Aims: The health effects of drinking may be related to psychological characteristics influencing both health and drinking habits. This study aims to examine the relationship between intelligence, later beverage preference and alcohol intake.
Design: Prospective cohort study.
Setting: Zealand, Denmark.
Participants: A total of 900 obese men and a random population sample of 899 young men.
Measurements: Intelligence testing at the draft board examinations over a 22-year period between 1956 and 1977. Percentage of wine of total alcohol intake (wine pct), preference for wine (wine pct >50), heavy drinking (>21 drinks per week) and non-drinking (<1 drink per week), and vocational education from follow-ups of the initial study sample in 1981-83 and 1992-94.
Findings: A strong dose-response-like association was found between intelligence quotient (IQ) in young adulthood and beverage preferences later in life in both the obese and the random population sample. At the first follow-up a 30-point advantage in IQ [2 standard deviations (SD)] was found to be associated with an odds ratio (OR) for preferring wine over beer and spirits of 1.7 (1.3-2.4). At the second follow-up the corresponding OR was 2.8 (2.0-3.9). A 30-point advantage in IQ was found to be associated with an OR for being a non-drinker of 0.5 (0.3-0.8) at the first follow-up and second follow-up. We examined whether, at the second follow-up, the association between IQ, beverage preferences and non-drinking could be explained by socio-economic position (SEP). The association between IQ and non-drinking disappeared when controlling for SEP. The association between IQ and beverage preferences was attenuated, but remained statistically significant. IQ was not associated with heavy drinking.
Conclusion: Irrespective of socio-economic position, a high IQ was associated with preference for wine to other beverages, but IQ was not related similarly to alcohol consumption.