No strong evidence linking varenicline to depression or criminality

New research provides reassurance for smokers using medication to help them quit.

There is no strong evidence that use of the smoking cessation drug varenicline (Pfizer’s Champix) leads to an increased risk of suicidal behaviours, criminal offending, transport accidents, traffic-related offences or psychoses, according to research

There is no strong evidence that a drug used by smokers attempting to quit leads to an increased risk of suicidal behaviour, criminal activity, road accidents or psychoses, according to research published in The BMJ
on 2 June 2015.

Varenicline (Pfizer’s Champix) is associated with a small increased propensity for anxiety in individuals who have a pre-existing mental health issue. However, it is unknown whether this is linked to the process of giving up smoking or the drug, say the authors of the research.

The findings, according to lead researcher Seena Fazel, senior research fellow in clinical science at the University of Oxford, provide reassurance to smokers about the risks and benefits of taking the drug in their attempt to quit. They will also be useful for people working in the transport industry — including pilots and bus drivers — where varenicline has either been restricted or banned because of previous reports of an associated risk of traffic accidents.

But Fazel stopped short of saying that the official Europe-wide safety warnings attached to the product — which recommend that patients seek advice if they develop specific mental health problems and should be monitored if they already have a mental health condition — should be amended. “I would be very careful, on the basis of this study, to say that there should be a safety review,” says Fazel. “But I think it should go into a future review when the guidelines are next looked at.”

The research, an observational study, comes just two months after another clinical trial, also published in The BMJ, boosted confidence in prescribers and users of varenicline regarding its association with specific mental health problems.

The latest varenicline research was based on an analysis of the adult population of Sweden including 69,757 people who were treated with varenicline between 2006 and 2009.

Using national registers, the researchers compared the incidence of new psychiatric conditions, suicidal behaviour, suspected and convicted criminal offences, road accidents and suspected and convicted traffic offences between the two population groups.

In the varenicline-treated population, the proportion suspected of a crime was 5.4% compared with 4.0% in the non-treated population; 4.6% were diagnosed with a new psychiatric condition compared with 2.2% in the non-treated population; the rate of serious traffic-related incidents was 1.4% in the varenicline group — the same percentage for non-treated patients; and 0.9% of varenicline patients received treatment for suicidal disorders compared with 0.3% in the non-treated population.

When the researchers compared the data, they found increased risks for all the adverse events investigated. But when they compared periods of treatment with non-treatment for the same person, thereby adjusting for any confounding by indication, they found no evidence of increased risk for these parameters.

The negative findings for a link between varenicline use and increased risk of traffic accidents and offences are at odds with previous post-marketing surveillance data reported to medicines regulators, but the researchers suggest that the signal identified previously may have resulted from the overall higher rates of traffic accidents among smokers.

Varenicline was associated with an increased risk for anxiety (hazard ratio [HR] 1.27, 95% confidence interval [CI] 1.06–1.51) and mood (HR 1.28, 95% CI 1.07–1.52) conditions for those patients who already had a pre-existing mental health disorder. The associations were not significant for psychoses.


[1] Molero Y, Lichtenstein P, Zetterqvist J et al. Varenicline and risk of psychiatric conditions, suicidal behaviour, criminal offending, and transport accidents and offences: population based cohort study. The BMJ 2015;350:h2388. doi: 10.1136/bmj.h2388.

Last updated
The Pharmaceutical Journal, PJ, 27 June/4 July 2015, Vol 294, No 7868/9;294(7868):DOI:10.1211/PJ.2015.20068752

You may also be interested in