Does Chantix Cause Mental Health Problems?
Some of the details of this tragedy can be found online via the ABC news story but the key points were that Mr Albrecht was actually killed by being shot in the head by a neighbor as he banged on the neighbor’s door, but that his girlfriend felt that his bizarre behavior may have been caused by the Chantix he was taking at the time.
Regular readers of this blog will know that this issue has come up before – see:
“Chantix: how does this new stop smoking medicine work?” 4/15/07
“Chantix and mental illness: what are the facts?” 08/12/07
“Two new studies of Chantix (varenicline)”. 08/19/07
There have been a number of comments from people who experienced frightening dreams, anxiety attacks, depression and suicidal thoughts while on the medicine, as well as the suicide of a family member. The number and pattern of these comments were sufficient to cause me some concern and so I decided to take another look at the published reports of the clinical trials of Chantix and also speak to a number of colleagues who treat many patients with Chantix, in order to try to get a sense of whether these experiences may be caused by Chantix, rather than other potential causes (including nicotine withdrawal effects).
On looking at the evidence from the clinical trials, it is more consistent with the idea that Chantix reduces depressive thoughts, rather than increases them. For example, one large study was published in JAMA on July 2006 comparing the outcomes of 352 smokers treated with Chantix (varenicline), 329 people treated with Zyban (bupropion) and 344 people treated with identical placebo pills. This was a randomized double-blind trial meaning that no-one knew which type of pills they received until the end. 22% quit completely for a year on Chantix, as did 16% on Zyban and 8% on placebo. The paper reported on changes in “negative affect” (a combination of unpleasant mood symptoms including depression and irritability). Patients on Chantix reported a significantly SMALLER increase in these symptoms than patients taking placebo. Zyban had a similar effect of reducing negative affect compared with placebo pills. The paper also listed adverse events reported by participants. The main symptom that was clearly reported more frequently by Chantix users was nausea, reported by 28% of Chantix users, compared with 13% on Zyban and 8% on placebo. Of the psychiatric disorders mentioned, only “abnormal dreams” appeared to be more common on Chantix (10%), compared with 6% on Zyban and 6% on placebo. There was no clear difference in reports of serious irritability (6%, 5%, 6%) and fewer patients on Chantix reported insomnia (14%) than did patients on Zyban (22%). In terms of “serious” adverse events, these were no more common for Chantix than placebo and the single case of a serious psychiatric event (acute exacerbation of schizophrenia) occurred in a patient taking placebo pills.
Another almost identical trial was reported by Jorenby and colleagues in the same issue of JAMA, with very similar results (i.e. higher quit rates with Chantix, along with lower reported negative affect [mood] than placebo, but higher rates of nausea.). Serious adverse events again were rare and scattered evenly across the different types of pills with little clear pattern, but there was one report of “acute psychosis, emotional lability” in the Chantix group (out of 344 taking Chantix). This study did, however, find a higher rate of “abnormal dreams” on Chantix (13%) than Zyban (6%) or placebo (4%). The earlier studies designed to identify the best dose of Chantix also had similar findings (dose-dependent increase in nausea and abnormal dreams) but no real evidence of other mental health symptoms. For example, Nides and colleagues found 10% on placebo and 12% on the high dose of Chantix reported serious “irritability’, and that “depression was not observed as an adverse event with varenicline (Chantix) treatment.”
The data sheet for prescribers of Chantix notes that 4500 people were exposed to Chantix during its premarketing development and that discontinuation of treatment due to adverse events was rare. The most frequent reason was nausea (3% for Chantix versus 0.5% for placebo). 0.3% reported discontinuing Chantix because of abnormal dreams as did 0.2% on placebo pills. As with all medications, the data sheet has a long list of symptoms experienced by participants in the trials, including “Psychotic disorder, suicidal ideation” as “rare”. Note – this does not imply that the drug caused these events – just that they occurred rarely in people taking the drug. Overall, the pattern of results from trials of Chantix suggest that with the exception of abnormal or vivid dreams, psychiatric symptoms such as depression or negative affect are LESS likely to occur in people taking Chantix to quit smoking, than in people taking placebo pills while quitting smoking.
However, one has to bear in mind that early clinical trials typically exclude patients currently being treated for mental health and other serious health problems. So the possibility remains that the drug may cause problems in types of patients that were not included in the initial trials. That’s where post-marketing surveillance is important. This is something that the pharmaceutical companies and doctors routinely carry out. For my part, I simply asked a large group of colleagues who are experienced in treating “real patients” with Chantix and other treatments, whether they had noticed any signs of worsening mental health associated with Chantix use. The clinicians I spoke to estimated that they had been involved in the treatment of over 2000 patients with Chantix, including patients with co-occurring serious mental health and other medical problems. There was a pretty clear consensus that while there were a few isolated cases (a couple) of patients reporting mental health problems, these were not noticeably more frequent than one normally encounters with other treatments (e.g. nicotine replacement or Zyban, or counseling with no medication).
So overall I am somewhat reassured that Chantix is a safe medicine that is effective at helping smokers to quit. But why the rash of reports on the internet of depression and bizarre behavior? Firstly, I don’t doubt that these people’s experiences are real and in some cases, very serious. I also think it is plausible that some (probably a minority) could be directly linked to Chantix. In some cases it could be an unusual interaction between the individual, the medicine and maybe another drug (including alcohol) they are taking. But for most, I suspect the serious behavioral/psychiatric problems experienced are unlikely to be caused by Chantix. Here are my reasons:
1. For highly addicted smokers, mood disturbance and altered thinking is common when quitting smoking, even without taking any medication.
2. The evidence described above, indicates that with the exception of abnormal dreams, Chantix reduces the severity of mood/psychological disturbance experienced while trying to quit smoking.
3. Around 3 million Americans have taken Chantix to try to quit smoking. Among that many smokers trying to quit for a month or two, one would expect a few thousand or more to have serious symptoms of depression etc even if they were not taking a medicine to help them. But when someone has these symptoms while taking a new drug, it is perfectly natural to conclude that the drug may have caused the symptom. In these days of widespread internet access, chat-rooms etc, that easily turns into a few hundred patients reporting similar symptoms on the internet while taking the same drug.
It therefore appears that if Chantix causes any serious mental health problems at all (which remains unproven), it is extremely rare (perhaps in the order of one per thousand). So my advice is that if you are considering quitting smoking and are interested in taking an FDA-approved medicine, whether it be nicotine replacement therapy, bupropion or Chantix, then you should not be put off by relatively isolated reports of side effects. The highest quality of evidence (from randomized placebo-controlled trials) demonstrates that these medicines are safe and will roughly double your chances of successfully quitting smoking. However, everyone reacts to medicines differently, and if you start to experience a worrying symptom that you believe may be caused by the medicine you should consult your doctor immediately. Even better, when you see your doctor to obtain a prescription, you should arrange a follow-up visit within a week or so of starting the medicine in order to discuss your progress, side-effects etc. If you have any concerns between appointments, call your doctor. It is also wise to get as much additional support from friends, family, telephone quitlines etc as possible. There is a national (US) toll-free number for telephone counseling (1-877-448-7848) and in the case of Chantix users in the US, there is additional support available via http://www.chantix.com/ .
Finally, anyone who believes there to be a causal link between use of a medication and a severe adverse event (e.g. depression, suicidal ideation, suicide, or any other serious adverse event), whether it be in yourself, your patient or a family member, should report it to the MedWatch program at: http://www.fda.gov/medwatch/how.htm . This is one of the main mechanisms of post-marketing surveillance that can help identify rare or previously unknown risks from medicines.