Is the efficacy of detecting a drowning patient altered if the person conducting surveillance is a non-swimmer?

TitleIs the efficacy of detecting a drowning patient altered if the person conducting surveillance is a non-swimmer?
Publication TypeConference Paper
Year of Publication2007
AuthorsDoyle, B
Conference NameWorld Conference on Drowning Prevention
Date Published09/2007
PublisherInternational Life Saving Federation
Conference LocationPorto, Portugal
Other Numbers01-13

Context: In March of 2003 and 2004, our group conducted a study to investigate the safety of a proposed qualification titled ‘Patrol Support’. Candidates would train in every aspect of the national lifeguard award apart from the swimming and water-rescue components. The proposed qualification would allow such individuals to conduct surveillance of swimmers. A review of the literature revealed there is a paucity of evidence about the efficacy of surveillance conducted by traditionally trained lifeguards; there is very little known about the efficacy of non-swimmers.

Method: We conducted a controlled laboratory based blinded study to compare the time to detect a standardised simulated drowning victim by traditionally trained lifeguards, (n=18) compared to ‘Patrol Support’ personnel, (n=18) following a standardised scanning training programme. Both groups were similar in terms of experience, gender and age distribution. They received the same basic training; however the ‘Patrol Support’ (non-swimming) group received no instruction or testing on the swimming or rescue components. Results gathered prior and post training were analysed, and the impact of the standardised scanning training assessed as an independent variable. Several human and environmental factors were recorded and analysed for their impact on detection times.

Results and Discussion: Our results demonstrate that non-swimmers take significantly longer to detect a drowning victim in most instances regardless of training (47.7sec vs. 69.6sec) however one subgroup of non-swimmers (those aged 25-35 years) showed no statistical difference to the same subgroup of swimmers (even with training) and showed statistically better results than two younger subgroups (those aged 14-20 and 20-25 years ). This difference was present before and increased after training. The most significant human factor that affected all groups regardless of age or any other factor was sleep. Normalised to 8 hrs we identified that 2hrs or more sleep deficit resulted in a decrease in performance in both groups; however the difference was greatest in lifeguards aged 14-19yrs. In our study, the ‘non-swimmers aged over 25-35 years’ were statistically correlated (as a sub group) to faster detection and scanning times when analysed as an independent variable. The reasons for this are unexplained by our study. These results indicate it is possible to effectively train ‘non swimmers’ to detect the signs of drowning; however their performance may not be as good as swimmers. The sub group of individuals aged 25-35 years may be an exception. The size of this study (n=36) and the statistical differences shown although significant (p<0.05) are ‘clinically’ too small to be conclusive. A non-simulated, larger multi-centre study is proposed sometime in the near future. The impact of the standardised training provided is of particular interest and may be studied separately. In 2005, the ‘Patrol Support’ qualification was abandoned in New Zealand; however, our research has wider implications for any sector of aquatic safety involved in the supervision of swimmers by non-swimmers such as parental, teacher or caregiver supervision for example.

Learning Outcomes
  1. Recognise the need and importance for specific training in scanning and surveillance if involved in supervising patrons in the water if you are a non swimmer
  2. Gain an increased understanding of the implications of prolonged detection times of drowning victims
  3. Be able to list several physiological factors that drastically affect surveillance efficacy.
Digital Media: