AIDS Cohort Study Research Articles

Factors associated with injection cessation, relapse and initiation in a community‐based cohort of injection drug users in Chennai, India

To characterize factors associated with injection cessation, relapse and initiation. The Madras Injection Drug User and AIDS Cohort Study (MIDACS) is a prospective cohort of injection drug users (IDUs) recruited in 2005-06 with semi-annual follow-up to 2009. Discrete-time survival models were used to characterize predictors of time to first injection cessation and relapse. Chennai, India. A total of 855 IDUs who reported injecting in the 6 months prior to baseline and had >1 follow-up visit. Cessation was defined as the first visit where no injection drug use was reported (prior 6 months) and relapse as the first visit where drug injection (prior 6 months) was reported after first cessation. All participants were male; median age was 35 years. Over 3 years, 92.7% reported cessation [incidence rate (IR): 117 per 100 person-years]. Factors associated positively with cessation included daily injection and incarceration and factors associated negatively with cessation included marriage, alcohol and homelessness. Of those who reported cessation, 23.6% relapsed (IR: 19.7 per 100 person-years). Factors associated positively with relapse included any education, injection in the month prior to baseline, sex with a casual partner, non-injection drug use, incarceration and homelessness. Alcohol was associated negatively with relapse. The primary reasons for cessation were medical conditions (37%) and family pressure (22%). The majority initiated with non-injection drugs, transitioning to injection after a median 4 years. Injection drug users in southern India demonstrate a high rate of injection cessation over 3 years, but relapse is not uncommon. Compensatory increases in alcohol use indicate that cessation of injection does not mean cessation of all substance use. Family pressure, concerns about general health, fear of human immunodeficiency virus infection and a history of non-injection drug use are important correlates of cessation.

Platelet decline as a predictor of brain injury in HIV infection

An association between platelet decline and increased risk of progression to dementia has been observed in an advanced HIV infection cohort study. This investigation evaluated the prognostic significance of platelet decline for dementia, for psychomotor slowing, and for brain injury, as quantified in vivo, in a much larger population of HIV+ men. Platelet counts and neurocognitive data were available from biannual visits of 2,125 HIV+ men participating in the prospective, Multicenter AIDS Cohort Study from 1984 to 2009. Brain volumetric data were also available from an imaging substudy of 83 seropositive participants aged 50 and older. The association of platelet counts with neurocognitive outcome was assessed using Cox proportional hazard models where change in platelet count from baseline was a time-updated variable. Marked platelet decline was associated with increased risk of dementia in univariate analysis (hazard ratio [HR] = 2.5, 95% confidence interval [CI] = 1.8-3.5), but not after adjustment for CD4 cell count, HIV viral load, age, study site, hemoglobin, race, education, smoking, and alcohol use (HR = 1.4, 95% CI = 0.78-2.5). Platelet decline did not predict psychomotor slowing in either univariate (HR = 0.79, 95% CI = 0.58-1.08) or multivariate (HR = 1.10, 95% CI = 0.73-1.67) analysis. Analysis of brain volumetric data, however, indicated a relationship between platelet decline and reduced gray matter volume fraction in univariate (p = 0.06) and multivariate (p < 0.05) analyses. Platelet decline was not an independent predictor of dementia or psychomotor slowing, after adjusting for stage of disease. Findings from a structural brain imaging substudy of older participants, however, support a possible relationship between platelet decline and reduced gray matter.

The mediating role of pain in substance use and depressive symptoms among Multicenter AIDS Cohort Study (MACS) participants

Pain in human immunodeficiency virus (HIV) frequently co-occurs with substance use and depression. The complex associations among patient characteristics, pain, depression, and drug use in HIV suggests a role for testing models that can account for relationships simultaneously, control for HIV status, and also test for mediation. Using structural equation modeling, the current study examined associations among pain, sociodemographics, illicit drug use, and depressive symptoms in 921 HIV-seropositive and 1019 HIV-seronegative men from the Multicenter AIDS Cohort Study, an ongoing prospective study of the natural history of HIV infection among gay/bisexual men. Longitudinal repeated measures data collected over a 6-year period were analyzed using predictive path models in which sociodemographics, HIV status, and CD4+ cell counts predicted pain, which, in turn, predicted depressive symptoms and illicit drug use. The path models did not differ substantially between HIV-seropositive and -seronegative men. Analyses using the total sample indicated that pain served both as a mediator and as a predictor of more use of cannabis, cocaine, and heroin, as well as more depressive symptoms. HIV-seropositive status predicted more use of inhaled nitrites. In this cohort, having lower CD4+ cell counts (predicted by HIV status), being African American, less educated, and older were all associated with more pain, which, in turn, was associated with more illicit drug use and more depressive symptoms. The results underscore the need for adequate pain management, particularly among vulnerable subgroups of HIV-seropositive and HIV-seronegative men to reduce the risk of drug use and depression.

