Search

Antenatal progesterone prevents preterm birth (PTB) in women with a short cervix or prior PTB in daily vaginal or weekly injectable formulations, respectively. Neither has been tested for the indication of maternal HIV, which is associated with an elevated risk of PTB. The Vaginal Progesterone (VP) Trial was a pilot feasibility study of VP to prevent HIV-related PTB in Lusaka, Zambia. Using mixed methods, we concurrently evaluated the acceptability of the trial and the study product among participants. Over a 1-year period, we enrolled 140 pregnant women living with HIV into a double-masked, placebo-controlled, randomized trial of daily self-administered VP or placebo. We administered an endline questionnaire to all participants and conducted in-depth interviews with 30 participants to assess barriers and facilitators to uptake and retention in the trial and to study product adherence. All interviews were audiotaped, transcribed, translated into English as needed, and independently coded by two analysts to capture emerging themes. Of 131 participants who completed the questionnaire, 128 (98%) reported that nothing was difficult when asked the hardest part about using the study product. When given a hypothetical choice between vaginal and injectable progesterone, 97 (74%) chose vaginal, 31 (24%) injectable, and 3 (2%) stated no preference. Most interviewees reported no difficulties with using the study product; others cited minor side effects and surmountable challenges. Strategies that supported adherence included setting alarms, aligning dosing with antiretrovirals, receiving encouragement from friends and family, sensing a benefit to their unborn baby, and positive feedback from study staff. Participants who reported preference of a vaginal medication over injectable described familiarity with the vaginal product, a fear of needles and resulting pain, and inconvenience of a weekly clinic visit. Those who would prefer weekly injections cited fewer doses to remember. Perceived barriers to study participation included mistrust about the motivations behind research, suspicion of Satanism, and futility or possible harm from a placebo. We report key influences on acceptability of a randomized trial of VP to prevent PTB among HIV-infected women in Zambia, which should inform methods to promote uptake, adherence, and retention in a full-scale trial. [ABSTRACT FROM AUTHOR]

Antenatal vaginal progesterone (VP) reduces the risk of preterm birth (PTB) in women with shortened cervical length, and we hypothesize that it may also prevent PTB in women with HIV as their primary risk factor. We conducted a pilot feasibility study in Lusaka, Zambia to investigate uptake, adherence, and retention in preparation for a future efficacy trial. This was a double-masked, placebo-controlled, randomized trial of 200mg daily self-administered VP suppository or placebo. Pregnant women with HIV who were initiating or continuing antiretroviral therapy were eligible for participation. Potential participants underwent ultrasound to assess eligibility; we excluded those ≥24 gestational weeks, with non-viable, multiple gestation, or extrauterine pregnancies, with short cervix (<2.0cm), or with prior spontaneous PTB. Participants initiated study product between 20–24 weeks of gestation and continued to 37 weeks (or delivery, if sooner). The primary outcome was adherence (proportion achieving ≥80% study product use), assessed by dye stain assay of returned single-use vaginal applicators. Secondary outcomes with pre-defined feasibility targets were: uptake (≥50% eligible participants enrolled) and retention (≥90% ascertainment of delivery outcomes). We also evaluated preliminary efficacy by comparing the risk of spontaneous PTB <37 weeks between groups. From July 2017 to June 2018, 208 HIV-infected pregnant women were eligible for screening and 140 (uptake = 67%) were randomly allocated to VP (n = 70) or placebo (n = 70). Mean adherence was 94% (SD±9.4); 91% (n = 125/137) achieved overall adherence ≥80%. Delivery outcomes were ascertained from 134 (96%) participants. Spontaneous PTB occurred in 10 participants (15%) receiving placebo and 8 (12%) receiving progesterone (RR 0.82; 95%CI:0.34–1.97). Spontaneous PTB < 34 weeks occurred in 6 (9%) receiving placebo and 4 (6%) receiving progesterone (RR 0.67; 95%CI:0.20–2.67). In contrast to findings from vaginal microbicide studies in HIV-uninfected, non-pregnant women, our trial participants were highly adherent to daily self-administered vaginal progesterone. The study's a priori criteria for uptake, adherence, and retention were met, indicating that a phase III efficacy trial would be feasible. [ABSTRACT FROM AUTHOR]

