A Malaria Transmission Model Predicts Holoendemic, Hyperendemic, and Hypoendemic Transmission Patterns Under Varied Seasonal Vector Dynamics

Abstract

A model is developed of malaria (Plasmodium falciparum) transmission in vector (Anopheles gambiae) and human populations that include the capacity for both clinical and parasite suppressing immunity. This model is coupled with a population model for Anopheles gambiae that varies seasonal with temperature and larval habitat availability. At steady state, the model clearly distinguishes uns hypoendemic transmission patterns from stable hyperendemic and holoendemic patterns of transmission. The model further distinguishes hyperendemic from holoendemic disease based on seasonality of infection. For hyperendemic and holoendemic transmission, the model produces the relationship between entomological inoculation rate and disease prevalence observed in the field. It further produces expected rates of immunity and prevalence across all three endemic patterns. The model does not produce mesoendemic transmission patterns at steady state for any parameter choices, leading to the conclusion that mesoendemic patterns occur during transient states or as a result of factors not included in this study. The model shows that coupling the effect of varying larval habitat availability with the effects of clinical and parasite-suppressing immunity is enough to produce known patterns of malaria transmission.

A Process-based Model with Temperature, Water, and Lab-derived Data Improves Predictions of Daily Culex pipiens/restuans Mosquito Density

“…We diverge from our previous assumption of neglecting competition to include a density-dependent death rate for the larva/pupa stage, modeled by applying an approximation to a quadratic loss differential equation as in Ratti and Wallace (2020) and Wallace et al (2017) to the combined larvae/pupae population (denoted as LP ), and delta is calculated as…”

“…However, these studies have alluded to the fact that assessment of the influence of different rainfall regimens on mosquito populations needs further examination as the rainfall linkage to mosquito habitats, in particular the larval habitat, could depend on factors such as slope, river routing, and availability of potential habitats for mosquitoes. Some have tackled this by using a lag in precipitation measurements or simulated rainfall to produce more accurate results ( Shone et al 2014 , Valdez et al 2017 , Ratti and Wallace 2020 , Ratti et al 2022 ). We hypothesize that including water gauge measurements in addition to temperature will improve predictions over temperature and precipitation (see Supp Fig.…”

Larval flushing alters malaria endemicity patterns in regions with similar habitat abundance

“…In a prior modeling study, we were able to reproduce three out of these four patterns of transmission, along with expected immunity levels, simply by varying the extent to which larval habitat depends on rainfall ( Ratti & Wallace, 2020 ). A detailed vector population model was based on the temperature and rainfall pattern in the western Kenya highlands.…”

“…We extend the model developed in Ratti & Wallace (2020) , which includes both the temperature- and habitat-dependent vector dynamics of An. gambiae and the transmission of P. falciparum .…”

“…This simulation is used to select days when rainfall exceeds a given threshold in order to reduce larval populations at a given rate. Rainfall scenarios which produce hyperendemic and holoendemic transmission patterns with no flushing effects, taken from Ratti & Wallace (2020) are extended to include flushing. Numerical experiments that vary the threshold rainfall required to flush larvae, and the rate at which those larvae are removed, are designed to establish conditions under which the same rainfall pattern may lead to malaria transmission patterns in multiple WHO classifications.…”