AN EVALUATION OF THE EFFECTS OF MEDIATIONAL INTERVENTION FOR SENSITIZING CAREGIVERS (MISC) AND A HEALTH AND NUTRITION EDUCATION PROGRAM ON THE SUSTAINED ATTENTION OF UGANDAN CHILDREN WITH HIV By Kayla Ann Musielak-Hanold A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements  for the degree of  School Psychology - Doctor of Philosophy 2016!ABSTRACT AN EVALUATION OF THE EFFECTS OF MEDIATIONAL INTERVENTION FOR SENSITIZING CAREGIVERS (MISC) AND A HEALTH AND NUTRITION EDUCATION PROGRAM ON THE SUSTAINED ATTENTION OF UGANDAN CHILDREN WITH HIV By Kayla Ann Musielak-Hanold Evidence suggests that children with the human immunodeficiency virus (HIV) may experience difficulties with sustained attention, an important skill for academic success and daily functioning. While the exact cause of this difficulty is unknown, it likely arises from a combination of factors. Research suggests that effects of the virus, compromised child-caregiver interactions, and challenges surrounding food and basic healthcare may all influence the neurocognitive development of children with HIV and possibly result in difficulties with sustained attention. The present study aimed to evaluate the effects of an intervention to enhance caregiver-child interactions, the Mediational Intervention for Sensitizing Caregivers (MISC), and an intervention to improve knowledge about and access to healthcare and nutrition on the sustained attention skills of young Ugandan children with HIV.  Semi-extant data for 111 caregiver-child dyads were used in the present study. These dyads were grouped by geographic region of residence and then groups were randomly assigned to one of two interventions. One group received a health and nutrition education intervention and the other received MISC. Both groups received nutritional supplements in the form of monthly food packages. ChildrenÕs sustained attention was measured at baseline, six months into the intervention, and at the conclusion of the yearlong intervention using the Early Childhood Vigilance Task (ECVT). Mixed-effects modeling was used to examine the effects of the two interventions while controlling for childrenÕs age, childrenÕs gender, childrenÕs viral load, childrenÕs HAART status, childrenÕs height and weight, childrenÕs exposure to video screens, use of a translator during the intervention, caregiver depression, and caregiver functioning. Mixed-effects modeling was also used to examine how the length of the interventions influenced sustained attention skills. Finally, a path analysis was used to evaluate how caregiver depression, caregiver functioning, and child viral load were related to childrenÕs sustained attention before intervention. The results indicated that both intervention groups made significant gains in sustained attention across the yearlong intervention. However, there was no significant difference between the gains made by the two groups. The gains made in the second six months of the intervention were significantly different than the gains made in the first six months of the intervention, such that significant gains were made during the second half of the intervention and not during the first half. This finding did not differ between groups. Finally, the proposed conceptual model that was tested to examine the relation of child and caregiver variables to childrenÕs sustained attention skills at baseline was found to have good overall fit. Evaluation of the specific paths in the model indicated that the proposed pathways do not explain baseline performance in childrenÕs sustained attention skills. Overall, this study highlights the potential benefits of interventions addressing health, nutrition, and caregiver-child interactions for young children with HIV. This study also provides a foundation for future research to examine additional questions surrounding interventions to promote sustained attention growth in children with HIV and to better understand the factors that contribute to sustained attention skills in children with HIV.!ACKNOWLEDGMENTS  I must express my gratitude to the individuals who helped to make not only this dissertation, but also my development as a psychological practitioner and researcher, possible. I want to thank Dr. Jodene Fine for the countless ways she has facilitated my development. You have never hesitated to share clinical and research opportunities with me, which has helped me to both broaden and deepen my interests and skills. You have pushed me to edges that I never would have approached on my own, and I have been successful because you have provided continual support and guidance. I also need to thank Dr. Michael Boivin for welcoming me onto his research project without hesitation and for generously sharing his data with me. I would not have been able to complete a dissertation project of this scope without the opportunity to join your work. To Dr. John Carlson, your guidance during the development of this dissertation was invaluable and deeply enhanced the quality of my work. I am thankful for your enthusiasm about and confidence in my work. I am also immeasurably grateful for MohammedÕs calm and patient support and guidance on my dissertation.   I am appreciative for the support that I have received in other domains of my life that allowed me to be successful in graduate school. Mom, thanks for reading to me every day when I was young, even when I asked you to read the same books over and over. Dad, thanks for playing games like Math Blaster and Memory with me, and doing things like building cities out of paper with me. It is undeniable that these experiences contributed to my early language, visual-spatial, memory, and problem-solving skills, and are the foundation of all that I have !ivaccomplished academically. Staci, thanks for the laughs and for commiserating with me. To Patti and Lee, thank you for always expressing your enthusiasm for my work.   Jess, what could I possibly say to describe how much I appreciate you and how much you have truly contributed to my graduate school career? Thank you for your consistent love, support, encouragement, openness, and genuine interest in my work and goals. I also need to thank you for kindly tolerating the irrational fears and odd behaviors that graduate school often brought out of me. I am so excited for what is coming next for us. To Henri and Oxford, thank you for being by my side as I wrote each page of this document.   !!vTABLE OF CONTENTS LIST OF TABLES ix                                                                                                                           LIST OF FIGURES x                                                                                                                          CHAPTER 1: Introduction 1                                                                                                                The Current Study  4                                                                                                                            CHAPTER 2: Literature Review 6                                                                                                       Introduction to Human Immunodeficiency Virus (HIV) 6                                                                    Transmission 9                                                                                                                                       Treatment 9                                                                                                                                           Sustained Attention Problems of Children with HIV 11                                                                        Importance of sustained attention 13                                                                                                     Defining sustained attention 14                                                                                                             Measuring sustained attention 14                                                                                                           Interventions to improve sustained attention 16                                                                                   Factors Affecting ChildrenÕs Neurocognitive Development 17                                                            Issues Affecting Caregivers with HIV 20                                                                                             Purpose of the Present Study 23                                                                                                           Research Questions and Hypotheses 24                                                                                                Research Question 1 24                                                                                                                          Hypothesis 1a 24                                                                                                                                     Hypothesis 1b 24                                                                                                                                     Rationale 24                                                                                                                                           Research Question 2 25                                                                                                                          Hypothesis 2a 25                                                                                                                                     Hypothesis 2b 26                                                                                                                                     Rationale 26                                                                                                                                           Research Question 3 26                                                                                                                          Hypothesis 3 26                                                                                                                                       Rationale 27                                                                                                                                         CHAPTER 3: Method 30                                                                                                                      Present Study Design 30                                                                                                                       Uganda 30                                                                                                                                             Interventions 31                                                                                                                                     Nutrition and health education intervention 32                                                                                      TAU protocol 38                                                                                                                                    Mediational Intervention for Sensitizing Caregivers (MISC) 38                                                        !vi   MISC protocol 45                                                                                                                                  Intervention fidelity 45                                                                                                                         Participants 51                                                                                                                                       Variables and Measures 60                                                                                                                     Early Childhood Vigilance Task (ECVT) 60                                                                                          ECVT coding and analysis 63                                                                                                                ChildrenÕs age 64                                                                                                                                   ChildrenÕs gender 64                                                                                                                              ChildrenÕs HAART status 65                                                                                                                 ChildrenÕs HIV viral load 65                                                                                                                  ChildrenÕs height and weight 67                                                                                                            Caregiver depression 67                                                                                                                         Caregiver functioning 69                                                                                                                      Analyses 71                                                                                                                                            Preliminary analyses 72                                                                                                                         Variable transformations 72                                                                                                                   Mixed-effects modeling 73                                                                                                                    Path analysis 74                                                                                                                                      Missing data 75                                                                                                                                     Project Approval 76                                                                                                                             CHAPTER 4: Results 80                                                                                                                      Results of ECVT Coding 80                                                                                                                 Correlations Among Variables 81                                                                                                         Research Question 1: Do nutrition and health education and caregiver-child               interaction interventions improve sustained attention in children with HIV? 84                                Research Question 2: How does the length of the nutrition and health education               and caregiver-child interaction interventions affect changes in sustained attention               in children with HIV?  88                                                                                                                     Research Question 3: How are caregiver depression, caregiver functioning,               and child viral load related to children's sustained attention?  89                                                       CHAPTER 5: Discussion 93                                                                                                                 Effects of MISC and TAU 93                                                                                                                 Sustained attention growth 93                                                                                                                Similar improvements of MISC and TAU 96                                                                                       Effect of Time 100                                                                                                                                Factors Influencing Sustained Attention in Children with HIV 103                                                      Measurement 104                                                                                                                                   Functioning and caregiving 105                                                                                                             Method of data collection 105                                                                                                              Limitations and Future Research 107                                                                                                    Control group 108                                                                                                                                !vii  Unknown psychometric properties 108                                                                                                 Intervention fidelity 109                                                                                                                        Longitudinal data related to HIV and sustained attention 111                                                              Clinical Implications 112                                                                                                                      Conclusion 112                                                                                                                                    APPENDIX 114                                                                                                                                   REFERENCES 119!                                                                                                                             !viiiLIST OF TABLES Table 1 Topics and Content of TAU 34                                                                                                Table 2 Definitions and Examples of Mediational Techniques in MISC 42                                       Table 3 Descriptive Statistics of MISC Lessons 49                                                                             Table 4 Descriptive Statistics of TAU Lessons 50                                                                               Table 5 Demographic Information 58                                                                                                  Table 6 Descriptive Statistics of Covariates 71                                                                                   Table 7 Summary of Research Questions, Variables, and Analyses 78                                               Table 8 Descriptive Statistics of ECVT Scores 81                                                                              Table 9 Correlations Among Predictor Variables and ECVT Scores 83                                              Table 10 Results of Mixed-Effects Model 86                                                                                      Table 11 Results of Path Analysis 90                                                                                                   !ixLIST OF FIGURES Figure 1 Conceptual Model 29                                                                                                            Figure 2 Countries of Africa 52                                                                                                           Figure 3 Districts of Uganda 53                                                                                                           Figure 4 Participant Flow Chart 55                                                                                                      Figure 5 Viral Load Distribution 66                                                                                                     Figure 6 ECVT Scores Across the Intervention Period 85                                                                  Figure 7 Predicted ECVT Scores 87                                                                                                    Figure 8 Estimates of Paths in Conceptual Model 92                                                                          !xCHAPTER 1: Introduction  The human immunodeficiency virus (HIV) remains a global epidemic, as approximately 35 million people around the world are currently infected with HIV (World Health Organization, 2014a). Approximately 3.2 million of these people are children under the age of 15 (World Health Organization, 2014b).  Decades ago, children with HIV would have been unlikely to survive. With more recent advances in antiretroviral medication, these children may now have the opportunity to survive well into adulthood (Hazra, Siberry, & Mofenson, 2010). However, children with HIV may also face a lifetime of medical, psychosocial, and neurocognitive challenges, such as opportunistic infections including tuberculosis and pneumonia (U.S. Department of Health and Human Services, 2010), stigma and discrimination (Bogart et al., 2008), and motor difficulties (Abubakar,Van Baar, Van de Vijver, Holding, & Newton, 2008), respectively.    Important findings regarding children with HIV is that they may struggle with sustained attention and they are at risk for developing severe difficulties because attention may be the most vulnerable neurocognitive domain for adults with HIV (Hardy, 2009). Sustained attention is the ability to focus on a stimulus for a period of time (Anderson, 2008) and is thought to be an important skill for acquiring and demonstrating academic skills (e.g., Campbell, DÕAmato, Raggio, & Stephens, 1991) and learning and performing everyday tasks (Sarter, Givens, & Bruno, 2001). Children with HIV have been found to perform more poorly on direct measures of sustained attention than children without HIV (Ruel et al., 2012; Watkins et al., 2000). Given that children with HIV may have difficulties with sustained attention, may be at risk for developing !1severe attention problems, and that sustained attention seems critical for daily and academic functioning, this vulnerable population may benefit from interventions targeting this skill.   The exact cause of sustained attention difficulties in children with HIV remains unknown, but may be related to neurological consequences of HIV. HIV has been found to have a direct influence on the central nervous system. The virus can cause cell death and damage to specific components of various brain cells (Ellis, Calero, & Stockin, 2009). Researchers have found concentrations of HIV in areas of the brain that are associated with attention skills (Klarborg et al., 2013; Kumar, Borodowsky, Fernandez, Gonzalez, & Kumar, 2007; Sarter et al., 2001; Wiley, Schrier, Nelson, Lampert, & Oldstone, 1986). Neuroimaging findings show damage to, and atypical functioning of, areas of the brain involved in attention in individuals with HIV (Chang et al., 2001). The compromised immune system resulting from HIV may also have an effect on the central nervous system that could affect childrenÕs cognitive abilities (Ellis et al., 2009). Finally, HIV is now commonly treated with highly active antiretroviral therapy (HAART) (Ellis et al., 2009), but some medications associated with HAART have been found to be toxic to brain cells (Robertson, Liner, & Meeker, 2012). This has called into question whether HAART may also influence neurocognitive skills (Robertson et al., 2012). These issues speak to the need to understand the role of the virus and its treatment on childrenÕs sustained attention skills.   The experiential canalization theory by Gottlieb (1991) suggests that neurocognitive development is influenced by the interaction of biological and environmental variables. There are several universal factors thought to influence neurocognitive development in very young children. These include nutrition, health, and positive cognitive stimulation (Irwin, Siddiqi, & !2Hertzman, 2007). Adequate nutrition and overall health are fundamental for positive neurocognitive development (Irwin et al., 2007). However, effective caregiver-child interactions are necessary to provide appropriate cognitive stimulation that is central to further neurocognitive development (Walker et al., 2011). Stimulation interventions have resulted in improvements above and beyond those resulting from health and nutrition interventions (Grantham-McGregor, Walker, Chang, & Powell, 1997). Understanding the importance of these factors raises concerns for children with HIV because all three critical variables may be compromised in families with HIV.   Issues such as a lack of adequate health care (Tuller et al., 2010) and difficulties with obtaining appropriate food resources (Weiser et al., 2012) have been reported in populations with HIV. It has also been found that the quality of caregiver-child interactions may be disrupted (Klunklin & Harrigan, 2002). This may be a result of biological (e.g., medical concerns, neurocognitive impairments), psychological (e.g., depression), and environmental (e.g., unavailability of necessary resources) stressors affecting the functioning of caregivers of children with HIV (Hackl, Somlai, Kelly & Kalichman, 1997). As such, all three factors, health, nutrition, and positive caregiver-child interactions, may be critical targets for interventions to support the sustained attention growth of children with HIV, but caregiver-child interactions may be most beneficial.   An important finding of previous studies examining the health, nutrition, and caregiver-child interaction interventions is that the duration of the intervention is related to childrenÕs developmental outcomes. In one study (McKay, Sinisterra, McKay, Gomez, & Lloreda, 1978), !3undernourished but otherwise healthy children who participated in combined health, nutrition, and stimulation interventions for a longer amount of time had more benefits in both cognitive and growth outcomes than those who participated for a shorter amount of time. In a two-year study of children with stunted growth who received either a nutritional intervention, stimulation intervention, combined nutrition and stimulation intervention, or no intervention (Grantham McGregor, Powell, Walker, & Himes, 1991), benefits to growth occurred in the first six months and benefits to the other domains of development (e.g., cognitive, motor) occurred during the entire two year period. These findings highlight the need to evaluate the effects of the length of treatments on childrenÕs outcomes.   In sum, millions of children around the world are affected by HIV. Research suggests that these children may have difficulties with sustained attention, a cognitive skill that is critical for daily (Sarter et al., 2001) and academic functioning (Campbell et al., 1991). This particular difficulty may be a result of the influence of the virus on the central nervous system, limited access to necessary nutrition and health care resources by caregivers, and less effective caregiver-child interactions. Interventions targeting these areas may be beneficial in supporting the development of sustained attention skills in children with HIV. However, there have been no published studies to date that have specifically examined this idea.  The Current Study The primary purpose of the current study was to examine the effects of an intervention to improve caregiver-child interactions and an intervention to improve knowledge about and access to health and nutrition on the sustained attention skills of children with HIV. A secondary aim of !4the proposed study was to examine how the length of the interventions influenced growth of sustained attention skills. A final goal of this study was to better understand the how caregiver factors (i.e., depression, daily functioning) and a child factor (i.e., child viral load) were related to young childrenÕs sustained attention skills.   !5CHAPTER 2: Literature Review According to the World Health Organization (WHO; 2014a), approximately 35 million people were infected with the human immunodeficiency virus (HIV) at the end of 2013. Of these, approximately 25 million live in Sub-Saharan Africa and 2.9 million are Sub-Saharan African children under the age of 15 years old (WHO, 2014b). While treatment for HIV has improved substantially over the last few decades, which has led to dramatic decreases in the transmission of HIV from mothers to children, an estimated 700 children become perinatally infected with HIV each day (U.S. Department of Health and Human Services, 2013). These children may face extensive medical (e.g., U.S. Department of Health and Human Services, 2010), psychosocial (e.g., Bogart et al., 2008), and neurocognitive challenges (e.g., Abubakar et al., 2008) as a result of their HIV status.  Introduction to Human Immunodeficiency Virus (HIV)  The human immunodeficiency virus (HIV) is an infectious disease that affects oneÕs immune system. There are two types of HIV, HIV-1 and HIV-2, and multiple subtypes of HIV within each. HIV-1 is the most common strain around the world and the subtypes of HIV-1 are associated with specific regions of the world (Ellis et al., 2009). The evidence that subtype has an effect on outcomes is mixed and not fully understood (Liner, Hall, & Robertson, 2007), but some research suggests that different subtypes may be associated with different neurocognitive sequelae in adults (Sacktor et al., 2009) and children (Boivin et al., 2010). However, current understanding of HIV infection is such that, in general, HIV enters CD4+ T cells, which are blood cells that help the immune system protect the body. The virus then reproduces, destroying !6the CD4+ T cells in the process (Douek et al., 2002). According to the Center for Disease Control and Prevention (CDC; 2014), HIV infection is presently thought to progress through three distinct stages: acute infection, clinical latency, and acquired immunodeficiency syndrome.   The acute infection stage occurs within two to four weeks after becoming infected with HIV. During this stage, the virus reproduces in great volumes and kills many CD4+ T cells. This results in an impaired immune system and leads to flu-like symptoms. This is also the time during which the virus enters the central nervous system (Wiley et al., 1986). After several weeks of rapid reproduction of the virus, the immune system is generally able to reduce the amount of virus in the body to a low, stable level, resulting in the clinical latency stage.  The clinical latency stage, otherwise known as asymptomatic HIV infection, is associated with relatively stable health. This stage can last many years, and up to several decades if treated. Many people do not feel sick during this time. HIV continues to reproduce during this stage, but at a relatively low frequency. Eventually, the bodyÕs immune system is not able to keep up with the infection, resulting in acquired immunodeficiency syndrome (AIDS).   AIDS is the last stage of the HIV infection. It is diagnosed when 1) the immune system of a person with HIV is so compromised from the infection that they develop illnesses, known as opportunistic infections, or 2) a person with HIV demonstrates exceptionally low CD4+ T cell counts (< 200 cells/mm3; normal counts range from 500 to 1,600 cells/mm3). Examples of opportunistic infections include cancers, toxoplasmosis (a protozoal infection in the brain), cytomegalovirus (a viral infection that causes eye disease), and pneumocystis pneumonia (a !7fungal infection) (U.S. Department of Health and Human Services, 2010). Significant medical care is generally needed to sustain life once a person reaches the AIDS stage of infection.   In sum, HIV is a viral infection of which the body cannot cure itself. After the initial rapid proliferation of the infection in the first several weeks, the virus remains dormant. Eventually, the virus takes over and the personÕs immune system becomes severely impaired. This generally leads to severe medical complications.  The above stages describe the general process of HIV infection in humans as described by the Center for Disease Control and Prevention (2014). There are additional classification stages of HIV infection that describe the clinical progression (based on symptomology) or immunological progression (based on CD4+ T cell counts) of HIV for adults and children (WHO, 2007). In adults, it is has been found that the clinical stage of HIV is not clearly linked to specific neurocognitive outcomes. While it may be more common for AIDS to be associated with neurocognitive challenges than earlier stages of the HIV infection, neurocognitive difficulties have also been found in individuals with earlier stages of HIV infections who are not experiencing medical symptoms (Heaton et al., 2011). Furthermore, research has not yet clearly determined how stage of HIV infection relates to specific neurocognitive outcomes in children. However, children with HIV who are in the early stages of HIV infection and are medically asymptomatic or experiencing moderate symptoms have also been found to demonstrate neurocogitive impairments compared to their peers without HIV (Ruel et al., 2012). In general, neurocognitive difficulties may be experienced by individuals in all stages of HIV infection.  !8 Transmission. HIV is a disease that must be acquired and that is primarily spread through three modes. One mode is sexual activity with someone who has HIV (WHO, 2014a). Another is through sharing drug injection equipment (WHO, 2014a). The third major avenue for transmitting HIV is from mother to child, which is known as vertical transmission (WHO, 2014a). This form of transmission is responsible for nearly all cases of HIV in very young children (WHO, 2014c). Vertical transmission of HIV can occur during pregnancy, delivery, or breastfeeding, and can happen in 15-45% of cases without preventative treatment (WHO, 2014c). With preventative treatment, which largely consists of antiretroviral treatment provided to the mother, the rates of mother-to-child transmission have been reduced to less than 2% (Cooper et al., 2002; Ioannidis et al., 2001).   Treatment. Treatments that slow the progression of the infection are now available to individuals with HIV. Highly active antiretroviral therapy (HAART) is considered the standard of care for individuals with HIV. According to Ellis and colleagues (2009), HAART generally includes a combination of three medications, including two nucleoside analogue reverse transcriptase inhibitors and either a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor. HAART regimens include multiple types of antiviral medications so that mutations of the virus that are resistant to one of the drugs or classes of drugs cannot proliferate.   In clinical and field studies, HAART interventions have been found to prolong the clinical latency stage of the infection (Ellis et al., 2009), reduce transmission from mother to child (El Beitune et al., 2004), and extend the overall life expectancy of individuals with HIV (Palella et al., 1998). HAART does this by preventing new HIV virions from being created in the !9hostÕs body, which naturally involves the preservation of CD4+ T cells that compose the bodyÕs immune system and allows the person to fight-off illnesses. With proper adherence to HAART, HIV viral loads often reach undetectable levels (Gulick et al., 1997). With poor adherence to HAART, however, HIV viral loads rise and the disease progresses (Ellis et al., 2009).  The influence of antiretroviral treatments on neurocognitive functioning is mixed. Evidence suggests that HAART is associated with lower incidences of severe neurocognitive impairments, such as HIV-associated dementia (Liner et al., 2010). However, less severe forms of neurocognitive impairment remain prevalent even in the era of HAART medication (Liner et al., 2010). Questions about the potential negative influence of HAART medications on neurocognitive functioning have been raised as a result of this finding. Robertson and colleagues (2012) examined the neurotoxicity of 15 antiretroviral compounds and found evidence to suggest that there may be direct toxic effects of antiretrovirals on neurons. Although there is evidence to suggest that HAART medications may be neurotoxic, there are no clear findings about its influence on specific neurocognitive functions, especially in children.  Despite the recognized importance of HIV treatment, research suggests that not all individuals with HIV receive the appropriate medical care and adhere to their medical regimen. Not all individuals with HIV have access to medical care, including medications. The financial costs of transportation to medical appointments (Senkomago, Guwatudde, Breda, & Khoshnood, 2011; Tuller et al., 2010) and the costs of treatment (Byakika-Tusiime et al., 2009) impede many individuals from receiving appropriate care. Additionally, competing family, employment, and childcare demands have been found to be barriers to engagement in HIV care (Williams, Amico, !10& Konkle-Parker, 2011). Finally, psychosocial issues related to having HIV prevent individuals with HIV from seeking out and following through with medical care. Specifically, stigma (Byakika-Tusiime et al., 2009; Cavaleri et al., 2013; Williams et al., 2011), feelings of denial and shame (Williams et al., 2011), and issues with depression (Byakika-Tusiime et al., 2009) are commonly reported barriers to care. This has implications for families with HIV. If caregivers do not receive adequate care, their functioning and care for their children may be compromised. Sustained Attention Problems of Children with HIV  According to empirical literature, sustained attention skills may be impaired in children with HIV. Watkins et al. (2000) found that children with HIV between the ages of 7 and 19 years old living across the United States performed significantly worse than children without HIV on a direct measure of sustained attention, the Continuous Performance Test. Specifically, Watkins and colleagues found that these children had difficulty inhibiting their responses to non-target stimuli. Approximately half of the children in this study were on antiretroviral treatment and approximately 74% of the children had CD4+ counts greater than 200 at the time of participation, suggesting that the virus was largely suppressed. This finding is similar to that of Ruel and colleagues (2012) who found that treatment-naŁve children with HIV performed more poorly than children without HIV on a direct measure of sustained attention, the Test of Variables of Attention. In particular, the children with HIV scored significantly worse on reaction time and ADHD scores. Ruel and colleagues also found that children with HIV with viral loads above the median level performed worse than children with HIV with viral loads below the median on omissions errors (indicating that they failed to respond to the target stimuli), overall reaction !11time, and ADHD scores. In this study, 93 treatment-naŁve Ugandan children between the ages of 6 and 12 years old who had relatively intact immune system functioning (CD4+ counts of greater than 350) were compared to 106 uninfected Ugandan children. Together, these findings may indicate that HIV impairs neurocognitive functioning related to sustained attention even when on antiretroviral treatment or when the virus does not appear to be prolific within the childÕs body.   The etiology of difficulties with sustained attention in children with HIV is not yet clear. There is no literature specially examining the etiology of sustained attention difficulties in children with HIV. However, there is some research that has examined the role of the virus in creating neurocognitive impairment in adults with HIV.  Evidence suggests that the virus directly influences the cells of the central nervous system and that damages to the brain may occur as a result of having a suppressed immune system (Ellis et al., 2009). Concentrations of HIV have been found in areas of the brain that are associated with attention skills (Klarborg et al., 2013; Kumaret al., 2007; Sarter et al., 2001; Wiley et al., 1986), which supports the idea that the virus leads to injuries to the brain in areas that are relevant for sustained attention. Neuroimaging has also been used to examine the role of HIV on attention deficits. Chang and colleagues (2001) had individuals with HIV and matched controls without HIV complete tasks of varying cognitive and attentional demands. They found that both groups performed with similar accuracy on the tasks, but that individuals with HIV showed overactivation in areas of the brain associated with attention and in surrounding brain areas. Based on their findings, Chang and colleagues concluded that individuals with HIV have damage to areas of the brain used for attention, which results in saturation of neural activity in !12these areas and the need for individuals with HIV to recruit additional neuronal regions.  In sum, there are pieces of research that suggest that HIV may lead to central nervous system damages that result in sustained attention deficits. However, the exact mechanisms by which HIV leads to issues with sustained attention in children remain not well understood.   Importance of sustained attention. According to Sarter and colleagues (2001) sustained attention is necessary for the development of other cognitive skills. Sustained attention is thought to represent one of the basic attention skills that support more advanced attention skills, such as selective attention (attending to one stimuli while ignoring another) and divided attention (attending to two stimuli simultaneously). It also supports overall cognitive ability and functioning. Finally, although there is little empirical literature specifically examining the role of sustained attention in learning and daily functioning, it is generally understood that impairments in sustained attention influence oneÕs ability to perform most activities of daily living (Sarter et al., 2001).  Research studies examining the relation between performance on continuous performance tests, which are direct measures of sustained attention, and academic achievement provide evidence that sustained attention is important for academic skills. In general, the consensus among studies is that sustained attention is significantly correlated with academic achievement in children of various ages (e.g., Campbell et al., 1991; Lam & Beale, 1991). Edley and Knopf (1987) found that sustained attention as measured by a continuous performance test predicted academic readiness as rated by teachers in the first weeks of preschool. Together, these !13findings imply that sustained attention may play a role in the learning of academic skills and, therefore, may be an essential skill for successful school performance.   Defining sustained attention. Sustained attention is the ability to focus on a stimulus for a period of time (Anderson, 2008). It has also been commonly referred to as vigilance in the research literature, especially in older publications (Oken, Salinsky, & Elsas, 2006). Limited sustained attention appears to be evident in infants as young as a few months old. Infants have been found to be able to gaze at some stimuli over others and manipulate specific toys and objects for short periods of time (Ruff, 1990). The length of time that a child is able to attend to a specific stimulus is thought to increase substantially between infancy and preschool age. Children between the ages of three and five years old appear to have the ability to successfully complete tasks of sustained attention for several minutes (Dowsett & Livesey, 2000), but this skill typically continues to develop through the preadolescent and adolescent period (DeLuca et al., 2003). Sustained attention does not appear to vary according to sex in children when examined using direct measures of sustained attention (Levy, 2006). However, boys and girls do appear to differ in rates of attention deficit hyperactivity disorder (ADHD), in which one symptom may be difficulty with sustaining attention (American Psychiatric Association, 2013). It has been found that males have a higher rate of ADHD, but that females are more likely to primarily demonstrate inattention difficulties (American Psychiatric Association, 2013).  Measuring sustained attention. Given the importance of sustained attention in childrenÕs development, it is important to be able to measure childrenÕs sustained attention skills. Rating scales are often used to assess inattentive behaviors observed by specific raters. However, !14two major limitations have been identified in regards to behavior rating scales used to measure attention. First, rating scales may result in incorrect or biased information because they are indirect and subjective measures. In one study, teachers were found to rate children of ethnic minority groups more poorly than their Caucasian peers (Reid et al., 1998). In another study, teachers, school psychologists, and parents rated children of low socioeconomic status and of ethnic minority groups as more behaviorally active than children of middle socioeconomic status or Caucasian children (Stevens, 1981). Children who have multiple difficulties, such as reading and behavioral difficulties, may also be rated more poorly because teachers may have a tendency to use additional or more global information to inform their ratings rather than just information specific to attention (Abikoff, Courtney, Pelham, & Koplewicz, 1993). Second, rating scales tend to focus on a variety of symptoms of attention disorders, rather than measuring specific attention mechanisms, such as sustained attention (Riccio, Reynolds, & Lowe, 2001). As a result, behavior rating scales may not be the best assessment tool for measuring a childÕs ability to sustain attention to a task.     Continuous performance tests (CPTs) are direct measures of sustained attention and may represent a more objective measure of a childÕs sustained attention. Although there are many CPTs in existence, they all have the same general framework for assessing sustained attention. That is, examinees are typically presented with multiple stimuli, one of which is a target stimulus. CPTs vary in terms of stimuli modality (e.g., auditory or visual) and type of stimuli used (e.g., letters, shapes, tones, colors). Examinees are asked to indicate the presence of the target stimulus by responding in some manner. Assessments of sustained attention in very young children tend to involve observing the duration of eye gaze towards a stimulus (Ruff, 1990). !15Older children are generally expected to respond to target stimuli by pressing buttons on a computer or another device after they recognize a target stimulus. Given that a childÕs performance on a CPT is determined by their own behavior and is not dependent on the ratings of another, it is thought to be a potentially more valid measure of sustained attention (Riccio et al., 2001).  Interventions to improve sustained attention. Supporting sustained attention skills in children is critical given the importance of the skill for academic success and daily living. In general, interventions related to attention have focused on children with ADHD. There is little literature examining interventions for supporting sustained attention skills in typically-developing children or otherwise medically-involved children. One of the most common interventions for difficulties with attention in children is stimulant medication. While medication may reduce ADHD symptoms, it may not lead to sustained improvement in academic outcomes (Molina et al., 2009). Parent training and behavioral contingency interventions are also commonly used with children with ADHD. However, these largely help with managing behavior and may not improve or support the development of sustained attention skills. In fact, no study to date has demonstrated that these interventions lead to improved sustained attention skills that are maintained for an extended period of time (Molina et al., 2009). Cognitive training is a relatively new intervention for attention deficits. A recent meta-analysis suggests that training attention skills by way of computerized activities does not improve attention abilities (Rapport, Orban, Kofler, & Friedman, 2013). Despite the number of interventions for children with ADHD, there appear to be no empirically-supported interventions that result in generalized, long-term improvements in sustained attention specifically. Furthermore, there seems to be no research !16examining interventions for sustained attention in children with HIV. This speaks to a significant need within the clinical and research literature.  Factors Affecting Neurocognitive Development   As summarized by Mahone and Schneider (2012), brain systems, including those necessary for sustained attention, in young children are shaped concurrently by biology and experience. Shaping refers to the myelination and pruning of neuronal pathways. In healthy development, shaping leads to improvements in attention mechanisms. In compromised situations, shaping leads to disrupted development of attention mechanisms.   The general theory behind this biological and experiential shaping of brain systems is known as experiential canalization (Gottlieb, 1991). The major idea of this theory is that experiences lead to changes from a behavioral level down to a cellular level and that development can only be completely understood when biological and environmental interactions are understood (Blair & Raver, 2012). Blair and Raver (2012) provided an example of how experience and biology can interact to influence childrenÕs self-regulation skills. Drawing on empirical literature, they described how adversity associated with poverty, such as discrimination and limited access to resources, can influence childrenÕs development. That is, caregivers may experience significant stress and depression resulting from these stressors, which influences the hormones that are present in their body during pregnancy. These hormones can influence the prenatal development of a fetus. The child, once born, continues to be influenced by their biological predispositions. They are also continuously influenced by the quality of the caregiving, which may be impaired due to stress and mental health effects experienced by the !17caregiver. The experiences of the child further shape the neural pathways in the childÕs developing brain. Some of these pathways may be associated with self-regulation skills. In conclusion, this model describes the method by which many variables in the childÕs life can interact to influence neurocognitive development.   According to the child development literature, certain variables are universally important for childrenÕs neurocognitive development and neural shaping. These include adequate nutrition, health care, and exposure to stimulating interactions and environments (Engle, et al., 2011; Irwin et al., 2007; Walker et al., 2011). The extent to which children are exposed to these variables depends on caregivers. According to Irwin and colleagues (2007), inadequate nutrition and poor health has been associated with long-term negative effects on childrenÕs cognitive and psychological outcomes. While adequate nutrition and health care are foundational for supporting child development, the quality of caregiver-child interactions and learning environments are considered to be primary factors in determining the outcome of childrenÕs development (Irwin et al., 2007). In fact, in a review of studies (Grantham-McGregor, Fernald, Kagawa, & Walker, 2007) examining interventions targeting stimulating experiences (e.g., caregiver-child interactions, preschool) and nutrition and health (e.g., nutrition education, nutrient supplementation, health promotion), findings indicated that nutrition and health interventions occasionally benefited child development and stimulation consistently benefited child development. In studies that examined the influence of stimulation and nutrition and health interventions provided together, there was evidence to suggest that there were independent effects of each component and that stimulation had an added effect on childrenÕs development. Importantly, the review found that stimulation interventions did not influence nutrition and !18health status, indicating the need to provide both stimulation and nutrition and health interventions in order to target both factors necessary for development. In the studies reviewed, child development could refer to any of the following: cognition, language, behavior, social-emotional skills, or motor development. Additionally, in the majority of studies reviewed, stimulation interventions took the form of home visits to families to teach caregivers how to engage in responsive play with their children. In general, these findings indicate that both nutrition/health and stimulation interventions are important for improving various domains of child development. Unfortunately, none of the studies examined the influence of these interventions on sustained attention.   The review also highlighted an important factor, intervention duration, that may be important to consider when providing health, nutrition, and stimulation interventions. The findings of one study (Muennig et al., 2009) were such that children who participated in interventions for a longer amount of time had more benefits in both cognitive and growth outcomes. In another study (Lozoff et al., 2010), benefits to childrenÕs physical growth occurred in the first six months of the intervention and benefits to other developmental domains (i.e., cognitive, social-emotional, motor) occurred in the first and second six months of the intervention. Finally, in a two-year study (Grantham McGregor et al., 1991), benefits to growth occurred in the first six months, and benefits to the other domains of development occurred during the entire two year period.   Taken together, these findings indicate that nutrition, healthcare, and stimulating interactions and environments are essential for childrenÕs neurocognitive development. An !19important aspect to consider when providing such interventions is the duration of the interventions. For children with HIV and their caregivers, these factors may be potentially beneficial targets for intervention.  Issues Affecting Caregivers with HIV  Because childrenÕs neurocognitive development is influenced by the abilities of their caregivers to provide nutrition, healthcare, and stimulating interactions, it is important to examine the functioning of caregivers of children with HIV. Notably, much of the research that has examined the role of caregivers of children with HIV has focused on caregivers who also have HIV (Klunklin & Harrigan, 2002). Little is known about the experiences of caregivers of children with HIV who do not have HIV themselves.  Psychological challenges can be serious outcomes for many individuals with HIV. Perceived stigma and discrimination on the basis of HIV status has been shown to predict depression symptoms in many studies across the world. For example, in one study of 366 women with HIV in the southern United States by Wingwood and colleagues (2007), those who reported perceived discrimination on the basis of their HIV status also reported more depressive symptoms and suicidal ideation and less self-esteem. In a study in South Africa, men and women with HIV reported feeling dirty, ashamed, or guilty because of their HIV status (Simbayi et al., 2007). These individualsÕ experiences with stigma significantly accounted for 4.8% of the variance in depression scores.  