Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Sep 28.
Published in final edited form as: Epidemiology. 2022 May 1;33(3):422–430. doi: 10.1097/EDE.0000000000001465

Risk of adverse birth outcomes in two cohorts of pregnant women with HIV in Zambia

Joan T Price a,b,c, Yuri V Sebastião a, Bellington Vwalika a,b, Stephen R Cole d, Felistas M Mbewe c, Winifreda M Phiri c, Bethany L Freeman a, Margaret P Kasaro b,c, Marc Peterson a, Dwight J Rouse e, Elizabeth M Stringer a, Jeffrey S A Stringer a
PMCID: PMC9516482  NIHMSID: NIHMS1834875  PMID: 35067569

Abstract

Background:

A trial of progesterone to prevent preterm birth among HIV-infected Zambian women (IPOP) found no treatment effect, but the risk of the primary outcome was among the lowest ever documented in women with HIV. In this secondary analysis, we compare the risks of preterm birth (<37 weeks), stillbirth, and a composite primary outcome comprising the two in IPOP versus an observational pregnancy cohort (ZAPPS) in Zambia, to evaluate reasons for the low risk in IPOP.

Methods:

Both studies enrolled women before 24 gestational weeks, during August 2015 - September 2017 (ZAPPS) and February 2018 - January 2020 (IPOP). We used linear probability and log binomial regression to estimate risk differences and risk ratios (RR), before and after restriction and standardization with inverse probability weights.

Results:

The unadjusted risk of composite outcome was 18% in ZAPPS (N=1450) and 9% in IPOP (N=791) (RR=2.0; 95%CI: 1.6–2.6). After restricting and standardizing the ZAPPS cohort to the distribution of IPOP baseline characteristics, the risk remained higher in ZAPPS (RR=1.6; 95%CI: 1.0–2.4). The lower risk of preterm/stillbirth in IPOP was only partially explained by measured risk factors.

Conclusions:

Possible benefits in IPOP of additional monetary reimbursement, more frequent visits, and group-based care warrant further investigation.

Keywords: HIV, preterm birth, stillbirth, pregnancy, Zambia, marginal structural models, inverse probability weighting

INTRODUCTION

Nearly one in five pregnancies in parts of Africa and South Asia end in preterm birth, defined as delivery before 37 gestational weeks.1 The risk of stillbirth, or the delivery of a neonate without signs of life, exceeds 2.5% in many low- and middle-income countries, where an estimated 98% of global stillbirths occur.2 In sub-Saharan Africa, up to one in four pregnant women have HIV, which confers an almost two-fold increased risk of experiencing either stillbirth or preterm birth,3 particularly those that arise spontaneously.4 Despite the clear benefit of antenatal antiretroviral therapy (ART) for both prevention of infant infection and maternal health, these therapies do not appear to reduce the risk of preterm birth or perinatal death.5–7 African cohorts have reported estimates of preterm birth risk among women with HIV as low as 11%8,9 to as high as 30%.6,10–12 Similarly, the risk of stillbirth among women with HIV has been reported between 2% and 5%.6,11–15 These wide ranges of estimates may be attributable to variations in HIV disease severity, ART regimen, and inconsistent methodologies for gestational age estimation and outcome ascertainment across cohorts.5,8

Considerable investment has been made to test strategies to reduce preterm birth, but few are broadly or consistently effective.16–20 Antenatal administration of intramuscular 17-hydroxyprogesterone caproate (17P) has reduced the risk of preterm birth among women with a history of spontaneous preterm birth in some populations.21 Because HIV infection is associated with systemic and vaginal inflammation,22,23 both common antecedents of spontaneous preterm birth,24,25 we hypothesized that the anti-inflammatory properties of progesterone could minimize HIV-mediated immune activation and inflammation and thus reduce the risk of preterm birth.26 However, despite complete data and high adherence, we found that 17P did not reduce the primary outcome of preterm birth or stillbirth at any gestational age in the recent Improving Pregnancy Outcomes with Progesterone (IPOP) randomized trial in Lusaka, Zambia.27 Combining the IPOP treatment groups, we found a surprisingly low 8% risk of preterm birth, which is among the lowest ever reported among women with HIV.3 In this study, we investigate the difference in preterm birth and stillbirth in the IPOP trial (treating both study groups as a single cohort) with participants in the ZAPPS study, a prospective pregnancy cohort conducted concurrently and in the same clinics as IPOP.

METHODS

IPOP Trial Design and Participants

The Improving Pregnancy Outcomes with Progesterone (IPOP) study was a phase III, randomized, double-masked, placebo-controlled trial conducted between February 2018 and August 2020 at the Women and Newborn Hospital of the University Teaching Hospitals and at the Kamwala District Health Centre in Lusaka, Zambia. Its primary objective was to evaluate whether weekly antenatal 17P would reduce the risk of delivery prior to 37 weeks of gestation or stillbirth at any gestational age.

The following were eligibility criteria for IPOP: (1) at least 18 years of age; (2) confirmed HIV-1 infection; (3) viable intrauterine singleton pregnancy; and (4) less than 24 0/7 weeks of gestation. Women with any of the following characteristics were excluded: (1) multiple gestation; (2) major uterine or fetal anomaly; (3) planned or in situ cervical cerclage; (4) indication for planned delivery before 37 weeks at enrollment; (5) threatened abortion, preterm labor, or ruptured membranes at enrollment; and (6) known allergy or medical contraindication (e.g., uncontrolled hypertension) to 17P. The trial also excluded (7) those who reported a prior spontaneous preterm delivery because these women would have been eligible to receive 17P for that indication. For the present secondary analysis, we excluded nine additional participants who reported a prior provider-initiated preterm birth. We did this to allow better comparability to a restricted ZAPPS cohort, where information to reliably distinguish between previous spontaneous versus provider-initiated preterm birth phenotypes was unavailable.

ZAPPS Cohort

The Zambian Preterm Birth Prevention Study (ZAPPS) is an ongoing cohort that began enrollment in September 2015. The primary goal of the ZAPPS cohort is to better understand epidemiologic and biologic determinants of preterm birth and other adverse outcomes. Phase I of ZAPPS enrolled women with and without HIV between August 2015 – September 2017.

