Back

Relative fertility of HIV-positive women in the ART era: updated estimates from national household survey data

Imai-Eaton, J. W. W.; Glaubius, R.; Mahy, M.; Johnson, L. F.; Stover, J.; Marston, M.

2026-08-12 epidemiology
10.64898/2026.08.11.26360197 medRxiv
Show abstract

Objectives: Estimate fertility rate ratios (FRR) of HIV-positive relative to HIV-negative women in sub-Saharan Africa (SSA) by age group, CD4 stage, ART status, and country. Design: Analysis of nationally representative household surveys with HIV serological testing. Methods: We analysed current pregnancy and births in the past three years by HIV status from 72 nationally-representative household surveys in SSA between 2003 and 2017. Spectrum 2018 estimates for the distribution by CD4 stage and ART status were used to infer fertility of women on ART from changes in fertility of all HIV-positive women as ART coverage increased. We allowed regional differences in the age pattern of relative fertility and estimated country-specific random effects. Results: The ratio of fertility in untreated HIV-positive women with CD4 [≥]500 to HIV-negative women was 1.6 to 1.8 for age 15-19, relatively similar to 10% times lower for age 20-29, and 15-50% lower above age 30. Among age 15-19, each 15-point increase in percent sexually active reduced relative excess fertility by 24%. Fertility decreased with lower untreated CD4 count stages, consistent with previous estimates. Women on ART >6 months had fertility closer to that of HIV-negative women for ages 15-29, but still 25-40% lower above age 30. There was substantial variation across countries. Conclusions: Fertility differences for HIV-positive women compared to HIV-negative women are smaller than previous estimates, but vary substantially across countries. Recent data suggest fertility of women on ART is greater than that of untreated HIV-positive women, but remains lower than HIV-negative women. This conclusion should be reviewed as new evidence becomes available.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.