Weintrob, Amy C MD*†; Grandits, Greg A PhD*‡; Agan, Brian K MD*; Ganesan, Anuradha MD*§; Landrum, Michael L MD*¶; Crum-Cianflone, Nancy F MD, MPH*‖; Johnson, Erica N MD*¶; Ordóñez, Claudia E MA**; Wortmann, Glenn W MD*†; Marconi, Vincent C MD*¶; and the IDCRP HIV Working Group
Several observational studies have demonstrated decreased virologic response to highly active antiretroviral therapy (HAART) in African Americans (AA) compared with European Americans (EA) infected with the HIV.1-4 A study of 961 HIV-infected women in the Women's Interagency HIV Study found that AA were 30% less likely than EA to achieve viral suppression after HAART initiation and were 30% more likely to experience viral rebound after suppression.1 After adjusting for continued antiretroviral use; however, there were no longer significant differences between the ethnicities in virologic suppression or rebound.1 Similarly, another analysis of 385 HIV-infected men and women from the southeastern United States demonstrated that minorities (89% of whom were AA) initiating antiretroviral therapy (ART) discontinued the therapy and experienced virologic failure more rapidly than EA.3 A recent prospective randomized trial conducted by the Adult AIDS Clinical Trials Group comparing class-sparing regimens for initial HIV therapy in 753 subjects also found AA ethnicity to be one factor associated with virologic failure in their analyses.5
Disparate virologic responses to HAART between ethnicities are likely multifactorial. Factors that have been proposed to contribute include differences in the following: duration of HIV infection before HIV diagnosis or HAART initiation,4-10 access to medications and to the health care system,1,3 rates of adherence,4,11,12 and tolerability of HAART regimens.13,14 We evaluated virologic response to HAART between ethnicities in a large US military cohort of HIV-infected persons where many of these factors are minimized: where there is equal access to free health care and medications, where dates of HIV seroconversions are known, and where active duty members are required to follow-up with HIV specialists at regular intervals. Although potential psychosocial reasons for differences in response to HAART may still exist in this cohort (including physician or patient bias or issues of mistrust), removing some of the other confounders may provide a clearer understanding of whether a differential response to HAART between ethnic groups is influenced by other factors that are not as well characterized.
The US Department of Defense (DoD) HIV Natural History Study (NHS) is an ongoing, prospective, continuous enrollment cohort study of consenting military personnel and beneficiaries with HIV infection and includes participants from the Army, Navy/Marines, Air Force, and their dependents. Since 1985, routine HIV testing (enzyme-linked immunosorbent assay and confirmatory western blot) has been used to exclude HIV-infected persons from enlisting for military service or from overseas deployment. Routine testing among active duty members occurs every 1-5 years, resulting in a defined seroconversion window for incident HIV infection. Subjects with HIV infection are referred to military medical centers where they receive evaluation, and ongoing care by a multidisciplinary team including an HIV specialist, and are invited to enroll as participants in the DoD HIV NHS. All care, medications, and travel are provided free of charge regardless of whether they are on active duty. HIV-infected service members can stay on active duty as long as they remain fit for duty. After they serve for at least 20 years, they can retire with full medical benefits.
Those who consent to enroll in the DoD HIV NHS are seen every 6 months by a HIV specialist as part of the study, in addition to receiving routine clinical care. Data are collected on demographic characteristics, markers of HIV disease progression, medication use, and clinical events with medical record confirmation. Institutional Review Boards at each medical center and centrally reviewed and approved the DoD HIV NHS. The central Institutional Review Board approved this analysis.
Information was extracted from the database on HIV-infected individuals who initiated HAART between 1996 and 2007 and had a viral load (VL) result available within 6 months before starting HAART and 6 months (±3 months) after HAART initiation. The information included dates of the last negative and first positive HIV tests, age, sex, self-reported ethnicity, military rank, active duty status, antiretroviral use before HAART, date of HAART initiation along with first regimen, CD4 counts, and VLs at HIV diagnosis and before and after HAART initiation, AIDS defining illnesses before HAART initiation, coinfection with chronic viral hepatitis or other sexually transmitted diseases (STD), clinical diagnosis of major depression, and serum levels of hemoglobin and creatinine at the time of HAART initiation.
For those subjects with documented negative HIV tests, the date of seroconversion was estimated as the midpoint between the last documented negative HIV test and the first documented positive HIV test.