Finally, in a study by Murphy, Austin, and Greenwell (2013) that included 118 mothers with HIV from Los Angeles, reports of experiences with stigma were positively associated with reports of depression symptoms. This finding remained after !20controlling for other factors known to be related to depression, such as social support, symptoms of illness, ethnicity, marital status, and education level. Altogether, these findings suggest that having a diagnosis of HIV may be a risk factor for experiencing stigma and discrimination, which may lead to difficulties with depression.  Additional literature has shown that depression may result in impaired daily functioning (e.g., maintaining household responsibilities, dressing, bathing) in adults with HIV. Individuals with higher depression scores have been found to have higher levels of impairment on activities of daily living (Jin et al., 2006). Furthermore, depression was found to predict impairment in daily functioning over and above the effect of having HIV alone (Jin et al., 2006). Depression and apathy have also been found to be the only significant predictors of difficulties with daily functioning even when examined alongside of cognitive impairment, history of substance use, HIV status, and immune system functioning (Kamat et al., 2012). This suggests that depression may have a significant role in the functioning of adults with HIV. What remains unclear is the extent to which daily functioning is related to aspects of caregiving in adults with HIV.   Although there is little empirical research examining the parenting practices of individuals with HIV, caregivers with HIV have been found to have difficulties with certain aspects of caregiving. In terms of caregiver-child interactions, Forehand and colleagues (1996) found that caregivers with HIV have been found to be less nurturing towards their children (as cited in Klunklin & Harrigan, 2002). Kotchick and colleagues (1997) found that mothers with HIV reported poorer relationship quality than mothers without HIV. In this study, the quality of mother-child relationships was assessed with a questionnaire that included items such as ÔI enjoy !21spending time with my childÕ and ÔYou think you and your child get along well.Õ It has also been found that caregivers with HIV have poorer communication with their children (Murphy, Marelich, Armistead, Herbeck & Payne, 2010). Communication in this study was assessed with a questionnaire with items such as, ÔDoes ____ insult you when he/she is angry with you?Õ, ÔIf _____ is upset, is it difficult for you to figure out what he/she is feeling?Õ, and ÔCan you discuss your beliefs with ____ without feeling restrained or embarrassed?Õ Caregivers responded to items using a five-point likert scale ranging from Ôalmost neverÕ to Ôalmost always.Õ They have also demonstrated less monitoring of and knowledge about their childrenÕs behavior (Kotchick et al., 1997), as well as less consistent discipline (Murphy et al., 2010). These findings speak to the various challenges that have been identified in caregiver-child interactions in families with HIV. They also raise the question about whether difficulties with caregiving are related to or are a part of difficulties with overall functioning in this population.   Finally, research has found that families with HIV have limited knowledge about or difficulties acquiring necessary resources. Adults with HIV often have limited access to adequate food (Weiser et al., 2012) and health care (Tuller et al., 2010). This results from a variety of factors, such as lack of money and transportation (Tuller et al., 2010). Additionally, in one study (Schable, 1995), caregivers with HIV were found to have limited knowledge about assistance programs available for their children. It was also found that few caregivers took advantage of the programs that were available for their children even when they knew of them. Pupradit found that caregivers of children with HIV self-reported that they made mistakes in starting certain food regimens with their children and that they had not washed their hands or the containers used to serve their children food (as cited in Klunklin & Harrigan, 2002). These findings indicate that !22caregivers of children with HIV may experience environmental challenges limiting their access to healthcare and nutrition, but that they may also lack some knowledge about health and nutrition and may not seek out related services even when are aware of them. Again, it may be possible that some of these difficulties are related to or are a part of the caregiversÕ challenges with overall functioning. Drawing upon experiential canalization theory, these factors likely influence the shaping of childrenÕs neural pathways.   In the context of the experiential canalization theory, it is likely that the various challenges experienced by caregivers with HIV could influence childrenÕs neurocognitive development. That is, caregiversÕ experiences with depression could lead to impaired caregiver functioning, which could result in various caregiving disruptions. These may include failure to provide adequate nutrition, a lack of appropriate health care, disruption to the childÕs medication regimen, and a lack of or comprised caregiver-child interactions. In turn, these factors could influence childrenÕs neurocognitive development, including their sustained attention skills.  Purpose of the Present Study The reviewed body of literature indicates that children with HIV may have impaired sustained attention compared to children without HIV (Ruel et al., 2012; Watkins et al., 2000). The exact etiology of this difficulty remains unclear, but potentially may be related to direct and indirect effects of the virus on the childÕs central nervous system, less effective caregiver-child interactions, and compromised nutrition and health care. As such, the purpose of the present study was to examine the effects of two interventions, one targeting nutrition and health education and one targeting caregiver-child interactions, on childrenÕs sustained attention. Given !23that no published study to date has specifically examined the effects of interventions on sustained attention in children with HIV, this study may provide clinically-relevant information about potential treatment options for this population. Another aim of this study was to attempt to better understand the associations between child and caregiver variables on childrenÕs sustained attention skills. This is important because a better understanding of the nature of the sustained attention difficulties observed in this population may also help inform prevention and treatment efforts.  Research Questions and Hypotheses  Research question 1. Do nutrition and health education and caregiver-child interaction interventions improve sustained attention in children with HIV?  Hypothesis 1a. It was hypothesized that both nutrition and health education and the caregiver-child interaction interventions would improve sustained attention in children with HIV.  Hypothesis 1b. It was hypothesized that receiving an intervention targeting caregiver-child interactions would result in greater improvements in sustained attention than would receiving a health and nutrition education program. Rationale. It was hypothesized that both interventions would lead to improvements in sustained attention in children with HIV because previous research has found that adequate nutrition, health care, and exposure to stimulating interactions and environments are important for neural shaping and for neurocognitive development (Engle et al., 2011; Irwin et al., 2007; Walker et al., 2011). Additionally, it was hypothesized that a caregiver-child interaction !24intervention would result in greater improvements to sustained attention than a health and nutrition education intervention because previous research had found that the quality of caregiver-child interactions and learning environments are primary factors in the development of childrenÕs neurocognitive skills. Studies that have examined health and nutrition interventions and caregiver-child interaction interventions have found that health and nutrition interventions occasionally benefited child development, but interventions involving caregiver-child interactions or similar cognitive stimulation consistently benefited child development (Grantham-McGregor et al., 2007). Finally, there is empirical evidence to suggest that interventions involving cognitive stimulation produced an added effect on childrenÕs development above and beyond the effect of nutrition and health care (Grantham-McGregor et al., 2007). Given that all caregivers in the present study were provided with food packages, it was hypothesized that the group receiving the caregiver-child interaction intervention would demonstrate greater growth in sustained attention skills.   Research question 2. How does the length of the nutrition and health education and caregiver-child interaction interventions affect changes in sustained attention in children with HIV?  Hypothesis 2a. It was hypothesized that one year of health and nutrition education intervention would lead to significantly greater gains in sustained attention than six months of intervention.  !25Hypothesis 2b. It was hypothesized that one year of an intervention targeting caregiver-child interactions would lead to significantly greater gains in sustained attention than six months of intervention.  Rationale. It was hypothesized that one year of both interventions would lead to significantly greater gains in sustained attention than six months of intervention because the result of a previous study (McGregor et al., 1991) indicated that interventions targeting health and interventions targeting stimulation resulted in physical health benefits during the first six months and benefits to cognitive development throughout the entire two-year intervention period. Additionally McKay and colleagues (1978) found that children who participated in health, nutrition, and stimulation interventions for longer periods of time experienced more cognitive benefits than children who participated for shorter periods of time. Taken together, these findings suggest that children in both groups may experience a benefit of participating in the intervention for the full year.   Research question 3. How are caregiver depression, caregiver functioning, and child viral load related to childrenÕs sustained attention?   Hypothesis 3. It was hypothesized that caregiver depression would directly influence caregiver functioning, which would directly influence child viral load and child sustained attention. Caregiver functioning would also indirectly influence sustained attention by way of child viral load. Please see Figure 1 for a visual depiction of the conceptual model that was tested.  !26Rationale. Because childrenÕs neurocognitive development is influenced by the ability of their caregivers to provide nutrition, health care (including appropriate HIV medication), and stimulating interactions, it was believed that caregiver factors would influence childrenÕs sustained attention skills. A sizable body of literature suggests that adults with HIV experience stigma and discrimination that may result in symptoms of depression (Murphy et al., 2013; Simbayi et al., 2007; Wingwood et al., 2007). Additional literature has shown that depression may result in impaired daily functioning in adults with HIV (Jin et al., 2006; Kamat et al., 2012).  While there are no known studies empirically examining the link between daily functioning and quality of caregiving in individuals with HIV, it is argued here that the daily overall functioning of caregivers would include providing their children with health care. For children with HIV, providing health care would likely involve obtaining and appropriately distributing the childrenÕs antiretroviral medication, which would influence childrenÕs viral loads. ChildrenÕs viral loads would thereby possibly influence their neurocognitive functioning, specifically, sustained attention.  In addition to daily functioning influencing sustained attention by way of providing medications that target viral load, it is argued that caregiver functioning would also include providing effective caregiver-child interactions and adequate nutrition. Existing literature indicates that caregivers with HIV have difficulties with certain aspects of caregiving, including nurturance (Forehand et al., 1996, as cited in Klunkin & Harrigan, 2002), relationships (Kotchick et al., 1997), communication (Murphy et al., 2010), discipline (Murphy et al., 2010), and less monitoring of and knowledge about childrenÕs behavior (Kotchick et al., 1997). Because the quality of the caregiver-child interactions is known to be a primary factor in childrenÕs neurocognitive development (Engle et al., 2011; Irwin et al., 2007; Walker et al., 2011), it was !27hypothesized that caregiver functioning would influence the quality of the caregiver-child interactions and, therefore, influence childrenÕs sustained attention. !!28     Child viral load covariate: HAART status Figure 1 Conceptual Model !29Caregiver DepressionCaregiver FunctioningChild Viral LoadChild Sustained AttentionCHAPTER 3: Method Present Study Design  The present study used semi-extant experimental study data to assess the effects of two intervention programs on the sustained attention of young children with HIV. The existing data were previously collected under the National Institute of Health RO1 Grant: HD070723. All of the data that were used in the present study were collected, processed, and scored prior to the proposal of this study, except for the data corresponding to the primary dependent variable of the proposed study. The data for the dependent variable were collected, but remained in raw data files (videos) that needed to be processed (coded).  Uganda  The original study from which the data for the present study were derived was conducted in Uganda. Uganda is an important place to examine the effects of interventions that are likely to support the neurocognitive development of children with HIV because there are many children in Uganda who could potentially benefit from such intervention. Uganda has the tenth highest percentage of individuals with HIV in the world (Central Intelligence Agency, 2014), which corresponds to approximately 1,500,000 Ugandans with HIV (UNICEF, 2013). Furthermore, 190,000 of these cases of HIV represent children under the age of 15 (UNICEF, 2013). Another primary reason why Uganda was selected for the original study was because there was an existing organization to work with who had already developed a culturally-relevant health and nutrition education curriculum that they provided to some of their citizens.  !30 Many factors affecting children and adults with HIV, as well as those affecting neurocognitive development in general, are thought to be consistent for populations around the world. Studies have been conducted with individuals with HIV across the world, ranging from rural America (Williams et al., 2011) to urban America (Murphy et al., 2013) to China (Jin et al., 2006) and across Africa (Simbayi et al., 2007; Thurman et al., 2012). Similar findings regarding depression (Murphy et al., 2013; Simbayi et al., 2007), neurocognitive effects (Ruel et al., 2012; Watkins et al., 2000), and impairments in functioning (Jin et al., 2007; Kamat et al., 2012) have been reported across these regions. It is also recognized that appropriate stimulation, nutrition, and health are important for the neurocognitive development of children worldwide (Irwin et al., 2007). As such, the factors discussed in the literature review likely apply to the individuals and children of Uganda. In turn, some of the findings of this study may be relevant for populations outside of Uganda. Interventions  Two yearlong interventions were provided to individuals participating in the original study. One intervention was a nutrition and health education intervention. The other was an intervention known as Mediational Intervention for Sensitizing Caregivers (MISC) that aims to enhance caregiver-child interactions and the stimulation provided to children (Klein & Rye, 2004).  Trained research assistants who had bachelors degrees from the local university, Makerere University, carried out both interventions. Different interventionists were assigned to each treatment group to ensure that only content from the assigned intervention was provided to !31the caregivers. Interventionists were fluent in Luganda and were accompanied by translators who spoke the language of the caregiver, when necessary. Translators were used more often in the group receiving the caregiver-child interaction intervention than the health and nutrition education intervention, "2(1, n = 117) = 13.39, p = .000. Approximately 57% of those in the MISC group required the use of a translator at least one time, whereas approximately 23% of the health and nutrition education group needed a translator at least one time. It is recognized that, while translators who are viewed as helpful community resources may help families from different cultures more readily accept intervention recommendations (Raval & Smith, 2003), using translators during interventions introduces potential barriers and challenges. These may include changes in meaning, issues related to the style of translating used by the translator, and issues with establishing rapport, among others (Lopez, 2008). As such, the potential effects of using a translator were controlled for in the analyses examining intervention effects.   The interventions were matched on length of intervention and time spent during each session. That is, 24 intervention sessions were planned for the year, which corresponded to two sessions per month. Each session was designed to last approximately 30 to 45 minutes. One session per month was to be held in the study office and the other was to be held in the caregiversÕ homes.   Nutrition and health education intervention. Caregivers of children infected with HIV in Uganda may receive a nutrition and health care curriculum developed by the non-governmental organization known as Uganda Community Based Association for Child Welfare (UCOBAC), which is supported by the United Nations ChildrenÕs Fund (UNICEF). This !32program is a free-standing program that may be provided to families with HIV in Uganda regardless of their participation in the original research study. Research assistants who were trained in this curriculum by the Director of UCOBAC implemented it with the participants in the study. Because this program is a free-standing program that may be provided to families with HIV in Uganda, it is referred to as the treatment as usual (TAU) in the present study.    The TAU is largely informational in nature and teaches caregivers about nutrition and health care as it relates to themselves and their children. It teaches caregivers how to provide health care related specifically to HIV. It does not provide the caregivers with information about cognitive development or how to provide appropriate cognitive stimulation to children. The intervention content covers the following 12 topics: Basic facts on HIV/AIDS; Positive living/coping mechanisms; Concept home-based care; Communication skills; Basic counseling; Basic nursing care and pain management; Hygiene; Nutrition; Facts on family planning; Child growth and development; Herbal medicine and HIV/AIDS; and Will making. The lessons are accomplished by verbally providing information, discussing facts and existing knowledge, providing written materials, and modeling. Please see Table 1 for a description of the contents of each topic. Knowledge about and access to health care and nutrition are some of the most basic and necessary components of healthy development, yet they represent common challenges facing individuals with HIV (Senkomago et al., 2011; Tuller et al., 2010). Given that general health care and nutrition are necessary for childrenÕs cognitive development (Irwin et al., 2007), the UCOBAC program may help improve the neurocognitive development of children, including sustained attention.  !33Table 1 Topics and Content of TAU TopicContentBasic facts on HIV/AIDS-What are HIV and AIDS -Transmission of HIV -Signs and symptoms of AIDS -Prevention of HIV -Condom use -Sexually transmitted infections (STIs) -STIs and HIVPositive living/coping mechanisms-Physical care (medical care for opportunistic infections, proper sleep habits, proper nutrition, avoiding substance use, home hygiene, avoiding unprotected sex, reducing sexual activity, family planning, herbal medicine, mosquito nets, using antiretroviral medication) -Psychological care (counseling, spirituality, creating fellowships, developing  positive attitude) -Caring for children (immunizations, balanced diet, good hygiene, staying warm, medical care for opportunistic infections, giving love and affection) -Challenges to living positively (poverty, societal expectations, structural limitations, ignorance, failure to cope, cultural expectations, gender differences)!34Table 1 (contÕd)Concept of home-based care-Importance of home-based care (creating a supportive environment, accepting the disease, reducing stigma and discrimination, participating in decision-making) -Who can provide home-based care -Advantages of providing home-based care -Challenges of home-based care -Types of home-based care -Record keeping of home-based careCommunication skills-Definition of communication -Process of communication -Types of communication -Important communication skills -Factors that promote effective communicationBasic counseling-Definition of counseling -Importance of counseling -Types of counseling -Levels of counseling -Who may need counseling -Where should counseling occur -Qualities of a good counselor!35Table 1 (contÕd)Basic nursing care and pain management-Oral/mouth hygiene (brushing teeth, rinsing with warm salt water, cleaning mouth with clean cloth) -Bathing (required materials, process of bathing) -Bed (cleaning linens, making a bed to support a personÕs medical condition) -Feeding (drinking fluids, giving frequent small meals, type of food) -Managing urine/bowel problems (using plastic sheet, feeding soft foods, drinking fluids) -Bed sores (massage body to aide circulation, keep rooms ventilated, change linen, clean sores)Hygiene-Definition of hygiene -Types of hygiene (personal, home, environment) -Common poor hygiene practices -Consequences of bad hygieneNutrition-What is good nutrition -Why to have a balanced diet -Types of foods (vitamins, proteins, carbohydrates, fats, sugar, water, mineral salts) -Important issues in nutrition (avoiding infections, does not require lots of money, properly preparing food, nutrition needs of children, nutrition needs of children with HIV)Facts on family planning-Definition of family planning -Types of birth control -Importance of family planning to women with HIV -What to consider when choosing a family planning method!36TAU was compared to MISC in a previous study (Boivin et al., 2013b) conducted in another region of Uganda. In this study, participants included caregiver-child dyads with HIV. Children were between the ages of sixteen months and five years. Children who received TAU showed improvements in the domains of memory, gross motor, fine motor, receptive language, Table 1 (contÕd)Child growth and development-What is a child -Stages of child growth and development (fetus versus a child) -Needs of a fetus via mother (water, balanced diet, prenatal care, rest, light exercise, avoiding stress, avoiding substance use, vaccinations, hygiene, prevention of transmission of HIV, love and care to the mother) -Needs of a newborn (feeding, bathing, breathing, immunizations, clinics, love, care) -Needs of a child to grow well (balanced diet, good hygiene, love, care, medical attention, play, education)Herbal medicine and HIV/AIDS-What is herbal medicine -Advantages of herbal medicine -Disadvantages of herbal medicine -Management of opportunistic infections using herbal medicineWill making-What is a will -Why write a will -When and how should a will be written -Contents of a will -How does a valid will look? -What happens if one does not leave a will!37visual reception, and overall cognitive skills over the course of the yearlong intervention. However, children in the MISC group performed better than children in the TAU group on visual reception. No study to date has examined the effects of TAU on childrenÕs sustained attention skills. However, the preliminary evidence suggests that this intervention may lead to improved neurocognitive performance, which may include sustained attention skills. Sarter and colleagues (2001) suggested that sustained attention is necessary for the development of other cognitive skills. As such, TAU may have promoted growth in sustained attention, which resulted in the improvements observed in other cognitive domains.   TAU protocol. The TAU protocol was provided to only those in the TAU group. The UCOBAC curriculum manual includes 12 trainings on topics related to health and nutrition. Given that the intervention period for the study was one year, the 12 topics were intended to be divided into two sessions each, totaling 24 sessions. In addition to covering the lesson plans outlined in the manual, nutritional support in the form of food packages were given to caregivers. The food packages were provided on every other intervention session, or once per month. These food packages contained two kilograms worth of food.   Mediational Intervention for Sensitizing Caregivers (MISC). The MISC treatment may help improve the sustained attention of children infected with HIV because it teaches caregivers how to have appropriately stimulating and enriching interactions with their children. MISC was designed to enhance caregiver-child interactions by training caregivers to be sensitive to the developmental needs of children, to foster appropriate behavior in children in positive ways, and to enhance the thinking and cognitive processes in children (Klein, 2000).  !38MISC was created with an eco-cultural framework; it was designed to incorporate behaviors, goals, and values that are appropriate for the context or culture in which it is being utilized (Klein, 2003), which may make it a culturally sensitive intervention appropriate for use in Uganda. MISC trainers teach the mediational techniques in ways that apply to the culture of the families with whom they are working. Moreover, MISC trainers use examples and model behaviors that are relevant to the individuals with whom they are working. In addition, MISC uses a developmental framework to meet the current needs of the children, while taking into account the needs and objectives of their caregivers (Klein, 2003). That is, MISC trainers teach caregivers to engage with their children in developmentally appropriate ways. For example, MISC might teach caregivers to use stimuli such as bottles with very young children and cups for children who are no longer drinking from bottles.  