The Design and Analysis of Partner Studies of HIV Transmission

Common goals in epidemiologic studies of infectious diseases include identification of the infectious agent, description of the modes of transmission and characterization of factors that influence the probability of transmission from infected to uninfected individuals. In the case of AIDS, the agent has been identified as the Human Immunodeficiency Virus (HIV), and transmission is known to occur through a variety of contact mechanisms including unprotected sexual intercourse, transfusion of infected blood products and sharing of needles in intravenous drug use. Relatively little is known about the probability of IV transmission associated with the various modes of contact, or the role that other cofactors play in promoting or suppressing transmission. Here, transmission probability refers to the probability that the virus is transmitted to a susceptible individual following exposure consisting of a series of potentially infectious contacts. The infectivity of HIV for a given route of transmission is defined to be the per contact probability of infection. Knowledge of infectivity and its relationship to other factors is important in understanding the dynamics of the AIDS epidemic and in suggesting appropriate measures to control its spread. The primary source of empirical data about infectivity comes from sexual partners of infected individuals. Partner studies consist of a series of such partnerships, usually heterosexual and monogamous, each composed of an initially infected "index case" and a partner who may or may not be infected by the time of data collection. However, because the infection times of both partners may be unknown and the history of contacts uncertain, any quantitative characterization of infectivity is extremely difficult. Thus, most statistical analyses of partner study data involve the simplifying assumption that infectivity is a constant common to all partnerships. The major objectives of this work are to describe and discuss the design and analysis of partner studies, providing a general statistical framework for investigations of infectivity and risk factors for HIV transmission. The development is largely based on three papers: Jewell and Shiboski (1990), Kim and Lagakos (1990), and Shiboski and Jewell (1992).

Prognosis of HIV-1-infected patients up to 5 years after initiation of HAART: collaborative analysis of prospective studies

OBJECTIVE: To estimate the prognosis over 5 years of HIV-1-infected, treatment-naive patients starting HAART, taking into account the immunological and virological response to therapy. DESIGN: A collaborative analysis of data from 12 cohorts in Europe and North America on 20,379 adults who started HAART between 1995 and 2003. METHODS: Parametric survival models were used to predict the cumulative incidence at 5 years of a new AIDS-defining event or death, and death alone, first from the start of HAART and second from 6 months after the start of HAART. Data were analysed by intention-to-continue-treatment, ignoring treatment changes and interruptions. RESULTS: During 61 798 person-years of follow-up, 1005 patients died and an additional 1303 developed AIDS. A total of 10 046 (49%) patients started HAART either with a CD4 cell count of less than 200 cells/microl or with a diagnosis of AIDS. The 5-year risk of AIDS or death (death alone) from the start of HAART ranged from 5.6 to 77% (1.8-65%), depending on age, CD4 cell count, HIV-1-RNA level, clinical stage, and history of injection drug use. From 6 months the corresponding figures were 4.1-99% for AIDS or death and 1.3-96% for death alone. CONCLUSION: On the basis of data collected routinely in HIV care, prognostic models with high discriminatory power over 5 years were developed for patients starting HAART in industrialized countries. A risk calculator that produces estimates for progression rates at years 1 to 5 after starting HAART is available from www.art-cohort-collaboration.org.

HAART and the heart: changes in coronary risk factors and implications for coronary risk in men starting antiretroviral therapy

OBJECTIVES: To estimate changes in coronary risk factors and their implications for coronary heart disease (CHD) rates in men starting highly active antiretroviral therapy (HAART). METHODS: Men participating in the Swiss HIV Cohort Study with measurements of coronary risk factors both before and up to 3 years after starting HAART were identified. Fractional polynomial regression was used to graph associations between risk factors and time on HAART. Mean risk factor changes associated with starting HAART were estimated using multilevel models. A prognostic model was used to predict corresponding CHD rate ratios. RESULTS: Of 556 eligible men, 259 (47%) started a nonnucleoside reverse transcriptase inhibitor (NNRTI) and 297 a protease inhibitor (PI) based regimen. Levels of most risk factors increased sharply during the first 3 months on HAART, then more slowly. Increases were greater with PI- than NNRTI-based HAART for total cholesterol (1.18 vs. 0.98 mmol L(-1)), systolic blood pressure (3.6 vs. 0 mmHg) and BMI (1.04 vs. 0.55 kg m(2)) but not HDL cholesterol (0.24 vs. 0.32 mmol L(-1)) or glucose (1.02 vs. 1.03 mmol L(-1)). Predicted CHD rate ratios were 1.40 (95% CI 1.13-1.75) and 1.17 (0.95-1.47) for PI- and NNRTI-based HAART respectively. CONCLUSIONS: Coronary heart disease rates will increase in a majority of patients starting HAART: however the increases corresponding to typical changes in risk factors are relatively modest and could be offset by lifestyle changes.

