or search evidence on Discover Portal

Dissemination Centre

Evidence at a glance on Obsessive Compulsive Disorder

NIHR has funded two major studies on OCD treatment. The first, a review of medical and psychological / behavioural treatments, explored which might work best for young people or adults. The second was a large trial involving self-help with support, called OCTET (Obsessive Compulsive Treatment Efficacy Trial). It looked at whether using self-help materials with some support could help while waiting for more intensive treatment with a therapist. Both studies add to what is already known about effectiveness, cost and acceptability of treatments.

 This Highlight considers some of the key points arising from the studies, but it is not a complete review of all the evidence in this area (see About the research for more information).

Which treatments work best according to this research?

Overall, psychological and drug treatments were similar in cost effectiveness. Looking across different trials, the most effective psychological / behavioural treatments seemed to work better than drug treatments but were more expensive.

There were some limitations to the research which mean we cannot be absolutely certain of the findings. For instance, some participants in psychological trials may have continued to take existing drug treatments. However, this is the best research available at the moment.

Psychological / behavioural treatments

In the NIHR review of treatment trials, all of the following psychological / behavioural treatment options worked better than not receiving active treatment:

  • Behavioural therapy - exposure and response prevention, meaning a person is taught to progressively face their fears without carrying out their compulsive rituals.
  • Cognitive behavioural therapy (CBT) - changing how you think and behave.
  • Cognitive therapy - changing your reaction to intrusive thoughts and correcting unhelpful beliefs, like constant self-blame.


  • The review included trials of drugs which affect the neurotransmitter serotonin. Drugs which boost low serotonin levels (such as SSRIs and the tricyclic antidepressant clomipramine) were previously thought to help treat OCD. All the drugs investigated (except venlafaxine and hypericum) worked better than no drug treatment for relieving symptoms of OCD.
  • Most evidence was found on antidepressant drugs, including fluvoxamine, fluoxetine, sertraline, and clomipramine.
  • Although effective, clomipramine seemed to have worse side effects than the other drugs, which may limit its use in practice.

Psychological compared with medical treatments

Overall, in the review the differences between effectiveness of psychological and drug treatments were small.

  • For adults, behavioural therapy and cognitive therapy were more effective than medication. Combinations of drugs with psychotherapy were the next most effective treatments, followed by CBT with sertraline. However, psychological treatments were more expensive than drug treatments.
  • For children, CBT and behavioural therapy worked better than medication, but CBT combined with sertraline was also effective.
  • It was difficult to compare these treatments because patients on psychological treatments were also taking medications in most instances.

Combinations of treatments

  • Taking medication plus receiving psychological therapy seemed to be among the most effective treatments. Combined approaches are recognised in practice as helpful for some people with OCD, generally those with more severe symptoms. However, there wasn’t much evidence on receiving two types of treatment at the same time, compared with the amount of evidence on single treatments.
  • The review did not look at the effect of changing from one type of treatment to another.
  • An NHS pilot trial, Optimal Treatment for OCD (OTO), is currently looking at CBT or treatment with sertraline, compared with combining CBT and taking sertraline.

Supported self-help treatments

  • The OCTET trial found that guided self-help with a workbook or supported self-help with an online CBT programme made no significant improvement to people’s symptoms compared with simply being on a waiting list.
  • However, by a year later fewer people had gone on to CBT treatment with a therapist if they had used self-help with support first.

What does research say about cost effectiveness?

  • In general, the most effective treatments (psychological / behavioural) tended to cost more than drug treatments. So, overall, there wasn’t much to choose between the cost effectiveness of all treatments - more expensive treatments tended to work better whereas cheaper treatments didn’t work as well.
  • The supported self-help treatments were more cost-effective than being on a waiting list for specialist therapy without supported self-help, because people who had received supported self-help were less likely to go on to receive specialist therapy.

Key questions for patients

  • What are my treatment options?
  • How long are waiting lists for treatment?
  • What are the side effects of medication, when might they start, and when will my symptoms start to improve?
  • What will I need to do to get the best out of psychological treatment?
  • Will a therapist work with me to achieve a level of exposure and response prevention that works but isn’t too overwhelming for me?
  • Will it help if I take medication as well as having psychological treatment?
  • What else is available locally, or online, in terms of support or other treatment?

Key questions for health professionals

  • Am I clear on OCD guideline recommendations, referral criteria and service options?
  • Do I understand the importance of shared goals with the person undergoing Exposure Response Prevention treatment?
  • Are there local outreach or support services that help make services accessible?
  • How long are local waiting lists for specialist therapy?
  • Is low intensity supported self-help available for people with OCD in the meantime?

Key questions for commissioners

  • How many people in my area, and who, find it hard to access and engage with services? (for example, people in a multi-ethnic community, people with particular social or health problems, those living in need or experiencing deprivation)
  • Can we learn from successful outreach initiatives in other areas?
  • How do waiting times for psychological treatment and recovery rates for my area compare with other areas, as shown in the NHS Digital IAPT data set?


National Institute for Health Research, Room 132, Richmond House, 79 Whitehall, London, SW1A 2NS

© 2016 - 2019 NIHR, all rights reserved