg, & J. R. Eiser (Eds.), Researpn in psycholpgy and nsgipins, (Vol. 2). London: Academic Press. 120 Ley, P. (1977). Psychological studies of doctor-patient communication. In S. Rackman's (Ed.), Contr but 0 t d a1 5 c 0 (Vol. 1). Oxford: Pergamon Press. Ley, P., Goldman, M., Bradshaw, P. W., Kincey, J. A., & Walker, C. (1972). The comprehensibility of some x-ray leaflets. 0 na of Instrncripn fpr flsalrh Eduparipn, IQ, 47-53. Ley, P., & Spellman, M. S. (1967). Connunicating with the patient. London: Staples Press. Linder-Pelz, S. (1982). Toward a theory of patient satisfaction. Social Scienpe and Medicine, IQ, 577-582. Linn, L. S., & DiMatteo, M. R. (1983). Humor and other communication preferences in physician-patient encounters. Medical Care, 21 (12), 1223-1231. Linn, L. S., DiMatteo, M. R., Chang, B. L., & Cope, D. W. (1984). Consumer clues and subsequent satisfaction ratings of physician behavior. Medical Care, 22(9), 804-812. Liptak, G. S., Hulka, B. S., & Cassel, J. C. (1977). Effectiveness of well child care during infancy. Pediarriss 60 186-189. Locker, D., & Dunt, D. (1978). Theoretical and methodological issues in sociological studies of consumer satisfaction with medical care. Sosial §piencs snd Medicine, I2, 283-292. Lyons, J. (1977). Semantics (Vol I). Cambridge: Cambridge University Press. McKinley, J. B. (1975). Who is really ignorant--physician or patient? gpurnal of flealrn sng Social fishavior, IQ, 3-ll. McGuown, N. A. (Ed.) (1971). e atura t 0 an inte iew. Microfilm Collection of Manuscripts on Cultural Anthropology, 15th Series, University of Chicago, Joseph Regenstein Library, Department of Photoduplication. Mehrabian, A. (1972). anverpal pommnnipsrion. Chicago: Aldine-Atherton. Nagera, H. (1978). Children and hospitalization. Chi d s chiat and W. 2. 3-6. National Center for Health Statistics, (1979). Physician visits: Volume and interval since last visit, U. S., 1975. Vital and health statisriss, Series 10, #128 DHEW Publication No. (PHS) 79-1556. Washington, D.C.: Department of Health, Education and Welfare. 121 Pantell, R. H., Stewart, T. J., Dias, J. K., Wells, P., & Ross, W. R. (1982). Physician communication with children and parents. Eediatrics, 12, 396-402. Parsons, T. (1975). The sick role and the role of the physician reconsidered. MillbanM Memorial Enng Qusrrerly, 22, 257-77. Parsons, T. (1951a). The soc'a s e . New York: The Free Press Parsons, T. (1951b). Illness and the role of the physicians. qurnal of Qrthppsycniatry, 21, 452-459. Pendleton, D. (1983). Doctor-patient communication: A review. In D. Pendleton & J. Hasler (Eds.), Doctor-Patient Communication (pp.5-56). London: Academic Press. Pendleton, D. & Hasler, J. (Eds.) (1983). Doctor-Patient Communication. London: Academic Press. Perrin E., & Gerrity, S. (1981). There's a demon in your belly: Children's understanding of illness. Eediatrics, 21, 841-844. Putnam, S. M., Stiles, W. B., Jacob, M. C., & James, S. A. (1985). Patient exposition and physician explanation in initial medical interviews and outcomes of clinic visits. Medical Care, 22(1), 74-83. Quirk, M., & Babineau, R. A. (1982). Teaching interviewing skills to students in the clinical years: A comparative analysis of three strategies. Journsl pi Medical Edusation, 21, 939-41. Raudenbush, S., & Bryk, A. S. (1986). A hierarchical model for studying school effects. §opiology pi Eduparipn, 22, 1-17. Reber, A. S. (1985). Qictionsry of psyshology. New York: Viking Penguin, Inc. Reisinger, K. S., & Bires, J. A. (1980). Anticipatory guidance in pediatric practice. Eedistriss, 22, 889-892. Roter, D. L. (1983). Physician/patient communication: Transmission of information and patient effects. Msrylsnd State Medical Journal, 22(4), 260-265. Roter, D. L. (1977). Patient participation in the patient-provider interaction: the effects of patient question-asking on the quality of interaction, satisfaction, and compliance. Hsaltn Educarion HM. é. 281-315. Roghmann, K. J., Hengst, A., & Zastowny, T. R. (1979). Satisfaction with medical care: Its measurement and relation to utilization. Medical Care, 11(5), 461-477. 122 Ruesch, J., & Bateson, G. (1951). Commu a ' e oc a m o spciery. New York: Norton. Seay, T.A., & Altekruse, M.K. (1979). Verbal and nonverbal behavior in judgments of facilitative conditions. ourna 0 Co n W. 22. 