“Unpicked” by an Expert

I shared some of my reflections on the Cross Party Group on Mental Health and Older People with the senior Scottish Psychiatrist and Government advisor who had given a presentation at the meeting. This was the reply I received. Two years on it is my view that this should be in the public domain. I have not edited the reply other than redacting identifiable names. 

The above quote is that of the physicist Richard P Feynman

Thanks for your thoughts. I think there are a number of inter-related issues that often get conflated. I’d like to try and ‘unpick’ some of these.

  • At an individual patient-level, the benefits vs risks will be different so you’re correct in that it’s hard to translate research findings to every patient. Someone who’s illness has taken their job and relationships may be much keener to remain on antidepressant indefinitely, whereas someone with uncomplicated episodes and some sexual side effects may be keener to stop sooner. The evidence should guide and inform decision-making; rather than ride roughshod over it. The benefits vs risks will depend on the patient group, as well. Some people with a short episode of depression and no significant comorbidity may have a less favourable benefit:risk ratio. Someone who has only just responded to a 10-year-history and who has had 5 previous episodes will have a much better benefit:risk ratio.
  • I’m reluctant to equate antidepressants and antipsychotics. The evidence for each has some similarities but also some important differences. The risk/benefit ratio of long-term treatment will vary according to the type of drug, the dose, and the risks associated with becoming unwell again. In general, the evidence for maintenance therapy in depression is better than in psychosis. However, there is a danger in trying to generalise from the specific to the general. Just because there are people on longer-term treatment may get less benefit doesn’t mean that this is an appropriate response for someone with a high-risk of recurrence and frequent relapses. Many people often assume that because some people experience harms, all people experience more harm than benefit. That’s not the case. There are no reliable predictors of who will response and who won’t (although previous history informs). When making treatment recommendations, we’re trying to increase the likelihood that there is maximum benefit and minimum harm; even though some people will undoubtedly get little benefit and lots of harm. The problem is not the drugs – it’s the system in which we work that reduces the likelihood that patients can have these discussions with their psychiatrist. No reasonable doctor would suggest that seeing a patient every nine months affords sufficient opportunity to optimise treatment and monitor harms.
  • I don’t think that the important issue is whether these drugs are safe/dangerous. The issue is that all treatments have harms but what doesn’t seem to happen frequently enough is that patients are given the opportunity to appraise those risks and benefits in order to make informed decisions. This is an issue that is arguably unrelated to the drugs themselves. The solution is not to conclude that the treatments are dangerous, but to recognise that how they’re given and monitored isn’t good enough for many people.
  • With regards to evidence, there is compelling evidence of benefit for maintenance treatment for antidepressants for at least 3 years (Geddes et al, 2003). For many people, there is some support for prophylactic treatment up to 5 years. However, the 15-year recurrence rate for depression is around 85% so the majority of people with depression will get another episode. Whether someone wants to take drugs to prevent another episode or to stop them and see if they’re in that 15% depends on their experience of recurrent depression and whether they’re getting side effects. However, many people when it’s explained that they’ve only got a one-in-six chance of non-recurrence and that maintenance therapy can reduce the risk of recurrence from 41% to 18%, many people are willing to remain on continuation therapy.
  • The issue about longer-term monitoring is related to, but not the same as, whether continuation therapy is beneficial. If someone is going to remain on medications, there should be some mechanism for monitoring treatment. This is where services often fall short, and if this is fixed, the risk/benefit ratio shifts since you can detect harms and maximise benefit earlier. The solution is not to declare that such drugs shouldn’t be prescribed, but that for people with a high risk of recurrence there should be monitoring in place. It’s a baby-bathwater issue that is complex but needs to be addressed.
  • With regards to the under-recognition of depression, the issue is complex. Like most disorders, there is simultaneous over-diagnosis and under-diagnosis. Most of the Scottish studies seen to have taken the approach of reviewing people on antidepressants, seeing that they’re not depressed, and concluding that the antidepressant was inappropriately prescribed. A good example is Johnson et al, 2012 which seemed to aim to stop antidepressants but didn’t attempt to follow any of the patients up. I’ve yet to find a study doing the same thing for hypertension, angina, or epilepsy: “Hmm, you haven’t had a fit for over a year. Perhaps you don’t have epilepsy and maybe we should stop the drugs.”
  • At a population level (and using the figures from Mitchell, 2009), the rate of false positives (15%) vs false negatives (10%) in detection of depression in primary care has implications. People who are put on antidepressants without a diagnosis of depression may experience adverse effects. In most cases, they’ll stop them. People with depression stop them and someone who doesn’t have depression will as well. In some cases, the GP may review and recommend non-pharmacological approaches instead. However, it means that those in whom depression is missed may experience impairment in occupational and relationship functioning that is much more significant. Can I prove this? No. Do the figures support it? Probably, yes.
  • As you’ll know (and much to the chagrin of the Scottish Government), introducing more psychological therapies (particularly in primary care) seems to result in either no change or increases in antidepressant prescribing (see references below). This is because once you improve one component of the pathway, all parts of the pathway get better. This is important because only two things have been shown to have population-level effects on suicide: 1) reducing access to lethal means; and 2) improving the recognition and treatment of depression (Mann et al, 2005).
  • I think you probably overestimate my ability to influence colleagues and key Opinion Leaders that have ties to industry. A bigger problem is that everyone has conflicts of interest. If I’ve just published a book on my new therapy and I run courses for training, I have a conflict of interest when it comes to promoting my therapy. One of the problems is that these conflicts are often undeclared. Who is most conflicted? A doctor who got £500 for speaking about a new antipsychotic, or someone whose company turns over £40,000/ year in book revenue, training materials, and courses? The problem is that there is only an expectation that the former should declare their interests. It’s akin to diagnostic overshadowing whereby we focus so much on pharma that we forget all the other conflicts. My biggest conflict is that I’m paid by the NHS to treat depression and OCD. If the Royal College of Psychiatrists were to handover old-age psychiatry to general psychiatrists and/or neurologists, you would presumably have a conflict should you attempt to extol the virtues of old-age psychiatry.
  • And yes, I do support something akin to a Sunshine Act. Most NHS Boards expect it, but they don’t enforce it. It should be easy to implement and report on, and it shouldn’t be voluntary. The biggest challenge relates to how it is compiled. If it’s too burdensome, it won’t be done by the individual and industry will be reluctant to openly disclose all payments. One of the concerns that a lot of people have is that industry funding for research is notionally allocated to an individual which makes it look as though they’re receiving large sums of money, whereas it’s probably paying for research assistants. The interesting thing is that the general public is much more excited about conflicts of interest relating to antidepressants (e.g. “doctors in the pay of big pharma”) but is often less excited about pharma sponsorship for anti-dementia drugs (e.g. “of course we need these drugs, don’t stand in their way”). This reflects stigma about some disorders.