Factors associated with the incidence of type 2 diabetes mellitus in HIV-infected participants in the Swiss HIV Cohort Study

BACKGROUND: Human immunodeficiency virus (HIV)-infected persons may be at increased risk for developing type 2 diabetes mellitus because of viral coinfection and adverse effects of treatment. METHODS: We studied associations of new-onset diabetes mellitus with hepatitis B virus and hepatitis C virus coinfections and antiretroviral therapy in participants in the Swiss HIV Cohort Study, using Poisson regression. RESULTS: A total of 123 of 6513 persons experienced diabetes mellitus during 27,798 person-years of follow-up (PYFU), resulting in an incidence of 4.4 cases per 1000 PYFU (95% confidence interval [CI], 3.7-5.3 cases per 1000 PYFU). An increased incidence rate ratio (IRR) was found for male subjects (IRR, 2.5; 95% CI, 1.5-4.2), older age (IRR for subjects >60 years old, 4.3; 95% CI, 2.3-8.2), black (IRR, 2.1; 95% CI, 1.1-4.0) and Asian (IRR, 4.9; 95% CI, 2.2-10.9) ethnicity, Centers for Disease Control and Prevention disease stage C (IRR, 1.6; 95% CI, 1.04-2.4), and obesity (IRR, 4.7; 95% CI, 3.1-7.0), but results for hepatitis C virus infection or active hepatitis B virus infection were inconclusive. Strong associations were found for current treatment with nucleoside reverse-transcriptase inhibitors (IRR, 2.22; 95% CI, 1.11-4.45), nucleoside reverse-transcriptase inhibitors plus protease inhibitors (IRR, 2.48; 95% CI, 1.42-4.31), and nucleoside reverse-transcriptase inhibitors plus protease inhibitors and nonnucleoside reverse-transcriptase inhibitors (IRR, 3.25; 95% CI, 1.59-6.67) but were not found for treatment with nucleoside reverse-transcriptase inhibitors plus nonnucleoside reverse-transcriptase inhibitors (IRR, 1.47; 95% CI, 0.77-2.82). CONCLUSIONS: In addition to traditional risk factors, current treatment with protease inhibitor- and nucleoside reverse-transcriptase inhibitor-containing regimens was associated with the risk of developing type 2 diabetes mellitus. Our study did not find a significant association between viral hepatitis infection and risk of incident diabetes.

Discordant responses to potent antiretroviral treatment in previously naive HIV-1-infected adults initiating treatment in resource-constrained countries: the antiretroviral therapy in low-income countries (ART-LINC) collaboration

OBJECTIVES: To assess the frequency of and risk factors for discordant responses at 6 months on highly active antiretroviral therapy (HAART) in previously treatment-naive HIV patients from resource-limited countries. METHODS: The Antiretroviral Therapy in Low-Income Countries Collaboration is a network of clinics providing care and treatment to HIV-infected patients in Africa, Latin America, and Asia. Patients who initiated therapy between 1996 and 2004, were aged 16 years or older, and had a baseline CD4 cell count were included in this analysis. Responses were defined based on plasma viral load (PVL) and CD4 cell count at 6 months as complete virologic and immunologic (VR(+)IR(+)), virologic only (VR(+)IR(-)), immunologic only (VR(-)IR(+)), and nonresponse (VR(-)IR(-)). Multinomial logistic regression was used to assess the association between therapy responses and clinical and demographic variables. RESULTS: Of the 3111 patients eligible for analysis, 1914 had available information at 6 months of therapy: 1074 (56.1%) were VR(+)IR(+), 364 (19.0%) were VR(+)IR(-), 283 (14.8%) were (VR(-)IR(+)), and 193 (10.1%) were VR(-)IR(-). OF THE 3111 patients eligible for analysis, 1914 had available information at 6 months of therapy: 1074 (56.1%) were VRIR, 364 (19.0%) were VRIR, 283 (14.8%) were (VRIR), and 193 (10.1%) were VRIR. Compared with complete responders, virologic-only responders were older, had a higher baseline CD4 cell count, had a lower baseline PVL, and were more likely to have received a nonstandard HAART regimen; immunologic-only responders were younger, had a lower baseline CD4 cell count, had a higher baseline PVL, and were more likely to have received a protease inhibitor-based regimen. CONCLUSIONS: The frequency of and risk factors for discordant responses were comparable to those observed in developed countries. Longer follow-up is needed to assess the long-term impact of discordant responses on mortality in these resource-limited settings.

