Date Published: February 27, 2007
Publisher: Public Library of Science
Author(s): Peter Wilson
Abstract: Wilson discusses a new study inPLoS Medicine that examined the effect of natural ventilation in eight hospitals in Lima, Peru upon risks of TB transmission.
Partial Text: Airborne transmission of infections such as tuberculosis (TB) can be a major problem in health care establishments. Health care workers infected with TB can be unwitting disseminators of the infection, and expensive retrospective screening (of other health workers and patients) is required when the diagnosis is made. In the past, the benefits of natural ventilation in the treatment of TB were implicit in the design of the sanatoria wards (Figure 1). More recently, expensive negative pressure facilities have been installed to accommodate patients with TB, particularly following outbreaks of TB among patients with HIV/AIDS .
In a new study in PLoS Medicine, Roderick Escombe and colleagues examined the effect of natural ventilation in eight hospitals in Lima, Peru . The study included 70 naturally ventilated clinical rooms, in five older and three newer hospitals, including respiratory and medical wards, clinics, and waiting rooms. The comparators were 12 negatively pressured rooms built in the last six years. A carbon dioxide tracer gas was used to establish the number of air changes in various sizes of rooms, with and without open windows and doors. Infection risk for TB was predicted using a model of airborne infection risk (the Wells-Riley equation). Larger rooms have higher absolute ventilation for a given number of air changes. To determine absolute ventilation, mechanically ventilated rooms were generally assumed to deliver 12 air changes per hour, but this may not be the case in many poorly monitored settings.
The study is a novel attempt to bring some scientific scrutiny to old practices. However, the Wells-Riley model that the authors used has some limitations. It assumes a steady state, and that one event leads to infection, while disregarding the effect of proximity to an infected case. Removal of infectious particles from the air by other processes including gravity is not considered.