For this study, HAART refers to a regimen of
1. Two or more nucleoside reverse transcriptase inhibitors (NRTIs) in combination with at least one protease inhibitor (PI) or one nonnucleoside reverse transcriptase inhibitor (NNRTI) or
2. One NRTI with at least one PI and one NNRTI or
3. An abacavir or tenofovir containing regimen of 3 or more NRTIs in the absence of both PIs and NNRTIs.
ART use refers to therapy with drugs not meeting the definition of HAART. As the DoD HIV NHS is an observational study, therapy was initiated by individual providers based on treatment guidelines and patient preferences.
VL at HIV diagnosis was defined as the earliest VL available within 6 months after first HIV-positive test. VL at HAART initiation was defined as the latest measurement within 6 months before initiating HAART. CD4 cell counts were similarly defined. Because VL assays with different lower limits of detection were used in the 1990s, an undetectable VL was defined as <400 copies/mL. Viral suppression refers to having an undetectable VL. The primary objective was to examine virologic response at 6-months post-HAART initiation as HIV treatment guidelines suggest that viral suppression should be achieved by 16-24 weeks posttherapy.15 Virologic response at 12 months and maintenance of viral suppression were also evaluated.
Standard t tests and chi-square tests were performed to compare AA and EA on characteristics at HIV diagnosis and at HAART initiation. The percent achieving VL suppression at 6 and 12 months were compared between ethnicities using chi-square tests, and changes in VL (log10) from HAART initiation were compared using t tests. Logistic regression was used to compare AA and EA (adjusted for age at HAART initiation, gender, military rank, VL and CD4 count at HAART initiation, AIDS diagnoses and ARV use before HAART, HAART regimen, hemoglobin at HAART initiation, hepatitis B coinfection, year of HAART start, STD after HIV infection, and diagnosis of major depression) on odds of achieving viral suppression at 6- and 12-months post-HAART initiation. Kaplan-Meier (KM) methods were used to estimate the time from virologic suppression to virologic failure (defined as the first of 2 consecutive detectable VLs) for (1) subjects who achieved an undetectable VL at 6 months post-HAART initiation and (2) subjects who achieved an undetectable VL more than 6 months after HAART initiation. The log-rank test was used to compare the KM curves between AA and EA for each analysis. Cox-regression was used to estimate the corresponding hazard ratio (HR), AA vs EA. All analyses were performed using SAS software, Version 9.1 (Cary, NC).
As of December 31, 2007, 4926 subjects had enrolled in the DoD HIV NHS and 2273 had initiated HAART. Of the 2273, 999 (44%) identified themselves as AA, 1003 (44%) as EA, and 271 (12%) as other ethnicities. This study was limited to AA and EA due to the paucity of subjects belonging to other ethnicities. There were 401 subjects (20%) who were excluded from this study due to missing VLs before HAART initiation and 238 (12%) additional subjects who were excluded due to missing VLs 6 months after HAART initiation. On average, the excluded subjects were diagnosed with HIV infection during an earlier calendar year (1992 vs 1995), started HAART during an earlier calendar year (1998 vs 1999), had a longer duration between HIV infection and HAART initiation (68 vs 50 months), had lower CD4 counts at HAART initiation (281 vs 362 cells/mm3), and were more likely to start a PI-based regimen (77% vs 61%) than those included in the study (all P values <0.001). There was no difference between the percentage of AA and EA excluded because of missing VLs (31% of AA and 33% of EA).
Table 1 shows the baseline characteristics of the 1363 AA and EA who started HAART and had 6-month VL results available. Compared with EA, AA were more likely to be younger at HAART initiation, women, enlisted, coinfected with hepatitis B or hepatitis C virus, diagnosed with an STD after HIV infection, and have lower hemoglobin and higher creatinine concentrations at HAART initiation (Table 1). There was no difference between ethnicities in percentage with prior AIDS diagnoses or prior ART use, in the number of visits per year, or in initial HAART regimens used. Approximately, 21% of subjects in both groups started a nelfinavir-based regimen, 26% started indinavir, and 29% started efavirenz. Over 75% of both AA and EA had dates of negative HIV tests before first positive HIV test providing an estimated date of seroconversion (90% of the first positive HIV tests were within 44 months of the last negative test). There was no difference between ethnicities in time from seroconversion to HIV diagnosis, time from HIV diagnosis to HAART initiation, or time from CD4 nadir to HAART initiation (Fig. 1). Despite the similar durations of time between seroconversion and diagnosis, AA had lower CD4 counts than EA at HIV diagnosis and at HAART initiation. These results remained the same when CD4 percentage was analyzed instead of absolute CD4 count (Table 1). There was no significant difference in the rate of CD4 decline from HIV diagnosis to HAART initiation between AA and EA (Fig. 1, P = 0.21).
