|Home | About | Journals | Submit | Contact Us | Français|
Adolescents of parents infected with HIV/AIDS (HIV-affected adolescents) were recruited from August 1993 to March 1995 to examine their depression as a function of their social support. Adolescent depression and the size, frequency of contact, satisfaction, negative role model influence, and positive social support were examined. Adolescent depression was associated with higher parent depression score, parental HIV disclosure, conduct problems, and was inversely related to higher self-esteem, being male, and a positive living situation. Negative social support was significantly associated with higher adolescent depression, indicating negative role model influence may be more salient than size, frequency of contact, perception, or positive social support. The findings underscore the complex relationships between social support and mental/behavioral outcomes among HIV-affected adolescents as well as the need to examine HIV disclosure in more detail within the context of adolescent social support.
Depression is one of the most common psychiatric disorders in the United States; it is estimated that between 8% and 18% of the population will experience at least one major depressive episode over the course of their lifetime.1 Approximately twice as many women as men are affected by depression.2,3 Furthermore, depression is a recurrent disorder, with over 80% of depressed patients having more than one depressive episode, and over 50% of depressed patients relapsing within 2 years of recovery.4-6 Depression is associated with suicide,7 an increased mortality risk from other causes of death,8 and other negative outcomes including substance abuse.9 Adolescence is an important developmental period for understanding depression, as significant increases in depressive symptoms occur during adolescence.10
Depression among adolescents is associated with a twofold higher risk in females; high comorbidity with anxiety disorders; substance abuse; and suicidal behaviors. In addition, adolescent depression results in high social costs as well as impairment of school performance.11
Social support is defined as the existence or availability of people who care about, value, and love us.12 Although the link between a lack of social support and reduced psychological well-being is well established,13-15 empirical research on social support among adolescents has barely begun, and is characterized by the lack of a reliable, general, and convenient index of social support.12 Existing indices of social support among adolescents treat social support as a homogeneous mass.16-21
While much more empirically derived evidence is needed to provide a basis for theoretical advances in the area of adolescent social support,22 several studies have examined the size of social support,23 the frequency of contact of social support,24 perception of social support,22 negative social support,25 and positive social support.26 A study by Bal and colleagues27 examined the role that social support plays in well-being and coping after a stressful event in a group of nonclinical adolescents. They found that a highly perceived availability of social support is directly associated with fewer trauma-related symptoms.
Adolescents whose parents are infected with HIV/AIDS (HIV-affected adolescents) are more vulnerable to depression, compared to adolescents in the general population.28 HIVaffected adolescents have elevated risks for developing emotional distress29,30 and are more likely to engage in high-risk behaviors.31,32 In addition, parents living with HIV (PLH), compared to parents with cancer, are more likely to have preexisting and long-lasting stressors that affect family relationships.28,33
HIV-affected adolescents are an important group to examine, as empirical research on social support for HIV-affected adolescents is extremely limited.34 Our primary objective was to examine the association between depression among HIV-affected adolescents and their social support, adjusting for potential correlates of adolescent depression. Five dimensions of social support (size, frequency of contact, perception, negative support, and positive support) were considered as potential predictors of depression. Additionally, we examined the association between parental HIV disclosure and adolescent depression. Studies have shown that although HIV disclosure serves as an acute and ongoing stressor, it also facilitates emotional support for PLH, which may lead to more effective coping and enhanced psychological adaptation.35 For example, one study of women, which included approximately 40% injection drug users (IDUs), found less depression among those who were able to disclose their HIV status. However, there is limited research examining the association between parental HIV disclosure36 and their adolescent children’s depression.37,38 Few studies discuss parental HIV disclosure in association with adolescent problem behaviors and negative family life events,39,40 and only one indicated that over time parental HIV disclosure was significantly associated with more adolescent problem behaviors and more negative family life events.37
This study uses baseline data collected as part of a randomized controlled intervention study.41 The New York City Division of AIDS Services (DAS) maintains a log of 95% of persons with AIDS who qualify for social welfare benefits. From August 1993 to March 1995, 619 persons were logged by DAS. The eligibility criteria for the study were: being HIV-infected; having at least one adolescent child aged 11 to 18 years; and obtaining permission of the case and the case managers to enroll in the study. Of the 619 potential participants, 155 (25%) died prior to being approached, 11 (1.8%) were ineligible because of severe illness or incarceration, and 35 (5.7%) were excluded by case managers who felt the research project could potentially harm the family. Therefore, of the 619 PLH logged, 418 (68%) were eligible; of these 418 PLH, 65 (16%) were untraceable and 46 (11%) refused participation. Thus, 73% (n = 307/418) of eligible PLH were recruited, reflecting 87% (n = 307/353) of the traceable eligible PLH. With Institutional Review Board approval, informed consent and permission to recruit adolescents were obtained from PLH and then from their adolescent children. Some PLH chose not to participate but allowed their children to participate (n = 14). There were six cases in which the adolescent child aged out of the study (over 18 years) prior to the assessment; these adolescents were eliminated from the analysis. A total of 413 adolescents were recruited from 307 families (average n per family = 1.5, standard deviation [SD] = 0.7; range, 1–5).
