Clinical Investigation

February 1, 2007

The CDC estimates that more than 850,000 people are living with HIV/AIDS in the United States today,1 with African Americans and Latinos being disproportionately affected.2 Over the past decade, the Latino population in the United States has been growing. In fact, in 2002 for the first time, Latinos surpassed African Americans as the largest US minority group. However, the term "Latino" is applied to a very heterogeneous group from different countries and different cultures.3 In addition, a large proportion of Latinos is undocumented in the United States and thus marginalized from the health care system.

Latinos represent a growing number of persons with HIV/AIDS in the United States. Reports of clinical features and therapeutic outcomes of this population are limited. We performed a retrospective cohort study of 75 foreign-born Latinos seen at the Grady Infectious Disease Program in Atlanta between January 2000 and August 2002. Of these, 72% were men with a median age of 38.5 years. The most common risk factor for HIV among the men was having sex with men. Median CD4 count at the time of diagnosis was 119/µL. Most were receiving antiretroviral therapy, with 91% achieving an undetectable HIV RNA level. In a multivariable analysis, men were more likely than women to have a history of sexually transmitted infections, to have received a diagnosis at a lower CD4 count, and to have virologic failure after achieving an undetectable HIV RNA level while receiving antiretroviral therapy. In foreign-born Latinos, HIV disease is usually diagnosed at an advanced stage, but when antiretroviral therapy is administered, many achieve an undetectable HIV RNA level. However, a sustained virologic response is difficult to achieve in men. [AIDS Reader. 2007;17:73-74, 78-80, 85-88]

The CDC estimates that more than 850,000 people are living with HIV/AIDS in the United States today,1 with African Americans and Latinos being disproportionately affected.2 Over the past decade, the Latino population in the United States has been growing. In fact, in 2002 for the first time, Latinos surpassed African Americans as the largest US minority group. However, the term "Latino" is applied to a very heterogeneous group from different countries and different cultures.3 In addition, a large proportion of Latinos is undocumented in the United States and thus marginalized from the health care system.

The existing disparities affecting the Latino population in the United States are also apparent with respect to HIV infection. In 2000, the AIDS incidence rate for Latinos in the United States was 22.5 per 100,000 population, more than 3 times the rate for whites. Among Hispanics/Latinos, males account for 81% of reported AIDS cases. However, the proportion of Latina women living with AIDS appears to be rising. Women represented 23% of cases reported among Hispanics in 2000, well above the 19% of female cases overall. Through 2002, men who have sex with men accounted for 42% of cumulative AIDS cases reported in the United States among Latino men. Injection drug use (IDU) and heterosexual contact each accounted for 17% of cases in Latino men. In comparison, most Latina women have been infected through heterosexual contact (80%) or IDU (17%).4

Inequalities are also seen in persons of color with respect to survival. Survival time after AIDS diagnosis is significantly lower for blacks and Latinos.5 In general, it has been shown that HIV disease is diagnosed at a late stage in Latinos and that these patients have lower CD4 counts, higher HIV RNA levels, more major opportunistic infections, and longer hospital stays than whites.6-8

Latino ethnicity was also associated with delay of initial HIV care for more than 3 months in 2 large cohorts and with decreased use of primary Pneumocystis carinii (or jiroveci) pneumonia (PCP) and Mycobacterium avium-intracellulare complex infection prophylaxis when compared with white patients.9,10 Reasons for delay of care include lack of access to transportation; being too sick to go to the doctor; and having 1 or more competing needs for expenditures, such as rent and food.11 However, there are data to show that the clinical progression of HIV disease in Latinos is similar to that in whites once the "playing field is leveled" (ie, once medical care is initiated and when health care access is not a factor).6,12