Pregnant women meeting the following criteria were eligible for enrollment in ZAPPS: (1) 18 years of age or older; (2) viable intrauterine singleton or twin pregnancy; (3) presentation to antenatal care prior to 24 gestational weeks.28

Ethical Considerations

The IPOP and ZAPPS studies were each approved by the University of North Carolina Institutional Review Board, the University of Zambia Biomedical Research Ethics Committee, and the Zambia Ministry of Health National Health Research Authority prior to initiation. All study participants provide written informed consent before enrollment.

Study Procedures

Detailed procedures for both the IPOP trial26 and ZAPPS cohort28 have been described previously. In brief, participants in both studies were recruited from public sector antenatal clinics and underwent initial ultrasound evaluation to establish eligibility and gestational age.29,30 Women found to be preliminarily eligible and interested in study participation were invited to undergo informed consent procedures. Participants in both cohorts returned to the study clinic for monthly antenatal visits, ultrasound cervical length assessment between 16 and 24 gestational weeks, and a repeat fetal biometry ultrasound at 32 weeks. Antenatal care followed standard Zambian guidelines and included screening for anemia, asymptomatic bacteriuria, and syphilis. All participants underwent testing for HIV at the screening visit. Participants in IPOP returned weekly for injection visits through 36 gestational weeks. At each weekly visit, IPOP midwives assessed adverse drug reactions; other signs, symptoms, or diagnoses; and concomitant medications. Prior to administering the study injection, midwives measured and documented fetal heart rate and maternal blood pressure. In both IPOP and ZAPPS, midwives collected key delivery information by medical record review, direct observation, and participant interview either at the time of delivery, or at a subsequent follow-up visit. All participants were scheduled to return to the study clinic for a single postpartum visit at 6 weeks following delivery. At every study visit, participants received standard monetary reimbursement for their time and transportation costs – in ZAPPS, participants received approximately $4 per monthly visit; IPOP participants received an additional $3 for intervening weekly injection visits.

Outcomes and Exposures of Interest

In this analysis of IPOP and ZAPPS data, we defined our primary outcome to mirror that of the IPOP trial: a composite of preterm birth before 37 gestational weeks or stillbirth at any gestational age. This composite primary outcome was chosen in IPOP because preterm birth and stillbirth are competing risks, often co-occur, and share many common risk factors and obstetrical antecedents. Secondary outcomes included the individual components of the primary outcome, namely (1) live birth before 37 weeks and (2) stillbirth at any gestational age.

Our primary exposure was defined as participation in ZAPPS cohort, relative to the IPOP trial. We considered the following baseline covariates for adjustment of confounding on the primary composite outcome, a priori, based on expert knowledge and availability in both cohorts: maternal age, body mass index (BMI), nulliparity, prior stillbirth, gestational age at screening, short cervix (mid-trimester cervical length < 2.5 cm), preconceptional ART, viral load, ART regimen, syphilis serostatus, bacteriuria (3+ leukocyte esterase or nitrites on urine dipstick), hypertension at screening (systolic ≥140 or diastolic ≥90mmHg), anemia at screening (hemoglobin <11 g/dL), alcohol use or smoking during pregnancy, educational attainment, marital status, and household assets (electricity, toilet facility, cooking fuel).

Of 800 participants randomized in the IPOP trial, 399 received 17P and 401 placebo. As reported in the primary intent-to-treat analysis, despite very high adherence and retention, the intervention had no effect on the primary outcome: 36 (9%) of 399 participants assigned to 17P had preterm birth or stillbirth, compared to 36 (9%) of 401 participants assigned to placebo.27 Retention at the delivery visit and ascertainment of the primary composite outcome was 100%. In the ZAPPS cohort, 1450 women were enrolled, of whom 1,216 (84%) were retained at delivery and 1,208 (83%) had complete gestational age and neonatal vital status information for the primary outcome of interest.

Statistical Analysis

We used frequencies and summary measures (median, ranges) to describe baseline covariates and unadjusted risk of outcomes by study cohort (ZAPPS and IPOP). IPOP participants had complete data for all outcome and baseline variables of interest with the exception of alcohol use. As ZAPPS participants had missing values for both outcome variables and several of the baseline variables of interest (eTable 1), we employed multiple imputation to account for the non-monotone missing data.31 We imputed missing covariate and outcome data 50 times using a Markov chain Monte Carlo algorithm that depended on all the above-listed covariates, in addition to the component outcome variables for live birth before 37 weeks and stillbirth at any gestational age. We performed all subsequent analyses for each imputed data set and then averaged estimates of frequencies and measures of association (using Rubin’s rule to combined standard errors).32

