Citation

Mines D, Hill D, Yu H, Novelli L. Pharmacoepidemiol. Drug Saf. 2005; 14(6): 367-372.

Affiliation

Global Safety Surveillance and Epidemiology, Wyeth Research, Collegeville, PA 19426-3930, USA. minesd@wyeth.com

Copyright

(Copyright © 2005, John Wiley and Sons)

DOI

10.1002/pds.1095

PMID

15924331

Abstract

PURPOSE: Three recent observational studies reported that the risk of fatal overdose is greater with venlafaxine than SSRI use. It is not clear whether patient factors could account for this finding. We evaluated whether risk factors for suicide are more prevalent among patients prescribed venlafaxine than patients prescribed fluoxetine or citalopram. METHODS: Using data from the UK General Practice Research Database (GPRD), we identified patients who were first prescribed any of the three drugs between January 1995 and April 2002. We ascertained risk factors for suicide documented in the 1 year before that first prescription. Separate analyses compared venlafaxine (N = 27 096) and fluoxetine (N = 134 996) cohorts, and venlafaxine and citalopram (N = 52 035) cohorts. RESULTS: Previous suicidal behaviors were documented for 1.0% of the venlafaxine cohort compared to 0.4% of the fluoxetine cohort (OR 2.8, 95%CI: 2.4, 3.2) and 0.4% citalopram cohorts (OR 2.4, 95%CI: 2.0, 2.9). 72.5% of venlafaxine patients had been prescribed at least one other antidepressant compared to 27.6% of fluoxetine (OR 6.9, 95%CI: 6.7, 7.1) and 39.5% of citalopram (OR 4.0, 95%CI: 3.9, 4.2) patients. Venlafaxine patients were also four to six times as likely to have been previously hospitalized for depression. CONCLUSION: In the UK, venlafaxine has been selectively prescribed to a patient population with a higher burden of suicide risk factors than patients prescribed fluoxetine and citalopram. Unless baseline population differences are accounted for, observational studies that compare the risk of suicide in patients receiving these agents may produce biased results.

Language: en