Case reports of adverse reactions with psychotropic drugs can be useful in raising hypotheses, to be tested with more rigorous study designs. However such reports have significant methodological drawbacks, making it hard to determine causality.
We undertook a systematic assessment of the quality of case reports and small case series published within Human Psychopharmacology over 25 years. For reports of adverse drug reactions, modified Bradford Hill criteria for causality (for consistency, strength, specificity, temporal relationship and plausibility) were used to ascertain the quality of the account. Reports which had been cited at least 10 times by December 2010 were examined in detail, to assess their overall contribution in extending understanding and influencing clinical practice.
Of 40 reports of adverse drug reactions, only seven were sufficiently robust for confidence in probable or possible causality. Nine reports had been cited more than 10 times: the five most frequently cited reports of adverse drug reactions described movement disorders, suicidal thoughts and discontinuation symptoms with SSRIs: clinical relevance was high, but their quality was not markedly greater than less frequently cited reports.
Nearly all reports of adverse drug reactions, published in a single journal over 25 years, were insufficiently robust to demonstrate probable causality. Reports that are cited frequently become influential because of their potential clinical relevance, rather than due to their methodological quality. Copyright © 2013 John Wiley & Sons, Ltd.