[Skip to Navigation]
Views 260
Citations 0
Comment & Response
February 2015

Multiple Sclerosis and Alcohol Misuse—Reply

Author Affiliations
  • 1Oxford University Medical School, John Radcliffe Hospital, Oxford, England
  • 2Unit of Health-Care Epidemiology, Nuffield Department of Population Health, University of Oxford, Oxford, England
JAMA Neurol. 2015;72(2):238. doi:10.1001/jamaneurol.2014.3879

In Reply We thank Dr Voci for his comments on our article1 reporting a positive association between alcohol misuse disorders and subsequent multiple sclerosis (MS), particularly in men. The possibility of reverse causality whereby alcohol misuse may be a consequence of as yet undiagnosed MS is a necessary consideration. Using the same method and reference cohort as in our original study, we analyzed the data set for the risk for alcohol misuse disorders following an admission for MS. Comparing the MS cohort with the control cohort, the rate ratios were 1.04 (95% CI, 0.59-1.69; P = .98), 1.13 (95% CI, 1.02-1.25; P = .02), and 1.01 (95% CI, 0.91-1.13; P = .82) for alcohol use, alcohol abuse, and alcohol dependence, respectively. In reducing the possibility of surveillance bias and reverse causality, including only cases of alcohol misuse disorders observed at least 1 year following the first admission for MS, the rate ratios were 0.84 (95% CI, 0.42-1.50; P = .65), 0.98 (95% CI, 0.86-1.11; P = .82), and 0.88 (95% CI, 0.77-1.00; P = .06) for alcohol use, abuse, and dependence, respectively. Although not significant, low rates of alcohol misuse disorders in the later years after first MS diagnosis lead us to speculate that perhaps, after a diagnosis of MS, people with it become more health conscious. These findings—combined with our previous results that there was a significantly elevated risk for MS within 1 year of first admission for alcohol abuse only and a significantly elevated risk for MS following all alcohol misuse disorders with an interval of more than 1 year between first recorded admission with the alcohol misuse disorder and first-recorded admission with MS—make us consider reverse causality to be unlikely to explain our findings. Of note, Dr Voci’s comment that reverse causality is likely seen in another study reported from this data set is not supported by the results of that study where, as we have previously reported, the positive association was similarly only significant in 1 direction.2

Add or change institution