We performed a systematic review evaluating timing of HIV diagnosis and survival after HIV/AIDS diagnosis for Hispanics in the US. Despite various definitions of delayed diagnosis, the majority of studies showed that Hispanics were more likely to be diagnosed later than Whites. This is particularly concerning in light of data from two large studies showing a survival benefit for patients initiating ART before the CD4 count declines to less than 500 cells/ll [34
], and recent changes in the DHHS guidelines recommending earlier initiation of antiretroviral therapy [36
]. Although there was a trend towards lower survival among Hispanics with HIV/AIDS compared to Whites, there were notable exceptions that highlight the heterogeneity of the Hispanic population in the US.
Almost all of the studies reviewed showed that Hispanics present later in the course of HIV disease compared to Whites, with males and foreign-born Hispanics at greatest risk for late diagnosis. Only one study, involving individuals with heterosexually-acquired HIV, showed no difference in late diagnosis between Whites and Hispanics [11
]. It did, however, show later presentation among Hispanic men compared to Hispanic women [11
]. Similarly, among foreign-born Hispanics presenting to a clinic in Atlanta, GA, men were 8.6 times more likely to present with AIDS than Hispanic women [18
]. Earlier detection of HIV among Hispanic women may be related to opportunities for HIV screening at routine gynecology or prenatal visits [11
]. In support of this hypothesis, the Atlanta study reported that that 25% of the women were diagnosed during prenatal visits [18
]. Even though Hispanic women are less likely than Hispanic men to present late, Hispanic women are still more than twice as likely as White women to present late in the disease [12
]. Therefore, HIV testing and prevention efforts need to be expanded to prevent late diagnosis among all Hispanics. A culturally acceptable method to increase testing among Hispanic women may involve expanding existing services offered at family planning and women’s health visits. Expanding testing among Hispanic males is more challenging, since many do not access the medical system until they are symptomatic with disease [37
]. The expansion of culturally-appropriate street and community outreach education and testing may be necessary to reach Hispanic men.
Delayed diagnosis among foreign-born Hispanics is not surprising given well-documented barriers in access to care, particularly for individuals without health insurance and limited English proficiency [38
]. Given the clear benefits associated with early initiation of ART, we expected that delayed diagnosis would decrease survival among Hispanics. However, the data on survival after HIV or AIDS diagnosis were less consistent than the data on late presentation.
Among 11 studies evaluating ethnic and racial differences in survival after HIV or AIDS diagnosis, six showed worse survival, three showed no difference in survival, and three showed better survival among Hispanics compared to Whites [11
]. One study found that Hispanics had worse survival than Whites after AIDS diagnosis, but equivocal survival after HIV diagnosis [17
]. Hispanics in two specific transmission risk categories (MSM and heterosexual) were reported to have better survival than Whites. An important caveat is that most studies adjusted risks and survival estimates for stage of HIV at diagnosis (CD4 count, AIDS diagnosis, or OI), thereby masking the impact of late diagnosis on survival. Other possible explanations for the discrepancy between late diagnosis and survival include inaccurate data collection, such as differential underreporting on death certificates, especially if a significant proportion of foreign-born Hispanics with HIV return to their country of origin, or misclassification of Hispanic ethnicity as White. Another explanation for the discrepancy between survival and late presentation/access to care is that once engaged in HIV care, Hispanics are receiving appropriate care. In fact, outcomes after initiation of ART were similar between Hispanics and Whites, suggesting adequate adherence to therapy among Hispanics engaged in care [14
It is important to note, however, that 50% of the studies found worse survival after HIV/AIDS diagnosis among Hispanics, and that specific Hispanic subgroups appear to be at highest risk for poor outcomes. In particular, even though Puerto Rican-born Hispanics do not present late compared to US-born Hispanics, Puerto Rican-born Hispanics with AIDS have significantly lower survival than US-born Hispanics [13
]. Puerto Ricans have the highest rates of HIV transmission through intravenous drug use, and data were adjusted for transmission category, but unknown factors such as adherence and co-morbidities associated with IDU can affect survival estimates. In addition, non-mainland Puerto Ricans may be primarily receiving HIV treatment in Puerto Rico, and differences in mortality may be secondary to system-level differences in the health care system in Puerto Rico compared to mainland US [13
Although foreign-born Hispanics were at greater risk for late diagnosis of HIV/AIDS than US-born Hispanics [12
], this delay in diagnosis did not always translate into worse survival. National surveillance data showed better survival among Hispanics with AIDS born in Mexico and Central America compared to US-born Hispanics [13
]. However, a higher risk of death in Central and South Americans compared to US born Hispanics was reported in a study from New York [26
]. Although HIV care is available to uninsured patients through the Ryan White Act, other factors such as language barriers, lack of familiarity with the US healthcare system, stigma, or fear of deportation often affect access to care. Both the Espinoza and Hanna studies analyzed survival after AIDS diagnosis, thereby controlling for late presentation to care [13
]. The reason for the conflicting results is unclear, but the discrepancies highlight the importance of evaluating local populations and specifically defining Hispanic subgroups to properly evaluate high risk groups that may be underrepresented in data from the heterogeneous Hispanic population as a whole.
Our study has several limitations. Many of the studies evaluated CDC name-based surveillance data that do not include HIV data from some states with large Hispanic populations, such as California. In addition, surveillance data and vital statistics data are subject to racial/ethnic misclassification, which is particularly problematic for Hispanics [41
]. Analysis of HIV reported cases or HIV cases engaged in care does not provide information about HIV-related mortality in individuals who were never diagnosed with HIV. Given significant delays in diagnosis among Hispanics, it is possible that HIV-related mortality was underestimated or underreported. Finally, and perhaps most importantly, the majority of the studies did not distinguish among Hispanic subgroups. Few studies reported key variables related to migration, such as country of origin, and none reported in-depth analysis of acculturation factors, although one did mention language of interview. Although systematic review of the data allowed us to discern some higher risk groups, such as Hispanic males, Puerto Ricans, and foreign-born Hispanics, the data were insufficient for a conclusive analysis.