Finally, the MISC intervention is one that teaches caregivers how to mediate the childÕs environment in positive ways that support cognitive growth. More specifically, MISC teaches caregivers how to respond to and intervene with their children in ways that will scaffold, instruct, or support their childÕs behaviors and learning (Klein, 2003). This mediational technique was developed from FeuersteinÕs (1980) theory of Structural Cognitive Modifiability (Klein, 2000). This theory suggests that intelligence can be shaped and that learning occurs by interacting with the environment. Feuerstein argued that a caregiver can be a mediator between environmental stimuli and the child by affecting the context, frequency, order of presentation, and other qualities of external stimuli. In turn, this may arouse the children and help them focus on the stimuli, which is thought to help them encode information and learn. Altogether, MISC has a !39theoretical basis that makes it a potentially useful intervention for children of various ages from a specific cultural group. It also has a theoretical foundation supporting its proposed role in enhancing cognitive development.  MISC teaches caregivers to 1) positively gain childrenÕs attention, 2) raise childrenÕs awareness of their feelings and emotions, 3) encourage learning and healthily praise competence, 4) help children expand their thinking to the future and to less concrete situations, and 5) demonstrate to children how to complete tasks and help children think about planning for completing tasks (Klein, 2000). In MISC, these mediational techniques have been termed: focusing, exciting, expanding, encouraging, and regulating, respectively (Klein & Rye, 2004). Please see Table 2 for definitions and examples of the mediational techniques of MISC based on those provided in Klein (2003) and Klein and Rye (2004).  Several cognitive mediational techniques, including focusing, regulating, and encouraging, may be particularly relevant for improving sustained attention. Focusing is designed to help children perceive and then attend to relevant stimuli (Klein & Rye, 2004). Improving this skill may benefit sustained attention because sustained attention involves attending to stimuli for an extended period of time (Anderson, 2008). Regulating techniques may also help improve sustained attention because they are designed to teach children expected behaviors for planning and carrying out tasks (Klein & Rye, 2004) and this explicit instruction may help children learn to plan for and follow-through on attending to necessary stimuli. Finally, encouraging may be a relevant technique for supporting sustained attention because caregivers may provide positive reinforcement for childrenÕs attention towards stimuli or tasks. In sum, it is !40possible that these techniques may involve explicitly teaching or reinforcing behaviors necessary for sustained attention.  The MISC program designed for use in Uganda also includes an emotional component designed to enhance nurturance and caregiver-child relationships. This aspect of the MISC program teaches caregivers about smiling to children, making developmentally appropriate vocalizations, making eye contact, the importance of physical closeness, the importance of appropriate touch, ways to share in joy with children, and how to engage in turn taking with children. These factors may enhance attachment, which may promote positive cognitive development (OÕConnor & McCartney, 2007).  According to Klein (2003), the MISC intervention begins with interviewing and collecting data from caregivers to understand their values, goals, beliefs, and expectations of behavior for their children. In the early training sessions, caregiver-child interactions are video recorded. These videos are then watched by the caregiver alongside of the trainer who uses the video to teach the lessons of the intervention. In later sessions, other mediational techniques and lessons that have not arisen naturally in the videos are taught. Because the intervention utilizes existing behaviors and interactions to teach mediational techniques and emotional components to parents and because it attempts to utilize each caregiverÕs goals, values, and beliefs, there is no script that is followed in each session. Instead, interventionists who are familiar with the techniques/lessons verbally instruct and model them as they see fit. !41Table 2 Definitions and Examples of Mediational Techniques in MISC TechniqueDefinitionExamplesFocusingBehaviors aimed at influencing a childÕs perception or behavior toward a stimulus-Caregiver brings an object, such as a bottle, toward the child -Caregiver alters the stimulus, such as tilts or shakes a bottle, until the child focuses on itExcitingBehaviors intended to convey emotion, excitement, or interest-Caregiver uses facial gestures or body language, such as a smile -Caregiver names a feeling or emotion -Caregiver emphasizes meaning (e.g., ÒLook at the tree. ItÕs BIG!Ó)ExpandingBehavior that demonstrates exploration of stimuli or thinking about objects/concepts past moment-Caregiver discusses the quality of objects, such as branches while gathering supplies for a fire -Caregiver makes statements that require reasoning (e.g., ÒI wonder if this tree is growing so fast because of the rain we received.Ó)!42Research examining the effects of MISC has found that MISC is associated with lasting improvements in certain cognitive abilities. Klein and Alony (1993) conducted an experimental study with randomized assignment into MISC and control groups with women in Israel who were of low socioeconomic status. The control group received information regarding developmental milestones and how to support child development. Maternal mediation behaviors were gathered at baseline, one year, and three years post intervention. ChildrenÕs cognitive abilities were assessed at baseline and then post-intervention with a battery of cognitive assessments. The results of the study were such that mothers demonstrated a significant increase in maternal mediation behaviors from the baseline to the one-year assessment. This finding Table 2 (contÕd)EncouragingBehavior that shows competence or success-Caregiver makes a nonverbal gesture indicating success (e.g., shakes head) -Caregiver makes a statement indicating competence (e.g., ÒNice job.Ó)RegulatingBehavior that demonstrates how to regulate behavior or be planful-Caregiver describes or models the order in which tasks should be completed to successfully complete a job -Caregiver models how to behave in an appropriate manner (e.g., demonstrates how to walk slowly through the home)!43remained significant at the three-year post-intervention assessment as well. Children in the MISC group demonstrated greater performance on a test of verbal reasoning, language comprehension, and receptive vocabulary than children in the control group. No differences were found on a test of auditory memory, visual reasoning, or visual-motor integration. Similar findings were found when MISC was used as an intervention for caregivers and children in Ethiopia (Klein & Rye, 2004). In this study, participants included caregivers and their children aged one to three years old. Families were from poor communities and were randomly assigned to the MISC or control group. The results were such that, compared to the control group, fewer harsh commands and orders were given by caregivers and caregivers were more sensitive and response to their children. Caregivers also were more optimistic about their ability to affect their childrenÕs development. The children in the study were found to have greater language and vocabulary skills than their peers in the control group. Boivin and colleagues (2013a) found that the MISC intervention resulted in significantly greater gains in receptive and expressive language and overall cognitive ability than a nutrition and health education intervention for Ugandan children who were perinatally exposed to HIV. Taken together, these findings suggest that MISC may be a useful intervention for supporting the development of childrenÕs cognitive skills, specifically those related to language. However, no published study to date has examined whether MISC influences childrenÕs ability to sustain attention. This is an important question to research because it may be that MISC fosters sustained attention, which allows children to develop in other cognitive domains. This hypothesis makes sense in light of the recognition that sustained attention is a skill that is necessary for the development and performance of more complex cognitive skills (Sarter et al., 2001).  !44 MISC protocol. MISC was provided to only those participants who were assigned to the MISC intervention group. Those in the MISC group, however, also received the same food packages as the TAU group. This was done in order to ensure that all families had the same access to necessary foundational nutrition. No educational content from TAU was provided to caregivers in the MISC group.   Ugandan interventionists completed three weeklong trainings provided by MISC consultants to learn how to implement the intervention prior to beginning training sessions with families. During the first session with the caregiver, the interventionist introduced the intervention and explained the positive benefits that can be expected from MISC. In subsequent home intervention sessions, interventionists taught the caregivers mediational techniques and recorded their behaviors in situations involving eating, feeding, and working. At the following office visits, a video analysis was completed in which the caregiver analyzed the video alongside of the caregiver demonstrating the strengths and weaknesses of the behaviors and how to use mediational techniques. A review of previous lessons was provided at the start of each new intervention session. It is noted that this intervention was designed to be relevant to each individual caregiver. As such, trainers were instructed to be flexible about which topics they covered at what time and to use their observations of caregivers to inform lessons that were taught. If a component was identified as needed because of an observed interaction, that lesson could be used instead of one that was tentatively planned for that day.  Intervention fidelity. The interventionists in Uganda maintained a log of each intervention session, including whether it occurred, why it did not occur (if applicable), what was !45planned for the session, the extent to which the plans were accomplished during the session, and the engagement of the caregiver in the intervention session. Please see Appendix A for examples of the log forms used for each intervention group. In general, participants in both groups completed the majority of intervention sessions. MISC dyads completed significantly more intervention sessions on average than TAU dyads, t(109) = -3.12, p = .002. Approximately 70% of the TAU group completed all 24 intervention sessions, whereas 92% of MISC dyads completed all 24 sessions. There were 17 dyads from the TAU group who did not complete 24 sessions. These dyads completed either 15 (n=2), 17 (n=2), 19 (n=2), 20 (n=1), 21 (n=3), 22 (n=5), or 23 (n=2) sessions. Only 4 dyads from the MISC group did not complete all 24 sessions. These dyads completed 17, 18, 20, and 23 sessions. Some trainings were not completed because caregivers moved out of the study location. For the rest of the dyads, it is not determinable from the data exactly why each dyad did not complete trainings. What is known is that individual training sessions were most often not conducted because the caregiver was absent for the training session (MISC = 46%; TAU = 58%) or the caregiver was too sick (MISC = 30%; TAU = 17%); However, some of these missed sessions were rescheduled.  By and large, dyads in both groups were highly engaged in the intervention sessions. Caregivers in the TAU group, however, were significantly more engaged than caregivers in the MISC group, t(109) = 3.36, p = .001. Caregivers in both groups were rated on a scale from Ô1Õ corresponding to ÔNot at allÕ to Ô4Õ corresponding to ÔVeryÕ engaged at the end of each intervention session by their trainers. These ratings were averaged for each caregiver and then the average for all caregivers in each group was also obtained. The average rating for caregivers !46in the MISC group across all interventions sessions was 3.90 (SD = .14). Average overall ratings for caregivers ranged from 3.17 to 4. The average rating for caregivers in the TAU group across all intervention sessions was 3.97 (SD = .06). Average overall ratings of engagement for caregivers in TAU ranged from 3.86 to 4.  For both MISC and TAU groups, there was a range of how many times each lesson was planned to be covered with a caregiver. Trainers recorded on their logs the lessons that were planned for each intervention session. Using these logs, the total number of times each lesson was planned to be covered with each caregiver was summed. An average of these sums was then calculated to better understand how many times caregivers in each intervention group were scheduled to receive each lesson. Please see Table 3 for averages, standard deviations, minimums, and maximums for the MISC group and Table 4 for the TAU group.   According to these data, individual lessons were planned to be taught to or reviewed with the MISC group many times throughout the yearlong intervention period. Once a session was planned to be taught, it tended to be planned to be reviewed or retaught during each subsequent session. The content on regulating childrenÕs behavior was the second to last lesson most frequently scheduled to be covered. On average, it was covered 12 times (SD = 2.46) per caregiver. Content on focusing was planned to be covered approximately 18 times (SD = 2.20) per caregiver, on average. Finally, content on encouraging was scheduled to be covered approximately 11 times (SD = 2.46) per caregiver, on average.   Given that this study was concerned with effects on sustained attention due to duration of intervention, another important aspect of the interventions to consider is when the lessons were scheduled to be covered.  In order to capture this information, the session number in which each !47lesson was first planned was determined. An average was then calculated for each lesson for the entire MISC group. The results are shown in Table 3. The findings indicate that lessons related to enhancing the mother-child relationship were scheduled to be taught first. Focusing content was first scheduled to be covered around session six (SD = 2.56), which corresponds to approximately three months into the intervention. Regulating, however, was first planned to be covered with caregivers around session 11 (SD = 4.34), which corresponds to approximately 6 months into or halfway through the intervention. Finally, content related to encouraging was first scheduled to be covered with caregivers around session 13 (SD = 4.77).  In the TAU group, participants generally received each lesson once or twice. Some caregivers never received certain lessons and some lessons were covered more than twice. It is unknown if different material was covered each time a topic was repeated or if the material was reviewed when the topic was reported to have been repeated because this level of detail was not recorded on the training logs. Issues related to child development were covered the most and counseling content was covered the least in the TAU group. Nutrition, family planning, herbal medicine, and will making were generally covered during the second half of the intervention period and all other lessons tended to be covered during the first half of the intervention period. This information is shown in Table 4.  !48Table 3 Descriptive Statistics of MISC Lessons LessonMinimumMaximumM (SD)First Session Covered M(SD)Encouraging41610.65(2.46)12.98(4.77)Regulating51712.44(2.46)11.31(4.34)Expanding61914.72(2.22)8.91(3.25)Exciting82016.63(2.22)7.00(2.93)Focusing92117.74(2.20)6.00(2.56)Touching132521.15(2.10)2.93(1.34)Take Turns112520.72(2.48)3.26(2.01)Physical Close132521.22(2.17)2.85(1.32)Eye Contact152622.43(2.02)1.69(.72)Smile152722.74(2.08)1.43(.60)Vocalize152723.09(1.93)1.11(.32)Share Joy112420.13(2.25)3.72(1.80)!49Table 4 Descriptive Statistics of TAU Lessons  Given that the data in the above tables represent the lessons that were planned for each session by the trainer, it was important to also examine the extent to which the goals for each intervention session were completed. For both intervention groups, each trainer rated the extent to which their goals were covered at the end of each intervention session on a scale from Ô1Õ representing ÔNoneÕ to Ô4Õ representing ÔAllÕ. An average of these ratings from the entire intervention period was calculated for each caregiver, and then an average of these averages was derived for each intervention group. The TAU group covered their goals significantly more often than the MISC group, t(109) = 3.75, p = .000. The average of the TAU group was 3.97 (SD = .LessonMinimumMaximumM(SD)First Session Covered M(SD)HIV/AIDS142(.46)1.07(.26)Positive Living and Coping242.09(.34)3.07(.56)Home Care132.04(.33)5.07(.84)Communication021.21(.49)7.78(2.43)Counseling021.19(.55)9.38(3.27)Nursing031.88(.47)9.75(2.48)Hygiene041.33(.64)11.35(1.66)Nutrition041.3(.71)12.62(1.36)Family Planning041.63(.86)15.16(4.56)Child Development063.56(1.63)14.49(2.47)Herbal Medicine021.6(.73)19.14(1.83)Will Making052.21(1.52)21.20(1.93)!5004) compared to 3.87 (SD = .19) for the MISC group. The average rating for each participant in the MISC group ranged from 3.17 to 4  and the average rating for each participant in the TAU group ranged from 3.86 to 4.  Acceptability is an important factor thought to facilitate intervention fidelity (Perepletchikova & Kazdin, 2005). Unfortunately, there are no data regarding the acceptability of the interventions as indicated by study participants. Therefore, caregiversÕ thoughts and attitudes about MISC and TAU in the present study are unknown, which limits our understanding of the extent to which they may have accepted the interventions and implemented intervention content with their children.  Participants  Participants whose data were used in the present study were recruited from the Tororo District of Uganda (See Figures 2 and 3). Eighteen sub-counties/regions of Tororo were randomly assigned to MISC and TAU groups, such that there were nine sub-counties assigned to MISC and nine assigned to TAU. Clustering into geographical regions was done to prevent spillover effects in which caregivers in one intervention share the information they learned with caregivers in the other intervention group. Child and caregiver dyads in these sub-counties were recruited to participate in the original study from local area HIV treatment clinics.  Caregiver-child dyads that met study criteria were eligible to enroll in the study. Children were required to be between the ages of two and five years old at the time of enrollment and HIV positive in order to participate. Children who had significant illnesses, serious birth complications, severe malnutrition, bacterial meningitis, encephalitis, cerebral malaria, seizures, !51or other known brain injuries or medical disorders were excluded from the study. Additionally, children who were attending or were scheduled to attend formal schooling (primary school) during the first eight months of the intervention period were excluded from the study because their attendance would limit the amount of caregiver-child interactions they could have. Additionally, children were excluded if they were participating in another research study. Finally, children and their primary caregivers were excluded from the study if the primary caregiver had a mental illness or disability that would preclude them from participating in the interventions. Five caregiver-child dyads left the study nearly immediately after enrolling. These dyads were removed from the final sample because of their lack of exposure to the interventions. These dyads received either 0 (n=2), 1 (n=1), 3 (n=1), or 4 (n=1) sessions out of the intended 24 sessions before they left the study. Reasons as to why they left the study are discussed below.  ! Figure 2 Countries of Africa. Image adapted from freeworldmaps.net. !52! Figure 3 Districts of Uganda. Image adapted from the Food and Agriculture Organization of the United Nations at http://www.fao.org/ag/AGP/AGPC/doc/counprof/uganda.htm.  Parental consent was given for 77 dyads in the MISC sub-counties and 70 in the TAU sub-counties. Of the 77 dyads in the MISC sub-counties, 5 children were HIV negative, 4 children were malnourished, 3 children were participating in another study, 5 children were not in the required age range, and 4 dyads did not live in the required sub-counties. In total, 58 dyads in the MISC sub-counties were eligible to enroll. All 58 dyads who enrolled in the study completed at least 1 baseline assessment. Only 54 dyads in the MISC group completed at least one six-month assessment because 2 children died, 1 dyad moved from the region, and 1 caregiver withdrew consent at the time of the 6-month assessment. Another caregiver moved prior to the 12-month assessment time, so 53 dyads completed at least 1 12-month assessment. Altogether, data for 54 dyads in the MISC group will be used in the present studyÕs analyses. !53Of the 70 dyads in the TAU sub-counties considered for eligibility, 2 children were HIV negative, 1 childÕs HIV status was unknown, 1 child was malnourished, 1 child was enrolled in another study, 2 children died prior to being enrolled, 1 child was out of the required age range, and 1 child had difficulties with tuberculosis. In total, 61 dyads were eligible for enrollment and 60 completed at least 1 baseline assessment. For one dyad, it is unknown why there are no data at any time point. Two caregivers moved prior to the 6-month assessment period and 1 moved prior to the 12-month assessment period. Altogether, data for 57 dyads in the TAU group will be used in the present studyÕs analyses. Please see Figure 4 for a summary of participant involvement across the study period.  !54Figure 4 Participant Flow Chart !55Please see Table 5 for a summary table of the caregiver and child characteristics described below. At the start of the study, the average age of caregivers in the MISC group was 36 years old and the average age of TAU caregivers was 35 years old. All of the caregivers were female across both groups. The majority of the caregivers in both groups were the childrenÕs biological mothers. Other caregivers included grandmothers, stepmothers, aunts, and a sister-in law. Caregivers primarily worked as peasants, but others worked in trading, as a teacher, in casual labor work, and other work. Some caregivers did not work. Most of the caregivers completed some formal schooling, including primary school or primary and secondary school. However, there were caregivers who did not complete any formal schooling. Most of the caregivers were HIV positive across both groups. Some caregivers had been tested and knew they were HIV negative, but others reported that their serostatus was unknown or not available. Of the caregivers who were HIV positive, the majority were receiving antiretroviral treatment. There were no significant differences between the MISC and TAU groups on any of these demographic factors.  At the start of the original study, the average age of the children in the MISC and TAU groups was approximately three years. Exactly half of the children in the MISC group were boys and there were slightly more boys than girls in the TAU group. All of the children were HIV positive. A potentially important variable given the computerized nature of the sustained attention test in the present study is the childrenÕs prior exposure to TV or video. Only 22 children from the MISC group had reportedly seen TV or video a couple of times and 20 children had seen it more than 3 times. Twelve children had not seen TV or video before. Almost half of the TAU group children had never seen TV or video (n = 45). Of the remaining children, many !56had seen it a couple of times (n=18) and several had seen it three or more times (n=12). This represents a moderately-sized, significant difference between the two groups, "2(3, n = 111) = 8.48, p = .04. Given this finding, prior exposure to TV and video were controlled for in the analyses examining the effects of the two interventions on sustained attention. Please see Table 5 for a summary of the MISC and TAU demographic information, as well as the results of the tests for significant differences between groups.    !57Table 5 Demographic Information CharacteristicMISCTAU n (%)pEffect sizeCaregiverAge (M(SD))36.44 (8.31)35.44 (8.70).5412aFemale54 (100)57 (100)Relation to child.18.24b     Mother40 (74.1)43 (75.4)     Grandmother12 (22.2)7 (12.3)     Stepmother2 (3.7)2 (3.5)     Aunt0 (0)4 (7.0)     Sister-in-law0 (0)1 (1.8)Occupation     .76.15b     Peasant45 (83.3)47 (82.5)     Teacher0 (0)1 (1.8)     Trading4 (7.4)2 (3.5)     Casual labor2 (3.7)2 (3.5)     None2 (3.7)2 (3.5)     Other1 (1.9)3 (5.3)Education level.19.17b     None13 (24.1)12 (21.1)     Primary School37 (68.5)34 (59.6)     Secondary School4 (7.4)11 (19.3)HIV status.80.06b     Negative10 (18.5)9 (29.8)     Positive41 (75.9)46 (50.9)!58Note. a = CohenÕs d. b = CramerÕs V. c = phi. *p < .05. Table 5 (contÕd)      Not Available3 (5.6)2 (3.5)ARV status (if HIV positive).37.13c     No10 (18.5)17 (29.8)     Yes31 (57.4)29 (50.9)ChildAge (M(SD))3.06 (.93)3.34 (.93).11.30aGender.11-.15c     Male27 (50)20 (35.1)     Female27 (50)37 (64.9)HIV positive54 (100)57 (100)Seen TV or video.04*.28b     No12 (22)26 (45.6)     1-2 times22 (40.7)18 (31.6)     3+ times20 (37.0)12 (21.1)     Unknown0 (0)1 (1.8)!59Variables and Measures  Assessments and clinical evaluations were completed prior to initiating the intervention, six months into the yearlong intervention period, and at the conclusion of the twelve month intervention period. Assessments and clinical evaluations were completed in project office spaces. Assessments were completed by trained research assistants.   Early Childhood Vigilance Task (ECVT).  Although CPT paradigms have been used to assess sustained attention in children of nearly all ages, it is important to consider a number of population and individual factors when selecting the appropriate CPT. In the case of young Ugandan children with HIV, there are several important factors that need to be considered. First, it is essential that tests be developmentally appropriate in terms of the length of the test. The test should likely only be several minutes long. Second, the tests used to measure sustained attention in children from Uganda should include stimuli that are understandable to the child. This precludes the use of tests that involve English letter names, letter sounds, and words. Third and finally, motor skill demands must be taken into consideration given young childrenÕs level of development and the potential influence of HIV infection on motor skill development. Therefore, a test with a non-motor response was thought to be appropriate for this population.  The Early Childhood Vigilance Task (ECVT; Romero, Yund, Zelinsky Goldman, Hughes, Jordan & Shapiro, 2012) is one test of sustained attention that may be appropriate for young Ugandan children with HIV. Romero and colleagues (2012) use the terms vigilance and sustained attention interchangeably throughout their manual and consider the ECVT a measure of sustained attention. The ECVT is a CPT that has only been used in research to date. According to !60Romero and colleagues (2012), it was created in order to explore sustained attention abilities in very young children for whom no other age-appropriate assessments exist. It was designed to be as similar to CPTs for older children as possible. That is, stimuli are presented intermittently on a computer screen and childrenÕs attention toward the screen is measured. Stimuli on the ECVT include cartoon creatures that appear from behind a rock for 10 seconds and then disappear for either 5, 10, or 15 seconds. A different creature appears each time. The length of time the creatures are absent is arranged such that each interval of time occurs six times. A camera on top of the computer screen records the childÕs face throughout the duration of the test. The test lasts seven minutes. Upon completion of the task, the examiner reviews the video keeping track of the amount of time the childÕs gaze appears to be on the screen. The ECVT was thought to be a useful test for measuring sustained attention in preschool-aged Ugandan children with HIV because it is relatively short, does not require a motor response, and has seemingly developmentally appropriate stimuli.   