AIDS incidence rates are higher for foreign-born Latinos, indicating that a disparity also exists within the Latino population with respect to place of birth.13 Data from the CDC show that 60% of Latinos with reported AIDS were born in the United States. However, 42% of those were actually born in Puerto Rico.2 Place of birth also affects the prevalence of HIV exposure risk factors in the United States. For example, in 2000, 48% of Puerto RicanÐborn AIDS patients were exposed through IDU, while only 9% of Mexican-born patients were exposed through IDU. In contrast, Mexicans (44%) were much more likely to be exposed through male homosexual contact than were Puerto Ricans (14%) or US-born Latinos (30%).2

In recent years, new regions of the United States, especially the Southeast and Midwest, have experienced a great influx of Latino residents. For example, Georgia experienced a greater than 100% increase in the Latino population from 1990 to 2000.14 The US Census Bureau reports that in 2000 approximately 435,000 Hispanics lived in Georgia, which is about 4 times the number counted in 1990. However, the number may be as high as 750,000 when undocumented residents are included.15

The presence of increasing numbers of Latinos affected by HIV highlights the need to better describe this population if advances in HIV prevention and treatment are to be accomplished. HIV prevention and early intervention efforts among Latinos must recognize and address the diverse makeup of the population to be successful.2,13,16

Our study was undertaken to describe the clinical and epidemiologic characteristics of the emerging epidemic of HIV/AIDS among foreign-born, HIV-infected Latino patients at an urban HIV clinic in the southeastern United States.

MATERIALS AND METHODS
We performed a retrospective cohort study of all foreign-born adult patients from Spanish-speaking countries in Latin America seen at the Grady Health System Infectious Disease Program (IDP) in Atlanta. Written and electronic medical records between January 2000 and August 2002 for all foreign-born Latino patients were reviewed. If at any point it was determined that the patient was, in fact, not Latino or not born in a Spanish-speaking Latin American country, the patient was excluded from the data collection process.

For each patient, a data collection form was used to record demographic characteristics, risk factors for HIV infection, HIV status and progression to AIDS, antiretroviral treatment history, clinical and virologic outcomes, and provider characteristics. Compliance with antiretroviral therapy was determined by review of pharmacy pickup records.

All data were analyzed using SAS version 9 software. General descriptive statistics of the population were generated. In addition, bivariate analysis was performed to evaluate differences between male and female patient populations. The Wilcoxon rank sum test and Student's t test were used, where applicable, to compare means between these groups, and the Fisher exact test was used to compare categorical variables. Logistic regression was used for multivariate analysis to determine gender differences. This study was approved by the Institutional Review Board of Emory University and the Grady Research Oversight Committee.

RESULTS
Retrospective review of patient records identified 109 possible foreign-born adult Latino patients. Of these, 34 patients were excluded: 20 were excluded because examination of their records revealed that they either did not identify themselves as Latino or were not born outside of the United States, and 8 were excluded because examination of their clinic records revealed that they had not, in fact, had a clinic visit at the IDP between January 2000 and August 2002. Finally, 6 patients were
excluded because their clinic chart could not be located.

Demographic data for the 75 patients included in the study are presented in Table 1. Overall, 72.0% of the sample were men with a mean age of 36.9 years (range, 26 to 61), with more than half born in Mexico (62.6%) and single (52.0%). The mean number of years spent in the United States was 10.6 (range, 1 to 40); however, 54.6% spoke no English and close to half did not have a secondary school education. At the time of data collection, roughly two thirds (69.4%) were employed, but only 17.3% had health insurance.

Risk factors for HIV infection in this cohort are also presented in Table 1. Male homosexual contact was the most common risk factor (42.6%). Close to half (49.3%) of the study cohort reported a previous, sexually transmitted infection (STI) diagnosis. In addition, this study cohort was characterized by a large proportion of patients reporting a history of substance abuse, psychiatric disorder, or hepatitis. A history of substance abuse was reported by 26.6% even though IDU was identified as a risk factor by only 8.0%.

