Indirect Maternal Mortality Due to Malaria: A Systematic Review and Meta-Analysis

Abstract

Background: Malaria is a major contributor to indirect maternal mortality in endemic regions. However, the magnitude of the effect of pregnancy on malaria-associated mortality and the role of malaria infection in maternal mortality remain uncertain. This study aimed to review the evidence and quantify the impact of pregnancy on the risk of malaria-associated death and the impact of malaria infection on the risk of pregnancy-associated mortality. Methods: A systematic review and meta-analysis were conducted using articles identified from PubMed database. A total of 10 studies met the inclusion criteria, with 5 studies contributing to each risk estimate. Common and random-effects models were used to estimate pooled RRs using traditional meta-analyses. We then applied the Meta-regression-Bayesian, Regularized, Trimmed (MR-BRT) model to derive bias-adjusted risk estimates with 95 % uncertainty intervals that account for between-study heterogeneity. In keeping with the Burden of Proof framework, we assigned the 5th quantile risk function as the Burden of Proof Risk Function (BPRF) or estimate of the magnitude of the risk closest to the null that is consistent with the evidence. We also calculated a star rating to summarize the strength of evidence for causal association. Results: Pregnancy was associated with a significantly higher risk of malaria-associated death among individuals with malaria (random-effects RR: 1.58; 95% CI: 1.11–2.25). The MR-BRT model-based RR estimate was 1.31 (95% UI: 1.08, 1.60) and (95% UI: 0.79, 2.19) for without and with between-study heterogeneity (γ), respectively, indicating that 36.7% of malaria-associated mortality in pregnant individuals is attributable to pregnancy's exacerbating effect. However, the BPRF (0.86) and ROS (-0.0770) suggested a weak strength of evidence for the causal relationship between pregnancy and malaria-associated mortality (star rating: 1). Malaria infection was also associated with a significantly increased risk of pregnancy-associated mortality, with a MR-BRT model-based RR of 8.25 (95% CI: 0.89, 75.68), which implies that 87.9 % of deaths in malaria-infected pregnant individuals are attributable to malaria infection. Similar to the pregnancy model, the BPRF estimate (1.28) and ROS score (0.1001) suggest weak evidence supporting the causal relationship between malaria and pregnancy-associated mortality (2-star rating). Conclusions: The findings suggest that pregnancy increases vulnerability to malaria-associated mortality and that malaria infection elevates the risk of pregnancy-associated death. Despite the weak strength of evidence for pregnancy's causal effect on malaria mortality, possibly due to heterogeneity of studies, the effect size estimates and attributable percentages for the exposures derived from this study could be employed in disease burden models to estimate country-level indirect maternal mortality attributable to malaria. Work in this area can inform policy and intervention strategies.