Vaginal dysbiosis has been shown to increase the risk of some adverse birth outcomes. HIV infection may be associated with shifts in the vaginal microbiome. We characterized microbial communities in vaginal swabs collected between 16–20 gestational weeks in the Zambian Preterm Birth Prevention Study to investigate whether HIV and its treatment alter the microbiome in pregnancy. We quantified relative abundance and diversity of bacterial taxa by whole-genome shotgun sequencing and identified community state types (CST) by hierarchical clustering. Associations between exposures—HIV serostatus (HIV+ vs HIV-) and preconceptional ART (ART+ vs ART-)—and microbiome characteristics were tested with rank-sum, and by linear and logistic regression, accounting for sampling by inverse-probability weighting. Of 261 vaginal swabs, 256 (98%) had evaluable sequences; 98 (38%) were from HIV+ participants, 55 (56%) of whom had preconceptional ART exposure. Major CSTs were dominated by: L. crispatus (CST 1; 17%), L.] iners (CST 3; 32%), Gardnerella vaginalis (CST 4-I; 37%), G. vaginalis & Atopobium vaginae (CST 4-II; 5%), and other mixed anaerobes (CST 4-III; 9%). G. vaginalis was present in 95%; mean relative abundance was higher in HIV+ (0.46±0.29) compared to HIV- participants (0.35±0.33; rank-sum p = .01). Shannon diversity was higher in HIV+/ART+ (coeff 0.17; 95%CI (0.01,0.33), p = .04) and HIV+/ART- (coeff 0.37; 95%CI (0.19,0.55), p < .001) participants compared to HIV-. Anaerobe-dominant CSTs were more prevalent in HIV+/ART+ (63%, AOR 3.11; 95%CI: 1.48,6.55, p = .003) and HIV+/ART- (85%, AOR 7.59; 95%CI (2.80,20.6), p < .001) compared to HIV- (45%). Restricting the comparison to 111 women in either CST 3 (L. iners dominance) or CST 1 (L. crispatus dominance), CST 3 frequency was similar in HIV- (63%) and HIV+/ART- participants (67%, AOR 1.31; 95%CI: (0.25,6.90), p = .7), but higher in HIV+/ART+ (89%, AOR 6.44; 95%CI: (1.12,37.0), p = .04). Pregnant women in Zambia, particularly those with HIV, had diverse anaerobe-dominant vaginal microbiota. [ABSTRACT FROM AUTHOR]

Background: Globally, preterm birth is the leading cause of neonatal death with estimated prevalence and associated mortality highest in low- and middle-income countries (LMICs). Accurate identification of preterm infants is important at the individual level for appropriate clinical intervention as well as at the population level for informed policy decisions and resource allocation. As early prenatal ultrasound is commonly not available in these settings, gestational age (GA) is often estimated using newborn assessment at birth. This approach assumes last menstrual period to be unreliable and birthweight to be unable to distinguish preterm infants from those that are small for gestational age (SGA). We sought to leverage machine learning algorithms incorporating maternal factors associated with SGA to improve accuracy of preterm newborn identification in LMIC settings. Methods and findings: This study uses data from an ongoing obstetrical cohort in Lusaka, Zambia that uses early pregnancy ultrasound to estimate GA. Our intent was to identify the best set of parameters commonly available at delivery to correctly categorize births as either preterm (<37 weeks) or term, compared to GA assigned by early ultrasound as the gold standard. Trained midwives conducted a newborn assessment (<72 hours) and collected maternal and neonatal data at the time of delivery or shortly thereafter. New Ballard Score (NBS), last menstrual period (LMP), and birth weight were used individually to assign GA at delivery and categorize each birth as either preterm or term. Additionally, machine learning techniques incorporated combinations of these measures with several maternal and newborn characteristics associated with prematurity and SGA to develop GA at delivery and preterm birth prediction models. The distribution and accuracy of all models were compared to early ultrasound dating. Within our live-born cohort to date (n = 862), the median GA at delivery by early ultrasound was 39.4 weeks (IQR: 38.3–40.3). Among assessed newborns with complete data included in this analysis (n = 468), the median GA by ultrasound was 39.6 weeks (IQR: 38.4–40.3). Using machine learning, we identified a combination of six accessible parameters (LMP, birth weight, twin delivery, maternal height, hypertension in labor, and HIV serostatus) that can be used by machine learning to outperform current GA prediction methods. For preterm birth prediction, this combination of covariates correctly classified >94% of newborns and achieved an area under the curve (AUC) of 0.9796. Conclusions: We identified a parsimonious list of variables that can be used by machine learning approaches to improve accuracy of preterm newborn identification. Our best-performing model included LMP, birth weight, twin delivery, HIV serostatus, and maternal factors associated with SGA. These variables are all easily collected at delivery, reducing the skill and time required by the frontline health worker to assess GA. Trial registration: ClinicalTrials.gov Identifier: [ABSTRACT FROM AUTHOR]

Introduction: Zambia has made substantial investments in health systems capacity, yet it remains unclear whether improved service quality improves outcomes. We investigated the association between health system capacity and use of prevention of mother-to-child HIV transmission (PMTCT) services in Zambia. Materials and methods: We analyzed data from two studies conducted in rural and semi-urban Lusaka Province in 2014–2015. Health system capacity, our primary exposure, was measured with a validated balanced scorecard approach. Based on WHO building blocks for health systems strengthening, we derived overall and domain-specific facility scores (range: 0–100), with higher scores indicating greater capacity. Our outcome, community-level maternal antiretroviral drug use at 12 months postpartum, was measured via self-report in a large cohort study evaluating PMTCT program impact. Associations between health systems capacity and our outcome were analyzed via linear regression. Results: Among 29 facilities, median overall facility score was 72 (IQR:67–74). Median domain scores were: patient satisfaction 75 (IQR 71–78); human resources 85 (IQR:63–87); finance 50 (IQR:50–67); governance 82 (IQR:74–91); service capacity 77 (IQR:68–79); service provision 60 (IQR:52–76). Our programmatic outcome was measured from 804 HIV-infected mothers. Median community-level antiretroviral use at 12 months was 81% (IQR:69–89%). Patient satisfaction was the only domain score significantly associated with 12-month maternal antiretroviral use (β:0.22; p = 0.02). When we excluded the human resources and finance domains, we found a positive association between composite 4-domain facility score and 12-month maternal antiretroviral use in peri-urban but not rural facilities. Conclusions: In these Zambian health facilities, patient satisfaction was positively associated with maternal antiretroviral 12 months postpartum. The association between overall health system capacity and maternal antiretroviral drug use was stronger in peri-urban versus rural facilities. Additional work is needed to guide strategic investments for improved outcomes in HIV and broader maternal-child health region-wide. [ABSTRACT FROM AUTHOR]

Background: Very few efforts have been undertaken to scale-up low-cost approaches to cervical cancer prevention in low-resource countries. Methods: In a public sector cervical cancer prevention program in Zambia, nurses provided visual-inspection with acetic acid (VIA) and cryotherapy in clinics co-housed with HIV/AIDS programs, and referred women with complex lesions for histopathologic evaluation. Low-cost technological adaptations were deployed for improving VIA detection, facilitating expert physician opinion, and ensuring quality assurance. Key process and outcome indicators were derived by analyzing electronic medical records to evaluate program expansion efforts. Findings: Between 2006-2013, screening services were expanded from 2 to 12 clinics in Lusaka, the most-populous province in Zambia, through which 102,942 women were screened. The majority (71.7%) were in the target age-range of 25–49 years; 28% were HIV-positive. Out of 101,867 with evaluable data, 20,419 (20%) were VIA positive, of whom 11,508 (56.4%) were treated with cryotherapy, and 8,911 (43.6%) were referred for histopathologic evaluation. Most women (87%, 86,301 of 98,961 evaluable) received same-day services (including 5% undergoing same-visit cryotherapy and 82% screening VIA-negative). The proportion of women with cervical intraepithelial neoplasia grade 2 and worse (CIN2+) among those referred for histopathologic evaluation was 44.1% (1,735/3,938 with histopathology results). Detection rates for CIN2+ and invasive cervical cancer were 17 and 7 per 1,000 women screened, respectively. Women with HIV were more likely to screen positive, to be referred for histopathologic evaluation, and to have cervical precancer and cancer than HIV-negative women. Interpretation: We creatively disrupted the 'no screening' status quo prevailing in Zambia and addressed the heavy burden of cervical disease among previously unscreened women by establishing and scaling-up public-sector screening and treatment services at a population level. Key determinants for successful expansion included leveraging HIV/AIDS program investments, and context-specific information technology applications for quality assurance and filling human resource gaps. [ABSTRACT FROM AUTHOR]