ECVT scores were the primary outcome variable of this study. The principal score of the ECVT is the proportion of time the child looked at the video screen during the assessment. That is, the total number of seconds the child appeared to be looking at the video screen was divided by the total number of seconds in the video (420). A coder who scores a video of the child watching the animation determines the proportion of time the child spent looking at the screen.   Only one published study to date has examined the psychometric properties of the ECVT. Zelinsky Goldman, Shapiro, and Nelson (2004) administered the ECVT and several attention and cognitive measures to 51 typically developing children ranging in age from 24 months to 30 !61months. The children were reportedly socioeconomically and ethnically diverse, having been recruited from two sites, a registry of research participants and an inner-city community clinic, in the Midwest United States. The results of the study suggested that the ECVT had adequate internal consistency, demonstrated by product-moment correlations for each third of the test ranging from .56 to .81. Interscorer reliability was also found to be high, as demonstrated by a product-moment correlation of .98. The validity of the ECVT was demonstrated through examining correlations between the ECVT and traditional measures of attention. The ECVT was found to have a significant, moderate, and positive correlation (r2 = .33) with the Behavior Rating Scale from the Bayley Scales of Infant Development-Second Edition (BSID-II; Bayley, 1993), which examines attention, persistence, motor regulation, adaptability, and negative affect. Additionally, the ECVT was found to significantly correlate (r2 = -.35) with a free play session coded according to the Self-Regulation scale of Minnesota Preschool Affect Rating Scale (MN-PARS; Shapiro, McPhee, Abbott, & Sulzbacher, 1994). The Self-Regulation scale of the MN-PARS provides ratings about attention, orientation to objects, persistence, and adaptability. Good self-regulation is associated with lower scores on the scale and poor self-regulation is associated with higher scores, which explains the negative correlation to the ECVT. Finally, the ECVT was significantly and positively associated (r2 = .35) with the proportion of time the child played with toys during an eight-minute free-play session. The ECVT was not found to correlate significantly with a parent rating of attention from the Colorado Childhood Temperament Inventory (CCTI; Rowe & Plomin, 1977). The authors of this study hypothesized that this may be due to parents lacking knowledge about normative behavior, parents not being accurate reporters of their childrenÕs behavior, or that parentsÕ ratings reflected other aspects of attention than those !62measured by the ECVT. Descriptive statistics obtained during this study indicated that the average time spent looking at the video was 67%, which corresponds to just over 4.5 minutes. The median score was 70%. Finally, the standard deviation of time spent looking at the screen was 19%, which corresponds to 1.33 minutes. According to Zelinsky Goldman and colleagues (2004), these descriptive statistics are highly consistent with those obtained during pilot testing with children of the same age. In sum, initial psychometric testing of this measure suggests that the measure may be a reliable and valid tool for measuring sustained attention in very young children.  ECVT coding and analysis. Raw ECVT data (video files) collected in Tororo were transported to Michigan State University. The videos were coded by a trained graduate-level researcher as instructed by the ECVT manual, which involved tracking the amount of time in seconds the child was looking at the screen and turning that amount into a percentage representing the amount of time spent attending to the screen out of the total time of the animation. ProCoder for Digital Video (ProCoderDV; Tapp, 2003), a software program developed at Vanderbilt University for gathering observational data from video and/or audio files, was used to code the ECVT videos.  In order to prevent bias in coding the ECVT videos, the researcher was blind to intervention group and assessment time point. This was accomplished by relabeling the videos, which were previously labeled with an identification number and assessment time point, with nondescript identifiers prior to coding. A log containing the current and nondescript labels was !63kept but not viewed by the researcher during the coding phase. The log was used after coding was completed to match each video with its previous identification.   It is important that the coder was reliable in the coding of the ECVT videos. Reliability of the coder was, therefore, assessed by recoding 17 randomly collected videos. A paired samples t-test was then conducted to determine whether the scores of the 17 videos coded initially are significantly different from the scores of the 17 videos coded a second time.  ChildrenÕs age. The age of the children at the time of baseline testing was included as a control variable in the analysis examining the effects of the interventions because sustained attention skills tend to increase with age (DeLuca et al., 2003) and this analysis incorporates longitudinal data. Age was not used as a covariate in the analysis testing the conceptual model because there was no theoretical reason for or evidence suggesting that age would affect variables in the model other than sustained attention. At the start of the original study, the average age of the children in the MISC and TAU groups was approximately three years. Descriptive statistics for MISC and TAU are shown in Table 5.  ChildrenÕs gender. Sustained attention skills have not been shown to vary according to gender as measured by direct assessments (Levy, 2006). However, disorders that may involve impaired sustained attention have been shown to differ according to gender (American Psychiatric Association, 2013). As such, gender was a covariate in the analysis examining the effects of the interventions. Gender was not used as a covariate in the analysis testing the conceptual model because there was no theoretical reason for or evidence suggesting that gender would affect variables in the model other than sustained attention. Exactly half of the children in !64the MISC group were boys and there were slightly more boys than girls in the TAU group. Descriptive statistics for MISC and TAU are shown in Table 5.  ChildrenÕs HAART status. Whether or not children were receiving HAART at baseline was included in both analyses addressing all three research questions. This is because HAART directly affects childrenÕs viral load, which likely affects neurocognitive functioning, and because there is a possibility that HAART could influence neurocognitive functioning by way of its neural toxicity (Liner et al., 2007).   Descriptive statistics for MISC and TAU are shown in Table 6. Over half of the children in both intervention groups were on HAART at baseline testing. Approximately 70% of the MISC group and 58% of the TAU group were on HAART at that time. There was no significant difference between the two groups based on HAART status.   ChildrenÕs HIV viral load.  Viral load count data were collected from child participants at baseline. Viral load is the number of copies of HIV RNA present in one milliliter of blood. This measure primarily reflects the amount of virus present in the body at the time the blood sample was collected. It may be an indicator of treatment status and/or and adherence to treatment because HIV viral loads remain low if on medication and if the medication is taken as directed. It also may be an indicator of overall health of the children because high viral loads may correlate with impaired immune system functioning. As such, child viral load was used as a control variable in the analysis addressing the effects of the intervention. Child viral load was also a variable in the conceptual model to be tested.  !65Please see Table 6 for the descriptive statistics for child viral load by intervention group at baseline. Figure 5 shows the distribution of pre-transformed viral loads for all children in the present study at baseline. Overall, children in the present study tended to have low viral loads at baseline. The majority of children had viral loads at or below 400, which was the lowest detectable level. There were no significant differences between the viral loads of children in TAU from those in MISC. Generally low viral loads are to be expected given that the majority of children in this study were on antiretroviral medication and, of those on antiretroviral medication, the majority were on HAART (See Table 6). Figure 5 Viral load distribution !66 ChildrenÕs height and weight. Height and weight are basic anthropomorphic measures that indicate overall health. Based on the theory of experiential canalization (Gobblieb, 1991), overall health may influence childrenÕs response to the interventions and sustained attention skills. As such, childrenÕs height, in centimeters, and weight, in kilograms, were included as covariates in the present studyÕs analyses examining the effects of the interventions. ChildrenÕs height and weight were not used as a covariates in the analysis testing the conceptual model because there was no theoretical reason for or evidence suggesting that height and weight would affect variables in the model other than sustained attention.   See Table 6 for childrenÕs height and weight by intervention group. Average height and weight were similar in both groups. Average height was around 92 centimeters and average weight was around 13 kilograms.   Caregiver depression. An adapted version of the 25-item Hopkins Symptom Checklist (HSCL-25) was used to assess depression in caregivers. The HSCL-25 is derived from the original Hopkins Symptom Checklist (HSCL; Derogatis, Lipman, Rickels, Uhlenhuth, & Covi, 1974). There are 15 items on this scale that measure depression symptoms; these 15 items were used in the present study. Examples of symptoms include: crying easily, difficulty with sleeping, feeling low energy and lack of strength, trouble concentrating, feeling sad, and having no hope.  The remaining items of the HSCL-25 assess anxiety and were not used in the present study. On the original HSCL-25, informants were asked to indicate their severity of symptoms on a four-point scale from Ônot at allÕ to Ôextremely.Õ Caregivers in the original study, however, were asked to indicate how often they have experienced each of the 15 symptoms in the last four weeks. !67Caregivers responded using a four-point likert scale: Ô0Õ = Ônot at allÕ, Ô1Õ = Ôrarely,Õ Ô2Õ = Ôsometimes,Õ and Ô3Õ = Ôoften.Õ The responses to the items were summed and used to assess severity of depression in the caregivers. The items were translated into three languages used by the caregivers, Luganda, Dhopadhola, and Ateso, and then administered to the caregivers in a verbal interview format.  The original HSCL was found to have appropriate reliability and validity. Reliability estimates for the depression scale of the HSCL ranged from .64 (interrater) to .86 (internal consistency) (Derogatis et al., 1974). The HSCL was also found to demonstrate appropriate criterion-related and construct validity through a number of analyses and studies (Derogatis et al., 1974). The HSCL-25 has been translated and used with specific populations around the world (e.g., Kaaya et al., 2002; Ventevogel et al., 2007); however, the reliability and validity of this HSCL-25 as adapted and used with this specific population is unknown.   The depression summary scores were used in the analyses examining the effects of the interventions on childrenÕs sustained attention because it was hypothesized that depression in caregivers with HIV was associated with caregiver-child interactions. It was also used as a variable in the conceptual model that was tested. Caregivers in both groups reported experiencing symptoms of depression of varied frequencies over the four weeks prior to the assessment. There was no significant difference between caregiver depression reported in the TAU group as compared to the MISC group. Please see Table 6 for the descriptive statistics of caregiver depression scores. !68Caregiver functioning. Functional impairment was measured using a scale developed as part of the original study. Prior to the start of the original study, a brief qualitative study was conducted to identify the responsibilities of women caregivers in this area of Uganda. Based on the responses of women caregivers, the caregiver functioning scale described below was developed.  The scale assesses 16 daily tasks and responsibilities that women in this cultural group engage in across 4 domains: caring for self, caring for family, caring for community, and caring for their children. Examples of items in the Ôcaring for selfÕ domain include: cleaning your body and brushing your teeth. Examples of items in the Ôcaring for familyÕ domain include: digging in the garden, cooking, and fetching water. Examples of items in the Ôcaring for childrenÕ domain include: bathing your children and clothing children. Examples of items in the Ôcaring for communityÕ domain include: helping friends and neighbors when they have a problem and participating in groups with other women. Caregivers rated the degree of difficulty they had on each task using a five-point likert scale: Ô0Õ = Ônone,Õ Ô1Õ = Ôlittle,Õ Ô2Õ = Ômoderate,Õ Ô3Õ = Ôa lot,Õ Ô4Õ = Ôoften cannot do.Õ A summary score of the items indicates degree of impairment in daily functioning. The items were translated into the three languages used by the caregivers and then administered to the caregivers in a verbal interview format. Given that this assessment was developed by the original studyÕs researchers, reliability and validity information is not available. However, it was thought that this tool would be beneficial because it was designed to be culturally sensitive to the daily functions of caregivers enrolled in the present study.  !69Caregiver functioning was used as a covariate in the analyses examining the effects of the two interventions on childrenÕs sustained attention. This was because caregiver functioning was hypothesized to influence the caregiverÕs ability to carry out the interventions, which could influence performance on the ECVT. It was also a variable in the conceptual model that was tested.  Caregivers tended to report relatively high levels of functioning. Mean scores for MISC and TAU were 6.6 and 8.6, respectively, out of a total possible 64. MISC and TAU were significantly different only with a p-value equal to .05. Please see Table 6 for descriptive statistics on caregiver functioning summary scores. !70Table 6 Descriptive Statistics of Covariates Note. a = CohenÕs d. b = phi. *p # .05. Analyses Several preliminary and primary analyses were conducted as part of this study. The IBM Statistical Package for the Social Sciences (IBM SPSS; IBM Corp, 2013) version 22.0 was used to prepare the existing data and to calculate the descriptive statistics presented above. IBM SPSS was also used to prepare the data and conduct additional preliminary analyses on the final dataset containing ECVT scores. The mixed-effects modeling and path analysis were completed using CharacteristicMISCTAUM (SD)M (SD)pEffect sizeChildHeight91.46 (11.10)92.86 (11.11).51-.13aWeight13.54 (3.22)13.35 (3.30).76.06aViral load101,992.76 (185,375.25)79,368 (169,256.44).52.13aOn HAART.17.13b     No16 (29.60)24 (42.10)     Yes38 (70.40)33 (57.90)CaregiverCaregiver depression14.98 (7.68)14.07 (5.76).48.13aCaregiver functioning6.57 (6.01)8.61 (4.82).05*-.37a!71the statistical package R (R: A Language and Environment for Statistical Computing; R Core Team, 2014). Each analysis is described below. Please see Table 7 for a summary of research questions, variables, and analyses.  Preliminary analyses. Preliminary analyses were conducted in order to better understand the MISC and TAU groups and the data in general. The results of the preliminary analyses can be found in Tables 5 and 6. Means and standard deviations of continuous variables (caregiver and child age) were determined. Independent samples t-tests were conducted to determine whether there were significant differences on mean scores. For all other variables, Chi-square tests were conducted in order to examine whether there were significant differences between groups. Chi-square tests were necessary due to the nominal nature of these variables. Effect sizes were calculated for all of the between-group comparisons using either CohenÕs d, CramerÕs V, or phi. Phi was used when the data conformed to a 2 x 2 matrix and CramerÕs V was used when the data conformed to a matrix greater than 2 x 2.  These preliminary analyses revealed one significant between-group difference, exposure to video screens, which was accounted for in the primary analysis examining intervention effects.  Variable transformations. Transformations were completed on several variables prior to completing the primary analyses of the present study (data described above refer to pre-transformed scores) to ease the interpretation of results or because of statistical issues. Caregiver functioning scores were reverse coded for the primary analyses of the present study to ease the interpretation of the results. That is, greater caregiver functioning scores now represent higher reports of functioning and lower functioning scores represent more difficulty with functioning. !72Because regressions are simpler when ordinal predictors are converted to binary variables, exposure to video screens was transformed from one ordinal variable to two binary variables, watchtv1or2 and watchtvmany. For watchtv1or2, Ô0Õ corresponded to Ônever having watched tv or videoÕ and Ô1Õ corresponded to having Ôwatched tv once or twice.Õ For watchtvmany, Ô0Õ corresponded to having Ôwatched tv less than 3 timesÕ and Ô1Õ corresponded to having Ôwatched tv three or more times.Õ Finally, viral load, depression, and functioning scores had to be re-scaled because the variance of viral load was significantly greater than the variance of the other variables, which resulted in an ill-scaled covariance matrix. Viral loads were divided by 100,000 and caregiver depression and functioning summary scores were divided by 10 to ensure all variables had relatively similar variances.  Centering of variables was also completed simply to ease the interpretation of the regression intercept. Age, height, weight, and caregiver depression scores were centered around the means of each variable. The means did not accurately reflect the central tendency of the distribution for two variables, viral load and caregiver functioning, so these variables were centered around their medians.  Mixed-effects modeling. Mixed-effects modeling was used to examine the effects of the interventions on sustained attention in children, as well as the effect of duration of intervention on childrenÕs sustained attention. Mixed-effects modeling was used because it could control for the potential effects of the geographic clusters of caregiver-child dyads and because of its ability to handle repeated measurements/longitudinal data. This method was also useful because it used !73all available data and did not require that a dyad be dropped from the analysis if it was missing one or multiple assessments, as long as there was one dependent variable data point.   A mixed-effects model contains fixed and random effects. In the present study, the fixed factors included the treatment group and the baseline ECVT scores. Additional fixed factors included time, time by treatment interaction, age, gender, exposure to TV/video, use of translator, child viral load, HAART status, child height, child weight, caregiver depression, and caregiver functioning. The random effects included individual dyads and the clusters to which each dyad was assigned.  In mixed-effects models like the one described above, the degrees of freedom for the t-tests examining the statistical significance of the parameters in the model are not well-defined (Mirman, 2014). As such, the statistical significance of the parameters in the above model were examined using t-tests that rely on the Sattherthwaite method for approximating degrees of freedom. In addition to examining p-values, CohenÕs d was also calculated in order to evaluate the meaningfulness of intervention effects.  Path analysis. Path analysis was used to test the proposed conceptual model linking caregiver depression, caregiver functioning, child viral load, and baseline ECVT performance while controlling for the effect of HAART status on child viral load (Figure 1). Specifically, simultaneous confirmatory modeling was used to test the hypothesized causal relations among the caregiver and child variables. Moreover, the path analysis determined how well the conceptual model fit the observed assessment data.  !74The fit between the conceptual model and the observed data was determined using absolute fit indices because they determine how well a model specified a priori fits the sample data. More specifically, absolute fit indices measure the extent to which the hypothesized model reproduces the sample covariance matrix (Lei & Wu, 2007). It is considered best practice to evaluate several fit indices when determining model fit because of the sensitivities of each index (Lei & Wu, 2007). Chi-square is the traditional standard fit index (Hooper, Couglan, & Mullen, 2008). It was one of three indices used in the present study. If there is good model fit, the chi-square test statistic is not significant (p > .05) (Hooper et al., 2008). Another fit indicator used was the Root Mean Square Error of Approximation (RMSEA). RMSEA was useful because it is sensitive to the number of parameters and favors parsimony (Hooper et al., 2008). An RMSEA of 0 indicates perfect fit and an RMSEA of .06 or less is considered good fit (Hooper et al., 2008). Finally, the comparative fit index (CFI) was used to assess fit in the present study. It compared the hypothesized model with a null model. This index accounts for sample size and is understood to be useful for small samples (Fan, Thompson, & Wang, 1999). Values greater than .95 indicate good fit with at least .91 being necessary to ensure that incorrect models are not accepted (Hu & Bentler, 1999). If a model is acceptable according to these indicators, the specific paths in the model can be interpreted.  Missing data. Missing data were handled in both the mixed-effects model and the path analysis with maximum likelihood procedures. Specifically, full information maximum likelihood was used in the mixed-effects model and maximum likelihood estimation with robust standard errors was used for the path analysis. Full information maximum likelihood produces unbiased parameter estimates for missing data. It uses all available data and does not exclude !75cases that are missing one or multiple assessment points. Maximum likelihood estimation with robust standard errors also involves producing parameter estimates for missing data, but it is better suited for variables that have a non-normal distribution because it does not assume normality when producing standard errors (Duncan, Duncan, & Strycker, 2006). This is important for the path analysis because of the non-normal distribution of caregiver functioning and viral load variables. Data must be missing at random for maximum likelihood procedures (OÕConnell, Pentimonti, & McCoach, 2013). In the present study, viral load data are missing for seven children. It is unknown exactly why these data are missing, but it is reasonably hypothesized that there were difficulties with collecting the blood samples or difficulties obtaining results from the lab. There is no reason to believe that these data are missing due to factors related to other variables in the study. There is also one rating missing for amount of tv watched for one child in the study. The exact reason for this missing data is unknown, but it could be missing because the caregiver was unsure about the answer or the question was missed during the assessment. There is no reason to believe that it is missing due to factors related to other variables in the study. There are no missing data for any other predictor variables. Therefore, it is assumed that these data are missing at random. Missing data for the ECVT are discussed in the Results section.  Project Approval  The data used in the present study were collected after permission to conduct the research was obtained from Michigan State UniversityÕs and Makerere UniversityÕs Institutional Review Boards, as well as the Ugandan National Council for Science and Technology. Permission to conduct the research of the present study was approved under IRB# 11-1026. Permission to use !76the data in this dissertation was obtained from the primary investigators, Dr. Michael Boivin and Dr. Judith Bass. !77Table 7 Summary of Research Questions, Variables, and Analyses Research QuestionsVariablesMethod of Analysis1. Do nutrition and health education and caregiver-child interac-tion interventions im-prove sustained attention in children with HIV? 2. How does the length of the nutrition and health education and caregiver-child interac-tion interventions affect changes in sustained at-tention in children with HIV? Fixed effects:  Treatment Baseline ECVT  Time Time comparison Age Sex Exposure to TV/video Use of translator Child viral load HAART status Child height Child weight Caregiver depression Caregiver functioning Time x treatment Time comparison x treatment Time x age Time x sex Time x use of translator Time x child viral load Time x HAART status Time x height Time x weight Time x depression Time x functioning Random effects: Geographical cluster Caregiver-child dyads Dependent variables: 6-month ECVT & 12-month ECVTMixed-effects modeling!78Table 7 (contÕd) 3. How are caregiver depression, caregiver functioning, and child viral load related to childrenÕs sustained attention? Caregiver depression Caregiver functioning Child viral load Sustained attention (at baseline) Covariates:  HAART status (on child viral load)Path analysis!79CHAPTER 4: Results Results of ECVT Coding  ECVT scores for each child at each assessment point were calculated using the procedure described in Chapter 3. Please see Table 8 for the means, standard deviations, minimums, and maximums of the ECVT scores for both groups. Generally, there was a trend towards growth in ECVT scores across the yearlong intervention period.   Reliability was calculated using the procedure described in Chapter 3. There was no significant difference between the means of the two samples, t(16) = .10, p =.92.  Sufficient reliability was assumed.  