Stochastic agent-based modeling of tuberculosis in Canadian Indigenous communities

Authored by Amy L Greer, Ashleigh R Tuite, Victor Gallant, Elaine Randell, Annie-Claude Bourgeois

Date Published: 2017

DOI: 10.1186/s12889-016-3996-7

Sponsors: No sponsors listed

Platforms: No platforms listed

Model Documentation: Other Narrative Flow charts

Model Code URLs: Model code not found

Abstract

Background: In Canada, active tuberculosis (TB) disease rates remain disproportionately higher among the Indigenous population, especially among the Inuit in the north. We used mathematical modeling to evaluate how interventions might enhance existing TB control efforts in a region of Nunavut. Methods: We developed a stochastic, agent-based model of TB transmission that captured the unique household and community structure. Evaluated interventions included: (i) rapid treatment of active cases; (ii) rapid contact tracing; (iii) expanded screening programs for latent TB infection (LTBI); and (iv) reduced household density. The outcomes of interest were incident TB infections and total diagnosed active TB disease over a 10-year time period. Results: Model-projected incidence in the absence of additional interventions was highly variable (range: 33-369 cases) over 10 years. Compared to the `no additional intervention' scenario, reducing the time between onset of active TB disease and initiation of treatment reduced both the number of new TB infections (47\% reduction, relative risk of TB = 0.53) and diagnoses of active TB disease (19\% reduction, relative risk of TB = 0.81). Expanding general population screening was also projected to reduce the burden of TB, although these findings were sensitive to assumptions around the relative amount of transmission occurring outside of households. Other potential interventions examined in the model (school-based screening, rapid contact tracing, and reduced household density) were found to have limited effectiveness. Conclusions: In a region of northern Canada experiencing a significant TB burden, more rapid treatment initiation in active TB cases was the most impactful intervention evaluated. Mathematical modeling can provide guidance for allocation of limited resources in a way that minimizes disease transmission and protects population health.

Tags

Simulation Mathematical model tuberculosis transmission public health Mycobacterium-tuberculosis Impact Canada Household Benefits Latent tb infection Nunavut Tb