Predicting Drowning Mortality Risk Using Machine Learning Models; A Retrospective Cohort Study
Archives of Academic Emergency Medicine,
Vol. 14 No. 1 (2026),
1 October 2025
,
Page e21
https://doi.org/10.22037/aaem.v14i1.2963
Abstract
Introduction: Particularly in highly tourist-active coastal locations, drowning is still a serious international public health concern. This study investigates the predictive value of machine learning approaches in estimating drowning-related mortality risk.
Methods: This retrospective cohort study analyzed drowning incident data from the Emergency Management and Medical Urgency Center of Guilan Province, covering the period from 2018 to 2023. The data were preprocessed, missing values imputed using the K-Nearest Neighbors (KNN) algorithm, and balanced using the Synthetic Minority Over-sampling Technique (SMOTE). Three models including logistic regression, decision tree, and naïve Bayes were evaluated in predicting the risk of mortality following drowning and sensitivity, specificity, and accuracy of each model was calculated and compared.
Results: A total of 600 consecutive cases meeting the eligibility criteria were extracted for analysis, forming the final dataset. Logistic regression exhibited the highest predictive power, with an accuracy of 51.67% and an area under the curve (AUC) of 60.02%. The most influential variables in drowning-related mortality prediction were drowning location, drowning year, gender, and age. High-risk areas posed a 33-fold higher mortality risk than safe locations (p < 0.001). Age and gender were not statistically significant predictors of fatal drowning.
Conclusion: Given its superior interpretability and predictive capability, logistic regression was identified as the most effective model for assessing drowning mortality risk. Preventative measures should focus on identifying high-risk areas, installing warning signs, implementing lifeguard teams, educating tourists, and enforcing strict coastal safety regulations to mitigate drowning fatalities.
- Drowning
- Mortality
- Machine Learning
- Machine Learning Algorithms
- Iran
How to Cite
References
1. Davis CA, Schmidt AC, Sempsrott JR, Hawkins SC, Arastu AS, Giesbrecht GG, et al. Wilderness Medical Society Clinical Practice Guidelines for the treatment and prevention of drowning: 2024 update. Wilderness Environ Med. 2024;35(1_suppl):94S-111S.
2. Davoudi-Kiakalayeh A, Barshan J, Sigaroudi FE, Mirak HM, Alavi SA. The application of the Haddon matrix in identifying drowning prevention solutions in the north of Iran. Heliyon. 2023 Jun 1;9(6). DOI: 10.1016/j.heliyon.2023.e16958
3. Shahbazi F, Mirtorabi SD, Hosein Mahdavi SA, Hashemi Nazari SS. Trend of mortality rate due to drowning in Iran (2013–2018). Archives of Trauma Research. 2020 Jul;9(3):111-5. doi: 10.30491/tm.2025.519420.1831
4. Xie X, Li Z, Xu H, Peng D, Yin L, Meng R, Wu W, Ma W, Chen Q. Non-fatal drowning risk prediction based on stacking ensemble algorithm. Children. 2022 Sep 14;9(9):1383. doi.org/10.3390/children9091383
5. JA Adlin Layola, S. Saranya, Mabel Rose RA, S. Nickel, Manoj Kumar. To Detect Active Drowning Using Deep Learning Algorithms. In2023 9th International Conference on Smart Structures and Systems (ICSSS) 2023 Nov 23 (pp. 1-7). IEEE. DOI: 10.1109/ICSSS58085.2023.10407065
6. Kao WC, Fan YL, Hsu FR, Shen CY, Liao LD. Next-Generation swimming pool drowning prevention strategy integrating AI and IoT technologies. Heliyon. 2024 Sep 30;10(18). DOI: 10.1016/j.heliyon.2024.e35484
7. Kiakalayeh AD, Mohammadi R, Ekman DS, Chabok SY, Janson B. Unintentional drowning in northern Iran: a population-based study. Accid Anal Prev. 2008;40(6):1977-81.
8. Willcox-Pidgeon SM, Franklin RC, Leggat PA, Devine S. Identifying a gap in drowning prevention: high-risk populations. Inj Prev. 2020;26(3):279-88.
9. Xie Z, Huang Z, Ran Q, Luo W, Du W. Global burden of drowning and risk factors across 204 countries from 1990 to 2021. Sci Rep. 2025;15(1):10916.
10. Davoudi-Kiakalayeh A, Mohammadi R, Yousefzadeh-Chabok S. Prevention of drowning by community-based intervention: implications for low-and middle-income countries. Arch Trauma Res. 2012;1(3):112.
11. JA Adlin Layola, Saranya S, RA MR, Nickel S, Kumar M. To Detect Active Drowning Using Deep Learning Algorithms. In2023 9th International Conference on Smart Structures and Systems (ICSSS) 2023 Nov 23 (pp. 1-7).
12. Bi Q, Goodman KE, Kaminsky J, Lessler J. What is machine learning? A primer for the epidemiologist. Am J Epidemiol. 2019;188(12):2222-39.
13. Zeng Y, Zhang X, Wang J, Usui A, Ichiji K, Bukovsky I, et al. Inconsistency between human observation and deep learning models: assessing validity of postmortem computed tomography diagnosis of drowning. j imaging inform med. 2024;37(3):1-10.
14. Xu H, Zhu X, Zhou Z, Xu Y, Zhu Y, Lin L, Huang J, Meng R. An exploratory model for the non-fatal drowning risks in children in Guangdong, China. BMC public health. 2019 7;19(1):599.
15. Quan L, Bierens JJ, Lis R, Rowhani-Rahbar A, Morley P, Perkins GD. Predicting outcome of drowning at the scene: a systematic review and meta-analyses. Resuscitation. 2016;104:63-75.
16. Roberts K, Thom O, Devine S, Leggat PA, Peden AE, Franklin RC. A scoping review of female drowning: an underexplored issue in five high-income countries. BMC Public Health. 2021;21(1):1072.
17. Scarr J-P, Jagnoor J. Identifying opportunities for multisectoral action for drowning prevention: a scoping review. Inj Prev. 2022;28(6):585-94.
18. Woods M, Koon W, Brander RW. Identifying risk factors and implications for beach drowning prevention amongst an Australian multicultural community. Plos one. 2022;17(1):e0262175.
19. Peden AE, Franklin RC, Queiroga AC. Epidemiology, risk factors and strategies for the prevention of global unintentional fatal drowning in people aged 50 years and older: a systematic review. Inj Prev. 2018;24(3):240-7.
20. Quan L, Cummings P. Characteristics of drowning by different age groups. Inj Prev. 2003;9(2):163-8.
- Abstract Viewed: 28 times
- pdf Downloaded: 64 times
Download Statastics
Make a Submission
SJR
COPE