ECVT scores were missing for 8 children at baseline, 14 children at 6 months, and 22 children at 12 months. In many cases, it is clear that the data are missing at random. In five cases, the ECVT video file was present but was corrupt and/or would not work for unknown reasons. In five cases, the video was too dark to accurately and consistently see the eyes of the child. In another five cases, the video file was not in full and was only a portion of the assessment. In one case, the parent overtly and severely interfered with the assessment. In one other case, the child cried excessively for the duration of the video. For 2 other children, ECVT scores were missing at 12 months because the family moved after the 6 month assessment period. In the 25 remaining cases, it is unknown why the ECVT videos are missing except that there appears to be no ECVT file for those children at that time point. For these children, ECVT scores might be missing due to missed assessment appointments for unknown reasons. The randomness of this missing data makes it possible for the mixed effects model to use full !80information maximum likelihood to produce unbiased parameter estimates for missing data points.   Table 8 Descriptive Statistics of ECVT Scores Correlations Among Variables  Once ECVT scores were calculated, it was important to examine the relations among the predictor variables, as well as between the predictor variables and ECVT scores. Please see Table 9 for correlations between predictor variables and sustained attention scores. In relation to child anthropomorphic variables, it is noted that significant medium positive correlations were observed between height and weight and sustained attention scores. Height and weight are markers for overall health, which would suggest that overall health is related to sustained attention skills. Another notable finding is that age was significantly and positively correlated MinimumMaximumM(SD)MISCBaseline.28.91.57(.14)6 months.37.89.62(.12)12 months.47.92.69(.11)TAUBaseline.33.88.65(.12)6 months.25.95.66(.13)12 months.48.94.73(.11)!81with sustained attention scores. Additionally, significant medium to high correlations were found among ECVT scores at the three assessment points. None of these variables were related to caregiver depression or functioning.    In relation to HIV-specific variables, the correlations indicated that there was a significant but weak negative relation between viral load and HAART status, such that being on HAART was associated with lower viral loads. Viral load and HAART status were not related to any of the sustained attention scores. Viral load and HAART status were also not related to any of the overall health measures. Finally, none of the HIV-specific variables were related to caregiver depression or functioning.   ChildrenÕs prior exposure to visual media was not related to childrenÕs sustained attention scores. Likewise, the use of a translator during the intervention was not related to sustained attention scores at six months or the conclusion of the intervention. There was a weak but significant negative relation between use of a translator during the intervention and childrenÕs height, such that if a translator was used, childrenÕs height tended to be lower. Similarly, there was a weak but significant negative relation between the use of a translator and childrenÕs age, such that the use of a translator was associated with younger ages of the children. It is necessary to consider whether these two findings represent spurious relationships. The use of a translator was not related to sustained attention scores or caregiver depression or functioning.  !82Table 9 Correlations among predictor variables and ECVT scores Note. * p <.05. ** p <.01.   1234567891011121Age2Sex-.113Hgt.78**-.19*4Wgt.53**-.25**.82**5Load-.20*.10-.16-.176HAART-.02-.07-.13-.07-.2*7TV.03-.11-.02-.09-.02.0980mo  ECVT.35**-.15.48**.44**-.15-.14.0696mo ECVT.32**-.05.36**.34**-.12-.12-.01.60**1012mo ECVT.24*-.14.28**.35**-.15-.15.07.49**.53**11Dep-.09-.05-.12-.11.10-.06.00.02-.07-.0112Func.04-.02-.03-.07.10.03-.03.12.01.18.1213Trans-late-.19*-.11-.27**-.17.07.00.11-.23-.18-.14.16.09!83Research Question 1: Do nutrition and health education and caregiver-child interaction interventions improve sustained attention in children with HIV?   A piecewise mixed-effects model, specifically, was used to answer this question for two main reasons. First, the gains in ECVT scores were not best described by a traditional linear model because the gains in the first half of the intervention were less than the gains in the second half of the intervention. This can be seen by examining the descriptive ECVT data in Table 8, and in Figure 6 shown below, which shows the slopes of the baseline to six months ECVT scores and six months to one year ECVT scores for both intervention groups without controlling for covariates. The piecewise model allowed the slope of childrenÕs sustained attention scores to shift at the six-month period. Second, a piecewise mixed-effects model provided results that answered Research Question 2, which asked how time affected the growth in ECVT scores. That is, the slope of childrenÕs sustained attention scores from baseline to six months could be compared to the slope of their sustained attention scores from six-months to one year using a piecewise model. !84Figure 6 ECVT scores across the intervention period  The results of the piecewise mixed-effects model are shown in Table 10 and Figure 7. Baseline sustained attention scores were significantly different between TAU and MISC groups (p <.01), such that TAU had higher sustained attention scores at baseline (M = .65, SD =.12) than MISC (M = .57, SD = .14). None of the demographic or control variables significantly predicted sustained attention growth. Both MISC and TAU groups made significant gains in sustained attention skills across the intervention period (p < .05). There was no significant difference between the sustained attention growth made by the MISC group as compared to the TAU group.  !85Table 10 Results of Mixed-Effects Model EffectEstimatetdfpIntercept6.71E-0120.488.287E+01<2e-16Months1.031E-03.271.468E+02.78Months.lag9.883E-032.051.629E+02.04*Age-5.384E-03-.269.388E+01.79Sex-1.063E-02-.459.725E+01.65Watchtv1or2-1.273E-02-.702.638E+02.48Watchtvmany-4.782E-03-.282.531E+02.78Translator-5.623E-04-.028.334E+01.98Load-5.921E-03-.909.241E+01.37Onhaart-7.175E-03-.299.864E+01.78Height4.733E-031.789.698E+01.08Weight4.052E-03.629.358E+01.54Depression7.367E-04.459.494E+01.65Functioning-1.42E-03-.661.048E+02.51Arm-8.523E-02-3.003.896E+01.00**Months x Age1.701E-04.091.449E+02.93Months x Sex5.154E-05.021.484E+02.98Months x Translator-1.136E-03-.451.126E+02.66Months x Load-1.351E-04-.221.437E+02.83Months x Onhaart4.508E-031.831.445E+02.07Months x Height-4.565E-04-1.801.536E+02.08Months x Weight5.560E-04.881.572E+02.38Months x Depression-6.078E-05-.391.496E+02.70Months x Functioning-1.500E-04-.761.422E+02.45!86Note. *p<.05. **p<.01.  Figure 7 Predicted ECVT scores Table 10 (contÕd) Months x Arm3.052E-03.721.451E+02.48Months.lag x Arm-3.380E-03-.491.618E+02.63!87Research Question 2: How does the length of the nutrition and health education and caregiver-child interaction interventions affect changes in sustained attention in children with HIV?  Please see Table 10 for the results of the piecewise mixed-effects model. The results indicate that statistically significant growth occurred between six and twelve months (p < .05); however, the growth from baseline to six months is not statistically significant. Because there was no significant difference between the slopes of the two intervention groups, the slope of growth during the second half was compared to the slope of growth during the first half of the intervention for the entire sample in order to understand the how the slope from the second half of the intervention is different from that of the first half. The slope of growth during the second half of the intervention is approximately 2.5 times greater than the slope of the growth made during the first half of the intervention. At six months, the predicted sustained attention score for TAU is .067 above that of a child in MISC with 0s as values on the covariates. At twelve months, the predicted sustained attention for a child in TAU is .069 above that of a child in MISC with 0s as values on the covariates.   In order to determine whether there was a meaningful difference between the growth of ECVT scores in TAU and MISC, CohenÕs d of model-implied ECVT scores between the TAU and MISC groups at 6 and 12 months was calculated. CohenÕs d at 6 months was .14 and CohenÕs d at 12 months was .15. Although TAU scored higher than MISC on average, the MISC group experienced slightly greater (but not significant) growth in sustained attention than TAU. As such, these effect sizes indicate that a participant in MISC with covariate scores of 0 would !88be expected to have a higher ECVT score than a participant in TAU with the same characteristics by roughly one tenth of a standard deviation. Research Question 3: How are caregiver depression, caregiver functioning, and child viral load related to childrenÕs sustained attention?  All fit indices indicated good overall model fit. The Chi-square statistic was not significant (p = .35). This result indicates good model fit because p is greater than .05 (Hooper et al., 2008). RMSEA was the second fit indicator to be used in this study. An RMSEA of 0 indicates perfect fit and an RMSEA of .06 or less is considered good fit (Hooper et al., 2008). The results of the path analysis indicated that RMSEA was 0.03 for the present study. As such, this fit indicator also suggests that there was good overall model fit. The final fit indicator used in the present study was CFI. CFI values greater than .90 are necessary to ensure that incorrect models are not accepted. The CFI for the proposed model was 0.94. Because the fit indices indicated good overall model fit, the specific paths of the proposed model can be interpreted.   None of the paths in the proposed model were significant. However, whether the child was prescribed HAART medication was included as a covariate for child viral load and being on HAART medication significantly predicted childrenÕs viral loads (p = .04). This finding indicates that the overall proposed pathways of the conceptual model do not explain baseline performance in sustained attention for children with HIV. See Table 11 and Figure 8 below for the results of the path analysis. !89Table 11 Results of Path Analysis PredictorOutcomeTypeMediatorEstimateSEzpChild viral loadChild sustained attentionDirect-.01.01-1.40.16Caregiver functioningChild viral loadDirect-.37.32-1.17.24Caregiver functioningChild sustained attentionDirect-.04.02-1.62.11Child OnHaartChild viral loadDirect (covariate)-.75.36-2.07.04*Caregiver functioningChild sustained attentionIndirectChild viral load.01.01.84.40Caregiver functioningChild sustained attentionTotal-.03.02-1.45.15Caregiver depressionCaregiver functioningDirect-.13.07-1.83.07Caregiver depressionChild sustained attentionIndirectCaregiver functioning.01.001.30.19Caregiver depressionChild sustained attentionIndirectCaregiver functioning x Child viral load-.00.00-.76.45!90Note. *p < .05. Table 11 (contÕd)Caregiver depressionChild sustained attentionTotal.00.001.23.22!91 Figure 8 Estimates of Paths in Conceptual Model !92CHAPTER 5: Discussion  Extant literature suggests that children with HIV may be at risk for, and experience difficulties with, sustained attention (Ruel et al., 2012; Watkins et al., 2000). To date, there are no known interventions that are effective at improving the sustained attention skills of young children with HIV. The present study aimed to address this issue by examining the efficacy of two interventions that theoretically could support the development of sustained attention skills. In addition to examining the efficacy of two interventions to improve sustained attention, the present study also attempted to better understand how caregiver and child factors interact to influence the sustained attention skills of children with HIV. The following discussion will elaborate on the results of this study by order of research question.  Effects of MISC and TAU  Sustained attention growth. Research Question 1 asked whether MISC and TAU were associated with improvements to the sustained attention skills of young children with HIV. It was first hypothesized that improvements in sustained attention would be observed in both intervention groups. The findings of the present study indicated that sustained attention skills increased significantly across the intervention period for children in both intervention groups. The finding that both groups made significant gains makes theoretical sense when considered in relation to the existing literature on factors influencing the general neurocognitive development of children.   According to the body of literature concerned with child development, several universal factors are considered essential for positive neurocognitive growth and development. These !93include overall health, access to adequate nutrition, and positive and effective caregiver-child interactions (Irwin et al., 2007). MISC targeted caregiver-child interactions by teaching caregivers how to respond to their children in developmentally appropriate ways and how to engage in cognitively stimulating interactions (Klein, 2003). TAU addressed issues related to health and nutrition by educating caregivers on HIV and AIDS, how to care for those who are ill, important aspects of nutrition, and how to feed and promote health in infants and young children, among others. The means by which these factors promote childrenÕs neurocognitive development can be understood in terms of GottliebÕs (1991) experiential canalization theory.  Experiential canalization, which refers to the complex interaction of biological and environmental factors supporting development (Gottlieb, 1991), can be used to understand the complex mechanisms by which TAU and MISC interventions could promote childrenÕs sustained attention. The following are examples of how TAU and MISC could lead to gains in sustained attention given the experiential canalization theory and relevant empirical literature. TAU could theoretically lead to improvements in sustained attention by enhancing caregiversÕ knowledge of how to properly feed a young child, thereby increasing the nutrition status of the child, which would provide necessary resources for brain growth and development. Evidence suggests that malnutrition during early life periods is negatively associated with brain weight (Brown, 1965) and that malnutrition may prevent healthy brain cell division (Winick & Rosso, 1969) and myelination (Fish & Winick, 1969). Certain compounds present in various types of food support these different aspects of brain development. For example, DHA3 fatty acids are necessary during early childhood brain development because of their roles in building brain cell membranes and apoptosis (Innis, 2007). In sum, TAU could have a relatively direct influence on !94childrenÕs neurocognitive development by improving caregiversÕ knowledge of nutrition, which improves the nutrition of the children.   TAU could theoretically also promote sustained attention growth through more complex biological and environmental interactions. For example, caregivers in TAU may learn how to better care for their own health and mental health needs. Research suggests that impaired caregiver health can have negative behavioral and health outcomes for children (World Health Organization, 2004), particularly those with chronic illnesses (Kelly & Hewson, 2000). These negative outcomes observed in children of caregivers who have their own health challenges may be a result of poor parenting practices (World Health Organization, 2004). Thus, improving caregiversÕ health and well-being may result in caregivers providing higher quality nutrition, healthcare, and cognitive stimulation to their children (World Health Organization, 2004), which is known for influencing neurocognitive development (Irwin et al., 2007). Although there is no evidence to suggest that these are actual mechanisms by which TAU specifically influences sustained attention, these two examples are possible pathways by which TAU could have resulted in improvements to childrenÕs sustained attention considering relevant theory and existing literature.   Similar to TAU, the MISC group may have experienced gains in sustained attention because of a complex interaction of biological and environmental variables. MISC could have theoretically resulted in improvements to sustained attention by teaching caregivers how to interact with their children in developmentally appropriate ways and in ways that promote positive attachment, learning, and cognitive stimulation. MISC may have taught caregivers ways !95to interact with their children that promote positive attachment, such as by smiling and vocalizing to the child. Positive attachment between caregiver and child is associated with higher overall cognitive performance in young children, as well as with higher performance in certain cognitive domains such as language (Main, 1983). Attachment can influence cognitive development because poor attachment may result in low levels of exploration and engagement in tasks by children, as well as poorer communication skills and low levels of assistance by caregivers (OÕConnor & McCartney, 2007). Low exploration and task engagement has been found to influence cognitive skills because children learn from their interactions with things and others (Carpenter, Nagell, & Tomasello, 1998; Schunk, 1985). The learning that occurs when children interact with their environments is a biological process involving the creation and strengthening of connections between neurons (Bear, Connors, & Paradiso, 2006). It is also possible that teaching caregivers the MISC lessons related to sustained attention may simply result in caregivers directly instructing their child on how to focus and regulate. The exact pathways and mechanisms by which MISC could result in improvements to neurocognitive development are unknown; however, the experiential canalization model suggests that it could be through a complex chain of environmental, behavioral, and biological interactions.  Similar improvements of MISC and TAU. The second hypothesis of Research Question 1 was that MISC would be associated with greater gains in sustained attention than TAU. This was hypothesized because previous studies examining the effects of health, nutrition, and various cognitive stimulation interventions found that health and nutrition interventions occasionally benefited child development, but interventions involving child-caregiver interactions or similar cognitive stimulation consistently benefited child development !96(Grantham-McGregor et al., 2007). Additionally, studies have found that interventions involving cognitive stimulation produced an added effect on childrenÕs development above and beyond the effect of nutrition and health care (Grantham-McGregor et al., 2007). Because children in the MISC group received nutritional food packages, it was hypothesized that the combined effect of nutrition and child-caregiver interventions would result in greater improvements to sustained attention than just the interventions for health and nutrition. The findings of this study, however, were such that MISC and TAU experienced similar levels of growth in sustained attention. Four plausible explanations for the findings of similar growth in sustained attention are presented next.   The simplest explanation for similar growth being observed in both groups is that the health and nutrition intervention resulted in similar gains in sustained attention as this particular caregiver-child interaction intervention for this particular population. That is, the possible mechanisms for the growth in MISC and TAU, which are described in the section above, may be different, but the effects on sustained attention may appear the same. Evidence for this explanation is that in previous studies of HIV-infected and HIV-exposed children, MISC was associated with greater gains in specific cognitive areas as compared to TAU. In a study of HIV-infected children, positive growth was observed in all cognitive areas examined, but MISC was associated with greater gains in visual perception and memory (Boivin et al., 2013b). Likewise, in a study of HIV-exposed children, positive growth was observed in nearly all cognitive areas examined, but MISC was associated with greater gains than TAU in expressive and receptive language (Boivin et al., 2013a). This suggests that MISC and TAU have both been associated with positive neurocognitive growth and that MISC has demonstrated it has a stronger influence !97only on certain skills. Given that both groups in the present study made significant gains, sustained attention may be another area in which MISC and TAU result in similar improvements.   Another possibility is that the positive growth in sustained attention observed in this study represents typical development of sustained attention skills and is not a result of the interventions. Support for this explanation may come from the correlational analyses that indicated that age, height, and weight, which are indicators of overall development, were the only variables that were significantly related to ECVT scores over time. It may be that sustained attention skills were developing as a result of overall maturation. Additionally, attention development is thought to be nonlinear and rather to develop as specific brain regions supporting specific attention skills develop (Halperin & Schulz, 2006). At first, this seems to align with the finding that significant growth was observed during the second half of the intervention. However, because the children in this study were different ages, it seems unlikely that they would be experiencing growth in the same brain regions that support sustained attention in the same six-month period.   It is also possible that the sustained attention growth observed in this study was due to the nutritional supplements provided to all caregivers. Because both intervention groups experienced similar growth in sustained attention skills, it seems possible that the nutritional packages that were the commonality in both groups is what accounts for the growth observed. Evidence for this explanation is that access to adequate nutrition is a challenge experienced by many families with HIV (Weiser et al., 2012), and receiving adequate nutrition promotes childrenÕs global neurocognitive development (Irwin et al., 2007). However, some evidence suggesting that this !98may not be the source of growth is that, on average, the children in this study appeared to already be at broadly normal heights and weights for children of their age and gender according to height and weight growth standard tables (World Health Organization, 2015). It is unclear from the literature whether additional nutritional supplements would further promote neurocognitive development.  Finally, another plausible explanation for the growth observed in both intervention groups is that the MISC modules that were theorized to directly influence childrenÕs sustained attention skills, focusing, regulating, and encouraging, were not provided to caregivers as often as the other modules.  Regulating was only planned approximately 12 times on average, focusing was planned approximately 18 times on average, and encouraging was planned approximately 11 times on average. In fact, regulating was the second to last most commonly planned lesson out of all the MISC lessons, focusing was the fifth to last most, and encouraging was the least most commonly planned lesson. Therefore, the lack of greater growth in the MISC group may be due to the limited number of times the components of MISC relating to attention were given. Given that MISC trainers were to provide the lessons that they determined necessary based on observing the caregivers, it is important to consider whether MISC trainers did not provide these lessons more because they did not observe problems with caregiversÕ behaviors regarding regulation or focusing or if MISC trainers were not fully attuned to noticing problems with these behaviors in caregivers. Because there is no data to indicate that those two lessons actually support sustained attention, it is difficult to determine whether this is in fact the cause of the growth observed. !99Altogether, the data point to multiple possible explanations for observing similar growth in the intervention groups. Some explanations, however, seem more solid than others. The two explanations with the most support appear to be that MISC and TAU have similar effects on sustained attention and that the MISC lessons thought to directly influence sustained attention may not have been provided as often as necessary to lead to significantly greater gains than TAU.  While more research is necessary in order to determine the exact nature of the observed similarities in MISC and TAU, examining the findings related to time may help elucidate the reasons for the finding.  Effect of Time  The findings of this study suggest that time may be an important factor to consider when thinking about interventions for sustained attention. The results of the mixed-effects model indicated that one year of MISC and TAU led to significantly greater gains than six months of intervention. In fact, while both groups made gains during the first six months of the yearlong intervention period, the sustained attention scores at six months were not significantly different from the sustained attention scores at baseline. These findings are somewhat consistent with hypothesis 2a and 2b and align with the existing research on other health, nutrition, and cognitive stimulation/caregiver-child interaction interventions. Other studies have found that longer intervention durations lead to more cognitive benefits (McKay et al., 1978). Furthermore, a two-year longitudinal study of stimulation and nutritional interventions for children with stunted growth showed that the interventions were associated with improvements in cognitive skills throughout the entire two-year intervention period (Grantham McGregor et al., 1991). In sum, !100this finding suggests that receiving one year of MISC and TAU may be necessary for significantly improving the sustained attention skills of children with HIV.   Although the findings of the present study align with the current research such that one year of intervention was associated with more growth than six months of intervention, they raise the question as to why significant growth occurred during the second half of the intervention but not during the first. The data suggest a few possible explanations. One possible explanation is that the improvements observed in the second half of the intervention were related to the accumulation of intervention knowledge. Other studies have found that those who participate in similar types of interventions for longer receive greater benefits (Grantham-McGregor et al., 1991; McKay et al., 1978). As such, it may be that the knowledge gained in the first half of  MISC and TAU was not enough to produce statistically significant results in time for the six month intervention, but that the accumulation of knowledge over time was enough to produce statistically significant results.  A similar hypothesis is that there may have been a lag effect, such that time was needed for caregivers to implement the interventions with their children and for those interventions to result in observable and significant neurocognitive changes. There is little evidence for this hypothesis though because other studies of nutrition, health, and cognitive stimulation interventions have been associated with significant, positive cognitive changes after six months with a variety of compromised populations (e.g., Gardner et al., 2005; Lozoff et al., 2010; Nahar et al., 2012 ). Furthermore, it should not be the case that sustained attention would take longer to respond to intervention because it has been argued that sustained attention is necessary for the !101development and functioning of other cognitive skills (Mahone & Schneider, 2012). This suggests that six months of intervention could result in changes to neurocognitive performance if the interventions are targeting the correct areas and are being carried out with fidelity.   