Background: Population-based evaluations of programs for prevention of mother-to-child HIV transmission (PMTCT) are scarce. We measured PMTCT service coverage, regimen use, and HIV-free survival among children ≤24 mo of age in Cameroon, Côte D'Ivoire, South Africa, and Zambia. Methods and Findings: We randomly sampled households in 26 communities and offered participation if a child had been born to a woman living there during the prior 24 mo. We tested consenting mothers with rapid HIV antibody tests and tested the children of seropositive mothers with HIV DNA PCR or rapid antibody tests. Our primary outcome was 24-mo HIV-free survival, estimated with survival analysis. In an individual-level analysis, we evaluated the effectiveness of various PMTCT regimens. In a community-level analysis, we evaluated the relationship between HIV-free survival and community PMTCT coverage (the proportion of HIV-exposed infants in each community that received any PMTCT intervention during gestation or breastfeeding). We also compared our community coverage results to those of a contemporaneous study conducted in the facilities serving each sampled community. Of 7,985 surveyed children under 2 y of age, 1,014 (12.7%) were HIV-exposed. Of these, 110 (10.9%) were HIV-infected, 851 (83.9%) were HIV-uninfected, and 53 (5.2%) were dead. HIV-free survival at 24 mo of age among all HIV-exposed children was 79.7% (95% CI: 76.4, 82.6) overall, with the following country-level estimates: Cameroon (72.6%; 95% CI: 62.3, 80.5), South Africa (77.7%; 95% CI: 72.5, 82.1), Zambia (83.1%; 95% CI: 78.4, 86.8), and Côte D'Ivoire (84.4%; 95% CI: 70.0, 92.2). In adjusted analyses, the risk of death or HIV infection was non-significantly lower in children whose mothers received a more complex regimen of either two or three antiretroviral drugs compared to those receiving no prophylaxis (adjusted hazard ratio: 0.60; 95% CI: 0.34, 1.06). Risk of death was not different for children whose mothers received a more complex regimen compared to those given single-dose nevirapine (adjusted hazard ratio: 0.88; 95% CI: 0.45, 1.72). Community PMTCT coverage was highest in Cameroon, where 75 of 114 HIV-exposed infants met criteria for coverage (66%; 95% CI: 56, 74), followed by Zambia (219 of 444, 49%; 95% CI: 45, 54), then South Africa (152 of 365, 42%; 95% CI: 37, 47), and then Côte D'Ivoire (3 of 53, 5.7%; 95% CI: 1.2, 16). In a cluster-level analysis, community PMTCT coverage was highly correlated with facility PMTCT coverage (Pearson's r = 0.85), and moderately correlated with 24-mo HIV-free survival (Pearson's r = 0.29). In 14 of 16 instances where both the facility and community samples were large enough for comparison, the facility-based coverage measure exceeded that observed in the community. Conclusions: HIV-free survival can be estimated with community surveys and should be incorporated into ongoing country monitoring. Facility-based coverage measures correlate with those derived from community sampling, but may overestimate population coverage. The more complex regimens recommended by the World Health Organization seem to have measurable public health benefit at the population level, but power was limited and additional field validation is needed. Please see later in the article for the Editors' Summary [ABSTRACT FROM AUTHOR]

Objectives: Mortality in patients starting antiretroviral therapy (ART) is higher in Malawi and Zambia than in South Africa. We examined whether different monitoring of ART (viral load [VL] in South Africa and CD4 count in Malawi and Zambia) could explain this mortality difference. Design:: Mathematical modelling study based on data from ART programmes. Methods: We used a stochastic simulation model to study the effect of VL monitoring on mortality over 5 years. In baseline scenario A all parameters were identical between strategies except for more timely and complete detection of treatment failure with VL monitoring. Additional scenarios introduced delays in switching to second-line ART (scenario B) or higher virologic failure rates (due to worse adherence) when monitoring was based on CD4 counts only (scenario C). Results are presented as relative risks (RR) with 95% prediction intervals and percent of observed mortality difference explained. Results: RRs comparing VL with CD4 cell count monitoring were 0.94 (0.74–1.03) in scenario A, 0.94 (0.77–1.02) with delayed switching (scenario B) and 0.80 (0.44–1.07) when assuming a 3-times higher rate of failure (scenario C). The observed mortality at 3 years was 10.9% in Malawi and Zambia and 8.6% in South Africa (absolute difference 2.3%). The percentage of the mortality difference explained by VL monitoring ranged from 4% (scenario A) to 32% (scenarios B and C combined, assuming a 3-times higher failure rate). Eleven percent was explained by non-HIV related mortality. Conclusions: VL monitoring reduces mortality moderately when assuming improved adherence and decreased failure rates. [ABSTRACT FROM AUTHOR]