Plenary XII The multicenter aids cohort study: 28+ years of collaboration, collegiality and science

The Multicenter AIDS Cohort Study (MACS) was initiated with a request for proposals from National Institute on Allergies and Infectious Diseases to study AIDS among men who have sex with men. Five grants were awarded, four of which joined to form the MACS. These were Johns Hopkins University (P.I Frank Polk), Pittsburgh (P.I. Charles Rinaldo), Northwestern University (P.I. John Phair) and UCLA (P.I. Roger Detels). Two years later, the investigators, recognising the need for a data management center, convinced the NIH to fund the Johns Hopkins center, CAMACS (P.I. Alvaro Munoz). The MACS recruited 5000 men at the four centers from 1984 to 1985. The decision was made to examine the men every 6 months with an extensive interview, physical examination and collection of specimens (blood, saliva, semen) and to establish both a central repository and a local repository at each of the sites including serum, plasma, and cells). In 1989 and 2001 the MACS was expended to include primarily minority-Americans, the most involved group. The original group of investigators has been expanded to include several hundred investigators from elsewhere in the U.S., Canada and Europe. Any investigator submitting a request for data and/or specimens has access to the MACS data base and national repository upon approval of a concept by the MACS Executive Committee. The MACS has over 90 000 person years of follow-up and a repository of over 1 400 000 specimens, facilitating a wide range of virologic, immunologic, genetic, neurologic, treatment and psychosocial studies. To date the MACS has published almost 1300 papers in refereed scientific journals and has contributed significantly to our understanding of this tragic disease.

Comparison of HIV-Infected Patients' Characteristics and Clinical Outcomes Between Cohort-Dependent and Independent Patients: Korean HIV Cohort Study

Comparison of HIV-Infected Patients' Characteristics and Clinical Outcomes Between Cohort-Dependent and Independent Patients: Korean HIV Cohort Study

A particular diffusion model for incomplete longitudinal data: application to the multicenter AIDS cohort study

Longitudinal studies, in which individuals are measured repeatedly in time, are often incomplete. We model continuous-time longitudinal data from the Multicenter AIDS Cohort Study using a diffusion model in which the diffusion parameters are functions of the covariates. These data are jointly modeled with the process of time-to-death due to AIDS. We show that, even for large data sets with a large number of missing variables, a Bayesian analysis is feasible using Gibbs sampling and compare a complete case analysis with a Bayesian treatment of missing values.

Prevalence of XMRV Nucleic Acid and Antibody in HIV-1-Infected Men and in Men at Risk for HIV-1 Infection

Xenotropic MLV-Related Virus (XMRV) was recently reported to be associated with prostate cancer and chronic fatigue syndrome (CFS). Infection was also reported in 3.7% of healthy individuals. These highly reported frequencies of infection prompted concerns about the possibility of a new, widespread retroviral epidemic. The Multicenter AIDS Cohort Study (MACS) provides an opportunity to assess the prevalence of XMRV infection and its association with HIV-1 infection among men who have sex with men. Reliable detection of XMRV infection requires the application of multiple diagnostic methods, including detection of human antibodies to XMRV and detection of XMRV nucleic acid. We, therefore, tested 332 patient plasma and PBMC samples obtained from recent visits in a subset of patients in the MACS cohort for XMRV antibodies using Abbott prototype ARCHITECT chemiluminescent immunoassays (CMIAs) and for XMRV RNA and proviral DNA using a XMRV single-copy qPCR assay (X-SCA). Although 9 of 332 (2.7%) samples showed low positive reactivity against a single antigen in the CMIA, none of these samples or matched controls were positive for plasma XMRV RNA or PBMC XMRV DNA by X-SCA. Thus, we found no evidence of XMRV infection among men in the MACS regardless of HIV-1 serostatus.