Six months after starting HAART, 63% of AA compared with 75% of EA had undetectable VLs (Table 2, P < 0.001). Decreases in VL at 6 months were also significantly smaller in AA than in EA (1.6 log10 vs 1.9 log10, P < 0.001). Limiting the analysis to subjects who were nonofficers did not change the results (data not shown). There was no significant difference between the 2 groups in the percentage of subjects who remained on the same HAART regimen at 6 months (71% and 70%, respectively, for AA and EA). Approximately, 10% in each group discontinued HAART altogether by 6 months and 19% changed their HAART regimen. At 12 months, a significant difference in virologic suppression between AA and EA remained (Table 2, P < 0.001). Restricting the analyses to patients remaining on their initial regimen showed similar differences with 71% of AA and 83% of EA achieving viral suppression at 6 months (P < 0.001) and 74% and 85%, respectively, at 12 months (P < 0.001). Table 3 shows differences in viral suppression at 6-months post-HAART by initial HAART regimen. The largest differences in viral suppression between AA and EA were with PI-based regimens.
Table 4 shows the odds ratios for selected factors of having an undetectable VL at 6-months post-HAART from a multivariate logistic regression model. Serum creatinine at HAART initiation and hepatitis C coinfection were not significant in the univariate analysis and were therefore not included in the multivariate model. The odds of obtaining an undetectable VL after 6 months of HAART was 0.6 for AA compared with EA [P < 0.001, 95% confidence interval (CI) 0.4-0.8] after adjusting for age and military rank at HAART initiation, sex, VL and CD4 count at HAART initiation, AIDS and ARV use before HAART, specific HAART regimen, serum hemoglobin at HAART initiation, hepatitis B coinfection, year of HAART start, STDs after HIV diagnosis, and diagnosis of major depression. Similarly, the odds ratio for AA at 12-months post-HAART was 0.6 (P = 0.002, 95% CI 0.4-0.8). The results did not significantly change when the analyses were limited to only subjects with estimated dates of seroconversion, only subjects who were not officers, only subjects who were diagnosed with HIV infection in the HAART era, or only subjects who were active duty (data not shown).
Of those AA and EA who achieved a VL < 400 c/mL 6 months after HAART initiation, there was no significant difference between AA and EA in time from suppression to virologic failure [HR (AA vs EA): 1.2, 95% CI 0.9-1.6, P = 0.18]. Of those who started HAART but did not achieve an undetectable VL until more than 6 months after HAART initiation, there was a trend toward AA experiencing virologic failure more quickly; however, this did not reach statistical significance (HR: 1.3, 95% CI 1.0-1.7, P = 0.07).
In this evaluation of over 650 AA and 650 EA initiating HAART in a military, longitudinal cohort study with open access to health care and free medications, AA had 40% lower odds of achieving viral suppression than EA at both 6 and 12 months after therapy initiation. Of AA achieving viral suppression at 6 months; however, their time to virologic failure was not significantly different than EA. Therefore, there seem to be factors that contribute to AA having decreased rates of initial virologic suppression, but those without these factors (ie, AA who achieve viral suppression at 6-months post-HAART) are able to maintain viral suppression over time similar to EA.
Several other studies have also found that AA do not have the same initial virologic response to HAART as EA.1-5 A study examining sociodemographic characteristics of persons receiving HIV-related care in the southeastern United States found that those of minority ethnicities discontinued ART more rapidly and demonstrated virologic failure more rapidly than whites. The authors noted that lack of private health insurance disproportionately affected the ethnic minorities and theorized that the reliance on state run AIDS drug assistance programs, clinical trials, or charity care programs may have made gaps in coverage more likely resulting in an inability to obtain medications.3 A study of 961 women (including 573 AA and 184 white) in the Women's Interagency HIV Study found that white women exhibited better immunologic and virologic responses to HAART and lower rates of virologic failure compared with AA. When the analysis was restricted to women who remained on ART after initiation, there were no significant differences in outcomes between the groups.1 Therefore, differential access to care or medications may explain at least part of the results of these studies.
