Chantix safety at the US Veterans Affairs Health Service
The report stated that between September 2006 and April 30, 2008 there were a total of 147,718 prescriptions for varenicline (69,765 unique patients) within the VA. As of May 26 2008 , VA healthcare providers had submitted 417 reports on possible adverse events associated with varenicline. The most common type of adverse event was vomiting (95), followed by nightmares/abnormal dreams (82), depression (68), agitation (33), suicidal ideation (31) and hallucinations (25). Other adverse events were reported much less frequently.
So the overall rate of adverse event reports appears to be around 6 per thousand, of which almost half consist of vomiting, abnormal dreams and other non-life threatening symptoms that are recognized as rarely occurring side effects of the drug. Interpretation of the rates of the other (primarily psychological) symptoms needs to bare in mind that the VA healthcare system provides fairly comprehensive and accessible mental health services and the rates of mental health problems in VA patients is therefore known to be higher than in the general population. For example, one recent study found that 40% of male VA patients were suffering from a current mental health disorder.
With this background, and an annual rate of suicide attempts of 1% among adult smokers, the numbers of patient being reported as suffering these adverse events in the VA system does not appear to be particularly high. One patient was known to have committed suicide in association with varenicline use, and although every case of suicide is tragic, one case in almost 70,000 patients in the VA system is almost certainly less than the background rate of suicide in the VA system among patients not using Chantix. I am not saying that Chantix may prevent these symptoms or behaviors. However, even after allowing for under-reporting of adverse events (which is normal), the frequency with which these events are being reported among Chantix users does not appear to be above the frequency that we would expect these types of events to be occurring in patients not using Chantix (in fact its much lower). What this means to me is that there is nothing in this new data that stands out as a clear signal that Chantix likely caused these events (other than perhaps vomiting and abnormal dreams, which are known to be increased by Chantix).
Of course it is very difficult to interpret adverse event reports, which are not necessarily recorded in a very systematic manner. The data we have so far suggest the following:
1. Prescribers should continue to use varenicline as one of the first-line medicines shown to help treat nicotine dependence.
2. Prescribers should make patients aware of the potential side effects, including the relatively frequent effects that have been shown to be caused by the drug (e.g. mild nausea in around 30% of patients), as well as the much less frequent symptoms that have not been proven to be caused by the drug (e.g. depression).
3. Providers should follow the US Public Health Service Guideline that recommends not only assisting smokers to quit by prescribing an effective medicine, but also arranging follow-up with patients to monitor progress and side-effects etc. Patients should also be told to notify their doctor immediately if they experience any serious side-effects between appointments.
4. Patients should continue to consider varenicline as a safe and effective treatment for nicotine dependence, but should be prepared to keep in contact with their healthcare provider and to report onset of unpleasant/serious symptoms to that provider if they occur.
5. Additional placebo-controlled trials in more typical patient samples should be carried out, to provide further high-quality data on efficacy and safety.