A final possible explanation is that the lessons that would theoretically influence sustained attention most directly in both interventions were generally not covered until the second half of the intervention period. When conceptualizing the possible effects of MISC and TAU using experiential canalization, it is conceivable that most lessons of both interventions could result in improvements to childrenÕs neurocognitive development via many indirect pathways. However, when thinking about the lessons that would seem to have the most direct effect, those related broadly to child development in TAU and the lessons on regulating, focusing, and encouraging in MISC seem to be the most influential. An analysis of each caregiverÕs training sessions revealed that caregivers in TAU generally began receiving lessons related to child development, including feeding, nutrition, and health specifically for children, in the seventh month of intervention. Likewise, caregivers in MISC received the lesson on focusing starting in the third month of the intervention and the lesson on regulating during the fifth month on average. Although there is no evidence that these particular lessons are actually related to sustained attention at all or more so than the other lessons, it seems logical that they might be related to sustained attention development in children and could account for the growth observed during the second half of the intervention.   In sum, there are a few potential explanations for why significant growth was observed during the second six months of intervention and not during the first. The exact cause, however, !102remains unclear and requires additional research. The finding does highlight the importance of considering intervention duration when providing health, nutrition, and stimulation interventions to children in this population.  Factors Influencing Sustained Attention in Children with HIV  In addition to examining the effects of MISC and TAU on sustained attention, a primary aim of this study was to examine the interaction of caregiver and child variables as they relate to the sustained attention skills of children with HIV.  The purpose of this research question was to attempt to better understand how environmental, behavioral, and biological variables interact to produce childrenÕs sustained attention skills before intervention. This aim was addressed through a path analysis examining the relations among child and caregiver variables. Although there was good overall model fit, the results of the path analysis indicated that the pathways of the proposed model do not explain childrenÕs sustained attention skills prior to intervention.  Before attempting to make sense of why the data did not result in any significant paths, it may be important to address why there was good overall model fit but no significant paths. The Chi-square test first uses the proposed path analysis model to produce an expected covariance matrix. It then compares the expected matrix to the observed matrix to determine if they are significantly different (Hooper et al., 2008). Good fit is when there is no significant difference between the two. In the present study, the observed covariance matrix showed no significant relations among the variables and the expected covariance matrix produced by the path analysis also showed little relations among the variables. Therefore, there was consistency between the covariance matrices and good fit based on the Chi-square index. RMSEA was the second fit !103index used. It compares the proposed model covariance matrix to the unknown population covariance matrix (Hooper et al., 2008). Again, good fit was found because of the similarity between the proposed model data and the population covariance matrix. CFI was the final fit index used in the present study. CFI compares the correlations between variables in the proposed model to an independence model in which there are no correlations (Hooper et al., 2008). Because there were no correlations in the proposed model, there was good fit between the correlations between proposed variables and the independence model. In sum, the fit indices found that the proposed model data were similar to the observed or unknown population data that indicated no significant relations among variables, which resulted in good overall fit. This allows for interpretation of the pathways in the model. In this case, because no paths were significant, it is important to explore why there were no relations among the variables.   Measurement. One reason why the paths were not significant may be that there were issues with the measures and reporting on the measures. As previously discussed, the depression scale was adapted from the original HSCL-25. It was translated for the participants and the response items were changed slightly. Additionally, there are no published reliability and validity studies on the HSCL-25 with this particular population. Therefore, it is reasonable to question whether this assessment is measuring depression in a reliable and valid way with this population. The caregiver functioning assessment has similar problems. It was developed from a qualitative study by the investigators of the original study from which the present data were drawn. There are no systematic psychometric studies detailing the reliability and validity of this measure. While it was hoped that the measure would reflect culturally-relevant behaviors of caregivers and assess their level of functioning related to daily living, the data from this measure indicate !104that caregivers generally reported relatively high functioning. This seems unusual given the endorsement of depression symptoms because depression symptoms often correlate with difficulties with functioning, a finding that has been observed in populations of adult individuals with HIV (Jin et al., 2006; Kamat et al., 2012). This raises questions about whether the measure was culturally-sensitive enough and whether caregivers were not comfortable with responding honestly to the measure.  Functioning and caregiving. The connection between adult depression and functioning has been clearly established in the literature. The connection between caregiver depression and parenting practices has also been examined. Studies have found that depressed mothers provide fewer stimulating experiences and lower quality stimulation to their children (Bettes, 1988). Depressed mothers also respond less and respond at a lower speed to their children (Livingood, Daen, & Smith, 1983). Impaired caregiving related to discipline and knowledge about relevant practices has also been found specifically within the HIV population (Kotchick et al., 1997; Murphy et al., 2010 ). What has not been addressed is the link between caregiver functioning and childcare practices. It was hypothesized that caregiver functioning would be related to child variables and outcomes in the present study, but the findings seem to suggest that caregiver functioning is not related to child outcomes. But, again, it is necessary to take into account the concerns regarding the caregiver functioning assessment. More research is needed to better under the relation between these variables.   Method of data collection. Children with HIV have been found to demonstrate difficulties with sustained attention (Ruel et al., 2012; Watkins et al., 2010). It has been suggested that the virus itself, as well as the immune system response to the virus, causes injuries !105to areas of the brain involved in sustained attention (Chang et al., 2001; Ellis et al., 2009). The injury is what causes observable behavioral difficulties with sustained attention. In general, the more prevalent the virus, the greater the disease severity and injury (Valcour, Sithianamsuwan, Letendre, & Ances, 2011) and the likelihood of developing neurocognitive disorders (Ellis et al., 2009). When these findings are taken into consideration, the lack of significant relations between viral load and sustained attention skills in the present study may also seem surprising. Those with higher viral loads did not necessarily have lower sustained attention scores. Likewise, those with low viral loads did not necessarily have higher sustained attention scores. However, this may be because viral load in the present study represented childrenÕs viral load at one specific point in time. Viral loads can change relatively quickly in response to medication and it takes time for the virus present in the body to result in neurocognitive changes. As such, there may have been no correlation between sustained attention scores and viral loads because the viral loads at baseline may not be an accurate representation of the childrenÕs viral loads over time. It may be more beneficial to measure viral load over an extended period of time and examine how those data relate to sustained attention scores.  Although the pathways of the proposed conceptual model do not appear to describe childrenÕs sustained attention skills, the correlations conducted as preliminary analyses may help to explain their baseline skills. Significant medium positive correlations were observed between height and weight and sustained attention scores. Height and weight are markers for overall health, which would suggest that overall health is related to sustained attention skills. The findings that overall health and development may be important for the sustained attention skills of children with HIV is consistent with existing literature that suggests that overall nutrition and !106health are important factors for supporting the neurocognitive development of children (Irwin et al., 2007). Several previous studies have found that weight and height, used as indicators of health and nutrition, are related to performance on cognitive assessments completed by preschool and kindergarten children (Karp, Martin, Sewell, Manni, & Heller, 1992). The finding that age was significantly and positively correlated with sustained attention scores is also consistent with existing literature suggesting that sustained attention skills increase with typical development in childhood (Mahone & Schneider, 2012). Together, these findings may suggest that overall health and development may be the most important factors contributing to these young childrenÕs sustained attention skills.   As discussed in the review by Abubakar and colleagues (2008), there is a lack of research examining both the biological (i.e., viral load, immune system functioning, disease progression, nutritional status) and psychosocial (i.e., caregiver depression, caregiver behavior, home environment) variables affecting the neurodevelopment of children with HIV. This study was the first known study to examine the interaction of these variables to this degree. However, this research suffers from the most consistent problem identified in the body of literature examining the neurodevelopment of children with HIV in other countries: a lack of reliable and valid measures for the relevant population. This limitation, among others, will be discussed next.    Limitations and Future Research  The present study was the first known study to examine the effects of a caregiver-child interaction intervention and a health and nutrition education intervention on the sustained attention of children with HIV. The study suggests that both interventions may have positive effects on sustained attention and that twelve months of intervention, but not six months, was !107associated with significant improvements to sustained attention. This study also aimed to examine how caregiver and child variables interact to influence childrenÕs sustained attention skills prior to intervention. The results suggested that there were no relations among caregiver depression, caregiver functioning, childrenÕs viral load, and childrenÕs sustained attention. The limitations of the present study are important to consider in light of these findings. They also provide potentially informative directions for future research.   Control group. A major limitation is the lack of control group. Without a control group who received no intervention, it is impossible to determine whether the gains observed in sustained attention in both intervention groups reflect growth that is likely the result of the interventions. This would have enhanced the clinical utility of this research. At this point, future research is needed to determine the exact cause of the growth observed in the children of this study. It is clear that future research should include a control group that receives no intervention to help disentangle growth in sustained attention from normal development from growth in sustained attention that may be due to MISC or TAU.   Unknown psychometric properties. Another limitation of this study is related to the assessments that were used, particularly the caregiver depression measure, caregiver functioning measure, and the ECVT. The depression assessmentÕs response items and scoring system were altered from their original forms in the HSCL-25 and there is no published research examining the reliability and validity of the original or adapted HSCL-25 with this particular population. The depression assessment was also translated to be administered to the caregivers in this study and it is unknown how this translation may have affected the assessment results. Additionally, !108the caregiver functioning assessment was developed specifically for the original study from which this studyÕs data were drawn. There is no published reliability and validity information on this assessment. Finally, the ECVT is a relatively new assessment with relatively little published data regarding its reliability and validity, and there is no reliability and validity data for the cultural group included in the present study. The ECVT was also used in this study with children who were slightly older than the children on whom it was initially tested and whose results are included in the manual (Romero et al., 2012) and in the published study that provides psychometric data (Zelinsky Goldman et al., 2004), which means that there is no existing data to demonstrate that the ECVT measures sustained attention or is reliable for all children in the present studyÕs age range. In sum, while the tools used in this study were selected because of their potential usefulness and cultural sensitivity to this Ugandan population (see Methods chapter), there is little empirical data to support their reliability and validity with the present studyÕs participants. As such, the findings of this study need to be interpreted with caution. It would also be helpful for future research in this area to dedicate efforts toward establishing the psychometric properties of the assessments to be used with this international population.   Intervention fidelity. A third major issue with the present study concerns intervention fidelity. An analysis of the fidelity data revealed that the TAU intervention was provided in such a way that participants in TAU received differing amounts of coverage of each lesson. Some caregivers never received certain lessons, and some caregivers received certain lessons many times. Caregivers also received lessons at different times in the intervention period. This makes it difficult to ascertain how the TAU intervention may have contributed to the growth observed in sustained attention in this group. Another issue to consider is that there was little coverage of the !109lesson on regulating in the MISC group. MISC trainers were instructed to provide the lessons that were relevant to the caregivers as determined by observations of the caregiver and child. However, regulating was covered with most caregivers towards the end of the yearlong intervention. It raises a question about why this may be. It is possible that MISC trainers did not observe difficulties with caregivers supporting childrenÕs self-regulation or that MISC trainers were not skilled in noticing and evaluating those behaviors. Given the consistency of MISC lesson coverage that was observed in the fidelity data, it may also be possible that trainers were not teaching MISC lessons in response to observed interactions, but rather to a fixed schedule. Regardless of the reason why regulation was not taught until the end of the intervention period, the fidelity data suggest that MISC caregivers did not receive as much coverage of this lesson that could theoretically influence childrenÕs sustained attention as other lessons. In future research examining the effects of these interventions, it may be beneficial to adhere to intervention fidelity to ensure that conclusions can be drawn about the effects of the interventions. Intervention fidelity may also be better assessed in future studies by having trainers record what was actually covered during session, rather than what was planned and the extent to which that goal was accomplished.  In a similar vein, the finding that the slope of sustained attention growth was greater during the second half of the intervention in which certain intervention lessons tended to be covered highlights the need for future research to systematically evaluate which components of the interventions are supporting sustained attention skills. That is, another question to be answered is how much, if any, sustained attention growth is due to specific lessons such as MISCÕs regulating and focusing and TAUÕs child development, and how much, if any, is related !110to enhancing more global factors such as better nutrition, greater health care, emotional attachment between caregiver and child, and higher quality cognitive stimulation. This information would likely have many clinical implications for improving the sustained attention of children with HIV.   Another issue in the present study related to intervention integrity is intervention acceptability. There are no data regarding the caregiversÕ feelings or thoughts about the interventions they received. This missing information is important because intervention acceptability is related to how likely interventions will be implemented (Perepletchikova & Kazdin, 2005). In this case, it is important to consider the extent to which caregivers actually implemented the interventions with their children outside of the intervention training sessions. Future research should include a measure of intervention acceptability, and also consider including a report of caregiver implementation.  Longitudinal data related to HIV and sustained attention. As previously discussed, the exact relation between HIV and sustained attention in young children remains unclear. The present study utilized viral load data collected at one time point, whereas a more appropriate method for understanding the relation between HIV and sustained attention in children may be to collect HIV disease indicators and sustained attention data over prolonged periods of time. In general though, much more research is needed to better understand the biological and environmental factors involved in sustained attention difficulties of children with HIV, as well as the interaction between them. !111Clinical Implications  There are possible clinical implications of this research. First, the findings of this study are such that both intervention groups saw improvements in sustained attention. While it is a limitation of this research that there was no control group with which to compare normal development, the results of this study suggest that these interventions may promote growth in sustained attention skills for children with HIV. If further research supports the conclusion that these interventions result in improvements to sustained attention, then it may be useful to provide health and nutrition and caregiver-child interaction interventions to caregivers of children with HIV in order to promote childrenÕs sustained attention skills. Another possible implication of this research is that these interventions may be useful for children and families with similar chronic medical conditions because health, nutrition, and caregiver-child interactions are considered universal protective factors that support neurocognitive development. Conclusion  The present study examined the effects of two caregiver interventions on the sustained attention skills of young Ugandan children with HIV. One caregiver intervention was designed to address caregiver-child interactions and the other targeted caregiverÕs knowledge about global health and nutrition. Additionally, this study attempted to examine the relation among caregiver factors, child factors, and childrenÕs sustained attention skills prior to intervention. Positive growth in sustained attention was observed in both intervention groups. However, there was no statistically significant difference in growth between the two intervention groups. Finally, the !112proposed conceptual model for how child and caregiver variables interact to influence sustained attention scores was unsupported. An important finding in this study was related to time. That is, children did not make statistically significant growth in the first six months of the interventions, but they did make statistically significant growth in the second six months of the intervention. Given the high number of children around the world living with HIV and the importance of sustained attention skills in daily functioning and academic achievement, it is important to find ways to support the development of this skill. While it remains unclear whether the growth observed in the present study was due to the interventions or to other factors discussed above, the results of this study suggest that additional research should be done to explore the possible benefits of these interventions. Because there are no known interventions for supporting the sustained attention skills of children with HIV, the findings of this study and the future research that is needed could have important clinical implications.  !113APPENDIX !114MISC Monitoring Form Trainer:_____________ Study Participant ID: ______________ Date: _________ "M01. Training site  ___ 1. Office  ___ 2. Home"M02. Child present  ___ 0. No ____ 1. Yes"M03. Video used  ___ 0. No ____ 1. Yes M04. Language of training: ______________________       M05. Translator involved   ___ 0. No ____ 1. Yes M06. Were you able to do MISC training today: __  0. No  __ 1. Yes  M06a.  If No, Why: ___ 1. Caregiver absent     ___ 2. Caregiver too busy     ___ 3. Caregiver too sick     ___ 4. Caregiver depressed Ð could only provide counseling     ___ 5. Child absent ___ 6. Child too sick ___ 7. Trainer unable to make it (any reason) M07. Which MISC Components were planned for today: M07a. Regulating    __  0. No  __ 1. Yes"M07b. Rewarding   __  0. No  __ 1. Yes"M07c. Expanding   __  0. No  __ 1. Yes"M07d. Affecting    __  0. No  __ 1. Yes M07e. Focusing                 __  0. No  __ 1. Yes M07f. Touch     __  0. No  __ 1. Yes M07g. Turn-Taking                __  0. No  __ 1. Yes M07h. Physical Closeness             __  0. No  __ 1. Yes M07i. Eye Contact   __  0. No  __ 1. Yes M07j. Smiles     __  0. No  __ 1. Yes"M07k. Vocalizations    __  0. No  __ 1. Yes"M07l. Sharing of joy   __  0. No  __ 1. Yes M08. Did you cover your goals: ___ 1. None ___ 2. A little ___ 3. Most ___ 4. All M09. How engaged was the caregiver: ___ 1. Not at all ___ 2. A little        ___ 3. Moderate  ___ 4. Very   M09a. Reasons for low (1-3) engagement: ___ 1. Depression/Mental Health ___ 2. Lack of interest         ___ 3. Sickness !115        ___ 4. Other, specify            _______________ "M10. How engaged was the child: ___ 1. Not at all  ___ 2. A little ___ 3. Moderate ___4. Very "___ 5. Not applicable: training in office"___ 6. Not applicable: child absent/sick"  M10a. Reasons for low (1-3) engagement: ___1.  Shyness         ___ 2. Fussy/crying ___ 3. Child is hungry "___ 4. Naughty/behavior problems         ___ 5. Sickness ___ 6. Absent"___ 7. Mother fails to engage the   child   ___ 8. Other, specify_____________                                                                                   M11. Was there interference by others:  ___ 0. No  ___ 1. Yes  M11a. If Yes:  ___ 1. Other adults   ___ 2. Other children ___ 3. Both M12. Did the participant have suicidal thoughts:   ___ 0. No ___ 1. Yes  M12a. If yes, what level of risk did you assess:  ___ 1. Low ___2.  Moderate           ___3.  High  M12b.  Did you set up a safety plan:   ___ 0. No ___ 1. Yes  M12c. Did you refer to the mental health center ___ 0. No ___ 1. Yes  M12d.  Did you contact Noeline/Paul:   ___ 0. No ___ 1. Yes  Plans (please describe):""M13. Is there any reason to suspect domestic violence   ___ 0. No ___ 1. Yes  M13a. Did you refer the caregiver to a support service  ___ 0. No ___ 1. Yes  Plans (please describe):  !116UCOBAC Monitoring Form Trainer:_____________ Study Participant ID: ______________ Date: _________ "M01. Training site  ___ 1. Office  ___ 2. Home"M02. Child present  ___ 0. No ____ 1. Yes M03. Language of training: ______________________       M04. Translator involved   ___ 0. No ____ 1. Yes M05. Were you able to do UCOBAC training today: __  0. No  __ 1. Yes  M05a.  If No, Why: ___ 1. Caregiver absent     ___ 2. Caregiver too busy     ___ 3. Caregiver too sick     ___ 4. Caregiver depressed Ð could only provide counseling     ___ 5. Child absent ___ 6. Child too sick ___ 7. Trainer unable to make it (any reason) M06. Which UCOBAC lessons were planned for today (please describe): _______________  M07. Did you cover your goals: ___ 1. None ___ 2. A little ___ 3. Most ___ 4. All M08. How engaged was the caregiver: ___ 1. Not at all ___ 2. A little        ___ 3. Moderate  ___4. Very   M08a. Reasons for low (1-3) engagement: ___ 1. Depression/Mental Health ___ 2. Lack of interest         ___ 3. Sickness         ___ 4. Other, specify ____________ M09. How engaged was the child: ___ 1. Not at all  ___ 2. A little ___ 3. Moderate ___4. Very "___ 5. Not applicable: training in office"___ 6. Not applicable: child absent/sick"  M09a. Reasons for low (1-3) engagement: ___1.  Shyness         ___ 2. Fussy/crying ___ 3. Child is hungry "___ 4. Naughty/behavior problems         ___ 5. Sickness ___ 6. Absent"___ 7. Mother fails to engage the child  ___ 8. Other, specify_______________   !117                                                                                               M10. Was there interference by others:  ___ 0. No  ___ 1. Yes  M10a. If Yes:  ___ 1. Other adults   ___ 2. Other children ___ 3. Both M11. Did the participant have suicidal thoughts:   ___ 0. No ___ 1. Yes  M11a. If yes, what level of risk did you assess:  ___ 1. Low ___2.  Moderate           ___3.  High  M11b.  Did you set up a safety plan:   ___ 0. No ___ 1. Yes  M11c. Did you refer to the mental health center ___ 0. No ___ 1. Yes  M11d.  Did you contact Noeline/Paul:  ___ 0. No ___ 1. Yes  Plans (please describe):""M12. Is there any reason to suspect domestic violence   ___ 0. No ___ 1. Yes  M12a. Did you refer the caregiver to a support service  ___ 0. No ___ 1. Yes  Plans (please describe): !118REFERENCES !119REFERENCES Abikoff, H., Courtney, M., Pelham, W. E., & Koplewicz, H. S. (1993). TeachersÕ ratings of disruptive behaviors: The influence of halo effects. Journal of Abnormal Child Psychology, 21, 519-533. doi: 0091-0627/93/1000-0519$07.00/0 Abubakar, A., Van Baar, A., Van de Vijver, F. J. R., Holding, P., & Newton, C. R. J. C. (2008). Paediatric HIV and neurodevelopment in sub-Saharan Africa : A systematic review. Tropical Medicine and International Health, 13, 880-887. doi : 10.1111/j.1365.2008.02079.x American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). Washington, DC : American Psychiatric Association.  