Cancer incidence in the multicenter aids cohort study before and during the HAART era

The incidence of Kaposi sarcoma (KS) and non-Hodgkin lymphoma (NHL) among human immunodeficiency virus (HIV)-infected individuals declined after the introduction of highly active antiretroviral therapy (HAART) in the mid-1990s, but the cancer risk associated with HIV infection during the HAART era remains to be clarified. Cancer incidence among HIV-infected and HIV-uninfected participants in the Multicenter AIDS (acquired immunodeficiency syndrome) Cohort Study (MACS) between 1984 and 2007 was compared with the expected incidence using US population-based data from the Surveillance, Epidemiology, and End Results (SEER) program. Age- and race-adjusted cancer incidence rates were also compared HIV by status and over time within the MACS. Exact statistical methods were used for all analyses. A total of 933 incident cancers were observed during 77,320 person-years of follow-up. Compared with SEER, MACS HIV-infected men had significantly (P<.05) elevated rates of KS (standardized incidence ratio [SIR], 139.10), NHL (SIR, 36.80), Hodgkin lymphoma (HL)(SIR, 7.30), and anal cancer (SIR, 25.71). Within MACS, HIV infection was found to be independently associated with each of these cancers across the entire follow-up period, and KS (incidence rate ratio [IRR], 54.93), NHL (IRR, 11.18), and anal cancer (IRR, 18.50) were each found to be significantly elevated among HIV-infected men during the HAART era. Among these men, the incidence of KS and NHL declined (IRR, 0.13 and 0.23, respectively), the incidence of anal cancer increased (IRR, 5.84), and the incidence of HL remained statistically unchanged (IRR, 0.75) from the pre-HAART to the HAART era. Cancer risk remains elevated among HIV-infected men who have sex with men, highlighting the continuing need for appropriate cancer screening in this population.

Association Between Human Immunodeficiency Virus Infection and Stiffness of the Common Carotid Artery

Individuals with human immunodeficiency virus (HIV) who use highly active antiretroviral therapy (HAART) may have an increased risk for cardiovascular-related events, although the underlying mechanism remains unclear. We tested the hypothesis that carotid arterial stiffness was higher among persons using HAART compared to HAART-naïve and HIV-uninfected persons. Between 2004 and 2006, we performed high-resolution B-mode ultrasound on 2789 HIV-infected and HIV-uninfected participants of the Women's Interagency HIV Study (1865 women) and the Multicenter AIDS Cohort Study (924 men) and determined carotid arterial distensibility, which is a direct measure of carotid arterial stiffness. We used generalized estimating equations to evaluate the association between distensibility and HIV infection, CD4(+) cell count, and exposure to HAART adjusted for demographic, behavioral, and clinical characteristics. In multivariable analysis, distensibility was 4.3% lower (95% confidence interval, -7.4% to -1.1%) among HIV-infected vs uninfected participants. Among HIV-infected participants with <200 CD4(+) cells, distensibility was 10.5% lower (95% confidence interval, -14.5% to -6.2%) than that among HIV-uninfected participants, and this effect did not differ significantly by cohort or race. Concurrent HAART use was independently associated with lower distensibility among Multicenter AIDS Cohort Study participants but not among Women's Interagency HIV Study participants. Our finding that advanced HIV-related immunosuppression was associated with increased carotid arterial stiffness independent from the effects of traditional atherosclerosis risk factors suggests that the etiologic mechanism underlying reports of an increased cardiovascular disease risk among HIV-infected individuals might involve HIV-related immunosuppression leading to vascular dysfunction and arterial stiffening.

