Population Health


The Role of Health Behaviors in Social Differentials in Mortality

Kieron Barclay, Jo Mhairi Hale, Pekka Martikainen, Alyson van Raalte, Mikko Myrskylä, Maarten Jacob Bijlsma, Karen van Hedel, Yaoyue Hu

Detailed description:

Behavioral risk factors, such as harmful consumption of alcohol or smoking, are well-established causes of increased mortality. Although it is widely accepted that the unequal distribution of behavioral risk factors plays a role in health inequalities, the overall contribution of behavioral risk factors to the level of and changes in socioeconomic differences remains unclear. Using Finnish data, we show that between 1988 and 2007 alcohol- and smoking-attributable deaths reduced life expectancy by 4.5 years among men. As alcohol-attributable mortality increased, smoking-attributable mortality decreased in this period, leaving the joint contribution stable. At the same time, among women, the combined contribution of these two health-related behaviors to life expectancy increased from 0.7 to 1.2 years. In 2003-2007, the life expectancy differential between the top and bottom income quintile was 11.4 years for men and 6.3 years for women. Without alcohol- and smoking-attributable mortality, these differences would have been 60% smaller for men and 36% smaller for women. Life expectancy differentials increased rapidly in Finland; without alcohol and smoking, this increase would have been 69% smaller among men and 85% smaller among women. Overall, our results show that alcohol consumption and smoking significantly influence social patterns of mortality and their joint contribution is increasing (except with regard to men’s smoking).

Relatively little is known about the relationship between death from alcohol-related causes and the likelihood of hospital admission in the years before death. We traced back the timing and causes of all hospitalizations in the 10-year period before death among men and women who were employed at the start of the study period, and who died from alcohol-related causes in Finland between 1997 and 2007. We found that the individuals who ultimately died from alcohol-related causes had an average of seven hospital admissions and spent 56 days in hospital during the study period. By the fifth year before death, about three-fifths of these individuals had been hospitalized at least once due to any cause, and less than one-third were admitted to hospital with an alcohol-related diagnosis. Further results show that persistent problem drinkers, defined on the basis of having an alcohol-related death or a hospital admission, are marginalized from the labor market early in their life course. Overall, the results imply that enhanced patient management at hospitals may potentially reduce alcohol-related harm.

Research keywords: Ageing, Mortality and Longevity; Family Behavior; Life Course

Region keywords: Finland


Mäkelä, P.; Herttua, K.; Martikainen, P.:
Alcohol and Alcoholism 50:6, 661-669 (2015).
van Raalte, A. A.; Myrskylä, M.; Martikainen, P.:
Demographic Research 32:20, 589-620 (2015).