BackgroundDrowning deaths constitute a significant proportion of unnatural deaths globally. In Sweden and other high-income countries, drowning deaths have decreased. This study investigates the epidemiology and current trends of unintentional, intentional, and undetermined drowning deaths with emphasis on the presence of alcohol and other drugs.MethodsDuring an 18-years period, 5,125 drowning deaths were autopsied in Sweden. Data on cases including toxicological analysis on alcohol, pharmaceutical drugs, and illicit drugs were obtained from the National Board of Forensic Medicine.ResultsDuring the study period, the annual incidence of drowning deaths in Sweden was 3.1/100,000 inhabitants and decreased on average by about 2% each year (p<0.001). The highest incidence was found among males and in middle/older age groups. The incidence increased 3% for each year of age. Children/adolescents (≤18 years) constituted 5% of all drowning deaths. Of all drowned females in the study, 55% (847/1,547) committed suicide, which was a significantly higher proportion compared with males (21%, 763/3,578) (p<0.001). In total, 38% (1,656/4,377) of tested drowned persons had alcohol in their blood and the mean concentration was 1.8 g/l. In the unintentional drowning group, intentional drowning group, and the undetermined group, the proportion of alcohol positive was 44%, 24%, and 45%, respectively. One or several psychoactive drugs were present in the blood in 40% (1,688 /4,181) of all tested persons and in 69% (965/1,394) of tested persons who died from suicidal drowning. The most common drug was benzodiazepines (21%, 891/4,181). Illicit drugs were detected in 10% (82/854) of tested persons.ConclusionPresence of alcohol and drugs were frequent and may have contributed to the drowning deaths. The incidence of drowning deaths significantly decreased during the study period. Males and the middle/older age groups had a higher incidence compared to females and children. Suicidal drowning was common especially among women. Alcohol and drugs are significant contributors in drowning deaths in Sweden and should be considered as part of a comprehensive prevention program.

The disagreements in the reclassification of cause of death were equal but large in both directions. The official death certificates should be used with caution to evaluate CVD in small community intervention programmes, and restricted to the chapter level and total populations.

The main aim of the present study was to estimate total alcohol-related mortality in Sweden. For natural deaths, a meta-analysis carried out in Australia was updated to the end of March 1998, and pooled estimates of the relative risks were calculated for different diseases based on data from scientific studies that have been published in the international literature. The proportion of current alcohol drinkers from recent Swedish surveys, and the pooled relative risk estimates were used to estimate disease-specific alcohol-attributable fractions. Natural deaths 'caused' or 'prevented' by alcohol were estimated for the period 1992-1996. For unnatural deaths, all cases from 1992 through 1996 in Sweden were analysed (n = 23 132). Alcohol was regarded to attribute to the death: if the deceased was a 'known alcoholic'; if the underlying or contributing cause of death was alcohol-related; if the deceased had an alcohol-related in-patient diagnosis during a 3-year period prior to death; if the case tested positive for blood alcohol. Person years of life lost/gained (<70 years) due to alcohol were also assessed. The assumptions underlying the attributable risk methods used to analyse alcohol-related mortality due to natural causes need to be borne in mind when interpreting the results on natural deaths. Moreover, the preventive effect of alcohol on coronary heart disease and stroke is still controversial. The findings of alcohol-related mortality due to unnatural causes were much more reliable. About 3.5% of deaths were attributed to alcohol; alcohol involvement was more than twice as common in deaths of males (4.8%) than in those of females (2.0%). About one-quarter of the deaths in those aged below 50 years were attributed to alcohol. In those (

The present estimates are conservative; alcohol involvement in unnatural deaths probably is even higher, up to 44% of the total. The present estimation is an important step in policy-making to lower the number of alcohol-related deaths in Sweden.