The psychodynamic treatment tradition, for example PDT, is older than the cognitive. But as for quantitative research on treatment outcomes, it’s the opposite. The initial reluctance of PDT therapists towards outcome research was mainly because the changes sought in therapy were considered impossible to quantify and therefore unsuitable for such research. Therefore, there are more outcome studies of CBT than PDT, which may explain the often-heard statement that “only CBT has evidence” or alternatively “has the best evidence.” Since more studies have been published, it should rightfully be said that CBT has the “best quantity of evidence,” which is not the same as “best evidence.” The background to this misunderstanding is multifaceted. Approximately simultaneously with the birth of CBT, there was a paradigm shift in the medical science’s treatment recommendations. It was now required that doctors base their advice on “scientific evidence.” What did this concept mean?
In a famous editorial (Sackett et al., 1996), the authors refer to “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients”. They did not give exclusive priority to evidence from systematic research, but advised integrating it with “individual clinical expertise” as well as “individual patients’ predicaments, rights, and preferences in making clinical decisions about their care”. Regarding clinical expertise, they issued a warning: without it, medical practice risks being “tyrannised by evidence, for even excellent external evidence [RCT studies of larger samples] may be inapplicable to or inappropriate for an individual patient”.
The idea behind the recommendation of Empirically Supported Treatments or “evidence-based treatments” is commendable: to reduce the influence of the clinician’s personal opinions and “this is how we’ve always done it.” Thus, objectivity is pursued. This is reasonably straightforward in, for example, drug trials. Two drugs, A and B, are “double-blinded,” each half of the sample receiving A or B. If B is found to have a statistically significant better effect, it increases B’s evidence. Then, it is investigated whether renewed trials yield similar results. For the study to have the highest evidence score (“gold standard”), it should be randomized (RCT) and replicated, i.e., repeated by another research group(s). There are clear guidelines for how such studies should be conducted. Unfortunately, they are very difficult to apply in psychotherapy research. The patient soon realizes, for example, whether they are receiving CBT or PDT. And how do you quantify mental symptoms? And what significance does the therapist’s personality and collaboration with the patient have for the outcome?
Another problem concerns efficacy versus effectiveness. If drug B had better results than A in a well-controlled randomized study, then B has shown its efficacy. But the way the drugs have been administered has been artificial. Getting an unmarked tablet from a research nurse is not the same as when a trusted doctor explains
Another problem concerns efficacy versus effectiveness. If drug B had better results than A in a well-controlled randomized study, then B has shown its efficacy. But the way the drugs have been administered has been artificial. Getting an unmarked tablet from a research nurse is not the same as when a trusted doctor explains to the patient the effects and possible side effects of the drug. Such aspects are taken into account in effectiveness studies in naturalistic settings, where the clinical usefulness of the treatment method, the doctor’s and patient’s appreciation, etc., are investigated. Only such studies can “demonstrate that a form of therapy works in the field” (Leichsenring, 2004, p. 137) and demonstrate their “practice-based evidence” (Holmqvist et al., 2015, p. 20).
Here is a wise conclusion: “For treatments that have been evaluated in RCTs, studies are needed to investigate their effectiveness in real-life conditions” (Leichsenring et al., 2015, 648). Personally, I first explored the efficacy of mother-infant treatments (Salomonsson and Sandell, 2011 a, b), as well as the follow-up study (Winberg Salomonsson et al., 2015 a, b). Then we conducted an effectiveness study of mother-child treatment in a naturalistic environment, i.e., at the Child Health Centre (Kornaros, 2020, Salomonsson et al., 2021).