First, we estimated the risk difference (RD) and risk ratio (RR), along with Wald-type 95% confidence intervals (CI), comparing the composite outcome in the ZAPPS cohort to the IPOP trial using linear probability and log binomial regression models, respectively. Second, we restricted the ZAPPS cohort to those participants who met IPOP eligibility criteria (single gestation, no prior preterm birth, HIV seropositivity), and again report the RD and RR, along with 95% CI. Third, we standardized the subset of ZAPPS participants who met IPOP eligibility to the distribution of covariates in the IPOP trial using inverse probability weights.33 We estimated the conditional probability of being a participant in the IPOP trial compared to the ZAPPS cohort using a logistic regression model with study cohort as the outcome variable and the following characteristics as covariates: maternal age, nulliparity, prior stillbirth, gestational age at screening, short cervix, syphilis, and hypertension at screening, plus a product term for the linear component of age and syphilis. We selected these covariates from the initial set of covariates examined based on their distribution in the study population and their unadjusted association with the composite outcome in IPOP, overall ZAPPS, and restricted ZAPPS cohort before imputation (eTables 2 and 3). Further details on covariate selection are provided in the eAppendix. Continuous covariates (age, gestational age) were included in the model for study cohort using restricted quadratic splines.34 The standardization weights for ZAPPS participants were estimated as the odds of being an IPOP participant (i.e., the conditional probability of being an IPOP participant divided by the complement of the conditional probability of being an IPOP participant). IPOP participants received a weight of 1. This application has been referred to as standardized morbidity ratio (SMR) weights33 or average treatment effect in the treated (ATT) weights.35,36 Based on the distribution of both the estimated probability of IPOP participation (mean: 0.75; minimum: 0.12; 10th percentile: 0.62; 90th percentile: 0.85; maximum: 0.93) and the resulting weights (mean: 3.7; minimum: 0.14; 10th percentile: 1.6; 90th percentile: 5.8; maximum: 13.9) for ZAPPS participants, we assumed that any bias due to non-positivity would be minimal.37 Absolute standardized differences were calculated to aid with qualitative assessment of covariate balance between the two cohorts before and after standardization.35 We report the standardized RD and RR from weighted regression models along with 95% CI. Wald-type CI for the weighted results were based on a nonparametric bootstrap, where we resampled the observed data 200 times with replacement and took the standard deviation of the 200 risk differences (or log risk ratios) as the estimated standard error.38 We repeated the analysis steps for the individual components of the primary composite outcome. Finally, we report results from a complete case analysis excluding ZAPPS participants who had missing data (eTables 2 and 4) and constructed a directed acyclic graph showing the hypothesized structure of the association between IPOP participation and the composite outcome (eFigure 1). Analyses were performed using SAS software, version 9.4 (Cary, NC).

RESULTS

Compared to 791 participants in IPOP, women enrolled in ZAPPS (N=1450) were younger; had lower BMI; were more likely to be nulliparous; more likely to have prior stillbirth, short cervix, and hypertension at screening; and presented earlier in pregnancy (Table 1). As a criterion for inclusion in the trial, all IPOP participants were confirmed HIV-infected at screening compared to 24% of participants in ZAPPS. Among participants with HIV in both studies, undetectable viral load, ART timing, and primary ART regimen were similar (95% tenofovir difumarate, emtricitabine/lamivudine, efavirenz), while 62% had initiated ART prior to conception in ZAPPS compared to 73% in IPOP. Syphilis seropositivity in ZAPPS was 6% compared to 15% in IPOP. ZAPPS participants had higher socioeconomic status markers compared to IPOP participants, while the prevalence of alcohol and smoking was low in both cohorts (Table 1).

Table 1.

Baseline distribution of covariates in the IPOP and ZAPPS cohorts

IPOP, N=791 ZAPPS, N=1450 ZAPPS, Restricteda, N=216 ZAPPS, Restricteda, Standardizedb, N=216
Maternal age, yrs – median (IQR) 29 (25, 33) 27 (23, 31) 29 (24, 33) 29 (24, 33)
BMI, kg/m2 – median (IQR) 25 (23, 29) 24 (21, 27) 24 (20, 27) 24 (21, 27)
Parous – n (%) 80% (634) 68% (992) 76% (165) 79% (170)
 Prior stillbirth – n (%) 2% (15) 14% (139) 6% (10) 3% (4)
 Prior PTB – n (%) 0% (0) 41% (411) 0% (0) 0% (0)
Gestational age at screening, wks – median (IQR) 18 (15, 21) 16 (13, 18) 17 (15, 20) 18 (15, 21)
Short cervix – n (%) 1% (4) 4% (59) 2% (4) 0% (1)
Twin gestation – n (%) 0% (0) 3% (38) 0% (0) 0% (0)
HIV-infected – n (%) 100% (791) 24% (351) 100% (216) 100% (216)
Preconceptional ART – n (%) 73% (576) 62% (902) 61% (132) 63% (137)
Viral load undetectable – n (%) 54% (429) 55% (795) 53% (115) 55% (120)
Viral load among detectable (copies/mL) – median (IQR) 1165 (186, 21800) 7243 (1057, 24559) 7550 (1032, 25238) 8037 (853, 24186)
ART regimen: TDF+XTC+EFV – n (%) 95% (751) 95% (1,370) 95% (205) 95% (205)
Syphilis – n (%) 15% (119) 6% (83) 11% (24) 16% (35)
Bacteriuria – n (%) 12% (93) 4% (55) 1% (2) 1% (2)
Hypertension at screening – n (%) 1% (10) 4% (55) 5% (10) 1% (3)
Hemoglobin < 11g/dL – n (%) 10% (83) 16% (231) 24% (51) 25% (53)
Did not complete primary – n (%) 36% (285) 24% (345) 31% (66) 32% (68)
Neither married nor cohabitating – n (%) 16% (125) 16% (238) 21% (45) 19% (42)
Electricity in household – n (%) 91% (717) 91% (1,314) 91% (197) 92% (199)
Flush/pour toilet facility in household – n (%) 46% (367) 53% (769) 48% (105) 48% (103)
Electricity used for cooking – n (%) 33% (263) 70% (1,014) 73% (158) 75% (162)
Alcohol use during pregnancy – n (%) 10% (76) 9% (128) 14% (29) 11% (24)
Smoking during pregnancy – n (%) 0% (0) 1% (9) 1% (3) 1% (2)
Open in a new tab

Variables with missing data in the ZAPPS cohort were imputed and, as a result, some columns may sum to more than the total N due to rounding of estimated proportions.

