The results of our study suggest that the associations between serostatus disclosure to mothers, HIV-specific family support and disease status depend, in part, on men's ethnic background. Non-Hispanic White MLWH who had disclosed to their mothers and who were receiving high levels of HIV-specific family support experienced lower viral load and higher CD4+ cell counts, but disclosure coupled with low levels of HIV-specific family support was not associated with these disease status indicators in these men. In contrast, Latino men who had disclosed to their mothers and were receiving low levels of HIV-specific family support experienced poorer disease status (higher viral load); however, disclosure coupled with high levels of support had no significant effect on disease status in these men. Moreover, these associations remained significant despite controlling for medication adherence. Finally, psychological (perceived stress and depressive symptoms) and neuroendocrine (24-hr urinary norepinephrine and cortisol) variables known to be associated with virologic and immune status did not explain the significant interaction effects observed in this sample.
Although we expected to find that disclosure to both mothers and fathers would be associated with disease status as a function of variations in the HIV-specific support men received from family members, only combinations of disclosure to mothers and HIV-specific support from family members was important in explaining men's virologic and immune status. Prior research suggests that mothers are often disclosed to the most of any family member (Serovich, Esbensen, & Mason, 2005
). Consistent with this research, almost three-quarters of MLWH had disclosed their serostatus to their mothers, whereas only about half had disclosed their serostatus to their fathers. Further, compared to Latino MLWH, non-Hispanic White MLWH were more likely to disclose their serostatus to both their mother and father. It is possible that not only did men disclose to their mothers more often, but that mothers were primary sources of family support or gateways of support for other family members in both Latino and non-Hispanic White men (e.g., Kalichman et al., 2003
Although we did not find ethnic differences in the amount of HIV-specific family support that men perceived receiving, it is likely that Latino MLWH relied on their immediate family members as primary sources of support more so than non-Hispanic White MLWH (Keefe et al., 1979
; Raymond et al., 1990; Vernon & Roberts, 1985
). The White MLWH in our study may not have disclosed to their parents with the intention of gaining or receiving continued support from family members. However, the illness-specific support that they received, possibly in response to having disclosed their serostatus, may have further enhanced men's support network and promoted better disease status. Conversely, if men disclosed to mothers and did not receive adequate HIV-specific support from their families, they likely had other social networks to draw support resources from.
In contrast to the social network structure of White cultures, Latino cultural norms prescribe close familial ties and fewer connections with social contacts outside of the family relationship (Vernon & Roberts, 1985
). Thus, if men needed or desired support to help them manage their illness from family members, but received inadequate levels of support, they may have felt stigmatized, out of alternative support options, and socially isolated as a result of their illness. These feelings of stigma or social isolation in combination with a lack of social resources in coping with their illness may explain impairments in Latino MLWH's virologic and immune status.
Contrary to our expectations, our results did not find that disclosures to mothers coupled with high levels of HIV-specific social support from family members explained disease status in Latino MLWH. Although Latino MLWH were less likely than non-Hispanic White MLWH to disclose their serostatus to either parent, it is possible that Latino MLWH were more comfortable disclosing to their mothers than their fathers because their mothers were seen as being more understanding, accepting, and supportive. In Latin culture, men and women often adopt the behavioral gender roles of machismo
, which prescribe that men may hold various forms of social power over women, but women are morally superior to men and are therefore often revered by adult male children (Wood & Price, 1997
). It is possible that the Latino MLWH in our study disclosed to mothers with the expectation of receiving the same amount of acceptance and support from their family as they did prior to the disclosure. If these expectations were met, the continued displays of support as a result of knowing about men's illness may not have explained disease status because men were already reaping the benefits of being in a supportive family environment.
Our study also examined the possibility that the interactive associations between serostatus disclosure to parents, HIV-specific family support, and ethnicity in explaining disease status were mediated by men's psychological or physiological indicators of stress. However, we did not find support for this hypothesis. This suggests that there may be other behavioral or psychosocial variables accounting for the associations between disclosure, support, ethnicity and disease status. For example, although our study controlled for medication adherence and monitored men for recreational drug use, it is possible that disclosure and support processes and disease status were explained by ethnic differences in alcohol consumption. Beyond health behavior explanations there may have been cognitive appraisal variables that mediated these findings. For instance, Latino MLWH may have experienced increased feelings of guilt, shame, or embarrassment (i.e., simpatía
) as a result of disclosing their serostatus. Negative cognitions, particularly those that center on self-blame, have been linked to declines in immune function in MLWH (Kemeny & Dean, 1995
; Segerstrom, Taylor, Kemeny, Reed, & Visscher, 1996
). Unfortunately, our study did allow for us to test the possibility that shame and guilt may function as mediators of disclosure, support, ethnicity and disease status. It will be important for future research to examine the importance of how cognitive processes may accompany the process of serostatus disclosure.
There are several limitations to the current research that should be noted. Because of the cross-sectional design of the study, there is no way of determining if men's disease status was an antecedent or consequence of serostatus disclosure to parents and HIV-specific support from these network members. Some research indicates that individuals may only disclose their serostatus when their disease progresses to a point where they can no longer hide it (Babcock, 1998
); however, other research indicates that individuals disclose to close network members when the benefits of disclosing outweigh the costs (Serovich, 2001
). Moreover, social support consistently predicts well-being in HIV+ individuals (e.g., Catz, Gore-Felton, & McClure, 2002
; Cederfjäll, Langius-Eklöf, Lidman, & Wredling, 2001
; Gielen, McDonnell, Wu, O'Campo, & Faden, 2001
; Hudson, Lee, Miramontes, & Portillo, 2001
). We did not ask men to indicate when they had disclosed their serostatus to each social network member; given the range of years that men in our sample had been living with their HIV-diagnosis, it is likely that a great deal of heterogeneity existed in how long ago men had disclosed their serostatus. Although we did control for the number of months since men were diagnosed with HIV, it is still unclear how our results would differ in men who recently initiated the process of serostatus disclosure versus men who began the process some time ago.
Despite its limitations, our study strengthens and extends research on serostatus disclosure and social support processes in MLWH in at least two ways. First, prior research on serostatus disclosure in MLWH is limited in that it does not consider the social context of the disclosure and does not examine associations among disclosure, social processes, and both virologic and immune indicators of disease. Second, little research has examined ethnic differences in the process of serostatus disclosure or the ways disclosure and social support interact in predicting disease status in MLWH. Our work highlights the complexity of associations between direct disclosures to specific family members and perceptions of illness related family support in different ethnic groups.
Our results may have implications for interventions aimed at alleviating stress and improving disease status in MLWH. Cognitive behavioral stress management (CBSM) interventions can improve both psychological and physical health in persons living with HIV by changing the way individuals appraise stress, including social stress, and teaching interpersonal skills for assertively communicating concerns to others in their social network (Carrico, Antoni, Weaver, Lechner, & Schneiderman, 2005
; Antoni et al., 2005
). However all of this prior work focused the intervention on persons infected with HIV rather than members of their immediate social environment. Psychosocial interventions that include family members may be advantageous for men who have disclosed to family members but are not receiving adequate levels of social support. Our results also suggest that these interventions would likely benefit from being sensitive to cultural and ethnic differences in the social processes that accompany disclosure in the families of MLWH.