Language

English

Publication Date

10-14-2025

Journal

Injury Epidemiology

DOI

10.1186/s40621-025-00623-8

PMID

41088464

PMCID

PMC12522868

PubMedCentral® Posted Date

10-14-2025

PubMedCentral® Full Text Version

Post-print

Abstract

Background: It is difficult to study the epidemiology of drowning at the regional level because of multiple data sources, many of which have a high degree of unstandardized and missing data. We aimed to link multiple datasets to identify demographics and geographic locations of unintentional fatal drowning in a metropolitan region and compare linked data with vital statistic data.

Methods: This cross-sectional study included unintentional drowning fatalities among persons of all ages in metropolitan Houston between 2016 and 2022. Probabilistic linking was used to link multiple datasets and geographical mapping to identify drowning locations. The effectiveness of data linkage was assessed by the recall, precision, and F1-score (harmonic mean of precision and recall). Geographic location of drownings by aquatic body was studied. Drowning burden by demographics, county and aquatic body were compared between linked data and vital statistic data using Chi-square tests.

Results: Data from 8 datasets were linked. The linkage metrics were Recall (0.88-1.00), Precision (0.91-1.00), and F1-score (0.91-1.00) for datasets. There were 790 drowning fatalities. The median age was 40 years (IQR: 18.5,60); 71% were males. Children aged 0-17 years constituted 24% of drowning fatalities. Drownings occurred in swimming pools (27%), bathtubs (19%) natural water (27%), flood control structures (20%), and during flooding events (6%). Adults commonly drowned in natural water, flood control structures, bathtubs, and during flooding events whereas most toddlers drowned in swimming pools and hot tubs. No significant differences in counts by age group, sex, county of drowning and body of water were observed between linked and vital statistic data. Drowning locations were geocoded in 769 of 790 records (97%).

Conclusion: Probabilistic data linkage can accurately determine the epidemiology of fatal drowning in a metropolitan region. Identification of high-risk drowning subpopulations and locations can inform drowning countermeasures at the regional level.

Keywords

Probabilistic data linkage, Drowning burden, Geocoding, Drowning epidemiology

Published Open-Access

yes

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.