Linking mosquito surveillance to dengue fever through Bayesian mechanistic modeling

Abstract

Our ability to effectively prevent the transmission of the dengue virus through targeted control of its vector, Aedes aegypti, depends critically on our understanding of the link between mosquito abundance and human disease risk. Mosquito and clinical surveillance data are widely collected, but linking them requires a modeling framework that accounts for the complex non-linear mechanisms involved in transmission. Most critical are the bottleneck in transmission imposed by mosquito lifespan relative to the virus' extrinsic incubation period, and the dynamics of human immunity. We developed a differential equation model of dengue transmission and embedded it in a Bayesian hierarchical framework that allowed us to estimate latent time series of mosquito demographic rates from mosquito trap counts and dengue case reports from the city of Vitória, Brazil. We used the fitted model to explore how the timing of a pulse of adult mosquito control influences its effect on the human disease burden in the following year. We found that control was generally more effective when implemented in periods of relatively low mosquito mortality (when mosquito abundance was also generally low). In particular, control implemented in early September (week 34 of the year) produced the largest reduction in predicted human case reports over the following year. This highlights the potential long-term utility of broad, off-peak-season mosquito control in addition to existing, locally targeted within-season efforts. Further, uncertainty in the effectiveness of control interventions was driven largely by posterior variation in the average mosquito mortality rate (closely tied to total mosquito abundance) with lower mosquito mortality generating systems more vulnerable to control. Broadly, these correlations suggest that mosquito control is most effective in situations in which transmission is already limited by mosquito abundance.

Fig 1. Vitória data and model estimates.

A: weekly observed case reports (points), with corresponding posterior median (black line) and 80% posterior credible interval (gray band). B: weekly mosquito trap counts (points), with posterior median (black line) and 80% posterior credible interval (gray band). C: extrinsic incubation period (EIP; weeks), computed from weekly mean temperature data. D: estimated weekly mosquito mortality rate, with the posterior median (black line) and the 80% posterior credible interval (gray band).

Fig 2. Mosquito mortality and temperature.

Posterior median mosquito mortality rate as a function of weekly mean temperature (degrees Celsius).

Fig 3. The effect of mosquito control as a function of the week in which it was applied.

Summary of the posterior predicted effect of mosquito control implemented in a given week of the year on the number of cases in the following year (relative to the number of cases expected without control). Black lines indicate the posterior median, while gray ribbons indicate the 80% credible interval. The first three panels show the results for control implemented in the years 2008-2010, and the last panel shows the overall effect of control impelmented in a given week of the year, summing over all three years. For example, mosquito control applied in week 37 of 2008 would have prevented about 13% of the human cases over the following year (i.e., the caseload would have been 87% of the expectation without control).

Fig 4. Posterior correlation between system parameters and the effectiveness of control.

The y-axis represents the effectiveness of optimally timed control, i.e., the effect of control implmented in the 34th week of the year on the relative number of cases in the following year, summed over 2008, 2009, and 2010. Each point represets a single sample from the posterior distribution, giving the number of cases in the controlled simulation (relative to the number of cases expected without control) as a function of A: the mean mosquito mortality rate, d₀, and B: the case reporting probability, ϕ, from that posterior sample.