Inflammatory biomarkers and abacavir use in the Women's Interagency HIV Study and the Multicenter AIDS Cohort Study

To assess associations between abacavir (ABC) use and systemic inflammation. Nested case-control study. The Multicenter AIDS Cohort Study (MACS) and Women's Interagency HIV Study (WIHS) cohort participants who initiated ABC were matched, using propensity score methods, to ABC-unexposed persons. Levels of high-sensitivity C-reactive protein (hsCRP) (microg/ml), interleukin-6 (IL-6) (pg/ml), and D-dimer (microg/ml) were measured from pre-HAART and on-HAART plasma. Random-effects models compared markers by ABC exposure and by changes from pre-HAART levels. Biomarkers were measured in N = 508 matched pairs (328 women; 180 men). Pre-HAART levels did not differ by exposure group except that hsCRP levels were higher among WIHS women who subsequently used ABC (P = 0.04). Regardless of ABC use, mean hsCRP increases and D-dimer reductions were seen when comparing pre-HAART to on-HAART levels, in the overall group (28 and -27%), for MACS men (28 and -31%) and for WIHS women [29 and -24%, P < 0.01 for all]; IL-6 levels declined in MACS men (P = 0.02). No adjusted biomarker level differences existed by ABC exposure at the on-HAART visit. HIV RNA reductions correlated with D-dimer (r = 0.14, P < 0.01) and IL-6 (r = 0.12, P < 0.01) reductions. Associations between ABC use and mean biomarker levels were modified by pre-HAART antiretroviral therapy experience. Renal dysfunction was equally likely among non-ABC and ABC recipients. ABC use was not associated with plasma elevations in hsCRP, IL-6, and D-dimer. Mechanisms other than increased systemic inflammation may account for ABC's reported association with increased cardiovascular disease. HAART-associated reductions in D-dimer and IL-6 were apparent regardless of ABC use and were correlated with HIV RNA reductions.

A Bayesian MCMC approach to survival analysis with doubly-censored data

Doubly-censored data refers to time to event data for which both the originating and failure times are censored. In studies involving AIDS incubation time or survival after dementia onset, for example, data are frequently doubly-censored because the date of the originating event is interval-censored and the date of the failure event usually is right-censored. The primary interest is in the distribution of elapsed times between the originating and failure events and its relationship to exposures and risk factors. The estimating equation approach [Sun et al. (1999). Regression analysis of doubly censored failure time data with applications to AIDS studies. Biometrics 55, 909–914] and its extensions assume the same distribution of originating event times for all subjects. This paper demonstrates the importance of utilizing additional covariates to impute originating event times, i.e., more accurate estimation of originating event times may lead to less biased parameter estimates for elapsed time. The Bayesian MCMC method is shown to be a suitable approach for analyzing doubly-censored data and allows a rich class of survival models. The performance of the proposed estimation method is compared to that of other conventional methods through simulations. Two examples, an AIDS cohort study and a population-based dementia study, are used for illustration. Sample code is shown in Appendix.

Semiparametric Analysis for Recurrent Event Data with Time‐Dependent Covariates and Informative Censoring

Recurrent event data analyses are usually conducted under the assumption that the censoring time is independent of the recurrent event process. In many applications the censoring time can be informative about the underlying recurrent event process, especially in situations where a correlated failure event could potentially terminate the observation of recurrent events. In this article, we consider a semiparametric model of recurrent event data that allows correlations between censoring times and recurrent event process via frailty. This flexible framework incorporates both time-dependent and time-independent covariates in the formulation, while leaving the distributions of frailty and censoring times unspecified. We propose a novel semiparametric inference procedure that depends on neither the frailty nor the censoring time distribution. Large sample properties of the regression parameter estimates and the estimated baseline cumulative intensity functions are studied. Numerical studies demonstrate that the proposed methodology performs well for realistic sample sizes. An analysis of hospitalization data for patients in an AIDS cohort study is presented to illustrate the proposed method.

A Generalized Log‐Rank‐Type Test for Comparing Survivals with Doubly Interval‐Censored Data

In doubly interval-censored data, the survival time of interest is defined as the elapsed time between an initial event and a subsequent event, but the occurrences of both events cannot be observed exactly. Instead, only right- or interval-censored observations on the occurrence times are available. Our purpose is to develop a generalized log-rank-type test for comparing survival functions of several groups. For the same problem, Sun (The Statistical Analysis of Interval-censored Failure Time Data, Springer, New York) suggested a nonparametric test based on the estimated marginal survival functions of the two related events. We consider a new method using uniform weights, which depend only on the size of the risk set at each observed time. The proposed method does not require the estimation of marginal survival functions and furthermore can reduce to the log-rank test for right-censored data. Results from a simulation study show that our test performs well in terms of size and power. We analyze the AIDS cohort study data taken from Kim et al. (Biometrics 49, 13-22).