The most common reason for HIV testing was seeking treatment for an acute illness in 31 patients (41.3%), with 21 (67.7%) of the 31 tested while hospitalized. Of those who received a diagnosis of HIV infection during an acute illness, 16 also had an AIDS-defining opportunistic infection or neoplasia. Among these, PCP was the most common, found in 7 patients, followed by tuberculosis, cerebral toxoplasmosis, and disseminated histoplasmosis in 2 patients each.

Patients presented to the IDP for HIV care very late in the course of their disease, with a median CD4 count of 119 cells/µL and a median HIV RNA level of 5.2 log10 copies/mL. Nearly three fourths presented with a CD4 count of less than 200 cells/µL, meeting the definition of AIDS. After HIV diagnosis and during follow-up at the IDP, an STI was diagnosed in 23.0% and hepatitis was diagnosed in 5.0%.

Antiretroviral therapy had been prescribed for more than 93.0% of the patients in this cohort at some time during follow-up, and 88.0% were on an antiretroviral regimen at the time of data collection. The most common reason for not receiving antiretroviral therapy was the providerÕs decision that therapy was not indicated because of high CD4 counts. The most common first antiretroviral regimen prescribed was a protease inhibitor (PI)-based regimen (67.0%), with the next most common being an NNRTI-based regimen (12.0%). Data were similar when examining current antiretroviral regimens. The median number of months patients had been taking antiretroviral therapy was 39.5, and the mean number of regimens prescribed was 2.0.

For the 71 patients with available HIV RNA data in follow-up, 91.5% achieved an undetectable HIV RNA level (< 400 copies/mL) at some point during antiretroviral therapy; the median number of months on therapy until an undetectable HIV RNA level was reached was 3.5. However, 70.0% of patients experienced a rebound to a detectable HIV RNA level after achieving an undetectable level at some point during follow-up. When antiretroviral-naive patients were examined at 3, 6, 9, and 12 months after initiation of antiretroviral therapy, about half had an undetectable HIV RNA lev-el. The median CD4 count recorded at the last visit before data collec-
tion was 340 cells/µL, and the median HIV RNA level was 2.6 log10 copies/mL.

In a bivariate analysis, men and women did not differ with respect to age, country of origin, education level, number of years in the United States, or ability to speak English. However, men were more likely to be employed than women (79.6% vs 40.0%; P = .002) and to have a his-tory of substance abuse (48.2% vs 19.1%; P = .03).

Important differences emerged between men and women when clinical and virologic variables were examined (Table 2). Women were much more likely than men to present for follow-up at the IDP earlier in the course of HIV disease (median CD4 count 224.5 vs 55.5 cells/µL; P = .0004). Only 10.0% of women presented with a CD4 count less than 200 cells/µL compared with 57.7% of men (P = .0004). Of note, 12 of 21 women were tested for HIV because of pregnancy (23.8%) or because a spouse had tested HIV positive (33.3%). No significant differences were seen in type of antiretroviral regimen prescribed for men and women. However, women were much less likely than men to rebound to a detectable HIV RNA level after achieving an undetectable level while receiving antiretroviral therapy (41.2% vs 80.9%; P = .004).

Multivariate logistic regression analysis comparing Latino men with Latina women demonstrated that when controlling for age, adherence, and the number of months on antiretroviral therapy, men were more likely to have a history of an STI before their first appointment at the IDP clinic (odds ratio [OR], 10.5; 95% confidence interval [CI], 1.2 - 95.1); to present later in the course of disease with a CD4 count less than 200 cells/µL (OR, 8.6; 95% CI, 1.3 - 58.7); and to rebound to a detectable HIV RNA level after achieving an undetectable viral load on antiretroviral therapy (OR, 31.8; 95% CI, 2.3 - 443.0) (Table 3).

DISCUSSION
This retrospective study of 75 HIV-infected patients born in Latin America and seen at the Grady IDP in Atlanta between January 2000 and August 2002 sought to better describe the epidemiologic and clinical characteristics of this emerging patient population. Because an increasing number of foreign-born Latinos are affected by HIV/AIDS, information gained from studies such as this will be necessary to adequately address the needs of HIV-infected Latinos as well as to prevent further infections in their communities. In our largely male Mexican population, we found that while patients were presenting late in the course of disease, once they were given care and were started on an antiretroviral regimen, they did quite well. However, for women, there are clear gender differences with regard to presenting earlier in the course of disease and having sustained virologic responses while receiving antiretroviral therapy.

The CDC reports that the proportion of Hispanic women living with HIV/AIDS appears to be rising, indicating the spread of a generalized heterosexual epidemic.2 In the Atlanta foreign-born population, this phenomenon appears to be reality. From the data in this study, heterosexual sex emerged as the predominant risk factor among this population. It is therefore not surprising that compared with national statistics, a greater proportion of the patients seen at the IDP are women (23% vs 28%).

It can also be inferred that many persons in this population were infected within the marital relationship: 27% reported being married and having only 1 sexual partner.
As discussed below, this concept is essential in the development of future prevention interventions. In addition, a significant proportion (over 13%) of men reported having sex with both men and women. This was not unexpected, but it reaffirms previous data, which suggest that bisexuality is common among Latino males.17

Other demographic indicators were generally similar to those for the US Latino population. These include a majority being from Mexico, having a low level of education, and having an inability to speak English, and very few having health insurance.18,19 While exact details could not be obtained from the medical records, this information is an important reminder that many of the foreign-born Latinos infected with HIV in Atlanta are also facing some of the multiple societal obstacles, such as documentation status, poverty, and ability to obtain health care, that plague the US Latino immigrant community. These societal obstacles may lead to a late diagnosis in these persons; more than 40% of Latinos receive an HIV diagnosis during an acute illness, as seen in this study.

For the majority of Latino patients seen at the IDP, antiretroviral therapy had been prescribed, and 88% were still receiving treatment at the time of data collection. This attrition rate of 6% is quite impressive when one considers a recent study in which 35% of Medicaid beneficiaries discontinued antiretroviral therapy after initiation over a 3-year period.20 These numbers indicate that foreign-born Latinos at the IDP clinic are receiving the current standard of care with respect to antiretroviral treatment and are achieving good virologic response. This is in contrast to a previous study done in an urban HIV clinic with mostly African American patients in which over 26% did not receive antiretroviral therapy.21 When the regimen was examined by type, almost all patients were receiving either a PI in combination with 2 NRTIs or an NNRTI in combination with 2 NRTIs. Once again, these regimens were consistent with the standard of care in this country at the time the patients began antiretroviral therapy, and the data do not support previous data showing that the use of PIs is decreased among foreign-born Latinos compared with whites.22,23

In this study, 91% of all patients were able to achieve an undetectable HIV RNA level (defined as less than 400 copies/mL) after initiation of antiretroviral therapy. This impressive response to treatment, however, is overshadowed when one considers that more than 70% of these patients rebounded to detectable levels and that at 3-month intervals after beginning antiretroviral treatment, only about half of the population had an undetectable HIV RNA level. This shows that while initial response to antiretroviral therapy may be quite good, a sustained response is difficult to achieve. Because intermittent detectable HIV RNA levels are associated with poor outcomes, further studies are necessary to determine the obstacles to long-term virologic response among foreign-born Latinos.24 Socioeconomic factors, cultural factors, and adherence are all likely to play important roles in sustaining virologic response and, as has been shown in other studies, HIV knowledge and health literacy may also be important.25,26

HIV infection was diagnosed earlier in the course of disease in foreign-born Latina women in this cohort than it was in men. This is likely the result of the success of
such programs as prenatal HIV testing, in which 23.6% of women were tested. Also of note, an additional 23.6% of women were tested because a spouse tested positive. Assuming these women were asymptomatic at the time of testing, this data supports early testing programs to identify those who are infected but are asymptomatic and provide them with care.

Another significant variable in the logistic regression analysis associated with male gender was having a history of an STI. Unfortunately for women, this could indicate that HIV infection is the first STI that they are exposed to and, thus, valuable health care contact before infection with HIV is missed. For men, this implies that opportunities could exist for prevention education at first diagnosis of an STI. Finally, Latina women at the IDP seem to have better outcomes with respect to maintaining undetectable HIV RNA levels. This is likely related to gender difference within the Latino culture that would benefit from further exploration.

There are several limitations of this study that must be taken into account when attempting to generalize its results for the overall population of foreign-born Latinos and Latinas living with HIV/AIDS in the United States. First, since this was a retrospective study, we were unable to collect all variables on each study participant because some charts did not include such data. Second, certain providers may not record information in the medical record with as much detail as other providers, so the data from their patients might be lacking. Third, the sample size of this study is small and limited to an urban Atlanta HIV clinic that typically serves uninsured or underinsured patients. Thus, the generalization of results beyond this setting is not possible.

However, despite these limitations, we believe that in the absence of large cohort studies or detailed clinical and epidemiologic reports of this particular population in states that have not had until recently a significant Latino immigrant population, the results of our study provide important preliminary information. The results may also provide important insight for practicing clinicians in the southeastern United States who are dealing with the growing population of foreign-born Latino patients in this region. Finally, understanding differences in cause-specific mortality between Latinos and whites is important for targeting future public health interventions and research aimed at eliminating health disparities.27

In summary, our study describes several clinical and epidemiologic features of foreign-born Latinos living with HIV/AIDS in the southeastern United States. Given the fact that in the United States Hispanics are an emerging at-risk group for HIV/AIDS, we believe that the preliminary results of our study provide useful information for HIV care providers by helping to delineate the best possible approaches to target both early diagnostic and therapeutic strategies specific for the US Latino community.

Acknowledgments:

This study was supported in part by the NIH/FIC/AIDS International Training and Research Program of Emory University (D43 TW01042) and the NIH/NIAID Center for AIDS Research of Emory University (2P30 AI 50409-04A1).

No potential conflict of interest was reported by the authors.

References:


1. Fleming PL, Byers RH, Sweeney PA, et al. HIV Prevalence in the United States, 2000. 9th Conference on Retroviruses and Opportunistic Infections; February 24-28, 2002; Seattle; Abstract 11.
2. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Reports. Cases of HIV/AIDS in the United States and dependent areas. 2005;17:10-46. Available at:

http://www.cdc.gov/hiv/topics/surveillance/resources/reports/2005report/pdf/2005SurveillanceReport.pdf

Accessed January 10, 2007.
3. Ramirez RR, de la Cruz GP. The Hispanic population in the United States: March 2002. Current Population Reports. Washington, DC: US Census Bureau; 2003:20-545.
4. Centers for Disease Control and Prevention. HIV/AIDS among Hispanics in the United States. Available at: http://www.cdc.gov/hiv/Pubs/facts/Hispanic.htm. Accessed January 10, 2007.
5. Murrain M. Differential survival in blacks and Hispanics with AIDS. Ethn Health. 1996;1:373-382.
6. Swindells S, Cobos DG, Lee N, et al. Racial/ethnic differences in CD4 T cell count and viral load at presentation for medical care and in follow-up after HIV-1 infection. AIDS. 2002;16:1832-1834.
7. Menke TJ, Giordano TP, Rabeneck L. Utilization of health care resources by HIV-infected white, African American, and Hispanic men in the era before highly active antiretroviral therapy. J Natl Med Assoc. 2003;95:853-861.
8. Sherer R, Pulvirenti J, Stieglitz K, et al. Hospitalization in HIV in Chicago. J Int Assoc Physicians AIDS Care. 2002;1:26-33.
9. Turner BJ, Cunningham WE, Duan N, et al. Delayed medical care after diagnosis in a US national probability sample of persons infected with human immunodeficiency virus. Arch Intern Med. 2000;160:2614-2622.
10. Asch SM, Gifford AL, Bozzette SA, et al. Underuse of primary Mycobacterium avium complex and Pneumocystis carinii prophylaxis in the United States. J Acquir Immune Defic Syndr. 2001;28:340-344.
11. Altman D, Lillie-Blanton M. Racial/ethnic disparities in medical care. BMJ. 2003;3:300-301.
12. Anastos K, Gange SJ, Lau B, et al. Association of race and gender with HIV-1 RNA levels and immunologic progression. J Acquir Immune Defic Syndr. 2000;24:218-226.
13. Klevens RM, Diaz T, Fleming PL, et al. Trends in AIDS among Hispanics in the United States, 1991-1996. Am J Public Health. 1999;89:1104-1106.
14. Kent MM, Pollard KM, Haaga J, Mather M. First glimpses from the 2000 US Census. Population Bulletin. 56(2). Available at:

