CBT for gamblers: new Cochrane review finds evidence promising but problematic

How many of you buy lottery or raffle tickets, or perhaps chuck a quid into the office sweepstake on Derby Day? It’s every day stuff, with up to 85% of adults in some Western countries reporting some gambling activity in the past year. The Elf husband likes to complain that in all the years of being ‘encouraged’ to buy raffle tickets for various worthy causes his only rewards have been a gold lamé clutch bag and a yoghurt. But for some, gambling can spiral out of control. Pathological and problem gambling is the subject of a new Cochrane review which, like the one I blogged about on Monday on pyschosocial therapies designed to reducing problem drinking, aimed to evaluate the effectiveness of talking treatments.

This review has avoided the problem encountered by the alcohol review, namely the difficulty of using ‘treatment as usual’ as a comparison with the intervention of interest because they generally share key components. Studies making such comparisons were excluded and instead the reviewers looked for randomised controlled trials (RCTs) with control groups who were not receiving treatment but were perhaps on a waiting list, or who were referred to Gamblers Anonymous (GA). They regarded GA as a ‘real world’ control which accounts for recovery over time. Typically there is no requirement to attend sessions and attendance is either low or unreported. Fouteen RTCs with 1245 people were included, eleven of the studies evaluating cognitive behavioural therapy (CBT).

Here’s what they found:

  • People having CBT showed significant improvements in how often they gambled, gambling symptom severity and financial loss from gambling, up to three months after treatment
  • Motivational interviewing for less severe gambling showed some benefit in terms of financial loss and gambling frequency
  • No benefit was seen in two studies of integrative therapies for people with low gambling severity scores
  • One study of 12-step therapy showed some short-term benefit in most outcomes


  • There were few studies suitable for inclusion in the review and they are of variable quality, many of them with multiple limitations. The reviewers caution that the beneficial effects of therapy may have been overestimated
  • There was too little evidence on therapies others than CBT for conclusions to be drawn
  • There is a lack of evidence on longer-term effects of treatment

The reviewers conclude that the current evidence supports CBT for the treatment of pathological and problem gambling and is consistent with recommendations that CBT should be classified as best practice, but also that it has highlighted important caveats to this advice, with a lack of high quality evidence and continuing uncertainty about whether any short term benefits are sustained.

It’s important to know what we don’t know and I can only hope the frequent findings in reviews about the design and reporting of trials are having an impact on the trials that follow in their wake.


Cowlishaw  S, Merkouris  S, Dowling  N, Anderson  C, Jackson  A, Thomas  S. Psychological therapies for pathological and problem gambling. Cochrane Database of Systematic Reviews  2012, Issue 11. Art. No.: CD008937. DOI: 10.1002/14651858.CD008937.pub2 Cochrane summary