In the current study, AA and EA had equal access to free care and medications. Approximately, 60% of both AA and EA enrolled were on active duty and therefore their command should have required them to seek medical care at regular intervals. There was no difference between active duty AA and active duty EA in the average number of visits attended per year (1.53 and 1.52, respectively). There was also no difference between nonactive duty AA or EA in the average number of visits attended per year (1.33 and 1.26, respectively) although active duty members of both ethnicities had significantly more visits than nonactive duty members. When the multivariate analysis was limited to only active duty members, the results were similar. Therefore, in this cohort, AA and EA accessed care similarly; despite this, however, there were large differences between the 2 groups in virologic response to HAART.
Besides access to care, viral suppression also depends on high levels of adherence to HAART.16-18 The DoD HIV NHS did not start collecting self-reported adherence until 2006 and we were not able to access refill information therefore we could not evaluate whether adherence differences account for the differences in virologic suppression. Other investigations into the association between ethnicity and adherence have shown mixed findings. A study of 2088 HIV-infected children and adolescents conducted by the Pediatric AIDS Clinical Trials Group did not find any association between self-reported medication adherence and ethnicity.11 Likewise, a study of 861 veterans taking HAART did not find any differences by ethnicity in adherence as measured by 3 questions taken from a validated instrument used by the Adult AIDS Clinical Trials Group.4 In contrast to these studies, a prospective, randomized, controlled trial of 3- vs 4-drug ART regimens for the initial treatment of HIV infection in 765 patients (approximately, 35% black and 40% white) found that black patients had lower self-reported adherence rates at weeks 4 and 12 post-HAART but not at weeks 24 and 48.12
Evaluating potential adherence differences between ethnicities in a population with free access to medications is important in understanding differences in HAART response but it still may not fully explain why differences exist. If virologic outcome differences disappear when adherence is adjusted for, as was seen in a study of intravenous drug users with free access to medications,19 one would then need to evaluate why there are differences in adherence between ethnicities. It may be due to differences in absorption, distribution, metabolism, or elimination of the HAART causing different side effects or toxicities in subjects of different ethnicities or it may be due to differences in beliefs about medications or differences in education received about the medications.
The largest differences in response to HAART between AA and EA were seen with protease inhibitor based regimens with or without ritonavir, and the smallest difference was seen with NNRTI-based regimens. NNRTIs have a lower genetic barrier to resistance, and therefore one might expect that if differential adherence between the ethnic groups that was not due to side effects or toxicities was the main driver of the observed difference in virologic response, that the largest difference in response would have been between subjects using NNRTI-based regimens. The majority of PIs in this study were older, less potent ones than are available today. Slight differences in absorption, distribution, metabolism, or elimination of these older regimens between ethnicities may have lead to different side effects, toxicities, or potencies and may have contributed to the differences in virologic suppression.
Having a college degree has been associated with a favorable response to HAART.20 In the military, rank is a rough surrogate for education. Although the vast majority of both AA and EA in this study were not officers, EA were significantly more likely to be officers than AA. In the multivariate model, rank was not significantly associated with viral suppression at 6 or 12 months but to ensure that rank was not contributing to the outcome, we repeated the analyses limiting the subjects to only nonofficers and the results remained unchanged.
AA in this cohort had lower CD4 counts and CD4 percentages at diagnosis compared with EA. Although non-HIV infected AA have lower total white blood cell counts than EA, in the general population there is no difference between these 2 ethnicities in percentage of total lymphocytes that would explain these findings.21-23 One study of CD4 counts within 24 months after seroconversion conducted from 1987 to 1991 did not find a significant difference in CD4 counts between ethnicities, but did in CD4 percentages.24 To further explore this, we examined the rate of CD4 decline from HIV diagnosis to HAART initiation to determine whether CD4 cells decline faster in AA. We did not find a difference in the slopes between AA and EA. There is some evidence that immune response may play a role in the virologic response to HAART.25,26 If this is true, whatever factors contribute to lower CD4 counts in AA at HIV diagnosis may also contribute to differences seen in virologic response to HAART between ethnicities. This finding needs to be confirmed with additional studies.