Anderson, P. (2008). Towards a developmental model of executive function. In V. Anderson, R. Jacobs, & P. Anderson (Eds.), Executive functions and the frontal lobes (pp. 3-22). New York, NY: Psychology Press.  Bayley, N. (1993). Bayley Scales of Infant Development Manual, Second Edition. San Antonio, TX: The Psychological Corporation.  Bear, M. F., Connors, B. W., & Paradiso, M. A. (2006). Neuroscience: Exploring the brain (3rd Ed.). Philadelphia, PA: Lippincott Williams & Wilkins. Bettes, B. A. (1988). Maternal depression and motherese: Temporal and intonational features. Child Development, 59, 1089Ð1096. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/3168616 Blair, C., & Raver, C. C. (2012). Child development in the context of adversity: Experiential canalization of brain and behavior. American Psychologist, 67, 309-318. doi: 10.1037/a0027493 Bogart, L. M., Cowgill, B. O., Kennedy, D., Ryan, G., Murphy, D. A., Elijah, J., & Schuster, M. A. (2008). HIV-related stigma among people with HIV and their families: A qualitative analysis. AIDS and Behavior, 12, 244-254. doi: 10.1007/s10461-007-9231-x Boivin, M. J., Bangirana, P., Nakasujja, N., Page, C. F., Shohet, C., Givon, D., ...Klein, P. S. (2013a). A year-long caregiver training program to improve neurocognition in preschool Ugandan HIV-exposed children. Journal of Developmental & Behavioral Pediatrics, 34, 269-278. doi: 10.1097/DBP.0b013e318285fba9 Boivin, M. J., Bangirana, P., Nakasujja, N., Page, C. F., Shohet, C., Givon, D., ...Klein, P. S. (2013b). A year-long caregiver training program improves cognition in preschool !120Ugandan children with Human Immunodeficiency Virus. Journal of Pediatrics, 165 (5), 1409-1416. doi: 10.1016/j.peds.2013.06.055 Boivin, M. J., Ruel, T. D., Boal, H. E., Bangirana, P., Cao, H., Eller, L. A., ...Wong, J. K. (2010). HIV-subtype A is associated with poorer neuropsychological performance compared to subtype D in ART-naive Ugandan children. AIDS. 24 (8), 1163-1170. Retrieved from http://journals.lww.com/aidsonline/pages/default.aspx?desktopMode=true Brown, R. E. (1965). Decreased brain weight in malnutrition and its implications. East African Medical Journal, 11, 584-595. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/5861857 Byakika-Tusiime, J., Crane, J., Oyugi, J. H., Ragland, K., Kawuma, A., Musoke, P., & Bangsberg, D. R. (2009). Longitudinal antiretroviral adherence in HIV+ Ugandan parents and their children initiating HAART in the MTCT-Plus Family Treatment Model: Role of depression in declining adherence over time. AIDS and Behavior, 13, S82-S91. doi: 10.1007/s10461-009-9546-x  Campbell, J. W., DÕAmato, R. C., Raggio, D. J., & Stephens, K. D. (1991). Construct validity of the computerized continuous performance test with measures of intelligence, achievement, and behavior. Journal of School Psychology, 29, 143-150. doi : 10.1016/S0022-4405(05)80006-8 Carpenter, M., Nagell, K., & Tomasello, M. (1998). Social cognition, joint attention, and communicative competence from 9 to 15 months of age. Monographs of the Society for Research in Child Development, 63(4), 1$143. http://www.eva.mpg.de/psycho/staff/carpenter/pdf/Carpenter_et_al_1998_monograph1.pdf Cavaleri, M. A., Kalogerogiannis, K., Mckay, M. M., Vitale, L., Levi, E., Jones, S., ... Flynn, E. (2013). Barriers to HIV care: An exploration of the complexities that influence engagement in the utilization of treatment. Social Work in Health Care, 19 (10), 934-945. doi: 10.1080/00981389.2010.514563 Centers for Disease Control and Prevention. (2014). About HIV/AIDS. Retrieved from http://www.cdc.gov/hiv/basics/whatishiv.html Central Intelligence Agency. (2014). The World Factbook: HIV/AIDS Ð Adult Prevalence Rate. Retrieved from https:// www.cia.gov/library/publications/the-world/factbook/rankorder/2155rank.html Chang, L., Speck, O., Miller, E. N., Braun, J., Jovicich, J., Koch, C., É Ernst, T. (2001). Neural correlates of attention and working memory deficits in HIV patients. Neurology, 57 (6), 1001-1007. doi: 10.1212.wNL.57.6.1001 Cooper, E. R., Charurat, M., Mofenson, L., Hanson, I. C., Pitt, J., Diaz, C., ÉWomen and InfantsÕ Transmission Study Group. (2002). Combination antiretroviral strategies for the !121treatment of pregnant HIV-1 infected women and prevention of perinatal HIV-1 transmission. Journal of Acquired Immune Deficiency Syndromes, 29 (5), 484 Ð 494. Retrieved from http://journals.lww.com/jaids/pages/default.aspx?desktopMode=true DeLuca, C. R., Wood, S. J., Anderson, V., Buchanan, J., Proffitt, T. M., Mahony, K., & Pantelis, C. (2003). Normative data from the CANTAB I: Development of executive function over the lifespan. Journal of Clinical and Experimental Neuropsychology, 25, 242-254. doi: 10.1076/jcen.25.2.242.13639 Derogatis, L. R., Lipman, R. S., Rickels, K., Uhlenhuth, E. H., & Covi, L. (1974). The Hopkins Symptom Checklist (HSCL) : A self-report symptom inventory. Behavioral Science, 19, 1-15. doi : 10.1002/bs.3830190102 Douek, D. C., Brenchley, J. M., Betts, M. R., Ambrozak, D. R., Hill, B. J., Okamoto, Y., ... Koup, R. A. (2002). HIV preferentially infects HIV-specific CD4+ T cells. Nature, 417, 95-98. doi: 10.1038/417095a Dowsett, S. M., & Livesey, D. J. (2000). The development of inhibitory control in preschool children: Effects of Òexecutive skillsÓ training. Developmental Psychobiology, 36(2), 161-174. doi: 10.1002/(SICI)1098-2303 Duncan, T. E., Duncan, S. C., & Strycker, L. A. (2006). An introduction to latent variable growth curve modeling: Concepts, issues, and application (2nd ed.). : Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Edley, R. S., & Knopf, I. J. (1987). Sustained attention as a predictor of low academic readiness in a preschool population. Journal of Psychoeducational Assessment, 5, 340-352. doi : 10.1177/073428298700500404 El Beitune, P., Duarte, G., Quintana, S. M., Figueiro-Filho, E. A., Marcolin, A. C., & Abduch, R. (2004). Antiretroviral therapy during pregnancy and early neonatal life: Consequences for HIV-exposed, uninfected children. The Brazilian Journal of Infectious Diseases, 8, 140-150. doi: 10.1590/S1413-86702004000200004 Ellis, R. J., Calero, P., & Stockin, M. D. (2009). HIV infection and the central nervous system. Neuropsychological Review, 19, 144-151. doi: 10.1007/s11065-009-9094-1 Engle, P. L., Fernald, L. C. H., Alderman, H., Behrman, J., OÕGara, C., Yousafzai, A., Éand the Global Child Development Steering Group. (2011). Strategies for reducing inequalities and improving developmental outcomes for young children in low-incomeand middle-income countries. The Lancet, 378, 1339-1353. doi : 10.1016/S0140-6736(11)60889-1  Fan, X., Thompson, B., & Wang, L. (1999). Effects of sample size, estimation methods, and model specification on structural equation modeling fit indexes. Structural Equation Modeling, 6, 56-83. doi : 10.1080/10705519909540119 !122Feuerstein, R., Rand, Y., Hoffman, M., & Miller, R. (1980). Instrumental enrichment. Baltimore (MD): University Park Press. Fish, I., & Winick, M. (1969). Cellular growth in various regions of the developing rat brain. Pediatric Research, 3(5), 407-412. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/5343536 Gardner, J. M., Powell, C. A., Baker-Henningham, H., Walker, S. P., Cole, T. J., & Grantham-McGregor, S. M. (2005). Zinc supplementation and psychosocial stimulation: Effects on the development of undernourished Jamaican children. American Journal of Clinical Nutrition, 82, 399Ð405. Retrieved from http://www.ajcn.org/cgi/pmidlookup?view=long&pmid=16087985 Gottlieb, G. (1991). Experiential canalization of behavioral development: Theory. Developmental Psychology, 27, 4-13. Retrieved from http://home.fau.edu/lewkowic/web/Gottlieb_Canalization%201991.pdf Grantham-McGregor, S. M., Fernald, L. C. H., Kagawa, R. m .C., & Walker, S. (2014). Effects of integrated child development and nutrition interventions on child development and nutritional status. Annals of the New York Academy of Sciences, 1308, 11-32. doi: 10.1111/nyas.12284 Grantham-McGregor, S. M., Powell, C. A., Walker, S. P., & Himes, J. H. (1991). Nutritional supplementation, psychosocial stimulation, and mental development of stunted children: The Jamaican study. The Lancet, 338 (8758), 1-5. doi: 10.1016/0140-6736(91)90001-6 Grantham-McGregor, S. M., Walker, S. P., Chang, S. M., & Powell, C. A. (1997). Effects of early childhood supplementation with and without stimulation on later development in stunted Jamaican children. American Journal of Clinical Nutrition, 66, 247-253.  Gulick, R. M., Mellors, J. W., Havlir, D., Eron, J. J., Gonzalez, C., McMahon, D., ÉCondra, J. H. (1997). Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy. The New England Journal of Medicine, 337 (11), 734Ð739. doi:10.1056/NEJM199709113371102. Hackl, K. L., Somlai, A. M., Kelly, J. A., & Kalichman, S. C. (1997). Women living with HIV/AIDS: The dual challenge of being a patient and caregiver. Health & Social Work, 22, 53-62. doi: 10.1093/hsw/22.1.53 Hazra, R., Siberry, G. K., Mofenson, L. M. (2010). Growing up with HIV: Children, adolescents, and young adults with perinatally acquired HIV infection. Annual Review of Medicine, 61, 169-185. doi: 10.1146/annurev.med.050108.151127 Heaton, R. K., Franklin, D. R., Ellis, R. J., McCutchan, A., Letendre, S. L., LeBlanc, S., ÉGrant, I. (2011). HIV-associated neurocognitive disorders before and during the era of !123combination antiretroviral therapy: differences in rates, nature, and predictors. Journal of Neurovirology, 17, 3-16. doi: 10.1007/s13365-010-0006-1 Hooper, D., Couglan, J., & Mullen, M. R. (2008). Structural equation modeling: Guidelines for determining model fit. Retrieved from http://arrow.dit.ie/libart/4 Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55. doi: 10.1080/10705519909540118 IBM Corp. (2013). IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp. Innis, S. M. (2007). Dietary (n-3) fatty acids and brain development. The Journal of Nutrition, 137, 855-859. Retrieved from http://jn.nutrition.org/cgi/pmidlookup?view=long&pmid=17374644 Ioannidis, J. P.,  Abrams, E. J., Ammann, A., Bulterys, M. Goedert, J. J., Gray, L., ÉWilfert, C. M. (2001). Perinatal transmission of human immuno-deficiency virus type 1 by pregnant women with RNA virus loads 1000 copies/ml. Journal of Infectious Diseases, 183 (4), 539 Ð545. doi: 10.1086/318530 Irwin, L. G., Siddiqi, A., & Hertzman, C. (2007). Early child development : A powerful equalizer. Retrieved from http://www.who.int/social_determinants/resources/ecd_kn_report_07_2007.pdf Jin, H., Atkinson, J. H., Yu, X., Heaton, R. K., Shi, C., Marcotte, T. P., ÉHNRC China Collaboration Group. (2006). Depression and suicidality in HIV/AIDS in China. Journal of Affective Disorders, 94, 269-275.  Kaaya, S. F., Fawzi, M. C. S., Mbwambo, J. K., Lee, B., Msamanga, G. I., & Fawzi, W. (2002). Validity of the Hopkins Symptom Checklist-25 amongst HIV-positive pregnant women in Tanzania. Acta Psychiatrica Scandinavica, 106, 9-19. doi : 10.1034/j.1600-0447.2012.01205.x Kamat, R., Woods, S. P., Marcotte, T. D., Ellis, R. J., Grant, I., & the HIV Neurobehavioral Research Program Group. (2012). Archives of Clinical Neuropsychology, 27, 520-531.  doi : 10.1093/arclin/acs055 Karp, R., Martin, R., Sewell, T., Manni, J., & Heller, A. (1992). Growth and academic achievement in inner-city kindergarten children. The relationship of height, weight, cognitive ability, and neurodevelopment level. Clinical Pediatrics, 31 (6), 336-340. doi: 10.1177/000992289203100604 !124Kelly, A. F., & Hewson, P. H. (2000). Factors associated with recurrent hospitalizations in chronically ill children and adolescents. Journal of Paediatrics and Child Health, 36, 13-18. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10723684 Klarborg, B., Madsen, K. S., Vestergaard, M., Kimminge, A., Jernigan, T. L., & Baare, W. F. C. (2013). Sustained attention is associated with right superior longitudinal fasciculus and superior parietal white matter microstructure in children. Human Brain Mapping, 34, 3216-3232. doi: 10.1002/hbm.22139 Klein, P. S. (2000). A developmental mediation approach to early intervention: Mediational intervention for sensitizing caregivers (MISC). Educational and Child Psychology, 17, 19-31. Retrieved from http://www.bpsshop.org.uk/Educational-Child-Psychology-Vol-17-No-3-2000-Psychological-influences-upon-educational-intervention-P3786.aspx Klein, P. S. (2003). Early intervention: Mediational intervention for sensitizing caregivers. In A. S. H. Seng, L. K. H. Pon, & O. S. Tan (Eds.). Mediated learning experience with children: Applications across countries. Singapore: McGraw-Hill Education. Klein, P. S., & Alony, S. (1993). Immediate and sustained effects of maternal mediating behaviors on young children. Journal of Early Intervention, 17, 177-193. doi: 10.1177/105381519301700208 Klein, P. S., & Rye, H. (2004). Interaction-oriented early intervention in Ethiopia: The MISC approach. Infants and Young Children, 17, 340-354. Retrieved from http://journals.lww.com/iycjournal/pages/currenttoc.aspx Klunklin, P., & Harrigan, R. C. (2002). Child-rearing practices of primary caregivers of HIV-infected children: An integrative review of the literature. Journal of Pediatric Nursing, 17, 289-296. doi:10.1053/jpdn.2002.126713 Kotchick, B. A., Forehand, R., Brody, G., & Armistead, L., Morse, E., Simon, P., & Clark, L. (1997). The impact of maternal HIV infection on parenting in inner-city African American families. Journal of Family Psychology, 11, 447-461. doi:10.1037/0893-3200.11.4.447 Kumar, A. M., Borodowsky, I., Fernandez, B., Gonzalez, L., & Kumar, M. (2007). Human immunodeficiency virus type 1 RNA levels in different regions of human brain: Quantification using real-time reverse transcriptase-polymerase chain reaction. Journal of Neurovirology, 13, 210-224. doi: 10.1080/13550280701327038 Lam, C. M., & Beale, I. L. (1991). Relations among sustained attention, reading performance, and teachersÕ ratings of behavior problems. Remedial and Special Education, 12, 40-47. doi : 10.1177/074193259101200208 !125Lei, P. & Wu, Q. (2007). Introduction to structural equation modeling : Issues and practical considerations. Educational Measurement : Issues and Practice, 26 (3), 33-43. doi : 10.111/j.1745-3992.2007.00099.x Levy, F. (2006). The development of sustained attention (vigilance) and inhibition in children: Some normative data. Journal of Child Psychology and Psychiatry, 21, 77-84. doi: 10.1111/j.1469-7610.1980.tb00018.x Liner, K. J., Hall, C. D., & Robertson, K. R. (2007). Impact of human immunodeficiency virus (HIV) subtypes on HIV- associated neurological disease. Journal of Neurovirology, 13 (4), 291Ð304. doi:10.1080/13550280701422383 Livingood, A. B., Daen, P. & Smith, B. D. (1983) The depressed mother as a source of stimulation for her infant. Journal of Clinical Psychology, 39, 369Ð375. Retrieved from: http://www.ncbi.nlm.nih.gov/pubmed/6874970 Lopez, E. C. (2008). Best practices in working with school interpreters. In A. Thomas & B. Grimes (Eds.), Best Practices in School Psychology V. Bethesda, MD : National Association of School Psychologists.  Lozoff, B., Smith, J. B., Clark, K. M., Perales, C. G., Rivera, F., & Castillo, M. (2010). Home intervention improves cognitive and social-emotional scores in iron-deficient anemic infants. Pediatrics, 126 (4), 884-894. doi : 10.1542/peds.2009.3535 Mahone, E. M., & Schneider, H. E. (2012). Assessment of attention in preschoolers. Neuropsychology Review, 22, 361-383. doi : 10.1007/s11065-012-9217-y  Main, M. (1983). Exploration, play, and cognitive functioning related to infant-mother attachment. Infant Behavior and Development, 6, 167-174. doi:10.1016/S0163-6383(83)80024-1 McKay, H., Sinisterra, L., McKay, A., Gomez, H., & Lloreda, P. (1978). Improving cognitive ability in chronically deprived children. Science, 200, 270-278. Retrieved from http://fundacionecologiahumana.org/TEXTOS/Science-magazine.pdf Mirman, D. (2014). Growth curve analysis and visualization using R. Boca Raton, FL: Chapman & Hall/CRC. Molina, B. S. G., Hinshaw, S. P., Swanson, J. M., Arnold, E., Vitiello, B., Jensen, P. S. É The MTA Cooperative Group. (2009). The MTA at 8 years: Prospective follow-up of children treated for combined-type ADHD in a multisite study. Journal of the American Academy of Child & Adolescent Psychiatry, 48 (5), 484-500. doi: 10.1097/CHI.0b013e31819c23d0 Murphy, D. A., Austin, E. L., & Greenwell, L. (2013). Correlates of HIV-related stigma among HIV-positive mothers and their uninfected adolescent children. Women & Health, 44, 19-42. doi: 10.1300/J013v44n03_02 !126Murphy, D. A., Marelich, W. D., Armistead, L., Herbeck, D. M., & Payne, D. L. (2010). Anxiety/stress among mothers living with HIV : Effects on parenting skills and child outcomes. AIDS care, 22 (12), 1449-1458. doi : 10.1080/09540121.2010.487085 Nahar, B., Hossain, M. I., Hamadani, J. D., Ahmed, T., Huda, S. N., Grantham-McGregor, S. M., & Persson, L. A. (2012). Effects of a community-based approach of food and psychosocial stimulation on growth and development of severely malnourished children in Bangladesh: A randomised trial. European Journal of Clinical Nutrition, 66, 701Ð 709. doi: 10.1038/ejcn.2012.13 OÕConnell, A. A., Logan, J. A. R., Pentimonti, J. M., & McCoach, D. B. (2013). Linear and quadratic growth models for continuous and dichotomous outcomes. In Y. Petscher, C. Schatschneider, & D. L. Compton (Eds.), Applied quantitative analysis in the education and the social sciences. New York, NY: Routledge. OÕConnor, E., & McCartney, K. (2007). Attachment and cognitive skills: An investigation of mediating mechanisms. Journal of Applied Development Psychology, 28, 458-476. doi:10.1016/j.appdev.2007.06.007  Oken, B. S., Salinsky, M. C., Elsas, S. M. (2006). Vigilance, alertness, or sustained attention: Physiological basis and measurement. Clinical neurophysiology, 117 (9) 1885-1901. doi: 10.1016/j.clinph.2006.01.017. Palella, F. J., Delaney, K. M., Moorman, A. C., Loveless, M. O., Fuhrer, J., Satten, G. A., ... The HIV Outpatient Study Investigators. (1998). Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. The New England Journal of Medicine, 338 (13), 853-860. doi: 10.1056/NEJM199803263381301 Perepletchikova, F., & Kazdin, A. E. (2005). Treatment integrity and therapeutic change: Issues and research recommendations. Clinical Psychology: Science and Practice, 12 (4), 365-383. doi: 10.1093/clipsy,bpi045 R Core Team. (2014). R : A Language and Environment for Statistical Computing. Vienna, Austria : R Foundation for Statistical Computing.  Rapport, M. D., Orban, S. A., Kofler, M. J., & Friedman, L. M. (2013). Do programs designed to train working memory, other executive functions, and attention benefit children with ADHD? A meta-analytic review of cognitive, academic, and behavioral outcomes. Clinical Psychology Review, 33 (8), 1237-1252. doi: 10.1016/j.cpr.2013.08.005 Raval, H., & Smith, J. A. (2003). TherapistsÕ experiences of working with language interpreters. International Journal of Mental Health, 32(3), 6-31. Retrieved from mesharpe.com/mall/results1.asp?ACR=imh Reid, R., DuPaul, G. J., Power, T. J., Anastopoulos, A. D., Rogers-Adkinson, D., Noll, M.-B., & Riccio, C. (1998). Assessing culturally different students for attention deficit !127hyperactivity disorder using behavior rating scales. Journal of Abnormal Child Psychology, 26, 187-198. doi: 0091-0627/98/0600-0187$15.00/0 Riccio, C. A., Reynolds, C. R., & Lowe, P. A. (2001). Clinical applications of continuous performance tests: Measuring attention and impulsive responding in children and adults. New York, NY: John Wiley & Sons, Inc.  Robertson, K., Liner, J., & Meeker, R. B. (2012). Antiretroviral neurotoxicity. Journal of Neurovirology, 18, 388-399. doi: 10.1007/s13365-012-0120-3 Romero, R., Yund, B., Zelinsky Goldman, D., Hughes, S., Jordan, C., & Shapiro, E. (2012). Revised Manual for the Early Childhood Vigilance Task (ECVT). Minneapolis, MN: University of Minnesota.  Rowe, D. C., & Plomin, R. (1977). Temperament in early childhood. Journal of Personality Assessment, 41, 150-156. doi: 10.1207/s15327752jpa4102_5 Ruel, T. D., Boivin, M. J., Boal, H. E., Bangirana, P., Charlebois, E., Havlir, D. V., ... Wong, J. K. (2012). Neurocognitive and motor deficits in HIV-infected Ugandan children with Hhigh CD4 cell counts. Clinical Infectious Diseases, 54 (7), 1001-1009. doi: 10.1093/cid/cir1037 Ruff, H. A. (1990). Individual differences in sustained attention during infancy. In J. Colombo, & J. Fagen (Eds.), Individual differences in infancy: Reliability, stability, prediction (pp.247-270). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.   Sacktor N., Nakasujja N., Skolasky R. L., Rezapour M., Robertson K., Musisi S., É Quinn, T. C. (2009). HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda. Clinical Infectious Disorders, 49, 780Ð786. doi: 10.1086/605284 Sarter, M., Givens, B., & Bruno, J. P. (2001). The cognitive neuroscience of sustained attention: Where top-down meets bottom-up. Brain Research Reviews, 35, 146-160. Retrieved from faculty.psy.ohio-state.edu/bruno/PDF%20files/top%20down%20ms.pdf Schable, B., Diaz, T., Chu, S. Y., Caldwell, M. B., Conti, L., Alston, O. M., ÉFrederick, M. (1995). Who are the primary caretakers of children born to HIV-infected mothers ? Reults from a multistate surveillance project. Pediatrics, 95, 511-515. Retrieved from http://go.galegroup.com/ps/i.do/id=GALE%7CA16899132&v-2.17u=msu_main&it=r&p=  AONE&sw=w&asid=23a265f24e6865b231ba4c27e5eb2122 Schunk, D. H. (1985). Self-efficacy and classroom learning. Psychology in the Schools, 22, 208$223. Retrieved from http://libres.uncg.edu/ir/uncg/f/d_schunk_self_1985.pdf !128Senkomago, V., Guwatudde, D., Breda, M., & Khoshnood, K. (2011). Barriers to antiretroviral adherence in HIV-positive patients receiving free medication in Kayunga, Uganda. AIDS Care, 23 (10), 1246-1253. doi: 10.1080/09540121.2011.564112 Shapiro, E. G., McPhee, J. T., Abbott, A. A., & Sulzbacher, S. I. (1994). Minneosta Preschool Affect Rating Scales: Development, reliability, and validity. Journal of Pediatric Psychology, 19, 325-345.  Simbayi, L. C., Kalichman, S., Strebel, A., Cloete, A. Henda, N., & Mqeketo, A. (2007). Internalized stigma, discrimination, and depression among men and women living with HIV/AIDS in Cape Town, South Africa. Social Science & Medicine, 64, 1823-1831. doi: 10.1016/j.socscimed.2007.01.006 Stevens, G., (1981). Bias in the attribution of hyperkinetic behavior as a function of ethnic identification and socioeconomic status. Psychology in the Schools, 18, 99-106. doi: 10.1002/1520-6807(198101)18:1<99::AID-PITS2310180119>3.0.CO;2-X Tapp, J. (2003). Procoder for Digital Video User Manual. Nashville, TN: Vanderbilt University.   Thurman, T. R., Jarabi, B., & Rice, J. (2012). Caring for the caregiver: Evaluation of support groups for guardians of orphans and vulnerable children in Kenya. AIDS Care, 24 (7), 881-819. doi: 10.1080/09540121.2011.644229 Tuller, D. M., Bangsberg, D. R., Senkungu, J., Ware, N. C., Emenyonu, N., & Weiser, S. D. (2010). Transportation costs impede sustained adherence to HAART in a clinic population in southwestern Uganda: A qualitative study. AIDS and Behavior, 14, 778-784. doi: 10.1007/s10461-009-9533-2 UNICEF. (2013). Uganda: Statistics. Retrieved from https.//www.unicef.org/infobycountry/uganda_statistics.html U.S. Department of Health and Human Services. (2010). Opportunistic infections. Retrieved from http://aids.gov/hiv-aids-basics/staying-healthy-with-hiv-aids/potential-related-health-problems/opportunistic-infections/ U.S. Department of Health and Human Services. (2013). Global AIDS overview. Retrieved from http:// www.aids.gov/federal-resources/around-the-world/global-aids-overview/ Valcour, V., Sithinamsuwan, P., Letendre, S., & Ances, B. (2011). Pathogenesis of HIV in the Central Nervous System. Current HIV/AIDS Reports, 8, 54-61. doi:10.1007/s11904-010-0070-4  Ventevogel, P., De Vries, G., Scholte, W. F., Shinwari, N. R., Faiz, H., Nassery, R., ÉOlff, M. (2007). Properties of the Hopkins Symptom Checklist-25 (HSCL-25) an the Self-Reporting Questionnaire (SRQ-20) as screening instruments used in primary care in !129Afghanistan. Social Psychiatry and Psychiatric Epidemiology, 42, 328-385. doi: 10.1007/s00127-007-0161-8 Walker, S. P., Wachs, T. D., Grantham-McGregor, S., Black, M. M., Nelson, C. A. Huffman, S. L., ÉRichter, L. (2011). Inequality in early childhood : Risk and protective factors for early child development. The Lancet, 378, 1325-1328. doi : 10.1016/S1040-6736911)60555-2 Watkins, J. M., Cool, V. A., Usner, D., Stehbens, J. A., Nichols, S., Loveland, K. A., É Neuchterlein, K. H. (2000). Attention in HIV-infected children: Results from the Hemophilia Growth and Development Study. Journal of the International Neuropsychological Society, 6, 443-454. Retrieved from http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=49089 Weiser, S. D., Tsai, A. C., Gupta, R., Frongillo, E. A., Kawuma, A., Senkungu, J., ÉBangsberg, D. R. (2012). Food insecurity is associated with morbidity and patterns of healthcare utilization among HIV-infected individuals in a resource-poor setting. AIDS, 26, 67-75. doi: 10.1097/QAD.0b013e32834cad37 Wiley, C. A., Schrier, R. D., Nelson, J. A., Lampert, P. W., & Oldstone, M. B. A. (1986). Cellular localization of human immunodeficiency virus infection within the brains of acquired immune deficiency syndrome patients. Proceedings of the National Academy of Sciences of the United States of America, 83(18), 7089-7093. Retrieved from http://www.pnas.org/content/83/18/7089.full.pdf+html Williams, B., Amico, K. R., & Konkle-Parker, D. (2011). Qualitative assessment of barriers and facilitators to HIV treatment. Journal of the Association of Nurses in AIDS Care, 22, 307-312. doi: 10.1016/j.jana.2010.11.001 Wingwood, G. M., DiClemente, R. J., Mikhail, I., Hubbard, D., Davies, S. L., Hardin, J. W., ... Saag, M. (2007). HIV discrimination and the health of women living with HIV. Women & Health, 46, 99-112. doi: 10.1300/J013v46n02_07 Winick, M., & Rosso, P. (1969). The effect of severe early malnutrition on cellular growth of human brain. Pediatric Research, 3(2), 181-184. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/5781822 World Health Organization. (2004). The importance of caregiver-child interactions for the survival and healthy development of young children: A review. Retrieved from http://whqlibdoc.who.int/publications/2004/924159134X.pdf World Health Organization. (2007). WHO case definitions of HIV for surveillance and revised clinical staging and immunological classification of HIV-related disease in adults and children. Retrieved from www.who.int/hiv/pub/guidelines/HIVstaging150307.pdf World Health Organization. (2014a). HIV/AIDS. Retrieved from http://www.who.int/gho/hiv/en/ !130World Health Organization. (2014b). HIV/AIDS: Data and statistics. Retrieved from http://www.who.int/hiv/data/en/ World Health Organization. (2014c). HIV/AIDS: Mother-to-child transmission of HIV. Retrieved from http://www.who.int/hiv/topics/mtct/en/ World Health Organization. (2015). The WHO Child Growth Standards. Retrieved from http://www.who.int/childgrowth/standards/en/ Zelinsky Goldman, D., Shapiro, E. G., & Nelson, C. A. (2004). Measurement of vigilance in 2-year-old children. Developmental Neuropsychology, 25, 227-250. doi: 10.1207/s15326942dn2503_ !131