108-119. Scott, D. E. (1985). Pediatrician Interpersonal Skills and Parent Satisfaction with the Medical Interview. Doctoral dissertation submitted to the California School of Professional Psychology. Berkeley. Scherer, K. R., & Ekman, P. (1982). Methodological issues in studying nonverbal behavior. In K. R. Scherer & P. Ekman (Eds.), Handbook pf nerhods in nonverbal researsn (pp. 1-44). Cambridge: Cambridge University Press. Schwartz, C. G. (1970). Strategies and tactics of mothers of mentally retarded children for dealing with the medical care system. In N. Bernstein (Ed.), Diminished people (pp. 73-105). Boston: Little and Brown. Seay, T.A., & Altekruse, M.K. (1979). Verbal and nonverbal behavior in judgments of facilitative conditions. 0 of u n Esysnology, 22, 108-119. Shapiro, A. K. (1960). A contribution to a history of the placebo effect. Behavioral Science, 2, 109-35. Shortell, S. M. (1976). Continuity of medical care: conceptualization and measurement. Medical Care, I2, 377-383. Simeonson, R., Buckley, L., & Monson L. (1979). Conceptions of illness causality in hospitalized children. Journa 0 Pediatric Psycnology, 2, 77-80. Stewart, T. J., Pantell, R. H., Dias, J. K., Wells, P. A., & Ross, A. W. (1981). Children as patients: A communication processs study in family practice. Ine Journal of Eamily Eractise, I2, 827-835. Stiles, W. B., Putnam, S. M., Wolf, M. H., James, S. A. (1979). Interaction exchange structure and patient satisfaction with medical interviews. Medical gare, I1, 667-681. Strecher, V. J. (1982). Improving physician-patient interactions: A review. Eatienr Counseling and Health Educarion, 2(3), 129-136. Szasz, T., & Hollender, M. (1956). A contribution to the philosophy of medicine: The basic models of the doctor-patient relationship. Journal of the American Medical Association, 21, 585-588. 123 Vaccarino, J. M. (1977). Malpractice: The problem in perspective. Ins WINE. 2__38. 861-63. Vaughn, G. F. (1957). Children in hospitals. Lancet, I, 1117-1128. Waitzkin, H. (1984). Doctor-patient communication. Jou a American Medical Association, 222, 2441-2446. Waitzkin, H., Stoeckle, J. D., Beller, E., & Mons, C. (1978). The informative process in medical care. Instrnctional Science, 1, 385-419. Ware, J. E., Davies-Avery, A., & Stewart, A. L. (1978). The measurement and meaning of patient satisfaction. flsslrh_sn2_Msgipsl_§srs §eryices Review, I, 3-15. Ware, J. E., Snyder, M. K., & Wright, W. R. (1976). eve me nd va idat'on 0 sea es to measur at ent sa sfaction with health care services; Volume I of a final report. Eart B; Results regarding scales constructed form the Patein; Satisfaction uestionna e and measu e f other ea t c e erce ion . Springfield, VA: National Technical Information Service. (NTIS No. PB 288-330). Weinberger, M., Green, J. Y., & Mamlin, J. J. (1981). The impact of clinical encounter events on patient and physician satisfaction. §ocisl Science and Msdicine, 12, 239-44. West, C. (1984). ou e com lica ' ' roubles with talk etween doctors and patienrs. Bloomington: Indiana University Press. Wexler, M. (1976). The behavioral sciences in medical education A view from psychology. American Psychologisr, 21, 275-283. Wolf, M. H., Putnam, S. M., James, S. A., & Stiles, W. B. (1978). The medical interview satisfaction scale: Development of a scale to measure patient perceptions of physician behavior. Journal of WW. 1. 391-401. Young, D. W. (1980). Meanings of counselor nonverbal gestures: Fixed or interpretive? Journal of Connssling Esypnplpgy, 21, 447-52. Young, P. C., Wasserman, R. C., McAullife, T., Long, J., Hagan, J. F., & Heath, B. (1985). Why families change pediatricians. American Jonrnal of Disableg gnildren, I22, 683-686. Zastowny, T. R., Roghmann, K. J., Hengst, A. (1983). Satisfaction with medical care: Replications and theoretic reevaluation. Medical Care, 21(3), 294-322. 124 Zuckerman, M., DePaulo, B. M., & Rosenthal, R. (1986). Humans as deceivers and lie detectors. In P. D. Blanck, R. Buck, & R. Rosenthal's (Eds.), Nonvsrbsl communication in the slinical context (pp. 13-35). University Park: The Pennsylvania State University Press. MICHIGAN STATE UNIV. LIBRnRIEs lllllHullW"llHI”WIllllWWIIHINIWNIWW 31293007678893