AA also had higher rates of hepatitis B, hepatitis C, and STDs than EA. Although these coinfections were not significantly associated with odds of viral suppression in the multivariate model and in other studies have not been found to be associated with response to HAART,27,28 they may be markers of a certain type of behavior or belief system which affects response to HAART. Another possibility is that complications of those coinfections themselves or side effects or drug interactions from their treatment, could affect HAART adherence. The DoD HIV NHS does not collect information on illicit substance use; however, because random mandatory drug testing was instituted in the military in 1992, rates of illicit drug use have remained low in the active duty population and the odds of illicit drug use do not differ between AA and EA.29 Furthermore, a study conducted in over 500 HIV-infected active duty members found that intravenous drug use was rare with a prevalence of <1%.30
Despite the large number of subjects included in this study, it is important to note that one-third of the AA and EA who started HAART were excluded from analysis because of missing VLs. The excluded individuals were diagnosed with HIV in earlier calendar years and because VL testing did not become routine in the military until 1997, this most likely explains why they did not have a VL around HAART initiation. Although excluded subjects were significantly different than those included (lower CD4 counts at HAART initiation and more PI-based regimens, both of which reflect the earlier calendar period of HIV diagnosis), the percentage of AA and EA excluded was the same and therefore it is not likely that these exclusions influenced the results of the study. Furthermore, the results of this study were unchanged after restricting the analysis to subjects diagnosed with HIV in the HAART era.
There may be additional factors that could contribute to the differences in virologic suppression between AA and EA. Perhaps physicians treat subjects of different ethnicities differently or perhaps patients respond to providers of discordant ethnic background or gender differently. The timing of HAART initiation and the regimens prescribed was not significantly different between the groups; however, there could be differences in education given to the subjects or other factors. There could be selection bias as to who was started on HAART; however, this may have lessened the chance of finding differences between the groups as physicians may have selected only those they thought would adhere to the HAART regimen.
This study adds to the body of evidence that AA do not have the same virologic response to HAART as EA. It is imperative that we understand why the HAART response rates are lower in African versus EA to optimize therapy for all HIV-infected individuals. More efforts need to be focused on psychological, social, and cultural factors and genetic differences between ethnicities and how these may impact response to therapy.
We would like to thank the patients without whom none of this work would be possible. We would also like to thank the research coordinators and support staff who diligently work on the DoD HIV NHS and the members of the IDCRP HIV Working Group (by site): National Institute of Allergy and Infectious Diseases, Bethesda, MD: M. Polis, J. Powers, J. Metcalf, E. Tramont; Naval Medical Center, Portsmouth, VA: J. Maguire, T. Lalani; Naval Medical Center, San Diego, CA: M. Bavaro, N. Crum-Cianflone, H. Chun, B. Hale; National Naval Medical Center, Bethesda, MD: C. Decker, A. Ganesan, T. Whitman; San Antonio Military Medical Center, San Antonio, TX: V. Marconi, M. Landrum, W. Bradley; Tripler Army Medical Center, Honolulu, HI: S. Fraser; Uniformed Services University of the Health Sciences, Bethesda, MD: S. Wegner, B. Agan, G. Martin, G. Quinnan; Walter Reed Army Institute of Research, Rockville, MD: L. Jagodzinski, N. Michael, M. Milazzo, R. O'Connell, S. Peel; Walter Reed Army Medical Center, Washington, DC: C. Hawkes, A. Weintrob, G. Wortmann.
1. Anastos K, Schneider MF, Gange SJ, et al. The association of race, sociodemographic, and behavioral characteristics with response to highly active antiretroviral therapy in women. J Acquir Immune Defic Syndr
2. Hartzell JD, Spooner K, Howard R, et al. Race and mental health diagnosis are risk factors for highly active antiretroviral therapy failure in a military cohort despite equal access to care. J Acquir Immune Defic Syndr
3. Pence BW, Ostermann J, Kumar V, et al. The influence of psychosocial characteristics and race/ethnicity on the use, duration, and success of antiretroviral therapy. J Acquir Immune Defic Syndr
