Objective: To determine whether the well-known U-shaped relationship betwee
n average alcohol intake and mortality also holds for other health measures
and for aspects of drinking other than weekly average alcohol intake, such
as frequency of heavy-drinking episodes. Method: This study was carried ou
r within the framework of a general population survey conducted in Eindhove
n, The Netherlands (N = 18,973). Apart from mortality, the following health
measures were considered: self-assessed health (based on perceived general
health and the Nottingham Health Profile questionnaire), a list of chronic
conditions and a list of health complaints. Respondents were categorized a
s abstainers, light (1-14 units/week), moderate (15-28 units/week) and exce
ssive drinkers (greater than or equal to 29 units/week). Information on the
frequency with which heavy-drinking episodes occurred was also available.
Results: Light or moderate drinkers had not only lower mortality but other
health burdens were lower than for either abstainers or heavier drinkers. F
requent heavy-drinking episodes were observed to be directly related to inc
reased mortality rates, although nor significantly. A trend was observed fo
r drinkers reporting seldom heavy-drinking episodes (once or twice in the p
revious 6 months) to report less health burdens and to have lower mortality
rates than those reporting no heavy drinking episodes. Conclusions: A U-sh
aped pattern was observed for mortality as well as for several other health
measures. Frequent heavy-drinking episodes were related to an increased li
kelihood of mortality (not significant but suggesting a J-shaped pattern) a
nd were not related to other health measures.