Scientific Publications by FDA Staff

Addiction 2016 Jul;111(7):1283-92

A comparison of neuropsychiatric adverse events during early treatment with varenicline or a nicotine patch.

Cunningham FE, Hur K, Dong D, Miller DR, Zhang R, Wei X, McCarren M, Mosholder AD, Graham DJ, Aspinall SL, Good CB

AIMS: We compared the risk of mental health episodes requiring hospitalization (primary aim) or outpatient clinic visits (secondary aim) associated with varenicline versus the nicotine patch (NP) in an era prior to psychiatric boxed warnings. DESIGN: Retrospective cohort SETTING: Department of Veterans Affairs (VA), USA PARTICIPANTS: VA patients with or without psychiatric comorbidities and a new prescription for varenicline (15,255) were propensity score matched (1:2) to new users of NP (123,054) between May 1, 2006 and September 30, 2007 resulting in 11,774 and 23,548 patients in the varenicline and NP groups, respectively. MEASUREMENTS: The primary outcomes were hospitalizations with a primary discharge diagnosis of a range of mental health disorders depression, schizophrenia, bipolar disorder, suicide attempt, post traumatic stress disorder, other psychosis, and drug-induced mental disorders. Secondary outcomes were outpatient clinic visits with a primary diagnosis of the above list of mental health disorders. FINDINGS: Background characteristics of the treatment groups were similar after matching. There was no statistically significant difference in risk of hospitalization for any of the studied mental health disorders with varenicline compared with NP. Among secondary outcomes there was an increased risk of outpatient clinic visits for schizophrenia among patients who received varenicline (HR 1.27; 95% CI 1.07,1.51), this increase being evident only in those with a pre-existing mental health disorder. CONCLUSION: In US VA patients studied prior to the boxed warning being implemented, use of varenicline for smoking cessation was not associated with a detectable increase compared with nicotine patches in hospitalization for any mental health outcomes. There was an increased rate of outpatient attendances with a primary diagnosis of schizophrenia amounting to 5 per 100 person years of treatment. This increase was only found in patients with a pre-existing mental health disorder.