Trends in unintentional drowning: the role of alcohol and medical care

JAMA. 1999 Jun 16;281(23):2198-202. doi: 10.1001/jama.281.23.2198.


Context: During the last few decades, mortality from drowning has decreased in the United States for unknown reasons. It has been hypothesized that this decline may be due to decreased use of alcohol in and around water or improved medical treatment after a submersion.

Objectives: To estimate changes in unintentional mortality due to submersion, estimate trends in drownings attributable to alcohol use, and assess the role of medical care in these mortality trends.

Design: A 21-year longitudinal study of case findings, from January 1, 1975, through December 31,1995.

Setting and participants: All residents of King County, Washington, who died unintentionally from submersion and 284 persons hospitalized for submersion who survived.

Main outcome measures: Changes in submersion-related mortality incidence over time, proportion of this mortality that could be attributed to alcohol use, changes over time in the case-fatality rate of treated patients, and estimate of deaths prevented in 1995 compared with projected estimates had there been no change in incidence since 1975.

Results: There were 539 deaths due to drowning in King County during 21 years. Mortality rates during this period declined by 59% (95% confidence interval [CI], -70% to -46%). The incidence of death attributable to alcohol use decreased by 81% (95% CI, -91% to -57%); this could account for 51% of deaths prevented in 1995. Among 249 comatose patients who received prehospital care, 205 died; the odds of survival decreased 40% over 21 years (P = .40). Among 101 comatose patients who were hospitalized, 63 died; the odds of survival decreased 29% (P = .75). The incidence of survival of comatose hospital patients decreased by 29% from 1975 to 1995 (95% CI, -78% to +125%). We found no evidence that trends in medical treatment prevented any deaths due to drowning in 1995.

Conclusions: Drowning incidence in King County, Washington, declined because of a decrease in severe submersion episodes rather than an increase in success of medical interventions. Our data support the theory that less use of alcohol around water prevents some deaths. About half of the decrease was unexplained.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Alcoholic Intoxication / epidemiology
  • Drowning / mortality*
  • Emergency Medical Services / statistics & numerical data
  • Humans
  • Longitudinal Studies
  • Survival Analysis
  • Washington / epidemiology