Influence of initial severity of depression on effectiveness of low intensity interventions: meta-analysis of individual patient data

In their outstanding paper “Influence of initial severity of depression on effectiveness of low intensity interventions: meta-analysis of individual patient data”, Bower et al. describe an individual patient data meta-analysis of depression severity as a moderator of the effect of low intensity interventions in depression (1).

Their insightful analysis of strengths and weaknesses of their data leads to the conclusion that there were no clinically meaningful differences in treatment effects between more and less severely ill patients receiving low intensity interventions. Authors also conclude that patients with more severe depression can be offered low intensity interventions as the first part of a stepped care model.

However, this last interpretation should be nuanced, especially because the authors directly refer to the concept of effectiveness in their title.

Effectiveness can be defined as generalized causal inferences derived from scientific studies. This concept encompasses the concept of efficacy (measured under optimal circumstances) and goes beyond by introducing the very pragmatic notion of what works in a “real life” setting (2). By extension, this concept could include “comparative effectiveness” since in “real life” many different kind of interventions are available. Recently the concept of effectiveness had known a great success in Evidence Based Medicine since it seems to be most relevant than efficacy for clinicians and for health authorities. Among the major papers recently published in psychiatry, there were indeed many which were interested in this issue (3-6).

The price of success is that the concept can be sometimes used wrongly since it could represent a guarantee for amplifying the buzz around each and every paper.

When Bower et al.’s speak of “what works for whom”, they touch on the subject but do not tackle it. Precisely, their meta-analysis could not measure effectiveness because the individual studies included were rather designed to study efficacy than effectiveness. Indeed, many authors have underlined the incompatibility of Randomised Controled Trials with a “real life” clinical practice, especially in the field of studies in major depressive disorder (7-9). In term of external validity, such studies are particularly difficult to generalize in routine care. For example, in the web appendix of their meta-analysis, we can see that in 12 (75 %) of the included studies, patients were recruited through advertisements, surveys or internet based screening. This represents probably a major selection bias since such recruitments do not obviously occur in day-to-day clinical practice.

From the “comparative effectiveness” perspective, it is not because there were no differences between more and less severely ill patients that severely ill patient should receive in a first step low intensity interventions rather than other approaches. The presence of the efficacy they demonstrated for severe depression is necessary but not sufficient to justify the use of such treatment in first line treatment in this population since other approaches could be more useful for these particular patients.

At last, authors propose to generalize their findings to the target population of patients involved in the first step of stepped care intervention (treatment-naïve patients). Nevertheless one should note that in their meta-analysis many patients have already passed this first step. Indeed, in the included studies, 19 % to 69 % of patients were taking antidepressants. This definitively restricts the scope for generalization of their findings.

To sum up, the paper deals with a promising issue, describes very important data but one could have the feeling that it omits to take into account “the long road from RCTs to effectiveness” (10). Effectiveness is not just a buzzword.

References

1. Bower P, Kontopantelis E, Sutton A, Kendrick T, Richards DA, Gilbody S, et al. Influence of initial severity of depression on effectiveness of low intensity interventions: meta-analysis of individual patient data. Bmj 2013;346:f540.

2. Falissard B, Izard V, Xerri B, Bouvenot G, Meyer F, Degos L. Relative effectiveness assessment of listed drugs (REAL): a new method for an early comparison of the effectiveness of approved health technologies. Int J Technol Assess Health Care 2010;26(1):124-30.

3. Leichsenring F, Rabung S. Effectiveness of long-term psychodynamic psychotherapy: a meta-analysis. Jama 2008;300(13):1551-65.

4. Lieberman JA, Stroup TS, McEvoy JP, Swartz MS, Rosenheck RA, Perkins DO, et al. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med 2005;353(12):1209-23.

5. Schneider LS, Tariot PN, Dagerman KS, Davis SM, Hsiao JK, Ismail MS, et al. Effectiveness of atypical antipsychotic drugs in patients with Alzheimer's disease. N Engl J Med 2006;355(15):1525-38.

6. Rush AJ, Trivedi MH, Wisniewski SR, Stewart JW, Nierenberg AA, Thase ME, et al. Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. N Engl J Med 2006;354(12):1231-42.

7. Moncrieff J, Kirsch I. Efficacy of antidepressants in adults. Bmj 2005;331(7509):155-7.

8. Ioannidis JP. Effectiveness of antidepressants: an evidence myth constructed from a thousand randomized trials? Philos Ethics Humanit Med 2008;3:14.

9. Naudet F, Maria AS, Falissard B. Antidepressant Response in Major Depressive Disorder: A Meta-Regression Comparison of Randomized Controlled Trials and Observational Studies. PLoS One 2011;6(6):e20811.

10. Cartwright N. A philosopher's view of the long road from RCTs to effectiveness. Lancet 2011;377(9775):1400-1.