The references provided to support  the above statements:

Mitchell, A. J., Vaze, A. & Rao, S. (2009) Clinical diagnosis of depression in primary care: a meta-analysis. The Lancet, 374, 609-619. http://dx.doi.org/10.1016/S0140-6736(09)60879-5

Rost, K., Zhang, M., Fortney, J., et al (1998) Persistently poor outcomes of undetected major depression in primary care. General Hospital Psychiatry, 20, 12-20. http://dx.doi.org/10.1016/S0163-8343(97)00095-9

Geddes, J. R., Carney, S. M., Davies, C., et al (2003) Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet, 361, 653-661. http://dx.doi.org/10.1016/S0140-6736(03)12599-8

Mueller, T. I., Leon, A. C., Keller, M. B., et al (1999) Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. American Journal of Psychiatry, 156, 1000-1006. http://dx.doi.org/10.1176/ajp.156.7.1000

Mann, J. J., Apter, A., Bertolote, J., et al (2005) Suicide prevention strategies: a systematic review. JAMA, 294, 2064-2074. http://dx.doi.org/10.1001/jama.294.16.2064

Robinson, P. (2009) Depression, ethnicity and talking therapy. HSJ Intelligence (5 February 2009), 9. http://www.hsj.co.uk/Journals/2/Files/2009/4/1/4-1986238.pdf

Sreeharan, V., Madden, H., Lee, J. T., et al (2013) Improving Access to Psychological Therapies and antidepressant prescribing rates in England: a longitudinal time-series analysis. British Journal of General Practice, 63, e649-e653. http://dx.doi.org/10.3399/bjgp13X671641

Williams, C., Wilson, P., Morrison, J., et al (2013) Guided Self-Help Cognitive Behavioural Therapy for Depression in Primary Care: A Randomised Controlled Trial. PLoS ONE, 8, e52735. http://dx.doi.org/10.1371%2Fjournal.pone.0052735

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s