Tenofovir use is associated with a reduction in calculated glomerular filtration rates in the Swiss HIV Cohort Study

BACKGROUND: A growing number of case reports have described tenofovir (TDF)-related proximal renal tubulopathy and impaired calculated glomerular filtration rates (cGFR). We assessed TDF-associated changes in cGFR in a large observational HIV cohort. METHODS: We compared treatment-naive patients or patients with treatment interruptions > or = 12 months starting either a TDF-based combination antiretroviral therapy (cART) (n = 363) or a TDF-sparing regime (n = 715). The predefined primary endpoint was the time to a 10 ml/min reduction in cGFR, based on the Cockcroft-Gault equation, confirmed by a follow-up measurement at least 1 month later. In sensitivity analyses, secondary endpoints including calculations based on the modified diet in renal disease (MDRD) formula were considered. Endpoints were modelled using pre-specified covariates in a multiple Cox proportional hazards model. RESULTS: Two-year event-free probabilities were 0.65 (95% confidence interval [CI] 0.58-0.72) and 0.80 (95% CI 0.76-0.83) for patients starting TDF-containing or TDF-sparing cART, respectively. In the multiple Cox model, diabetes mellitus (hazard ratio [HR] = 2.34 [95% CI 1.24-4.42]), higher baseline cGFR (HR = 1.03 [95% CI 1.02-1.04] by 10 ml/min), TDF use (HR = 1.84 [95% CI 1.35-2.51]) and boosted protease inhibitor use (HR = 1.71 [95% CI 1.30-2.24]) significantly increased the risk for reaching the primary endpoint. Sensitivity analyses showed high consistency. CONCLUSION: There is consistent evidence for a significant reduction in cGFR associated with TDF use in HIV-infected patients. Our findings call for a strict monitoring of renal function in long-term TDF users with tests that distinguish between glomerular dysfunction and proximal renal tubulopathy, a known adverse effect of TDF.

Predictors of optimal viral suppression in patients switched to abacavir, lamivudine, and zidovudine: the Swiss HIV Cohort Study

OBJECTIVE: To investigate predictors of continued HIV RNA viral load suppression in individuals switched to abacavir (ABC), lamivudine (3TC) and zidovudine (ZDV) after successful previous treatment with a protease inhibitor or non-nucleoside reverse transcriptase inhibitor-based combination antiretroviral therapy. DESIGN AND METHODS: An observational cohort study, which included individuals in the Swiss HIV Cohort Study switching to ABC/3TC/ZDV following successful suppression of viral load. The primary endpoint was time to treatment failure defined as the first of the following events: two consecutiveviral load measurements > 400 copies/ml under ABC/3TC/ZDV, one viral load measurement > 400 copies/ml and subsequent discontinuation of ABC/3TC/ZDV within 3 months, AIDS or death. RESULTS: We included 495 individuals; 47 experienced treatment failure in 1459 person-years of follow-up [rate = 3.22 events/100 person-years; 95% confidence interval (95% CI), 2.30-4.14]. Of all failures, 62% occurred in the first year after switching to ABC/3TC/ZDV. In a Cox regression analysis, treatment failure was independently associated with earlier exposure to nucleoside reverse transcriptase inhibitor (NRTI) mono or dual therapy [hazard ratio (HR), 8.02; 95% CI, 4.19-15.35) and low CD4 cell count at the time of the switch (HR, 0.66; 95% CI, 0.51-0.87 by +100 cells/microl up to 500 cells/microl). In patients without earlier exposure to mono or dual therapy, AIDS prior to switch to simplified maintenance therapy was an additional risk factor. CONCLUSIONS: The failure rate was low in patients with suppressed viral load and switch to ABC/3TC/ZDV treatment. Patients with earlier exposure to mono or dual NRTI therapy, low CD4 cell count at time of switch, or AIDS are at increased risk of treatment failure, limiting the use of ABC/3TC/ZDV in these patient groups.

