Intended for healthcare professionals

CCBYNC Open access
Research

Association between clinically recorded alcohol consumption and initial presentation of 12 cardiovascular diseases: population based cohort study using linked health records

BMJ 2017; 356 doi: https://doi.org/10.1136/bmj.j909 (Published 22 March 2017) Cite this as: BMJ 2017;356:j909

Association is not causation - even with big data

Using data on almost 2 million patients, experiencing more than 100, 000 cardiovascular events, Bell et al. nicely showed the complex association between alcohol intake and cardiovascular diseases (CVD) (1). It is however not clear what this type of research adds in solving the issue of the cardiovascular effects of alcohol. Many studies have already been published on the association between alcohol and the risk of CVD. For making clear public health recommendations about alcohol intake, we need proof of the causal relationship between alcohol and CVD; associations are useless (2). Reassessing the association, even if it is based on big observational data, will not help policy makers or future "precision medicine" physicians decide what to recommend.

1) Bell et al BMJ 2017
2) Glass TA, et al. Causal inference in public health. Annu Rev Public Health. 2013

Competing interests: No competing interests

14 April 2017
Arnaud Chiolero
Public health physician
Institute of Primary Health Care (BIHAM), University of Bern
Bern, Switzerland