Effective ventilation in general hospital wards is important for controlling the airborne transmission of infectious respiratory diseases. Experiments have been carried out to increase our understanding of the interaction of the breathing flows of two individuals in a full-scale experimental hospital ward with three ventilation systems, i.e. mixing, downward and displacement ventilation. Two life-size breathing thermal manikins were used to simulate a source patient and a receiving patient. The exhalation jet from a bed-lying manikin was visualized using smoke. N2O was used as tracer gas to simulate the droplet nuclei exhaled by patients; and the spatial distribution of its concentrations was measured. Our experimental results show that for both mixing and downward ventilation, the exhaled jet penetrates a short distance and is diluted quickly by ventilation air. The exhaled droplet nuclei are well mixed in the ward. Bed distance does not affect the personal exposure of the receiving patient. For displacement ventilation, the exhaled jet can penetrate a long distance. A high concentration layer of exhaled droplet nuclei because of thermal stratification locking has also been observed with displacement ventilation. This work is useful for identifying an appropriate ventilation method that can remove droplet nuclei more effectively and minimize the risk of cross-infections in a hospital ward environment.
Practical implications: As one of the major potential sources for infectious droplet nuclei in a hospital environment, exhalation flows of an infected patient can interact with the respiratory activities of other close individuals and with the room ventilation systems. Our latest results provide information on the penetration of exhalation jets into the ambient environment in different ventilation systems. This work is useful in identifying an appropriate and effective ventilation method for removing droplet nuclei more effectively, and thus minimizing the risk of cross-infections in hospital wards with multiple beds.