In the case of HIV+ ethnic minority women infected by a partner, HIV may be viewed as a disease of couple relationships. Based upon the stress/social support hypothesis (Burman & Margolin, 1992), this study aimed to (a) test if relationship functioning and conflict affected CD4 T-cell count, and whether this relationship was mediated by depression, social support, and social undermining in HIV-positive Latinas and African American women, and to (b) test whether there were significant differences between the HIV-positive Latinas and the HIV-positive African American women in the hypothesized model.

Secondary analyses were performed on a sample of HIV-positive Latinas (n=58) and African American Women (n=64), all of whom were in a romantic relationship, from the University of California-Los Angeles Charles Drew Medical Center Women and Family Project (Wyatt & Chin, 1992). Multilevel path analyses revealed that, among the Latinas, higher relationship functioning significantly but negatively predicted CD4 count (−0.019, p< 0.05), and that social support significantly mediated this association (0.012, p=0.012). Increased social support was associated with higher CD4 counts among the Latinas. Among the Black women, no relationship factors significantly predicted CD4 count, but social undermining significantly and negatively mediated the association between couple conflict and CD4 count (−1.227, p=0.032). Increased social undermining was associated with lower CD4 counts among the Black women. Between-group analyses demonstrated that social undermining was a stronger predictor of CD4 count among the African Americans, and depression was more strongly associated with social support among the Latinas.

These results suggest that poor relationship quality, couple conflict, and social support and undermining can both directly and indirectly affect immune health among HIV+ women of color. Implications are discussed.