4. McGinnis KA, Fine MJ, Sharma RK, et al. Understanding racial disparities in HIV using data from the veterans aging cohort 3-site study and VA administrative data. Am J Public Health
5. Riddler SA, Haubrick R, DiRienzo G, et al. Class-sparing regimens for initial treatment of HIV-1 infection. N Engl J Med
6. Schackman BR, Ribaudo HJ, Krambrink A, et al. Racial differences in virologic failure associated with adherence and quality of life on efavirenz-containing regimens for initial HIV-therapy. J Acquir Immune Defic Syndr
7. Tan R, Westfall AO, Willig JH, et al. Clinical outcome of HIV-infected antiretroviral-naïve patients with discordant immunologic and virologic responses to highly active antiretroviral therapy. J Acquir Immune Defic Syndr
8. Jensen-Fangel S, Pedersen L, Pederson C, et al. The effect of race/ethnicity on the outcome of highly active antiretroviral therapy for human immunodeficiency virus type 1-infected patients. Clin Infect Dis
9. Nellen JF, Wit FW, de Wolf F, et al. Virologic and immunologic response to highly active antiretroviral therapy in indigenous and nonindigenous HIV-infected patients in the Netherlands. J Acquir Immune Defic Syndr
10. Van den Berg JB, Hak E, Vervoort SC, et al. Increased risk of early virological failure in non-European HIV-1-infected patients in a Dutch cohort on highly active antiretroviral therapy. HIV Med
11. Williams PL, Storm D, Montepiedra G, et al. Predictors of adherence to antiretroviral medications in children and adolescents with HIV infection. Pediatrics
12. Gulick RM, Ribaudo HJ, Shikuma CM, et al. Three- vs four-drug antiretroviral regimens for the initial treatment of HIV-1 infection. JAMA
13. Haas DW, Ribaudo HJ, Kim RB, et al. Pharmacogenetics of efavirenz and central nervous system side effects: an Adult AIDS Clinical Trials Group Study. AIDS
14. Haas DW, Smeaton LM, Shafer RW, et al. Pharmacogenetics of long-term responses to antiretroviral regimens containing efavirenz and/or nelfinavir: an Adult AIDS Clinical Trials Group Study. J Infect Dis
16. Paterson DL, Swindells S, Mohr J, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med
17. Wood E, Hogg RS, Yip B, et al. Effect of medication adherence on survival of HIV-infected adults who start highly active antiretroviral therapy when the CD4+ cell count is 0.200 to 0.350 × 10(9) cells/L. Ann Intern Med
18. Nieuwkerk PT, Oort FJ. Self-reported adherence to antiretroviral therapy for HIV-1 infection and virologic treatment response: a meta-analysis. J Acquir Immune Defic Syndr
19. Wood E, Montaner JSG, Yip B, et al. Adherence and plasma HIV RNA responses to highly active antiretroviral therapy among HIV-1 infected injection drug users. CMAJ
20. Gifford AL, Bormann JE, Shively MJ, et al. Predictors of self-reported adherence and plasma HIV concentrations in patients on multidrug antiretroviral regimens. J Acquir Immune Defic Syndr
21. Hsieh MM, Everhart JE, Bryd-Holt DD, et al. Prevalence of neutropenia in the US population:age, sex, smoking status and ethnic differences. Ann Intern Med
22. Reed WW, Diehl LF. Leukopenia, neutropenia, and decreased hemoglobin levels in healthy American blacks. Arch Intern Med
23. Nowicki MJ, Karim R, Mack WJ. Correlates of CD4+ and CD8+ lymphocyte counts in high risk immunodeficiency virus (HIV)-seronegative women enrolled in the Women's Interagency HIV Study (WIHS). Hum Immunol
24. Gorham ED, Garland FC, Mayers DL, et al. CD4 lymphocyte counts within 24 months of human immunodeficiency virus seroconversion. Findings in the US Navy and Marine Corps. Arch Intern Med
25. Mildvan D, Bosch RJ, Kim RS, et al. Immunophenotypic markers and antiretroviral therapy (IMART): T cell activation and maturation help predict treatment response. J Infect Dis
26. Ahuja SK, Kulkarni H, Catano G, et al. CCL3L1-CCR5 genotype influences durability of immune recovery during antiretroviral therapy of HIV-1 infected individuals. Nat Med
27. Hoffmann CJ, Charalambous S, Martin DJ, et al. Hepatitis B virus infection and response to antiretroviral therapy (ART) in a South African ART program. Clin Infect Dis
28. Omland LH, Weis N, Skinhoj P, et al. Impact of hepatitis B virus co-infection on response to highly active antiretroviral therapy and outcome in HIV-infected individuals: a nationwide cohort study. HIV Med
29. Bray RM, Hourani LL. Substance use trends among active duty military personnel: findings from the United States Department of Defense Health Related Behavior Surveys, 1980-2005. Addiction
30. Brodine SK, Starkey MJ, Shaffer RA, et al. Diverse HIV-1 subtypes and clinical, laboratory and behavioral factors in a recently infected US military cohort. AIDS
© 2009 Lippincott Williams & Wilkins, Inc.