Smoking Enhances Risk for New External Genital Warts in Men

Repeat episodes of HPV-related external genital warts reflect recurring or new infections. No study before has been sufficiently powered to delineate how tobacco use, prior history of EGWs and HIV infection affect the risk for new EGWs. Behavioral, laboratory and examination data for 2,835 Multicenter AIDS Cohort Study participants examined at 21,519 semi-annual visits were evaluated. Fourteen percent (391/2835) of men reported or were diagnosed with EGWs at 3% (675/21,519) of study visits. Multivariate analyses showed smoking, prior episodes of EGWs, HIV infection and CD4+ T-lymphocyte count among the infected, each differentially influenced the risk for new EGWs.

Low CD4+ T-cell count as a major atherosclerosis risk factor in HIV-infected women and men

To assess the association of HIV infection, HIV disease parameters (including CD4+ T-cell counts, HIV viral load, and AIDS) and antiretroviral medication use with subclinical carotid artery atherosclerosis. Cross-sectional study nested within a prospective cohort study. Among participants in the Women's Interagency HIV Study (1331 HIV-infected women, 534 HIV-uninfected women) and Multicenter AIDS Cohort Study (600 HIV-infected men, 325 HIV-uninfected men), we measured subclinical carotid artery lesions and common carotid artery intima-media thickness using B-mode ultrasound. We estimated adjusted mean carotid artery intima-media thickness differences and prevalence ratios for carotid lesions associated with HIV-related disease and treatments, with multivariate adjustment to control for possible confounding variables. Among HIV-infected individuals, a low CD4+ T-cell count was independently associated with an increased prevalence of carotid lesions. Compared with the reference group of HIV-uninfected individuals, the adjusted prevalence ratio for lesions among HIV-infected individuals with CD4+ T-cell count less than 200 cells/mul was 2.00 (95% confidence interval, 1.22-3.28) in women and 1.74 (95% confidence interval, 1.04-2.93) in men. No consistent association of antiretroviral medications with carotid atherosclerosis was observed, except for a borderline significant association between protease inhibitor use and carotid lesions in men (with no association among women). History of clinical AIDS and HIV viral load were not significantly associated with carotid atherosclerosis. Beyond traditional cardiovascular disease risk factors, low CD4+ T-cell count is the most robust risk factor for increased subclinical carotid atherosclerosis in HIV-infected women and men.

HIV-Related Peripheral Neuropathy and Glucose Dysmetabolism: Study of a Public Dataset

Aims and Methods: The authors analyzed the public dataset of Multicenter AIDS Cohort Study to examine the association of glucose dysmetabolism with HIV-related peripheral neuropathy. Results: Among 5,622 participants in the data, 282 (5%) had peripheral neuropathy, of which 225 had sensory neuropathy. Of 225 participants with sensory neuropathy, 191 (84.8%) had exposure to highly active antiretroviral therapy. Of 225 with sensory neuropathy, 46 had fasting sugars measured; 26 of them (56.5%) had either impaired fasting glucose (fasting glucose ≥110 mg/dl) or diabetes (fasting glucose ≥126 mg/dl). Among 693 participants without neuropathy, who had fasting sugars measured, 441 (63.7%) had glucose dysmetabolism (p = 0.33). The mean HbA_1c levels in participants with sensory neuropathy (5.1 ± 1.02, n = 51) were significantly higher than in those without sensory neuropathy (4.8 ± 0.6, n = 164, p = 0.02), though its clinical relevance is unclear. Among 739 participants who had fasting sugars measured, 63.2% had glucose dysmetabolism, showing high prevalence of glucose dysmetabolism among participants with HIV infection. Conclusions: Except for slightly higher HbA_1c levels, there was no significant difference in prevalence of glucose dysmetabolism among participants with and without sensory neuropathy. The study brings up important aspects of using public datasets for epidemiologic studies including its limitations and a unique opportunity for pilot studies for new investigators.