Late presentation of HIV-infected individuals

Late presentation remains a major concern despite the dramatically improved prognosis realized by ART. We define a first presentation for HIV care during the course of HIV infection as 'late' if an AIDS-defining opportunistic disease is apparent, or if CD4+ T-cells are <200/microl. In the Western world, approximately 10 and 30% of HIV-infected individuals still present with CD4+ T-cells <50 and <200/microl, respectively; estimates are substantially higher for developing countries. Diagnosis and treatment of opportunistic diseases and intense supportive in-hospital care take precedence over ART. Benefits of starting ART without delay, that is, when opportunistic diseases are still active, include faster resolution of opportunistic diseases and a decreased risk of recurrence. The downside of starting ART without delay could include toxicity, drug interactions and immune reconstitution inflammatory syndrome (IRIS). Among asymptomatic or oligosymptomatic individuals presenting late, where ART and primary prophylaxis are initiated, approximately 10-20% will become symptomatic from drug toxicity or undiagnosed opportunistic complications, including IRIS, which require appropriate therapies. In this review we describe late presentation to HIV care, the scale of the problem, the evaluation of a late-presenting patient and challenges associated with initiation of potent antiretroviral therapy (ART) in the setting of acute opportunistic infections and other comorbidities.

Ecological study of the predictors of successful management of dyslipidemia in HIV-infected patients on ART: the Swiss HIV Cohort Study

PURPOSE: Antiretroviral therapy (ART) may induce metabolic changes and increase the risk of coronary heart disease (CHD). Based on a health care system approach, we investigated predictors for normalization of dyslipidemia in HIV-infected individuals receiving ART. METHOD: Individuals included in the study were registered in the Swiss HIV Cohort Study (SHCS), had dyslipidemia but were not on lipid-lowering medication, were on potent ART for >or= 3 months, and had >or= 2 follow-up visits. Dyslipidemia was defined as two consecutive total cholesterol (TC) values above recommended levels. Predictors of achieving treatment goals for TC were assessed using Cox models. RESULTS: Analysis included 958 individuals with median followup of 2.3 years (IQR 1.2-4.0). 454 patients (47.4%) achieved TC treatment goals. In adjusted analyses, variables significantly associated with a lower hazard of reaching TC treatment goals were as follows: older age (compared to 18-37 year olds: hazard ratio [HR] 0.62 for 45-52 year olds, 95% CI 0.47-0.82; HR 0.40 for 53-85, 95% CI 0.29-0.54), diabetes (HR 0.39, 95% CI 0.26-0.59), history of coronary heart disease (HR 0.27, 95% CI 0.10-0.71), higher baseline TC (HR 0.78, 95% CI 0.71-0.85), baseline triple nucleoside regimen (HR 0.12 compared to PI-only regimen, 95% CI 0.07-0.21), longer time on PI-only regimen (HR 0.39, 95% CI 0.33-0.46), longer time on NNRTI only regimen (HR 0.35, 95% CI 0.29-0.43), and longer time on PI/NNRTI regimen (HR 0.34, 95% CI 0.26-0.43). Switching ART regimen when viral load was undetectable was associated with a higher hazard of reaching TC treatment goals (HR 1.48, 95% CI 1.14-1.91). CONCLUSION: In SHCS participants on ART, several ART-related and not ART-related epidemiological factors were associated with insufficient control of dyslipidemia. Control of dyslipidemia in ART recipients must be further improved.

Predictive value of known and novel alleles of CYP2B6 for efavirenz plasma concentrations in HIV-infected individuals

To assess the association of CYP2B6 allelic diversity with efavirenz (EFV) pharmacokinetics, we performed extensive genotyping of 15 relevant single nucleotide polymorphism in 169 study participants, and full resequencing of CYP2B6 in individuals with abnormal EFV plasma levels. Seventy-seven (45.5%) individuals carried a known (CYP2B6*6, *11, *15, or *18) or new loss/diminished-function alleles. Resequencing defined two new loss-of-function alleles: allele *27 (marked by 593T>C [M198T]), that results in 85% decrease in enzyme activity and allele *28 (marked by 1132C>T), that results in protein truncation at arginine 378. Median AUC levels were 188.5 microg h/ml for individuals homozygous for a loss/diminished-function allele, 58.6 microg h/ml for carriers, and 43.7 microg h/ml for noncarriers (P<0.0001). Individuals with a poor metabolizer genotype had a likelihood ratio of 35 (95% CI, 11-110) of presenting very high EFV plasma levels. CYP2B6 poor metabolizer genotypes explain to a large extent EFV pharmacokinetics and identify individuals at risk of extremely elevated EFV plasma levels.