Groesbeck Parham and colleagues describe their Cervical Cancer Prevention Program in Zambia, which has provided services to over 58,000 women over the past five years, and share lessons learned from the program's implementation and integration with existing HIV/AIDS programs. [ABSTRACT FROM AUTHOR]

Background: Patients starting antiretroviral therapy (ART) for acquired immunodeficiency syndrome (AIDS) in sub-Saharan Africa have high rates of mortality in the initial weeks of treatment. We assessed the association of serum phosphate with early mortality among HIV-infected adults with severe malnutrition and/or advanced immunosuppression. Methodology/Principal Findings: An observational cohort of 142 HIV-infected adults initiating ART in Lusaka, Zambia with body mass index (BMI) ,16 kg/m² or CD4+ lymphocyte count <50 cells/μL, or both, was followed prospectively during the first 12 weeks of ART. Detailed health and dietary intake history, review of systems, physical examination, serum metabolic panel including phosphate, and serum ferritin and high-sensitivity C-reactive protein (hsCRP) were monitored. The primary outcome was mortality. Baseline serum phosphate was a significant predictor of mortality; participants alive at 12 weeks had a median value of 1.30 mmol/L (interquartile range [IQR]: 1.04, 1.43), compared to 1.06 mmol/L (IQR: 0.89, 1.27) among those who died (p,0.01). Each 0.1 mmol/L increase in baseline phosphate was associated with an incremental decrease in mortality (AHR 0.83; 95% CI 0.72 to 0.95). The association was independent of other metabolic parameters and known risk factors for early ART-associated mortality in sub-Saharan Africa. While participant attrition represented a limitation, it was consistent with local program experience. Conclusions/Significance: Low serum phosphate at ART initiation was an independent predictor of early mortality among HIV patients starting ART with severe malnutrition or advanced immunosuppression. This may represent a physiologic phenomenon similar to refeeding syndrome, and may lead to therapeutic interventions that could reduce mortality. [ABSTRACT FROM AUTHOR]

Background: The benefit of routine HIV-1 viral load (VL) monitoring of patients on antiretroviral therapy (ART) in resource-constrained settings is uncertain because of the high costs associated with the test and the limited treatment options. We designed a cluster randomized controlled trial to compare the use of routine VL testing at ART-initiation and at 3, 6, 12, and 18 months, versus our local standard of care (which uses immunological and clinical criteria to diagnose treatment failure, with discretionary VL testing when the two do not agree). Methodology: Dedicated study personnel were integrated into public-sector ART clinics. We collected participant information in a dedicated research database. Twelve ART clinics in Lusaka, Zambia constituted the units of randomization. Study clinics were stratified into pairs according to matching criteria (historical mortality rate, size, and duration of operation) to limit the effect of clustering, and independently randomized to the intervention and control arms. The study was powered to detect a 36% reduction in mortality at 18 months. Principal Findings: From December 2006 to May 2008, we completed enrollment of 1973 participants. Measured baseline characteristics did not differ significantly between the study arms. Enrollment was staggered by clinic pair and truncated at two matched sites. Conclusions: A large clinical trial of routing VL monitoring was successfully implemented in a dynamic and rapidly growing national ART program. Close collaboration with local health authorities and adequate reserve staff were critical to success. Randomized controlled trials such as this will likely prove valuable in determining long-term outcomes in resource-constrained settings. Trial Registration: Clinicaltrials.gov NCT00929604 [ABSTRACT FROM AUTHOR]