Early and delayed benefits of HIV-1 suppression: timeline of recovery of innate immunity effector cells

The kinetics of recovery for innate immune effectors following antiretroviral therapy are unknown. Multiple sequential cryopreserved samples (viremic and ART-suppressed) from 66 patients enrolled in the Women's Interagency HIV Study or Multicenter AIDS Cohort Study cohorts (median follow-up, 700 days) were analyzed to determine natural killer, dendritic and T-cell changes by flow cytometry. Functional parameters were also measured in a subset of samples. Changes over time were analyzed by mixed-effect modeling based on a linear spline with a single knot at 270 days. Following viral suppression, a rapid rise in CD4 and white blood cell counts and a decline in T-cell activation were confirmed. However, natural killer cell subsets increased after 270 days of therapy, with a negative effect by baseline CD4%. CD123+ plasmacytoid but not myeloid dendritic cells showed a trend to increase during the first 270 days with a positive effect of baseline CD4%; plasmacytoid dendritic cell-induced interferon-alpha production significantly increased by end of follow-up. The kinetics of natural killer and plasmacytoid dendritic cell recovery are markedly different from those of T-cell subsets, indicative of early and delayed benefits of suppressive regimens.

Prospective Accuracy for Longitudinal Markers

In this article we focus on appropriate statistical methods for characterizing the prognostic value of a longitudinal clinical marker. Frequently it is possible to obtain repeated measurements. If the measurement has the ability to signify a pending change in the clinical status of a patient then the marker has the potential to guide key medical decisions. Heagerty, Lumley, and Pepe (2000, Biometrics 56, 337-344) proposed characterizing the diagnostic accuracy of a marker measured at baseline by calculating receiver operating characteristic curves for cumulative disease or death incidence by time t. They considered disease status as a function of time, D(t) = 1(T<or=t), for a clinical event time T. In this article we aim to address the question of how well Y(s), a diagnostic marker measured at time s(s>or= 0, after the baseline time) can discriminate between people who become diseased and those who do not in a subsequent time interval [s, t]. We assume the disease status is derived from an observed event time T and thus interest is in individuals who transition from disease free to diseased. We seek methods that also allow the inclusion of prognostic covariates that permit patient-specific decision guidelines when forecasting a future change in health status. Our proposal is to use flexible semiparametric models to characterize the bivariate distribution of the event time and marker values at an arbitrary time s. We illustrate the new methods by analyzing a well-known data set from HIV research, the Multicenter AIDS Cohort Study data.

Predictors for Lower Quality of Life in the HAART Era Among HIV-Infected Men

In the era of highly active antiretroviral therapy (HAART), maximizing health-related quality of life (QOL) has become a high priority of long-term management of HIV-infected individuals. Modifiable determinants of lower QOL should be identified for interventions specifically targeted to the HAART-using individuals to improve their QOL. To identify the predictors for lower QOL among HAART-using study participants in the Multicenter AIDS Cohort Study, a longitudinal study of HIV infection among homosexual and bisexual men in 4 cities. In the Multicenter AIDS Cohort Study, 636 HAART-using subjects had QOL data before HAART initiation and at least 2 consecutive QOL measurements after HAART initiation to visit 40 (April 2004). Variables of sociodemographics, individual risk behaviors, social support, biological markers, HIV-related medication use and clinical outcome indicators preceding the study outcomes, the physical health summary score and the mental health summary score derived from the standard SF-36 QOL form, were assessed as possible predictors using random-effects mixed models. QOL before HAART initiation was a strong predictor of QOL subsequent to HAART initiation. Older age, lower socioeconomic status, less male sexual partners, no alcohol drinking, and more advanced HIV disease stage were significant predictors for lower physical health summary score. In addition, more outpatient visits, depression, amprenavir use, antiretroviral drug interruption, recreational drug use, and less social support were significantly associated with lower mental health summary score. Many predictors of lower QOL are alterable risk factors that can be effectively targeted for interventions to maximize patients' QOL. With appropriate treatment and management of HIV disease and depression, clinicians can help improve the QOL of their patients. Through modification of individual risk behaviors, HIV-infected individuals can enhance their own QOL with support from clinicians and the community. In addition, active social support can also be an effective way to improve mental health of the infected persons.

Risk Of Injury and Accident-Related Mortality Among Participants in the Multicenter Aids Cohort Study (Macs) And The Women's Interagency Hiv Study (WIHS)

Risk Of Injury and Accident-Related Mortality Among Participants in the Multicenter Aids Cohort Study (Macs) And The Women's Interagency Hiv Study (WIHS)