Adjunctive corticosteroids for Pneumocystis jiroveci pneumonia in patients with HIV infection: a meta-analysis of randomised controlled trials

BACKGROUND: The objective of this study was to review the effects of adjunctive corticosteroids on overall mortality and the need for mechanical ventilation in HIV-infected patients with Pneumocystis jiroveci pneumonia (PCP) and substantial hypoxemia (arterial oxygen partial pressure <70 mmHg or alveolar-arterial gradient >35 mmHg on room air). METHODS: We conducted a systematic search of the literature for randomised trials published up to December 2004. Selected trials compared adjunctive corticosteroids with placebo or usual care in HIV-infected patients with PCP and reported mortality data. Two teams of reviewers independently evaluated the methodology and extracted data from each primary study. RESULTS: Six studies were included in the meta-analysis. Risk ratios for overall mortality for adjunctive corticosteroids were 0.54 (95% confidence interval [CI], 0.38-0.79) at 1 month and 0.67 (95% CI, 0.49-0.93) at 3-4 months of follow-up. Numbers needed to treat, to prevent 1 death, are 9 patients in a setting without highly active antiretroviral therapy (HAART) available and 22 patients with HAART available. Only the 3 largest trials provided data on the need for mechanical ventilation with a risk ratio of 0.37 (95% CI, 0.20-0.70) in favour of adjunctive corticosteroids. CONCLUSION: The number and size of trials investigating adjunctive corticosteroids for HIV-infected patients with PCP is small, but our results suggest a beneficial effect for patients with substantial hypoxemia.

Unsafe sex and increased incidence of hepatitis C virus infection among HIV-infected men who have sex with men: the Swiss HIV Cohort Study

BACKGROUND: Data on the incidence of hepatitis C virus (HCV) infection among human immunodeficiency virus (HIV)-infected persons are sparse. It is controversial whether and how frequently HCV is transmitted by unprotected sexual intercourse. METHODS: We assessed the HCV seroprevalence and incidence of HCV infection in the Swiss HIV Cohort Study between 1988 and 2004. We investigated the association of HCV seroconversion with mode of HIV acquisition, sex, injection drug use (IDU), and constancy of condom use. Data on condom use or unsafe sexual behavior were prospectively collected between 2000 and 2004. RESULTS: The overall seroprevalence of HCV infection was 33% among a total of 7899 eligible participants and 90% among persons reporting IDU. We observed 104 HCV seroconversions among 3327 participants during a total follow-up time of 16,305 person-years, corresponding to an incidence of 0.64 cases per 100 person-years. The incidence among participants with a history of IDU was 7.4 cases per 100 person-years, compared with 0.23 cases per 100 person-years in patients without such a history (P<.001). In men who had sex with men (MSM) without a history of IDU who reported unsafe sex, the incidence was 0.7 cases per 100 person-years, compared with 0.2 cases per 100 person-years in those not reporting unsafe sex (P=.02), corresponding to an incidence rate ratio of 3.5 (95% confidence interval, 1.2-10.0). The hazard of acquiring HCV infection was elevated among younger participants who were MSM. CONCLUSIONS: HCV infection incidence in the Swiss HIV Cohort Study was mainly associated with IDU. In HIV-infected MSM, HCV infection was associated with unsafe sex.

Modeling the influence of APOC3, APOE, and TNF polymorphisms on the risk of antiretroviral therapy-associated lipid disorders

BACKGROUND: Single-nucleotide polymorphisms in genes involved in lipoprotein and adipocyte metabolism may explain why dyslipidemia and lipoatrophy occur in some but not all antiretroviral therapy (ART)-treated individuals. METHODS: We evaluated the contribution of APOC3 -482C-->T, -455T-->C, and 3238C-->G; epsilon 2 and epsilon 4 alleles of APOE; and TNF -238G-->A to dyslipidemia and lipoatrophy by longitudinally modeling >2600 lipid determinations and 2328 lipoatrophy assessments in 329 ART-treated patients during a median follow-up period of 3.4 years. RESULTS: In human immunodeficiency virus (HIV)-infected individuals, the effects of variant alleles of APOE on plasma cholesterol and triglyceride levels and of APOC3 on plasma triglyceride levels were comparable to those reported in the general population. However, when treated with ritonavir, individuals with unfavorable genotypes of APOC3 and [corrected] APOE were at risk of extreme hypertriglyceridemia. They had median plasma triglyceride levels of 7.33 mmol/L, compared with 3.08 mmol/L in the absence of ART. The net effect of the APOE*APOC3*ritonavir interaction was an increase in plasma triglyceride levels of 2.23 mmol/L. No association between TNF -238G-->A and lipoatrophy was observed. CONCLUSIONS: Variant alleles of APOE and APOC3 contribute to an unfavorable lipid profile in patients with HIV. Interactions between genotypes and ART can lead to severe hyperlipidemia. Genetic analysis may identify patients at high risk for severe ritonavir-associated hypertriglyceridemia.

