Background Frailty is recognised as an important public health concern among older people due to its detrimental consequences on health. Alcohol consumption may be a plausible risk factor for frailty. There is only limited evidence in the literature on association between alcohol and frailty, which is still controversial. Therefore we investigated the relationship between alcohol consumption and the risk of incident frailty.

Methods A total of 2544 community-dwelling older people aged >60 years from the English Longitudinal Study of Ageing were followed for incident frailty over four years according to alcohol consumption. Frailty was defined using Fried frailty phenotype criteria with slight modification according to data availability. Questionnaire data on alcohol consumption was grouped into non-drinkers, >0–7 units/week, >7–14 units/week, >14–21 units/week, >21 units/week. Multivariable logistic regression models were used to examine incident frailty risk.

Results Compared with those drinking >0–7 units per week (reference), those drinking >21 units per week were less likely to develop frailty (Odds ratio (OR)=0.45, 95% confidence interval (CI)=0.27–0.73, p<0.001) in the unadjusted model, however the association became non-significant in the fully adjusted model controlling for age, gender, smoking, education and wealth (OR=0.65, 95% CI 0.38 to 1.11, p=0.12). There was no significant difference in incident frailty risk in those drinking >7–14 and >14–21 units per week compared to >0–7 units per week in any models. Non-drinkers had worse health profile and higher incident frailty risk than drinkers (fully adjusted OR=1.63, 95% CI 1.07 to 2.48, p=0.02).

Conclusion In this cohort non-drinkers had a significantly higher risk of developing frailty compared to light drinkers (>0–7 units per week). No significant associations were observed among drinkers. The true association between alcohol and frailty may have been affected by a ‘sick-quitter’ effect (i.e. drinkers who are ill health tend to reduce or cease alcohol use), which should be addressed in future research.