Antidepressant Tx for anxiety disorders: How long?

CAVEATS

Potential bias … bias … and more bias

While the authors used standard and appropriate methodologies for this type of study, some significant threats to validity remained. All but 2 studies in the analysis were industry funded. Publication bias is another potential issue, even though the authors identified and included 6 unpublished studies, 4 of which had negative results.

This study provides evidence that stopping antidepressant treatment for anxiety disorders before 1 year increases the risk of relapse.

Additionally, the authors graded 11 of 28 trials as having a high likelihood of selective reporting bias, meaning that important portions of the original studies’ results may not have been published. Most studies were at high risk for attrition bias, resulting in loss of information when patients dropped out of the study. While this happened more often in the stopping groups, it is still possible that there are unidentified harms or unexpected outcomes in the medication groups.

While PTSD and OCD are no longer considered anxiety disorders, subgroup analyses found no difference in relapse rates between these diagnoses and the others included in the studies. Finally, treatment duration longer than 52 weeks has not been studied, so the optimal treatment duration is unknown.

CHALLENGES TO IMPLEMENTATION

Patients may resist continuing treatment once symptoms abate

Some patients may want to discontinue antidepressant treatment if their anxiety symptoms improve prior to 1 year. It may be difficult to convince them that continuing treatment will prevent relapse of their condition. Providing patients with information about the increased relapse rate with stopping medication early (with an estimated number needed to treat of 5) may help patients make a more informed decision.

ACKNOWLEDGMENT

The PURLs Surveillance System was supported in part by Grant Number UL1RR024999 from the National Center For Research Resources, a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center For Research Resources or the National Institutes of Health.