The spatial spread of HIV in Malawi: An individual-based mathematical model

Abstract

AbstractThe prevalence of HIV varies greatly between and within countries. We therefore developed a flexible individual-based mathematical model for HIV transmission, that comprises a spatial representation and individual-level determinants. We tested this model by calibrating it to the HIV epidemic in Malawi and exploring whether the heterogeneity in HIV prevalence could be caused without accounting for heterogeneity in behaviour. We ran the model for Malawi between years 1975-2030 with five alternative realizations of the geographical structure and mobility: (I) no geographical structure; 28 administrative districts including (II) only permanent relocations between districts, (III) permanent relocations and between-district casual sexual relationships, or (IV) permanent relocations between districts and to/from abroad and between-district casual sex; and (V) a grid of 10×10km2 cells, with permanent relocations and between-cell casual relationships. We assumed HIV was present in 1975 in the districts with >10% prevalence in 2010. We calibrated the models to national and district-level prevalence estimates.Reaching the national prevalence required all adults to have at least 20 casual sex acts/year until 1990. Models II, III and V reproduced the geographical heterogeneity in prevalence to some extent if between-district relationships were either excluded (Model II) or restricted to minimum (Models III, V). Long-distance casual partnership mixing (Models III-V) mitigated the differences in prevalence substantially; with international migration the differences disappeared completely (Model IV). National prevalence was projected to decrease to 4-5% by 2030. Our model sustained the major differences in HIV prevalence across Malawi, if casual relationships between districts were kept at sufficiently low level. An earlier introduction of HIV into the Southern part of Malawi may thus be one of the explanations to the present heterogeneity in HIV prevalence.Author summaryThe prevalence of HIV varies greatly across the settings, both globally and within countries. The ability of the commonly used compartmental models to account for the geographical structure and individual-level determinants that cause this heterogeneity is limited. In this project, we developed an individual-based simulation framework for modelling HIV transmission in a real setting. We built the model to take into account an unlimited number of individual-level characteristics, and a geographical representation of the setting that can be defined using an arbitrary resolution and distance matrices. We demonstrate the use of this model by simulating the HIV epidemic of Malawi 1975-2030 and exploring whether the observed heterogeneity could be preserved without taking into account any spatial heterogeneity in sexual behaviour. A relatively simple version of the model reproduced the broad-scale differences in HIV prevalence, but the detailed differences will need further investigation.