Body fat changes among antiretroviral-naive patients on PI- and NNRTI-based HAART in the Swiss HIV cohort study

BACKGROUND: Body fat changes are common in patients with HIV. For patients on protease inhibitor (PI)-based highly active antiretroviral therapy (HAART), these changes have been associated with increasing exposure to therapy in general and to stavudine in particular. Our objective is to show whether such associations are more or less likely for patients on non-nucleoside reverse transcriptase inhibitor (NNRTI)-based HAART. METHODS: We included all antiretroviral-naive patients in the Swiss HIV Cohort Study starting HAART after April 2000 who had had body weight, CD4 cell count and plasma HIV RNA measured between 6 months before and 3 months after starting HAART, and at least one assessment of body fat changes after starting HAART. At visits scheduled every 6 months, fat loss or fat gain is reported by agreement between patient and physician. We estimate the association between reported body fat changes and both time on therapy and time on stavudine, using conditional logistical regression. RESULTS: Body fat changes were reported for 85 (9%) out of 925 patients at their first assessment; a further 165 had only one assessment. Of the remaining 675 patients, body fat changes were reported for 156 patients at a rate of 13.2 changes per 100 patient-years. Body fat changes are more likely with increasing age [odds ratio (OR) 1.18 (1.00-1.38) per 10 years], with increasing BMI [OR 1.06 (1.01-1.11)] and in those with a lower baseline CD4 cell count [OR 0.91 (0.83-1.01) per 100 cells/microl]. There is only weak evidence that body fat changes are more likely with increasing time on HAART [OR 1.16 (0.93-1.46)]. After adjusting for time on HAART, fat loss is more likely with increasing stavudine use [OR 1.70 (1.34-2.15)]. There is no evidence of an association between reported fat changes and time on NNRTI therapy relative to PI therapy in those patients who used either one therapy or the other [OR 0.98 (0.56-1.63)]. CONCLUSION: Fat loss is more likely to be reported with increasing exposure to stavudine. We find no evidence of major differences between PI and NNRTI therapy in the risk of reported body fat changes.

Early loss of HIV-infected patients on potent antiretroviral therapy programmes in lower-income countries

OBJECTIVE: To analyse the early loss of patients to antiretroviral therapy (ART) programmes in resource-limited settings. METHODS: Using data on 5491 adult patients starting ART (median age 35 years, 46% female) in 15 treatment programmes in Africa, Asia and South America with (3) 12 months of follow-up, we investigated risk factors for no follow-up after treatment initiation, and loss to follow-up or death in the first 6 months. FINDINGS: Overall, 211 patients (3.8%) had no follow-up, 880 (16.0%) were lost to follow-up and 141 (2.6%) were known to have died in the first 6 months. The probability of no follow-up was higher in 2003-2004 than in 2000 or earlier (odds ratio, OR: 5.06; 95% confidence interval, CI: 1.28-20.0), as was loss to follow-up (hazard ratio, HR: 7.62; 95% CI: 4.55-12.8) but not recorded death (HR: 1.02; 95% CI: 0.44-2.36). Compared with a baseline CD4-cell count (3) 50 cells/microl, a count < 25 cells/microl was associated with a higher probability of no follow-up (OR: 2.49; 95% CI: 1.43-4.33), loss to follow-up (HR: 1.48; 95% CI: 1.23-1.77) and death (HR: 3.34; 95% CI: 2.10-5.30). Compared to free treatment, fee-for-service programmes were associated with a higher probability of no follow-up (OR: 3.71; 95% CI: 0.97-16.05) and higher mortality (HR: 4.64; 95% CI: 1.11-19.41). CONCLUSION: Early patient losses were increasingly common when programmes were scaled up and were associated with a fee for service and advanced immunodeficiency at baseline. Measures to maximize ART programme retention are required in resource-poor countries.