Modeling the Impact of Low-Cost Ventilation Interventions on Nosocomial Tuberculosis Infection Risk in Liberian Hospitals

Healthcare facilities in low-and middle-income countries (LMICs) are high-risk environments for the spread of infectious diseases, and ventilation remains underutilized for preventing nosocomial (i.e., hospital-acquired) airborne infections. Liberia is among the countries with the highest burden of tuberculosis (TB). Previous work in 10 Liberian hospitals found that two thirds (46/70) of patient care spaces failed to meet World Health Organization ventilation guidelines ( ≥ 60 L/s/person), indicating elevated transmission risk in these spaces. Installation of louvered windows and whirlybird turbines in three emergency departments resulted in moderate improvements in the indoor ventilation. The aim of this analysis is to evaluate the anticipated impact of these interventions for averting nosocomial TB transmission in these settings. We applied the Rudnick-Milton equation within a Monte Carlo simulation to model the characteristics of the emergency departments and calculate the expected number of nosocomial TB infections averted due to the ventilation improvements. The model was parametrized using prevalence estimates drawn from active case-finding data in emergency department settings and a lognormal distribution for quanta emission rates (i.e., patient infectivity estimates). Parameter uncertainty was tested using one-way and probabilistic sensitivity analyses. Over the course of a year, we simulated a total of 175 nosocomial TB infections in the pre-installation conditions and 143 TB infections in the post-installation conditions (19% reduction in transmission). Presuming that 5% of infections would result in active cases, these interventions would prevent 30 latent TB infections and 2 active cases of TB. One-way sensitivity analyses showed that the results were sensitive to the prevalence of active TB among patients presenting to the emergency department, the relative infectivity of the patients, and the annual patient volume. The probabilistic sensitivity analysis suggests that results from our main analysis fall within a reasonable range of parameter uncertainty (median number of cases averted: 2.5, IQR: 1.4 - 4.5). Overall, these findings demonstrate that low-cost, sustainable ventilation interventions are anticipated to meaningfully reduce nosocomial TB transmission in LMIC healthcare facilities and would likely extend to other diseases. Further research exploring the cost-effectiveness, optimization, and application of these interventions at a larger scale is warranted.