Intended for healthcare professionals

Editorials

NICE guidelines for the management of depression

BMJ 2005; 330 doi: https://doi.org/10.1136/bmj.330.7486.267 (Published 03 February 2005) Cite this as: BMJ 2005;330:267

Lack of clinical significance of antidepressants

I am not as convinced as Middleton et al1 that the issue of severity of depression and response to antidepressant treatment is as clear as the NICE guideline makes out. The full guideline2 gives the following evidence statement about serotonin specific reuptake inhibitors (SSRIs):-

There is evidence suggesting that there is a statistically significant difference favouring SSRIs over placebo on reducing depression symptoms as measured by the HRSD but the size of this difference is unlikely to be of clinical significance.

If antidepressants do not produce clinically significant change, why are they recommended? This is especially a question when this evidence statement still applies specifically to both moderate and very severe depression, although for severe depression there is said to be some evidence suggesting that there is a clinically significant difference. Middleton et al support the use of SSRIs in moderate depression, despite the apparent lack of clinical significance, and make no mention of the complication of the lack of evidence for very severe depression compared to severe depression.

Maybe the NICE guideline is relying on another evidence statement:-.

There is strong evidence suggesting that there is a clinically significant difference favouring SSRIs over placebo on increasing the likelihood of patients achieving a 50% reduction in depression symptoms as measured by the HRSD.

Why is there this apparent inconsistency in the evidence statements - a clinically significant difference when 50% reduction in symptoms is measured but not when reducing depressive symptoms is measured? Response rates, such as 50% reduction, may in fact be misleading, especially when the criterion for response is close to the mean improvement rate, which does seem to be the case in antidepressant trials. Thus a person with a 12 point improvement on the Hamilton scale may be classed as a responder, whereas an 11 point improvement would not.3 The difference is still only one point - not clinically significant.

Confounding of severity statements by regression to the mean also needs to be considered. More severely depressed patients may well show greater improvement, but this is true for patients treated with placebo as well as those treated with medication. Also, people treated as inpatients tend to be the most severely depressed and do not have as good outcomes as outpatients.4

The issue may therefore be more complicated than a decision about the degree of severity at which to start treatment, as Middleton et al suggest. The NICE data in fact shows a nonlinear relationship between severity and drug/placebo differences - not as simple as Middleton et al would like to think. The basic point is that even in studies with more severely depressed patients, there is a strong placebo response and a relatively small difference between drug and placebo (maybe on average less than two points on HRSD). This small difference may be explicable as an amplified placebo effect due to methodical bias in clinical trials.5 More should be made of this finding rather than debate being diverted into the issue of merely whether depression is severe enough to treat. There are basic issues about the efficacy of antidpressants that are being ignored by this manoeuvre.

 

  1. Middleton H, Shaw I, Hull S, Feder G. NICE guidelines for the management of depression. BMJ 2005;330:267-268 (5 February), doi:10.1136/bmj.330.7486.267 [Full text]
  2. National Institute for Clinical Excellence. CG23 Depression: management of depression in primary and secondary care - Full guidelinehttp://www.nice.org.uk/pdf/cg023fullguideline.pdf
  3. Kirsch, I., Scoboria, A., & Moore, T.J. Antidepressants and placebos: Secrets, revelations, and unanswered questions.Prevention and Treatment. 2002, Article 33 http://www.journals.apa.org/prevention/volume5/pre0050033r.html
  4. Moncrieff J (2003) A comparison of antidepressant trials using active and inert placebos. International Journal of Methods in Psychiatric Research 12: 117-127.
  5. Moncreiff J & Double DB. Double blind random bluff.Mental Health Today 2003; Nov: 24-26

Competing interests:
None declared

Competing interests: No competing interests

08 February 2005
D B Double
Consultant Psychiatrist
Norfolk and Waveney Mental Health Partnership NHS Trust