Background
Drowning is a leading and preventable cause of death that has suffered an attention deficit. Improving drowning data in countries would assist the understanding of the full extent and circumstances of drowning, to target interventions and evaluate their effectiveness. The World Health Organization identifies data collection as a key strategy underpinning effective interventions. This study compares unintentional fatal drowning data collection, management and comparison using the databases of Australia, Canada and New Zealand.
Methods
Cases of fatal unintentional drowning between 1-January-2005 and 31–December-2014 were extracted. Cases were combined into a single dataset and univariate and chi square analysis (
p
< 0.01) were undertaken. Location and activity variables were mapped and combined. Variables consistently collected across the three countries were compared to the ILCOR Drowning Data Guideline. The authors also recommend variables for a minimum core dataset.
Results
Of 55 total variables, 19 were consistent and 13 could be compared across the three databases. When mapped against the ILCOR Drowning Data Guideline, six variables were consistently collected by all countries, with five compared within this study. The authors recommend a minimum core dataset of 11 variables including age, sex, location, activity, date of incident, and alcohol and drug involvement).
There were 8176 drowning deaths (Australia 34.1%, Canada 55.9%, New Zealand 9.9%). All countries achieved reductions in crude drowning rates (Australia − 10.2%, Canada − 20.4%, New Zealand − 24.7%). Location and activity prior to drowning differed significantly across the three countries. Beaches (X
2
= 1151.0;
p
< 0.001) and ocean/harbour locations (X
2
= 300.5;p < 0.001) were common in Australia and New Zealand, while lakes/ponds (X
2
= 826.5;
p
< 0.001) and bathtubs (X
2
= 27.7;p < 0.001) were common drowning locations in Canada. Boating prior to drowning was common in Canada (X
2
= 66.3;p < 0.001).
Conclusions
The comparison of data across the three countries was complex. Work was required to merge categories within the 20% of variables collected that were comparable, thus reducing the fidelity of data available. Data sources, collection and coding varied by country, with the widest diversity seen in location and activity variables. This study highlights the need for universally agreed and consistently applied categories and definitions to allow for global comparisons and proposes a core minimum dataset.