BMI, body mass index; PTB, preterm birth; ART, antiretroviral therapy; TDF, tenofovir difumarate; XTC, emtricitabine or lamivudine; EFV, efavirenz

a

Restricted to IPOP eligibility criteria: HIV-infected, single gestation, no prior preterm birth; 38 of the initial 1450 participants were excluded due to twin pregnancy; 398 of the remaining 1412 were excluded due to prior preterm birth; and 798 of the remaining 1014 were excluded because they were not HIV-infected.

b

Standardized, using inverse probability weights, to the distribution of the following covariates in the IPOP population: age (restricted quadratic splines), parous, prior stillbirth, gestational age at screening (restricted quadratic splines), short cervix, syphilis, hypertension, and a product term for the linear component of age and hypertension; weighted N=793

Restricting the ZAPPS cohort to IPOP eligibility criteria resulted in the exclusion of 38 (3%) of the initial 1450 participants due to twin gestation, 398 (28%) of the remaining 1412 due to prior preterm birth, and 798 (79%) of the remaining 1014 because they were not HIV seropositive (Figure 1). As a result, the distributions of some baseline characteristics in the restricted ZAPPS cohort were more closely aligned with IPOP (e.g., maternal age, nulliparity, gestational age at screening, syphilis) while others diverged (e.g., marital status, electricity, alcohol use, smoking). Standardization of the restricted ZAPPS cohort to distributions of key baseline covariates in IPOP (i.e., age, nulliparity, prior stillbirth, gestational age at screening, short cervix, syphilis, hypertension at screening, and a product term for the linear component of age and syphilis) further improved the balance, with all covariates included in the model for estimating standardization weights bearing absolute standardized differences less than or equal to 9% in ZAPPS compared to IPOP (Table 1).

Figure 1.

Figure 1.

Open in a new tab

Flowchart of participants in IPOP and ZAPPS included in analysis.

Overall, an estimated 18% (n/N: 254/1450) of ZAPPS participants experienced the primary composite outcome compared to 9% in IPOP (69/791), yielding a RR of 2.0 (95% CI: 1.6, 2.6), and RD of 8.8 (95% CI: 5.9, 11) (Table 2). After restricting the ZAPPS cohort to IPOP eligibility criteria, the estimated risk of composite outcome in ZAPPS lowered to 14% (31/216), the RR for the comparison to IPOP was 1.6 (95% CI: 1.1, 2.5) and RD was 5.6 (95% CI: 0.10, 11). Finally, after standardizing the restricted ZAPPS cohort to the distribution of key baseline covariates in the IPOP population, the estimated risk of composite outcome in ZAPPS was 14% (weighted n/N: 109/793), yielding standardized RR for the comparison to IPOP of 1.6 (95% CI: 1.0, 2.4) and RD of 5.0 (95% CI: −0.12, 10), respectively. (Figure 2).

Table 2.

Risk of composite outcome and its components in the IPOP and ZAPPS cohorts.

IPOP, N=791 ZAPPS, N=1450 ZAPPS, Restricteda, N=216 ZAPPS, Restricteda, Standardizedb, N=216
Live birth before 37 gestational weeks or stillbirth at any gestational age
Events, n 69 254 31 30
Risk, % (95% CI) 9 (7, 11) 18 (15, 20) 14 (10, 19) 14 (9, 18)
RR (95% CI) 1.0 2.0 (1.6, 2.6) 1.6 (1.1, 2.5) 1.6 (1.0, 2.4)
RD (95% CI) 0.00 8.8 (5.9, 11) 5.6 (0.10, 11) 5.0 (−0.12, 10)
Live birth before 37 gestational weeks
Events, n 50 177 20 18
Risk, % (95% CI) 6 (5, 8) 12 (10, 14) 9 (5, 13) 8 (4, 13)
RR (95% CI) 1.0 1.9 (1.4, 2.6) 1.4 (0.83, 2.4) 1.3 (0.76, 2.3)
RD (95% CI) 0.00 5.8 (3.3, 8.3) 2.7 (−1.8, 7.3) 2.0 (−3.1, 7.2)
Stillbirth at any gestational age
Events, n 19 78 11 12
Risk, % (95% CI) 2 (1, 4) 5 (4, 7) 5 (2, 9) 5 (2, 8)
RR (95% CI) 1.0 2.2 (1.3, 3.7) 2.2 (0.98, 4.8) 2.2 (1.0, 4.8)
RD (95% CI) 0.00 3.0 (1.2, 4.7) 2.8 (−0.73, 6.4) 3.0 (−0.28, 6.3)
Open in a new tab

Variables with missing data in the ZAPPS cohort were imputed and results were averaged across 50 imputations; as a result, some sample size values (n) may sum to more than the total N due to rounding of estimated proportions.

CI, confidence interval; RR, risk ratio; RD, risk difference

a

Restricted to IPOP eligibility criteria: HIV-infected, single gestation, no prior preterm birth

b

Standardized, using inverse probability weights, to the distribution of the following covariates in the IPOP population: Age (restricted quadratic splines), parous, prior stillbirth, gestational age at screening (restricted quadratic splines), short cervix, syphilis, hypertension, and a product term for the linear component of age and hypertension; weighted N=793

Figure 2.

Figure 2.

Open in a new tab

Frequency of composite outcome and its components in the IPOP (N=791) trial, ZAPPS full cohort (N=1450), ZAPPS restricted cohort (N=216), and ZAPPS restricted and standardized cohort (N=216).

Before restriction, an estimated 12% of ZAPPS participants (177/1450) and 6% of IPOP participants (50/791) experienced live birth before 37 weeks (RR 1.9, 95% CI: 1.4, 2.6; RD: 5.8, 95% CI 3.3, 8.3), while 5% (78/1450) experienced stillbirth in ZAPPS compared to 2% (19/791) in IPOP (RR: 2.2; 95% CI: 1.3, 3.7; RD: 3.0; 95% CI: 1.2, 4.7) (Table 2). After restriction and standardization, the estimated risk of live birth before 37 weeks in ZAPPS participants reduced to 8% (RR: 1.3, 95% CI 0.76, 2.3; RD: 2.0; 95% CI: −3.1, 7.2), and the estimated risk of stillbirth remained at approximately 5% (RR: 2.2, 95% CI 1.0, 4.8; RD: 3.0; 95% CI: −0.28, 6.3).

In complete case analysis, the overall risk of the composite outcome in ZAPPS was lower, and the reduction in risk that resulted from standardization was not as pronounced as observed in our analysis on multiply imputed data (eTable 4).

DISCUSSION

In the recently concluded IPOP trial, despite no effect of the intervention, there was a low overall risk of preterm birth and stillbirth among participants with HIV.39 In this secondary analysis, we compared birth outcomes among participants in the prospective ZAPPS antenatal cohort to those in the IPOP trial (treatment groups combined) by restricting ZAPPS by the same eligibility criteria used in IPOP, and then standardizing the ZAPPS cohort to the distribution of key baseline characteristics in the IPOP trial using inverse probability weighting. We found that the initial difference in risk of preterm birth or stillbirth between ZAPPS and IPOP participants was reduced but remained nearly 60% higher in ZAPPS after applying the methods to adjust for confounding. Analysis of the individual components of the composite outcome revealed that the overall absolute difference in risk between ZAPPS and IPOP was initially higher for the outcome of preterm livebirth than for the outcome of stillbirth, but restriction and standardization reduced this difference more substantially for preterm livebirth than it did for stillbirth. While eligibility criteria in IPOP that excluded some high-risk women (e.g., prior spontaneous preterm birth and twins) contributed to the differential risk of preterm or stillbirth between the studies, the observed risk reduction associated with participation in IPOP after restriction and standardization of the ZAPPS cohort remains unexplained.

The IPOP trial was conducted at the same Lusaka hospital and nearby primary care clinic as the ZAPPS cohort. Participants were drawn from the same population and attended by the same clinical research staff. Tenofovir, emtricitabine/lamivudine, efavirenz was the predominant regimen in both studies. The study protocols were deliberately aligned from the outset, including procedures for screening and enrollment, ultrasound evaluation, point-of-care prenatal testing, specimen collection, and phenotyping of adverse outcomes.26,28 After restricting the ZAPPS cohort to those women who would have been eligible for IPOP participation and accounting for remaining differences in baseline covariates by standardization to the IPOP distribution using inverse probability weighting, the absolute risk of the composite outcome remained 5% higher and the relative risk 60% higher. Once these adjustments are made, only a few explanations for the observed differences seem likely. These include residual imbalance of risk factors, better retention and stricter adherence to protocols in a trial setting compared to a cohort, salutary effects of more frequent contact with providers, and differential reimbursement for participation in the studies. Although restriction and standardization narrowed the difference in the composite outcome between IPOP and ZAPPS overall, it remains unclear why the difference narrowed more with preterm live birth (RD from 5.83 to 2.05) than stillbirth (RD from 2.98 to 2.22) but may signal that IPOP trial procedures reduced stillbirth risk independent of participant characteristics.

Previous studies have reported that women facing poverty, psychosocial stress, poor nutrition, and substance use during pregnancy have a higher risk of adverse birth outcomes.40–45 Other studies have reported that both conditional46 and unconditional47 cash transfers can have a positive impact on the use of health services and certain health outcomes in low- and middle-income countries. Recent public programs have implemented unconditional cash transfer for pregnant Black and Pacific Islander women in San Francisco and for low-income women in Manitoba under the assumption that the benefit of cash on promoting financial stability and alleviating risk is independent of any conditionality imposed.48,49 Whereas no similar public programs have been reported from sub-Saharan Africa, both of our studies provided monetary reimbursement for time and transport to all women who attended study visits.26,28 Women in ZAPPS attended monthly antenatal visits and received approximately $4/month, while participants in IPOP, who traveled to the clinic for weekly injections, received approximately $14/month. The local IRB approved reimbursement figures to cover transportation expenses, but participants may have had residual cash if their costs were less than the reimbursement. In Zambia, where extreme poverty is endemic,50 a supplement of $14 per month can certainly benefit a family living in poverty. Although we cannot conclude that the additional cash provided in IPOP equated with a reduced risk of adverse outcomes, future studies should evaluate whether income supplementation could improve maternal stress indices and reduce adverse birth outcomes in Zambia.

Most antenatal care worldwide is provided through private one-on-one interactions between a provider and a patient. Group antenatal care models involving frequent visits, participatory health assessments, group learning, and peer support have been implemented successfully in hundreds of settings in the US and globally.51,52 Two meta-analyses have demonstrated that women randomly assigned to group care had lower risk of preterm birth (RR 0.71; 95% CI 0.52, 0.96)53 or low birthweight (RR 0.81; 95% CI 0.69, 0.96)54 compared to women who received individualized care, although data from randomized trials are limited.55 More recently, a large, matched cohort study reported that group care was associated with nearly 70% risk reductions of both preterm birth and low birthweight,56 particularly among women attending five or more visits compared to women who attended individualized care. In low- and middle-income countries, several studies have demonstrated improved antenatal attendance,57,58 facility delivery,57,58 postpartum contraceptive use,59 and breastfeeding59 among patients in group care, and randomized trials are underway in Africa to investigate the effect on birth outcomes.60,61 Whereas IPOP did not provide care in a group setting, participants had frequent contact with midwives and inadvertent cohorts resulted from weekly injection and monthly antenatal care appointments that occurred on the same day of the week throughout pregnancy.26 From anecdotal reports, women in IPOP gathered before and after their visits with the midwife, underwent health counseling together, and participated in chilimba, a Zambian practice in which a rotating member of the group collects small donations from the others. In contrast, ZAPPS participants were provided a window within which to attend their monthly visits such that they did not consistently encounter the same group. Together, more frequent contact with care providers and impromptu group activities in IPOP may have supported collective retention and agency during pregnancy and deserve further study as possible interventions for risk reduction in pregnant women facing social marginalization and poverty.