http://www.prb.org/Template.cfm?Section=Population_Bulletin1&template=/ContentManagement/ContentDisplay.cfm&ContentID=7502#hispanic

. Accessed January 10, 2007.
15. US Census Bureau. State and County Quick Facts: Georgia. Available at: http://quickfacts.census.gov/qfd/states/13000.html. Accessed May 19, 2006.
16. Murphy J, Mueller G, Whitman S. Epidemiology of AIDS among Hispanics in Chicago. J Acquir Immune Defic Syndr. 1996;11:83-87.
17. Lehner T, Chiasson MA. Seroprevalence of human immunodeficiency virus type 1 and sexual behaviors in bisexual African-American and Hispanic men visiting a sexually transmitted disease clinic in New York City. Am J Epidemiol. 1998;147:269-272.
18. Holmes M. Latinos, immigrants, and the uninsured. N C Med J. 2006;67:202-205.
19. Schmidley D. Profile of the foreign born population in the United States: Current Population Reports. Washington, DC: U.S. Census Bureau; 2001:23-206. Available at: http://www.census.gov/prod/2002pubs/p23-206.pdf. Accessed January 10, 2007.
20. Crystal S, Sambamoorthi U, Moynihan PJ, McSpiritt E. Initiation and continuation of newer antiretroviral treatment among Medicaid recipients with AIDS. J Gen Intern Med. 2001;16:850-859.
21. Giordano TP, White AC Jr, Sajja P, et al. Factors associated with the use of highly active antiretroviral therapy in patients newly entering care in an urban clinic. J Acquir Immune Defic Syndr. 2003;32:399-405.
22. Yeni PG, Hammer SM, Carpenter CC, et al. Antiretroviral treatment for adult HIV infection in 2002: updated recommendations of the International AIDS Society-USA Panel [published correction appears in JAMA. 2003;11:32]. JAMA. 2002;288:222-235.
23. Sorvillo F, Kerndt P, Odem S, et al. Use of protease inhibitors among persons with AIDS in Los Angeles County. AIDS Care. 1999;11:147-155.
24. Easterbrook PJ, Ives N, Waters A, et al. The natural history and clinical significance of intermittent viraemia in patients with initial viral suppression to < 400 copies/ml. AIDS. 2002;16:1521-1527.
25. Rojas-Guyler L, Ellis N, Sanders S. Acculturation, health protective sexual communication, and HIV/AIDS risk behavior among Hispanic women in a large midwestern city. Health Educ Behav. 2005;32:767-779.
26. Miller JE, Guarnaccia PJ, Fasina A. AIDS knowledge among Latinos: the roles of language, culture, and socioeconomic status. J Immigr Health. 2002;4:63-72.
27. Wong MD, Tagawa T, Hsieh HJ, et al. Differences in cause-specific mortality between Latino and white adults. Med Care. 2005;43:1058-1062.