Background: The effect of alcohol on bone mineral density (BMD) is unknown, with studies relating it to lower BMD when intake is high, and increased BMD when intake is moderate. Similarly conflicting results are found in studies relating alcohol consumption to fractures and fall risk. More reliable information would be useful because of the high fracture risk in men with low BMD. Cawthon and colleagues examined the relationship between alcohol consumption and BMD, falls, and fractures.

The Study: Eligible participants were recruited from a cohort of 5,995 community-dwelling men 65 years or older. Information about height, weight, body mass index (BMI), smoking, race, history of falling and fractures, and exercise habits, as well as additional demographic information and self-reported health status, were obtained.

Alcohol use was quantified into six categories based on number of drinks in a given period (per week for regular drinkers). Binge drinking was also determined by asking how often participants had five or more drinks on a single occasion. Those who reported this level of drinking on a daily basis were classified as excessive drinkers. In addition, the CAGE questionnaire was administered to all participants who had consumed 12 or more drinks in their lifetime.

BMD was ascertained by dual energy x-ray absorptiometry of the femoral neck, hip, and lumbar spine. Participants reported falls and fractures over the course of the study (an average of 3.65 years).

Results: More than one third of the cohort (2,121 or 35.5 percent) drank very little or not at all and were classified as abstainers. Slightly more than one half (3,156 or 52.8 percent) had light intake (one to 13 drinks a week). Another 697 (11.7 percent) were considered moderate to heavy drinkers (14 or more drinks per week). In terms of the CAGE questionnaire, 16.7 percent gave positive answers to two or more of the questions. Some binge drinking was found in 13 percent of the men, and more frequent binge drinking in 6 percent. There was an association between problem drinking and younger age, taller stature, heavier weight, single marital status, past or present smoking, and poorer health compared with those who did not have a history of problem drinking.

Although the differences were small, drinkers had higher BMD levels than non-drinkers, with BMD increasing as alcohol intake increased. Current heavy drinking and binge drinking were associated with higher BMD levels. There were no differences in fracture rates among categories of drinkers. Abstinence was associated with a higher fall risk than light drinking (relative risk [RR] = 0.77). There was no association between fall risk and moderate or heavy drinking. Those with a history of problem drinking had a higher fall risk than the group without problem drinking, with men who responded positively to two or more questions on the CAGE questionnaire having a 1.62 times (95% confidence interval [CI], 1.33 to 1.97) greater likelihood of falling than those who answered positively to one or none of the questions. This risk level applied to current and former problem drinkers. Similarly, heavy drinkers, irrespective of past or current status, were 50 percent more likely to fall than men with no history of heavy drinking. Binge drinking in the preceding year was not associated with increased fall risk.

Conclusion: In this study, greater alcohol consumption in the year before baseline was associated with greater BMD in older men. There was no difference in fracture risk. The lowest risk of falls was in the group of men who were light drinkers (fewer than 13 drinks per week). These findings suggest that the association between alcohol consumption and lower BMD in some studies is not direct but is likely because of the poorer health of alcoholics.