We acknowledge limitations to this analysis. While we attempted to exhaustively identify and account for imbalance of risk factors between participants in ZAPPS and IPOP, it is possible that there remains residual confounding. We were unable to fully address all exclusion criteria in IPOP, namely presence of cerclage, signs of threatened abortion or preterm labor, planned preterm delivery, and major uterine or fetal anomaly due to incomplete documentation in ZAPPS, although the frequency of these were exceedingly rare reasons for exclusion among women screened for IPOP.27 We chose not to standardize on all baseline characteristics because some were not associated with the outcome (e.g., anemia, preconceptional ART, viral load, marital status, electricity), had comparable distribution in IPOP compared to the restricted ZAPPS cohort (e.g., ART regimen), or their prevalence was rare (e.g., smoking, bacteriuria). Additionally, CD4 testing is not part of standard practice in Zambia and thus we were unable to compare immune status between the studies; it is possible that ZAPPS participants with HIV had more advanced disease despite viral suppression being comparable. Given sample size limitations, we were unable to perform analyses of subgroups by viral suppression, ART initiation timing, or duration of HIV disease. We also note that the ZAPPS study was conducted in the two years preceding the IPOP trial such that we cannot exclude the possibility that the reduced risk observed could be attributed to a temporal change in the background risk, but such a rapid and strong reduction in preterm birth would be at odds with worldwide experience. Similarly, while seasonality of adverse birth outcomes has been described in our region,62 it is unlikely to be a driving factor in the residual difference found in our cohorts since both enrolled women over 2-year periods. Finally, we note that our study procedures (e.g., enrollment before 24 gestational weeks, universal ultrasound, strong adherence and retention support, careful clinical classification of outcomes) in addition to the likelihood of self-selection of women with poor pregnancy history into the cohorts, represent important characteristics that limit both generalizability of our results to the broader population and comparability to other cohorts in the region.

In summary, despite restricting and standardizing ZAPPS cohort participants to the characteristics of IPOP trial participants, the risk of preterm birth and stillbirth remained lower among women in the trial. A differentially elevated risk of stillbirth compared to preterm birth may be partially related to stricter adherence to protocols for standardized obstetrical treatment and better retention at antenatal visits in the trial. Potential added benefits of income supplementation, more frequent visits, and group-based care in this high-risk population warrant further investigation.

Supplementary Material

Supplemental Tables

Acknowledgments:

The authors wish to acknowledge the invaluable contributions to this research from our participants and their families, the staff of UNC Global Projects Zambia, the local community advisory boards in Lusaka, and members of our Data and Safety Management Board and Scientific Advisory Committee.

Funding:

This work was supported by the Bill and Melinda Gates Foundation (grants OPP1033514 and OPP1172799 to JSAS) and the US National Institutes of Health (grant R01 HD087119 to JSAS). Additional support was provided by the US National Institutes of Health through the UNC Center for AIDS Research (grant P30 AI50410). Trainee support has been provided through the US National Institutes of Health for JTP (grants T32 HD075731 and K01 TW010857). The NC TraCS Institute (grant UL1TR002489) supported the development and maintenance of the REDCap system used for participant randomization in the IPOP trial. The conclusions and opinions expressed in this article are those of the authors and do not necessarily reflect those of the funders. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Abbreviations:

17P

17-alpha hydroxyprogesterone caproate

ART

Antiretroviral therapy

BMI

Body mass index

CI

Confidence interval

HIV

Human immunodeficiency virus

IPOP

Improving Pregnancy Outcomes with Progesterone

RD

Risk difference

RR

Risk ratio

ZAPPS

Zambian Preterm Birth Prevention Study

Footnotes

Conflicts of Interest: The authors have no conflicts of interest to declare.

Data Sharing:

Individual participant data that underlie the IPOP clinical trial results reported in this article, after de-identification, will be made available beginning 3 months and ending 5 years following article publication for researchers who provide a methodologically sound proposal to achieve aims in the approved proposal. Proposals should be directed to the corresponding author; to gain access, data requestors will need to sign a data access agreement. ZAPPS cohort data is freely available at Open Science Framework https://doi.org/10.17605/OSF.IO/WT6Q8. The SAS code used to generate this report is available upon request to the authors.

