Letters Antidepressants v cognitive behavioural therapies

Patients in trials of antidepressants are not typical of those in everyday practice

BMJ 2016; 352 doi: https://doi.org/10.1136/bmj.i577 (Published 10 February 2016) Cite this as: BMJ 2016;352:i577
  1. Ian Hamilton, lecturer
  1. Department of Health Sciences, University of York, York YO10 5DD, UK
  1. ian.hamilton{at}york.ac.uk

Amick and colleagues’ review should help clinicians and patients when choosing treatments for depression.1 This review is welcome given the prevalence of depression and the paucity of research on the effectiveness of treatment.

However, most of the evidence came from pragmatic randomised controlled trials, so how closely do such trials reflect everyday clinical practice? The inclusion and exclusion criteria used in the primary studies offer a clue.

Many people with depression will probably have other problems such as substance misuse.2 Yet people who use substances, including alcohol, are routinely excluded from trials. This is often justified by the higher dropout rates in this group and the fact that substance misuse can complicate the intervention and outcome. However, I’m not sure how useful it is for information to be based on atypical patients.

Compounding this mismatch between research and practice is the way people often use a range of substances to ameliorate their unwanted feelings, including low mood. Again in reality many people with depression use other substances along with prescribed drugs.3 We need more information about potential drug-drug interactions in these patients.

All these factors might pose difficulties for research, but for the evidence to be useful we need to find ways of including people in trials who also turn up in everyday practice.


  • Competing interests: None declared.


View Abstract

Sign in

Log in through your institution