Two-person interview teams conducted 2-hour in-home interviews with the PLH and all adolescents in the household. Similar to the participants, interviewers were predominantly African American or Latino (62%); 5 of the 15 interviewers were bilingual in Spanish and English. Interviewers were certified after being trained in ethics, confidentiality (particularly of PLH’s serostatus), child abuse, crisis protocols, HIV/AIDS, and conducting in-home assessments on laptop computers. Quality assurance was maintained by audio taping all interviews and routinely monitoring approximately 10% of randomly selected tapes.
Background characteristics were collected from participants at the baseline interview, including gender, age, living situation, and socioeconomic status.
Adolescent depression was assessed with the Brief Symptom Inventory (BSI)42 depression subscale (Cronbach α = 0.76), consisting of the adolescents’ rating of the level of severity across six items reflecting symptoms of depression from 0 (not at all) to 4 (extremely).
Size of social support was measured by the number of people providing support (social support provider) as reported by the adolescents (range, 0–15 providers).
Frequency of contact captured the number of times per month adolescents were in contact with social support providers; this was asked by each reported support provider. By summing across support providers and dividing by the number of providers, we determined the average frequency of contact (range = 0–22 times per month).
Perception of social support captured perceived satisfaction with support, measured in a 5-point Likert scale (5 = very satisfied to 1 = very dissatisfied). Adolescents reported their satisfaction for each support provider, which was then summed across providers and divided by the number of providers to determine the average satisfaction of social support.
Positive social support captured adolescents’ report of the number of positive behaviors among support providers (e.g., helps when in trouble, provided crucial support, was supportive about parent’s illness). It was assessed for each provider and averaged across the number of providers to determine the average positive social support, ranging from 0 (no positive support) to 4 (four counts of positive support).
Negative social support captured adolescents’ perception of the number of negative behaviors among support providers. Negative social support reflected adolescents’ perception of the number of negative behaviors among their social support providers (e.g., having unprotected sex, criminal activity). It was assessed for each provider and averaged to determine the negative social support, ranging from 0 (no negative support) to 4 (four counts of negative role model behaviors).
Multiple problem behaviors, reflecting adolescents’ risk behaviors, were measured by counting the presence (1) or absence (0) of unprotected sexual intercourse, alcohol use, drug use, contact with the criminal justice system, trouble with peers, trouble at school, and not being enrolled in school (potential range, 0–7).
Conduct problems captured adolescents’ level of delinquent behaviors (e.g., robbed someone, threatened someone in order to steal from them), calculated by summing across 18 delinquent behaviors (Cronbach α = 0.61).
Adolescent self-esteem was captured by the Rosenberg Self-Esteem Scale, a 10-item measure validated and found reliable with normative samples of adolescents of many ethnic groups and ages (Cronbach α = 0.85).43
Family life stressors were measured by counting the presence (1) or absence (0) of 10 stressful events within the family, such as arguments between parents or mother absent from home (Cronbach α = 0.57).
