Health Insurance and Use of Medical Services by Men Infected with HIV

Effect of race and/or ethnicity in use of antiretrovirals and prophylaxis for opportunistic infection: a review of the literature

“…[33][34][35][41][42][43][49][50][51]57 A more detailed look at these studies yields patterns that help place these results in perspective. For example, among the 10 articles that failed to find any association across all outcomes examined, seven relied on cross-sectional survey data, [33][34][35]41,43,50,51 and four had sample sizes less than 350. 33,34,43,50 These two study characteristics may have limited the ability of the The cohort for OI prophylaxis was different from the cohort for PCP prophylaxis ( …”

“…37,45,48,[52][53][54] Three studies found a positive association with at least one antiretroviral use measure, 40,44,48 two of which found either no association 44 or a negative association 48 across a different measure of antiretroviral use. Sixteen studies found no association between race/ethnicity and at least one measure of antiretroviral use; [33][34][35]37,[41][42][43][44][45][49][50][51][52][53][54]57 however, five of these also found a negative association on at least one antiretroviral use measure, 37,45,52-54 and one also found a positive association. 44 In four of these studies an association between race/ethnicity and antiretroviral use explicitly noted in bivariate analyses lost statistical significance after adjustment for other covariates.…”

“…Overall, 16 studies did not find an association between race/ ethnicity and at least one measure of antiretroviral use. [33][34][35]37,[41][42][43][44][45][49][50][51][52][53][54]57 Among them, seven found a statistically significant association between race/ethnicity and one of the study's measures of antiretroviral use, 37,44,45,[52][53][54] and therefore were discussed in above sections.…”

“…[33][34][35][41][42][43][44][49][50][51]57 Among these ten, five reported on sample sizes of Ͻ350. 30,31,40,47,54 Specifically, Weissman et al measured AZT use among 249 patients with AIDS from the Boston Health Study; 33 Katz et al evaluated use of AZT among 142 HIV-infected homosexual men with a CD4 cell count Ͻ500 cells/mm 3 in San Francisco; 34 Kass et al evaluated use of any antiretrovirals among 307 men enrolled in the Baltimore site of the MACS; 44 Sorvillo et al evaluated protease inhibitor use among 339 people with AIDS in Los Angeles County; 50 and Schwarz et al evaluated subsequent HAART use among 219 HIV-infected adolescents who were HAART-naive at study enrollment. 57 The remaining five studies were not substantially different from the studies which found a negative association in terms of sample size, population selection, or overall methodology.…”

“…The possibility of this limitation is further supported by the fact that seven [33][34][35]41,43,50,51 out of the 12 studies that performed cross-sectional analyses [31][32][33][34][35]39,41,43,45,46,50,51,57 and five 33,34,43,50,57 out of seven studies with analytic sample sizes less than 350 33,34,43,45,52,57 found no association. In contrast, only three 42,49,57 out of 14 studies that performed longitudinal analyses, [36][37][38]40,42,44,[47][48][49][52][53][54][55]57 and five 35,…”