REFERENCES

  • 1.Chawanpaiboon S, Vogel JP, Moller AB, et al. Global, regional, and national estimates of levels of preterm birth in 2014: a systematic review and modelling analysis. Lancet Glob Health. 2019;7(1):e37–e46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lawn JE, Blencowe H, Waiswa P, et al. Stillbirths: rates, risk factors, and acceleration towards 2030. Lancet. 2016;387(10018):587–603. [DOI] [PubMed] [Google Scholar]
  • 3.Wedi CO, Kirtley S, Hopewell S, Corrigan R, Kennedy SH, Hemelaar J. Perinatal outcomes associated with maternal HIV infection: a systematic review and meta-analysis. Lancet HIV. 2016;3(1):e33–48. [DOI] [PubMed] [Google Scholar]
  • 4.Price JT, Vwalika B, Edwards JK, et al. Maternal HIV infection and spontaneous versus provider-initiated preterm birth in an urban Zambian cohort. J Acquir Immune Defic Syndr. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Uthman OA, Nachega JB, Anderson J, et al. Timing of initiation of antiretroviral therapy and adverse pregnancy outcomes: a systematic review and meta-analysis. Lancet HIV. 2017;4(1):e21–e30. [DOI] [PubMed] [Google Scholar]
  • 6.Chen JY, Ribaudo HJ, Souda S, et al. Highly active antiretroviral therapy and adverse birth outcomes among HIV-infected women in Botswana. J Infect Dis. 2012;206(11):1695–1705. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Zash R, Jacobson DL, Diseko M, et al. Comparative safety of dolutegravir-based or efavirenz-based antiretroviral treatment started during pregnancy in Botswana: an observational study. Lancet Glob Health. 2018;6(7):e804–e810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Malaba TR, Newell ML, Madlala H, Perez A, Gray C, Myer L. Methods of gestational age assessment influence the observed association between antiretroviral therapy exposure, preterm delivery, and small-for-gestational age infants: a prospective study in Cape Town, South Africa. Ann Epidemiol. 2018;28(12):893–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.van der Merwe K, Hoffman R, Black V, Chersich M, Coovadia A, Rees H. Birth outcomes in South African women receiving highly active antiretroviral therapy: a retrospective observational study. J Int AIDS Soc. 2011;14:42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Li N, Sando MM, Spiegelman D, et al. Antiretroviral Therapy in Relation to Birth Outcomes among HIV-infected Women: A Cohort Study. J Infect Dis. 2015. [DOI] [PubMed] [Google Scholar]
  • 11.Zash R, Souda S, Chen JY, et al. Reassuring Birth Outcomes With Tenofovir/Emtricitabine/Efavirenz Used for Prevention of Mother-to-Child Transmission of HIV in Botswana. J Acquir Immune Defic Syndr. 2016;71(4):428–436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Price JT, Vwalika B, Rittenhouse KJ, et al. Adverse birth outcomes and their clinical phenotypes in an urban Zambian cohort. Gates Open Res. 2019;3:1533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Malaba TR, Phillips T, Le Roux S, et al. Antiretroviral therapy use during pregnancy and adverse birth outcomes in South African women. Int J Epidemiol. 2017;46(5):1678–1689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Stringer EM, Vwalika B, Killam WP, et al. Determinants of stillbirth in Zambia. Obstet Gynecol. 2011;117(5):1151–1159. [DOI] [PubMed] [Google Scholar]
  • 15.Kupka R, Kassaye T, Saathoff E, Hertzmark E, Msamanga GI, Fawzi WW. Predictors of stillbirth among HIV-infected Tanzanian women. Acta Obstet Gynecol Scand. 2009;88(5):584–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Medley N, Vogel JP, Care A, Alfirevic Z. Interventions during pregnancy to prevent preterm birth: an overview of Cochrane systematic reviews. Cochrane Database Syst Rev. 2018;11:CD012505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Dodd J, Jones L, Flenady V, Cincotta R, Crowther C. Prenatal administration of progesterone for preventing preterm birth in women considered to be at risk of preterm birth. Cochrane Database of Systematic Reviews. 2013(7). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Romero R, Conde-Agudelo A, Da Fonseca E, et al. Vaginal progesterone for preventing preterm birth and adverse perinatal outcomes in singleton gestations with a short cervix: a meta-analysis of individual patient data. Am J Obstet Gynecol. 2018;218(2):161–180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Norman JE, Marlow N, Messow CM, et al. Vaginal progesterone prophylaxis for preterm birth (the OPPTIMUM study): a multicentre, randomised, double-blind trial. Lancet. 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Blackwell SC, Gyamfi-Bannerman C, Biggio JR Jr., et al. 17-OHPC to Prevent Recurrent Preterm Birth in Singleton Gestations (PROLONG Study): A Multicenter, International, Randomized Double-Blind Trial. Am J Perinatol. 2019. [DOI] [PubMed] [Google Scholar]
  • 21.Meis PJ, Klebanoff M, Thom E, et al. Prevention of recurrent preterm delivery by 17 alpha-hydroxyprogesterone caproate. N Engl J Med. 2003;348(24):2379–2385. [DOI] [PubMed] [Google Scholar]
  • 22.Hileman CO, Funderburg NT. Inflammation, Immune Activation, and Antiretroviral Therapy in HIV. Curr HIV/AIDS Rep. 2017;14(3):93–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Belec L, Gherardi R, Payan C, et al. Proinflammatory cytokine expression in cervicovaginal secretions of normal and HIV-infected women. Cytokine. 1995;7(6):568–574. [DOI] [PubMed] [Google Scholar]
  • 24.Rinaldi SF, Hutchinson JL, Rossi AG, Norman JE. Anti-inflammatory mediators as physiological and pharmacological regulators of parturition. Expert Rev Clin Immunol. 2011;7(5):675–696. [DOI] [PubMed] [Google Scholar]
  • 25.Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. The Lancet. 2008;371(9606):75–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Price JT, Vwalika B, Freeman BL, et al. Intramuscular 17-hydroxyprogesterone caproate to prevent preterm birth among HIV-infected women in Zambia: study protocol of the IPOP randomized trial. BMC Pregnancy Childbirth. 2019;19(1):81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Price JT, Vwalika B, Freeman BL, et al. Weekly 17 alpha-hydroxyprogesterone caproate to prevent preterm birth among women living with HIV: a randomised, double-blind, placebo-controlled trial. Lancet HIV. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Castillo MC, Fuseini NM, Rittenhouse K, et al. The Zambian Preterm Birth Prevention Study (ZAPPS): Cohort characteristics at enrollment. Gates Open Res. 2018;2:25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Papageorghiou AT, Kemp B, Stones W, et al. Ultrasound-based gestational-age estimation in late pregnancy. Ultrasound Obstet Gynecol. 2016;48(6):719–726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Papageorghiou AT, Kennedy SH, Salomon LJ, et al. International standards for early fetal size and pregnancy dating based on ultrasound measurement of crown-rump length in the first trimester of pregnancy. Ultrasound Obstet Gynecol. 2014;44(6):641–648. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Perkins NJ, Cole SR, Harel O, et al. Principled Approaches to Missing Data in Epidemiologic Studies. Am J Epidemiol. 2018;187(3):568–575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Harel O, Mitchell EM, Perkins NJ, et al. Multiple Imputation for Incomplete Data in Epidemiologic Studies. Am J Epidemiol. 2018;187(3):576–584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Sato T, Matsuyama Y. Marginal structural models as a tool for standardization. Epidemiology. 2003;14(6):680–686. [DOI] [PubMed] [Google Scholar]