Parent depression was measured by the BSI depression subscale (six items reflecting symptoms of depression).
HIV disclosure, measured by the occurrence (1) or absence (0) of HIV disclosure to each adolescent in the family.
Time to HIV disclosure was captured by a dichotomous indicator to reflect the elapsed time between disclosure and the timing of the interview (over 12 months versus 12 months or less).
Prior to estimating statistical models, distributional assumptions and nonlinearity of the model were assessed. Furthermore, correlational analyses were performed to determine if any factors were highly correlated with each other, in order to avoid collinearity problems. After performing the exploratory analyses, we examined the association between adolescent baseline depression and the five dimensions of social support (size, frequency of contact, perception, negative support, and positive support), along with other potential correlates. Mixed effects models were used to take into account the intracluster correlation at the family level, because some families having more than one adolescent. Model diagnostics were performed for each model to assess model appropriateness. Other correlates of depression and various interaction terms, based on a priori knowledge and existing literature, were considered and included in the final analysis.
Table 1 presents the background characteristics of HIV-affected adolescents and PLH at the time of recruitment. The 413 adolescents had 307 PLH aged from 25 to 70 years (mean = 37.6 years; SD = 5.4). Over 80% of the adolescents had a PLH who was a mother. The sample was predominantly Latino (50.5%, mostly of Puerto Rican and Dominican descent) and African American (38.4%); 3.0% were white, and 8.1% reported other ethnic heritages. The distribution of PLH’s gender, age, employment status, and education were similar among male and female adolescents. However, a smaller proportion of female adolescents were satisfied with their current living situation.
Table 2 compares the social support and depression factors among male and female adolescents and PLH. A significantly higher proportion of female adolescents perceived experiencing negative social support and had a higher baseline depression score. Conversely, male adolescents reported significantly higher conduct problems. There were no gender differences in the size, frequency of contact, perception, or positive social support. In addition, we found no adolescent gender differences in baseline measures of multiple problem behavior, Rosenberg self-esteem score, or family life stressors. Furthermore, we found no adolescent gender differences in PLH’s social support, PLH depression score, or PLH’s average coping score. A similar proportion of PLH had disclosed their HIV status and made custody arrangements among female and male adolescents in the study.
Six different models were considered when examining the predictors of adolescent depression. Each of the following characteristics were considered the main predictors of depression, adjusting for other correlates of depression: size of social support, positive perception of support, frequency of contact of support, negative social support, positive social support, and PLH’s social support. Other correlates of depression were determined based on a priori knowledge and existing literature. For example, in addition to HIV disclosure, we assessed the effect of time since HIV disclosure as a potential correlate of adolescent depression. In addition, interactions between social support variables and other factors were considered in various models. The associations between the variables included in the final model are shown in Figure 1.
Table 3 outlines the results of the final mixedeffects model among HIV-affected adolescents, adjusting for the intracluster correlation at the family level. The final model included the following variables: negative social support, parent depression, gender, positive living situation, disclosure of HIV status, self-esteem, and conduct problems. Interactions between negative social support and other predictors of adolescent depression were considered. In addition, time since disclosure was considered in the final model. However, since time since disclosure and other interaction terms were not significantly associated with adolescent depression, they were not retained in the final model.
Negative social support reported by the adolescents was significantly associated with their level of depression. Adolescents whose parent disclosed their HIV status had a higher depression score. In addition, parent depression score was significantly associated with adolescent depression, and adolescents with more conduct problems had a higher depression score. Other significant correlates of adolescent depression included: adolescents reporting a positive living situation had a lower depression score; females reported significantly higher levels of depression compared to males; and adolescents with a lower self-esteem score had a higher adolescent depression score (Table 3).
