Disentangling patterns of community malaria transmission and burden using malaria prevalence among pregnant women attending antenatal care: a modelling study
File(s)
Published online version
Author(s)
Type
Journal Article
Abstract
Background
Malaria prevalence measured among pregnant women at the first antenatal care (ANC1) visit provides longitudinal estimates of malaria burden in pregnancy and correlates well with cross-sectional community prevalence, but additional analysis is required to estimate community incidence. We aimed to test whether ANC1-based malaria prevalence can, via an open-source, mechanistic, model-based framework, recover seasonal patterns of clinical incidence suitable for subnational programmatic decision making.
Methods
We conducted a modelling study using monthly ANC1 malaria prevalence data from six previously published studies of malaria in pregnancy in six sub-Saharan African countries between May, 2010, and August, 2014. An extended, validated, age-structured malaria transmission model was fitted to monthly ANC1 malaria prevalence using particle Markov chain Monte Carlo (pMCMC) to infer monthly clinical incidence and seasonality metrics. Agreement between model-derived incidence and independently observed time series was assessed using the Markham Seasonality Index (MSI) and peak timing with concordance correlation coefficients (CCCs) and 95% CIs.
Findings
Across the six intermittent screening and treatment in pregnancy (ISTp) datasets, total ANC1 sample sizes and positivity were Ghana, 622 (47·9%) of 1298; Burkina Faso, 592 (41·9%) of 1413; Mali, 284 (21·7%) of 1308; The Gambia, 105 (8·8%) of 1194; Kenya, 323 (21·1%) of 1528; and Malawi, 291 (15·9%) of 1825. Strong agreement was observed between model-derived incidence and independent cohort data for MSI (CCC 0·82 [95% CI 0·31–0·97]) and for peak timing (CCC 0·98 [95% CI 0·87–1·00]).
Interpretation
A mechanistic pMCMC framework applied to routine ANC1 data can recover clinically relevant seasonality in incidence for the broader community, enabling subnational timing of seasonal interventions (such as seasonal malaria chemoprevention). These capabilities are especially valuable when high-quality case surveillance is scarce and household surveys are underfunded. Our work operationalises WHO guidance, highlighting ANC1 as an opportunity to strengthen malaria surveillance.
Funding
Bill & Melinda Gates Foundation and MRC Centre for Global Infectious Disease Analysis.
Malaria prevalence measured among pregnant women at the first antenatal care (ANC1) visit provides longitudinal estimates of malaria burden in pregnancy and correlates well with cross-sectional community prevalence, but additional analysis is required to estimate community incidence. We aimed to test whether ANC1-based malaria prevalence can, via an open-source, mechanistic, model-based framework, recover seasonal patterns of clinical incidence suitable for subnational programmatic decision making.
Methods
We conducted a modelling study using monthly ANC1 malaria prevalence data from six previously published studies of malaria in pregnancy in six sub-Saharan African countries between May, 2010, and August, 2014. An extended, validated, age-structured malaria transmission model was fitted to monthly ANC1 malaria prevalence using particle Markov chain Monte Carlo (pMCMC) to infer monthly clinical incidence and seasonality metrics. Agreement between model-derived incidence and independently observed time series was assessed using the Markham Seasonality Index (MSI) and peak timing with concordance correlation coefficients (CCCs) and 95% CIs.
Findings
Across the six intermittent screening and treatment in pregnancy (ISTp) datasets, total ANC1 sample sizes and positivity were Ghana, 622 (47·9%) of 1298; Burkina Faso, 592 (41·9%) of 1413; Mali, 284 (21·7%) of 1308; The Gambia, 105 (8·8%) of 1194; Kenya, 323 (21·1%) of 1528; and Malawi, 291 (15·9%) of 1825. Strong agreement was observed between model-derived incidence and independent cohort data for MSI (CCC 0·82 [95% CI 0·31–0·97]) and for peak timing (CCC 0·98 [95% CI 0·87–1·00]).
Interpretation
A mechanistic pMCMC framework applied to routine ANC1 data can recover clinically relevant seasonality in incidence for the broader community, enabling subnational timing of seasonal interventions (such as seasonal malaria chemoprevention). These capabilities are especially valuable when high-quality case surveillance is scarce and household surveys are underfunded. Our work operationalises WHO guidance, highlighting ANC1 as an opportunity to strengthen malaria surveillance.
Funding
Bill & Melinda Gates Foundation and MRC Centre for Global Infectious Disease Analysis.
Date Issued
2026-07-15
Date Acceptance
2026-03-13
Citation
The Lancet Microbe, 2026
ISSN
2666-5247
Publisher
Elsevier
Journal / Book Title
The Lancet Microbe
Copyright Statement
© 2026 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
License URL
Publication Status
Published online
Article Number
101415
Date Publish Online
2026-07-15