Mental health

Mental health risks of obesity drug

The new anti-obesity drug rimonabant (brand name Acomplia) is associated with increasing the chances of depression and suicidal thoughts, reported the Daily Mail . The drug was launched in the UK 18 months ago and is currently taken by tens of thousands of people to help weight loss; but “evidence suggests that one in 10 of those using Acomplia may develop side effects such as low mood anxiety, irritability, nervousness and sleep disorders”, the newspaper said._ The Daily Telegraph_ said that patients taking the drug “are 2.5 times more likely to stop taking the medication because they become depressed and three times more likely to stop because of anxiety”.

The stories are based on the findings of a combined review of several studies that have looked into rimonabant’s safety and efficacy. Rimonabant has recently been assessed by the US Food and Drug Administration (FDA) in the wake of concerns about its mental health side-effects. The FDA has suggested that more information is needed about the long-term safety effects of rimonabant. The medical profession has been alerted to the increased risk of depression and anxiety with the use of rimonabant, and in the UK, doctors have been told it should not be prescribed for anyone suffering from depression or taking antidepressant drugs. The study behind today's headlines adds weight to these concerns.

Where did the story come from?

Robin Christensen and colleagues from Frederiksberg Hospital and University of Copenhagen, Denmark have carried out the research. Funding was provided by the Centre for Pharmacogenomics, University of Copenhagen, The Oak Foundation, The H:S Research Foundation, and Diabesity EC-FP6. It was published in the peer-reviewed medical journal The Lancet .

What kind of scientific study was this?

This was a systematic review and meta-analysis – a 'study of studies' – where the researchers combined the results of several studies examining the safety and efficacy of rimonabant, with particular emphasis on psychiatric adverse effects. It followed  concerns about the risk of depression and suicidal thoughts in people taking rimonabant.

The researchers looked through computer databases to identify all trials published up until November 2006 that had investigated the effects of rimonabant for weight loss compared with a placebo (a dummy pill). They only looked at double-blind randomised trials where the participants had been randomly allocated to rimonabant or a placebo and where neither the people in the trials nor the researchers knew whether they had received placebo or the active drug. The studies included only those people that were eligible for anti-obese drug treatment, that is, they had a body mass index (BMI) of 30; or a BMI of 27 or more as well as one or more medical problems that may be related to obesity (e.g. diabetes or heart disease).

The researchers looked at the average weight loss in the studies, and the number of participants that achieved at least a 10% weight reduction with treatment. They also looked at depression and anxiety effects by examining scores on the Hospital Anxiety and Depression Scale (HADS score – a standard scale for measuring depressive symptoms). They used statistical methods to account for the differences in the different methods used in each of the trials.

What were the results of the study?

The researchers identified only four randomised controlled trials that used reliable methods and had available information on the effects of rimonabant after one year of use. All of the studies were part of the RIO (Rimonabant In Obesity)  programme that aimed to investigate the effects of rimonabant upon obesity, including groups of patients with diabetes and cholesterol problems, and were carried out in multiple centres across the US and Europe. Together, they examined 4,105 overweight people over one to two years. All of the studies were similar. They looked at the same outcomes, and required participants to follow a four week programme of diet control before starting rimonabant or placebo treatment whilst continuing with diet control. Each of the studies was sponsored by the company that manufactures rimonabant.

Those treated with rimonabant achieved significantly greater weight loss (4.7kg) at one year compared with placebo, and were five times more likely to achieve the target of at least a 10% loss in their weight. There was no difference in depression score between the rimonabant and placebo groups at entry to the studies, but a slightly higher anxiety score in the rimonabant group compared with the placebo group at entry. However, there was a 40% increased risk of either any adverse effect, or any serious adverse event occurring whilst taking rimonabant compared with placebo (examples of these events are not detailed by the report).

The researchers also found that those taking rimonabant were two and a half times more likely to discontinue the drug due to depressive symptoms and three times more likely to discontinue due to anxiety, than those taking placebo. The studies did not examine the effects of the drug on suicidal thoughts.

What interpretations did the researchers draw from these results?

The researchers conclude that rimonabant was more effective than placebo in achieving weight loss when used to treat obesity. However, they agree with the FDA report that rimonabant increases the risk of psychiatric adverse effects such as depression and anxiety. They say that with these findings they “recommend increased alertness by physicians to these potentially severe psychiatric adverse reactions”.

What does the NHS Knowledge Service make of this study?

This is a reliable review, which suggests that further research balancing the beneficial effects of rimonabant against its potential harms is needed. Some points to consider:

  • Although rimonabant was demonstrated to be effective for achieving weight loss, these studies have not been able to assess the longer-term effects of the drug or whether weight could have been regained once discontinuing. They also could not examine whether mood and anxiety problems returned to normal once rimonabant was stopped. 
  • From this analysis, there is evidence for an increased risk of any adverse effects and withdrawal of the medication due to depression or anxiety only. This paper did not examine the effects upon suicidal thoughts, which was a concern raised by the US FDA. 
  • The studies cannot account for the fact that people being treated for obesity are known to be at greater risk of depression anyway. 
  • This review also has limitations introduced by the often inconsistent reporting of adverse effects by trials, the reliance upon the study participant’s self report of depression and anxiety symptoms (which may be under reported), and the inability to look in more detail at the type of mental health problems caused, other than a general classification of depression or anxiety.  
  • A healthy calorie-controlled diet is an essential part of obesity management, alongside drug treatment. Anti-obesity drug treatment is only prescribed to those who have failed to control their weight through diet and exercise, and who fulfil the criteria mentioned above.

The medical profession and regulatory bodies are aware of the increased risk of depression and anxiety with use of rimonabant. It is not licensed for use in anyone suffering from depression or taking antidepressant drugs, and should be used with caution in anyone with past depression or where there is concern about mood or anxiety problems. Based on these findings, it seems sensible that if rimonabant is prescribed for obesity, a review of the person’s psychological and emotional health should be as much a part of GP follow-up as assessment of their weight loss. Further research is needed into the safety and beneficial effects of this drug, to review its place in the management of obesity.

In addition, it is reassuring that the two meta-analysis involving obesity drugs reported today, from different authors and journals, report the same studies and give the same result.

Sir Muir Gray adds...

All medical intervention carries the risk of harm; the best intervention to lose weight is to walk an extra 60 minutes a day. To keep the weight off once you have achieved your target, walk an extra 30 minutes a day; this way there is little risk of side effects.


NHS Attribution