In contrast to existing literature,26,27 we found no significant associations between depression and the positive dimensions of social support. The HIV-affected adolescents in our study may be facing multiple challenges that are distinct from adolescents in the general population; therefore, the impact of social support in this highly marginalized group may be unique. One possible explanation of this finding may be that the negative social support perceived by the adolescents outweighs the potential protective effects of the positive dimensions of social support. Across the social support providers, adolescents reported, on average, less than one count of negative social support. However, over 39% of the adolescents reported at least one negative social support. The types of behaviors captured in the negative social support dimension (e.g., social support providers doing drugs, abusing alcohol, having unprotected sex, being engaged in criminal activities) may have had a much more serious impact on adolescent depression, compared to their positive dimensions of social support (e.g., size, frequency of contact, and positive social support). That is, negative social support may overwhelm the associations between adolescent depression and other dimensions of social support. Alternatively, the magnitude of positive social support was too low to show a difference. For example, adolescents on average reported over two counts of positive social support; however, the magnitude of the impact of positive social support may be offset by negative social support.
Consistent with the existing literature,41,42 significant associations were observed between adolescent depression and being female, parent depression, poor living situation, parent’s disclosure of HIV status, having lower self-esteem, and having more conduct problems. Past studies suggest that disclosure of HIV status might be protective against depression.37,44 In this study, we found no significant association between HIV disclosure and parent depression. Among adolescents, however, we observed a higher depression score if the parent had disclosed his/her HIV status. This is consistent with the study by Lee and Rotheram-Borus37 indicating that over time parental HIV disclosure was significantly associated with more adolescent problem behaviors and more negative family life events.
There are inherent limitations of this study. Because the study is a cross-sectional examination of associations, we cannot make definitive conclusions about causal effects for adolescent depression. First, we could not follow the trajectories of adolescent depression since the time of the disclosure. Instead, we considered time since disclosure as a predictor in our model. Second, although we considered the interactions between HIV disclosure and social support dimensions, we did not have the social support information at the time disclosure took place. Third, we found a significant association between low self-esteem and higher adolescent depression. However, we cannot make a causal inference about the effect of low self-esteem on adolescent depression. Despite the inherent limitations of the study design, our findings underscore important examinations of the associations between depression among HIV-affected adolescents and their social support, and the association between parent HIV disclosure and adolescent depression.
There has been limited empirical research on depression and social support among HIV-affected adolescents. This study highlights the need for future prevention programs to focus on the potential impact of social support in reducing depression among HIV-affected adolescents. Our findings underscore the importance of addressing social support in both ends of the spectrum. Given that HIV-affected adolescents are faced with various additional challenges that make them more vulnerable to depression, addressing and reducing negative social influences within their network may be as important as promoting positive social support and family relationships. In addition, our findings underscore the need to examine parental HIV disclosure in more detail within the context of the adolescent’s social support. Parental HIV disclosure may be beneficial to their adolescent children when it occurs within the context of nurturing social support. Disclosing HIV status while projecting negative social support may pose more significant adjustment and coping challenges for HIV-affected adolescents. Prior to disclosing, PLH should be prepared for the possible consequences of disclosing. In particular, post-test counseling should be effectively put in place in order to help PLH deal with their own feelings of anger, fear, and/or depression prior to disclosing their status to their children, as well as counseling for children before and after parent HIV disclosure.37
A longitudinal analysis of the same cohort of PLH and their adolescent children is underway in order to examine the potential causal impact of social support on adolescent depression over time. This may shed new light on the relationships between social support and adolescent depression.
This paper was completed with the support of National Institute of Mental Health grants #1ROI MH49958-04 and P30MH58107. We wish to thank the parents, caregivers, and adolescents who participated in this study, as well as those who assisted in it including Gricel Arredondo, Coleen Cantwell, Susan Cantwell, Tri Cisek, Linda Cisek, Betty Crenshaw, Barbara Draimin, Amy Elkavich, Jen Elliott, Marya Gwadz, Nionne James, Julie LeHane, Patrice Lewis, David Litke, Tanko Mohammed, Sanna Moore, Marian Reidel, Laura Rosen, Nim Tottenham, Julian Wang, Karen Wyche, the staff of Housing Works, and our interviewers and group leaders.