Background: Intrapartum and neonatal single-dose nevirapine (NVP) reduces the risk of mother-to-child HIV transmission but also induces viral resistance to non-nucleoside reverse transcriptase inhibitor (NNRTI) drugs. This drug resistance largely fades over time. We hypothesized that women with a prior single-dose NVP exposure would have no more than a 10% higher cumulative prevalence of failure of their NNRTI-containing antiretroviral therapy (ART) over the first 48 wk of therapy than would women without a prior exposure. Methods and Findings: We enrolled 355 NVP-exposed and 523 NVP-unexposed women at two sites in Zambia, one site in Kenya, and two sites in Thailand into a prospective, non-inferiority cohort study and followed them for 48 wk on ART. Those who died, discontinued NNRTI-containing ART, or had a plasma viral load $400 copies/ml at either the 24 wk or 48 wk study visits and confirmed on repeat testing were characterized as having failed therapy. Overall, 114 of 355 NVP-exposed women (32.1%) and 132 of 523 NVP-unexposed women (25.2%) met criteria for treatment failure. The difference in failure rates between the exposure groups was 6.9% (95% confidence interval [CI] 0.8%-13.0%). The failure rates of women stratified by our predefined exposure interval categories were as follows: 47 of 116 women in whom less than 6 mo elapsed between exposure and starting ART failed therapy (40%; p<0.001 compared to unexposed women); 25 of 67 women in whom 7-12 mo elapsed between exposure and starting ART failed therapy (37%; p = 0.04 compared to unexposed women); and 42 of 172 women in whom more than 12 mo elapsed between exposure and starting ART failed therapy (24%; p = 0.82 compared to unexposed women). Locally weighted regression analysis also indicated a clear inverse relationship between virologic failure and the exposure interval. Conclusions: Prior exposure to single-dose NVP was associated with an increased risk of treatment failure; however, this risk seems largely confined to women with a more recent exposure. Women requiring ART within 12 mo of NVP exposure should not be prescribed an NNRTI-containing regimen as first-line therapy. [ABSTRACT FROM AUTHOR]

In this article, the authors discuss the improvement of patient retention in the treatment programs for AIDS. They note that the introduction of AIDS care represented a change in the delivery of medical care. They point out that antiretroviral therapy (ART) interruptions are considered as a problem due to their associated risk of incomplete virologic suppression, immunologic decline and death.

Background: Accurate estimates of gestational age (GA) at birth are important for preterm birth surveillance but can be challenging to obtain in low income countries. Our objective was to develop machine learning models to accurately estimate GA shortly after birth using clinical and metabolomic data., Methods: We derived three GA estimation models using ELASTIC NET multivariable linear regression using metabolomic markers from heel-prick blood samples and clinical data from a retrospective cohort of newborns from Ontario, Canada. We conducted internal model validation in an independent cohort of Ontario newborns, and external validation in heel prick and cord blood sample data collected from newborns from prospective birth cohorts in Lusaka, Zambia and Matlab, Bangladesh. Model performance was measured by comparing model-derived estimates of GA to reference estimates from early pregnancy ultrasound., Results: Samples were collected from 311 newborns from Zambia and 1176 from Bangladesh. The best-performing model accurately estimated GA within about 6 days of ultrasound estimates in both cohorts when applied to heel prick data (MAE 0.79 weeks (95% CI 0.69, 0.90) for Zambia; 0.81 weeks (0.75, 0.86) for Bangladesh), and within about 7 days when applied to cord blood data (1.02 weeks (0.90, 1.15) for Zambia; 0.95 weeks (0.90, 0.99) for Bangladesh)., Conclusions: Algorithms developed in Canada provided accurate estimates of GA when applied to external cohorts from Zambia and Bangladesh. Model performance was superior in heel prick data as compared to cord blood data., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2023 Hawken et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Background: We estimated the unit costs and cost-effectiveness of a government ART program in 45 sites in Zambia supported by the Centre for Infectious Disease Research Zambia (CIDRZ)., Methods: We estimated per person-year costs at the facility level, and support costs incurred above the facility level and used multiple regression to estimate variation in these costs. To estimate ART effectiveness, we compared mortality in this Zambian population to that of a cohort of rural Ugandan HIV patients receiving co-trimoxazole (CTX) prophylaxis. We used micro-costing techniques to estimate incremental unit costs, and calculated cost-effectiveness ratios with a computer model which projected results to 10 years., Results: The program cost $69.7 million for 125,436 person-years of ART, or $556 per ART-year. Compared to CTX prophylaxis alone, the program averted 33.3 deaths or 244.5 disability adjusted life-years (DALYs) per 100 person-years of ART. In the base-case analysis, the net cost per DALY averted was $833 compared to CTX alone. More than two-thirds of the variation in average incremental total and on-site cost per patient-year of treatment is explained by eight determinants, including the complexity of the patient-case load, the degree of adherence among the patients, and institutional characteristics including, experience, scale, scope, setting and sector., Conclusions and Significance: The 45 sites exhibited substantial variation in unit costs and cost-effectiveness and are in the mid-range of cost-effectiveness when compared to other ART programs studied in southern Africa. Early treatment initiation, large scale, and hospital setting, are associated with statistically significantly lower costs, while others (rural location, private sector) are associated with shifting cost from on- to off-site. This study shows that ART programs can be significantly less costly or more cost-effective when they exploit economies of scale and scope, and initiate patients at higher CD4 counts.