Background Lifeguards are crucial in ensuring patron safety at public pools, through scanning and surveillance to detect drowning victims. Whilst there are guidelines as to lifeguard to patron ratios for adequate supervision, there is no empirical evidence for these ratios.

This research aimed to determine the ideal ratio of lifeguards to persons in the water to detect a potential drowning within an appropriate timeframe to prevent death or neurological damage, using virtual reality (VR) eye-tracking technology.

Methods Fifty qualified lifeguards in Victoria, Australia viewed nine 360-degree VR videos of one or two pools that lasted 120 seconds each. Six videos contained simulated drownings, and three had no drowning. Videos were randomised with different scenes and patron numbers (25–115). Lifeguards noted if they identified a drowning victim. Time of identification was coded using the VR eye-tracking.

Results With >75 patrons, a drowning victim was not identified 50% of the time (p<.01). There was a moderate, positive, linear relationship between the time a drowning victim was identified and the number of patrons at the pool (p<.000). With >50 patrons, lifeguards tended to take >10 seconds to identify a drowning victim.

Conclusion This study demonstrated that with higher numbers of patrons in a pool, the ability of lifeguards to detect a drowning victim within the recommended time to prevent long-term effects decreases. Therefore, current guidelines for the ratio of patrons per lifeguard require review.

Learning outcomes VR and eye-tracking technology provides a novel way to address research problems in a controlled safe environment.