Are Randomized Controlled Trials Relevant to Clinical Practice?

Abstract

Ithink that randomized controlled trials (RCTs) are highly relevant to clinical practice. If I had a serious medical illness, I would want to know that what I was about to receive actually worked and how it fared relative to other treatment options. Saying that something works means that it has a causal effect (relative to its absence), and saying that it works better than other options means that its impact is even larger than those other options. The most powerful method yet devised to determine whether something has a causal effect (or how it compares with other options) is to test it in an RCT. I think that people with psychological disorders deserve the same level of certainty as people with medical disorders; I want to know whether what I do works. I do not defend RCTs because I conduct them, I conduct RCTs because I can defend them.All causal inference starts with observation. If I want to know if a treatment works, I want to know what happens to people who get treated. If I had a serious medical illness, the first thing I would do is talk to other people who had the disorder and find out what they did and how it went. The problem with relying on such anecdotal observations is that I could never know whether what happened was a consequence of treatment. I would be glad if all the outcomes were benign, but that would not tell me that treatment was necessary to produce those happy outcomes. When my son was growing up he played soccer with teammates who were spanked by their parents when they misbehaved. When those parents observed that we did not spank our son, they were curious as to how we expected him to learn right from wrong. Their parents had spanked them when they were young and they had turned out well. Therefore, they believed it was their obligation as caring and responsible parents to spank their children if they wanted them to turn out well. The fact that something happens does not mean that is was caused by what you did.I may decide to talk to people who had been treated and others who had not (if any were still living) and see how each turned out. This is a quasi-experiment in which people are exposed to different treatments and their subsequent outcomes are observed. The problem with this strategy is that it confounds individual differences with differential treatment. Considered medical opinion for the last several decades has been that women going through menopause benefited from hormone replacement therapy. Women who were treated with estrogen did appear to do better than women who were not. The problem with this inference was that women were free to choose whether they received estrogen or not and those who choose estrogen typically took better care of themselves and had better access to health care. When the field finally got around to doing an RCT on adding estrogen, the study was stopped before completion because replacement therapy increased the risk for cancer and heart disease.1 It is not that the doctors providing hormone replacement therapy did not care about their patients or tha...

Good news for allegedly bad studies. Assessment of psychometric properties may help to elucidate deception in online studies on OCD

“…This objection is graver for basic research studies that usually impose quite rigid selection criteria and enroll ''pure'' samples. For intervention studies this is less of a problem considering that there is a recent trend in psychopathological research to target ''typical'' patients (Hollon & Wampold, 2009). Comorbid disorders that would result in exclusion from basic research studies are usually tolerated in intervention trials-nondetection of comorbid disorders in online studies can thus be considered a lesser evil.…”

Efficacy of Technology-delivered Cognitive Behavioural Therapy for OCD Versus Control Conditions, and in Comparison with Therapist-Administered CBT: Meta-Analysis of Randomized Controlled Trials

“…De`ttore, Pozza and Andersson COGNITIVE BEHAVIOUR THERAPY Generalizability of findings of our metaanalysis could be enhanced by the fact that the included trials were conducted in three different continents, and patients were recruited from a variety of settings, including not only university facilities but also community and specialist healthcare facilities. Variety of recruitment strategies across the included trials may be considered a strength of our work, as it is argued by some researchers and practitioners that the results obtained only from academic settings are of limited relevance to ordinary clinical practice, since in those settings treatments are often delivered by highly specialized therapists, often include only strongly motivated patients recruited through advertisements, who can be less likely to have comorbid psychological disorders (Hans & Hiller, 2013;Hollon & Wampold, 2009). …”

“…Furthermore, the added value of medication in addition to CBT is probably not substantial as CBT appears to be more effective overall (Cuijpers et al, 2013). Despite controlling for concurrent medication use may improve the internal validity of a study, it is likely to decrease the external validity of a study as several patients with OCD are on medication at the time of seeking psychological help (Hollon & Wampold, 2009). …”

Do it yourself! Evaluation of self-help habit reversal training versus decoupling in pathological skin picking: A pilot study

“…The following inclusion criteria were applied: participants had to suffer from skin picking (i.e., defined as excessive manipulation of the skin; the symptomatology of skin picking was explained to prevent that the invitation would attract subjects with skin conditions other than skin picking), had sufficient time to work through the manual, and had to provide consent to participate in two anonymous (internet-based) surveys. In view of the pilot character of the study, no special exclusion criteria were applied, mirroring a new trend in intervention research to collect representative (typical) samples (Hollon & Wampold, 2009). Moreover, our aim was to learn more about factors potentially moderating outcome.…”