  • 34.Howe CJ, Cole SR, Westreich DJ, Greenland S, Napravnik S, Eron JJ Jr. Splines for trend analysis and continuous confounder control. Epidemiology. 2011;22(6):874–875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate behavioral research. 2011;46(3):399–424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Morgan SL, Todd JJ. 6. A Diagnostic Routine for the Detection of Consequential Heterogeneity of Causal Effects. Sociological Methodology. 2008;38(1):231–282. [Google Scholar]
  • 37.Cole SR, Hernán MA. Constructing inverse probability weights for marginal structural models. American journal of epidemiology. 2008;168(6):656–664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Efron B, Gong G. A Leisurely Look at the Bootstrap, the Jackknife, and Cross-Validation. The American Statistician. 1983;37(1):36–48. [Google Scholar]
  • 39.Price JT, Vwalika B, Freeman BL, et al. Weekly 17 alpha-hydroxyprogesterone caproate to prevent preterm birth among women living with HIV: a randomised, double-masked, placebo-controlled trial. Lancet HIV. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Weber KA, Carmichael SL, Yang W, Tinker SC, Shaw GM, National Birth Defects Prevention S. Periconceptional stressors and social support and risk for adverse birth outcomes. BMC Pregnancy Childbirth. 2020;20(1):487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Staneva A, Bogossian F, Pritchard M, Wittkowski A. The effects of maternal depression, anxiety, and perceived stress during pregnancy on preterm birth: A systematic review. Women Birth. 2015;28(3):179–193. [DOI] [PubMed] [Google Scholar]
  • 42.Dadi AF, Miller ER, Bisetegn TA, Mwanri L. Global burden of antenatal depression and its association with adverse birth outcomes: an umbrella review. BMC Public Health. 2020;20(1):173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Fekadu Dadi A, Miller ER, Mwanri L. Antenatal depression and its association with adverse birth outcomes in low and middle-income countries: A systematic review and meta-analysis. PLOS ONE. 2020;15(1):e0227323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Lima SAM, El Dib RP, Rodrigues MRK, et al. Is the risk of low birth weight or preterm labor greater when maternal stress is experienced during pregnancy? A systematic review and meta-analysis of cohort studies. PLoS One. 2018;13(7):e0200594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Hobel C, Culhane J. Role of psychosocial and nutritional stress on poor pregnancy outcome. J Nutr. 2003;133(5 Suppl 2):1709S–1717S. [DOI] [PubMed] [Google Scholar]
  • 46.Lagarde M, Haines A, Palmer N. The impact of conditional cash transfers on health outcomes and use of health services in low and middle income countries. Cochrane Database Syst Rev. 2009(4):CD008137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Pega F, Liu SY, Walter S, Pabayo R, Saith R, Lhachimi SK. Unconditional cash transfers for reducing poverty and vulnerabilities: effect on use of health services and health outcomes in low- and middle-income countries. Cochrane Database Syst Rev. 2017;11:CD011135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Brownell M, Nickel NC, Chartier M, et al. An Unconditional Prenatal Income Supplement Reduces Population Inequities In Birth Outcomes. Health Aff (Millwood). 2018;37(3):447–455. [DOI] [PubMed] [Google Scholar]
  • 49.Brownell MD, Chartier MJ, Nickel NC, et al. Unconditional Prenatal Income Supplement and Birth Outcomes. Pediatrics. 2016;137(6). [DOI] [PubMed] [Google Scholar]
  • 50.Bank World, Development Research Group. PovcalNet: an online analysis tool for global poverty monitoring. Accessed 28 September 2021 at: http://iresearch.worldbank.org/PovcalNet/home.aspx.
  • 51.Rising SS. Centering pregnancy. An interdisciplinary model of empowerment. J Nurse Midwifery. 1998;43(1):46–54. [DOI] [PubMed] [Google Scholar]
  • 52.Rising SS, Kennedy HP, Klima CS. Redesigning prenatal care through CenteringPregnancy. J Midwifery Womens Health. 2004;49(5):398–404. [DOI] [PubMed] [Google Scholar]
  • 53.Ruiz-Mirazo E, Lopez-Yarto M, McDonald SD. Group prenatal care versus individual prenatal care: a systematic review and meta-analyses. J Obstet Gynaecol Can. 2012;34(3):223–229. [DOI] [PubMed] [Google Scholar]
  • 54.Carter EB, Temming LA, Akin J, et al. Group Prenatal Care Compared With Traditional Prenatal Care: A Systematic Review and Meta-analysis. Obstet Gynecol. 2016;128(3):551–561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Catling CJ, Medley N, Foureur M, et al. Group versus conventional antenatal care for women. Cochrane Database Syst Rev. 2015(2):CD007622. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Cunningham SD, Lewis JB, Shebl FM, et al. Group Prenatal Care Reduces Risk of Preterm Birth and Low Birth Weight: A Matched Cohort Study. J Womens Health (Larchmt). 2019;28(1):17–22. [DOI] [PubMed] [Google Scholar]
  • 57.Grenier L, Suhowatsky S, Kabue MM, et al. Impact of group antenatal care (G-ANC) versus individual antenatal care (ANC) on quality of care, ANC attendance and facility-based delivery: A pragmatic cluster-randomized controlled trial in Kenya and Nigeria. PLoS One. 2019;14(10):e0222177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Eluwa GI, Adebajo SB, Torpey K, et al. The effects of centering pregnancy on maternal and fetal outcomes in northern Nigeria; a prospective cohort analysis. BMC Pregnancy Childbirth. 2018;18(1):158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Lori JR, Chuey M, Munro-Kramer ML, Ofosu-Darkwah H, Adanu RMK. Increasing postpartum family planning uptake through group antenatal care: a longitudinal prospective cohort design. Reprod Health. 2018;15(1):208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Musange SF, Butrick E, Lundeen T, et al. Group antenatal care versus standard antenatal care and effect on mean gestational age at birth in Rwanda: protocol for a cluster randomized controlled trial. Gates Open Res. 2019;3:1548. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Kabue MM, Grenier L, Suhowatsky S, et al. Group versus individual antenatal and first year postpartum care: Study protocol for a multi-country cluster randomized controlled trial in Kenya and Nigeria. Gates Open Res. 2018;2:56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Caniglia EC, Abrams J, Diseko M, et al. Seasonality of adverse birth outcomes in women with and without HIV in a representative birth outcomes surveillance study in Botswana. BMJ Open. 2021;11(9):e045882. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Tables
NIHMS1834875-supplement-Supplemental_Tables.pdf (216.2KB, pdf)

Data Availability Statement

Individual participant data that underlie the IPOP clinical trial results reported in this article, after de-identification, will be made available beginning 3 months and ending 5 years following article publication for researchers who provide a methodologically sound proposal to achieve aims in the approved proposal. Proposals should be directed to the corresponding author; to gain access, data requestors will need to sign a data access agreement. ZAPPS cohort data is freely available at Open Science Framework https://doi.org/10.17605/OSF.IO/WT6Q8. The SAS code used to generate this report is available upon request to the authors.

RESOURCES