Some considerations on randomised clinical trials (RCTs)

In this critical analysis by Prof. Dr. Wolfgang Mertens, emeritus professor of clinical psychology and psychotherapy at the Department of Psychoanalysis and Psychodynamic Research at Ludwig Maximilian University in Munich, published in 2008, the central characteristics of randomised controlled trials (RCTs) are compared with arguments for and against their use as the sole or highest criterion of efficacy, and the methodological limitations of RCTs are highlighted. Their significance for certain research questions in therapeutic clinical practice involving complex forms of therapy (in which aspects such as relationship dynamics, context and the inclusion of changes in the course of treatment are inherent) is also critically examined. This is followed by introductory paragraphs and a link to the original paper.

"On evidence-based practice and the postulation of evidence levels:

In recent years, psychotherapy research methodology has increasingly aligned itself with the behavioural medicine and pharmacological standards of evidence-based medicine (EBM). This is welcomed by many officials in the healthcare sector, as it finally achieves the desired standardisation and thus comparability. Why should the various psychotherapy methods be given special treatment in terms of research methodology, especially since behavioural therapists have not expressed any concerns about this convergence? The Scientific Advisory Board has also joined this trend.

In analogy to basic laboratory research, psychotherapy research now also refers to internal and external validity. High internal validity is always present when the outcome criterion (success of therapeutic treatment) can be clearly attributed to the treatment variable (therapeutic treatment condition). This is best achieved through the use of controlled, randomised studies.

However, this presents a dilemma: the higher the internal validity that a researcher can achieve by eliminating or controlling all confounding variables, the lower the external validity. This means that the results can only be applied to practical conditions with difficulty. Although internal and external validity do not have to be completely mutually exclusive, the stricter the requirements for ensuring internal validity, the lower the generalisability tends to be.

The question, however, is whether the hierarchy of internal and external validity, as required by evidence-based practice, makes sense at all: Can experimental quality criteria derived from basic research be transferred to practice? Therapeutic treatment is, a priori, a complex process with many influencing variables that also interact with each other. Therefore, the previous postulation of evidence classes (studies with assured internal validity are those in which the main focus is on external validity, which is preferable for the scientific recognition of psychotherapy methods) would not be appropriate for the subject of psychotherapy research from the outset.

Closely related to this, but essentially of secondary importance, is the question of the usefulness of the randomisation strategy. ...." (Mertens, 2008)

Link to the original work cited here:

Mertens, W. (2008, January 13). Einige Überlegungen zur randomisierten klinischen Studie („RCT“). Ludwig Maximilians Universität München - Department Psychologie / Abt. Psychoanalyse und psychodynamische Forschung. Retrieved on 7 January 2026 from https://www.researchgate.net/publication/238725862_Einige_Uberlegungen_zur_randomisierten_klinischen_Studie_RCT_Zur_Evidenzbasierung_und_zur_Postulierung_der_Evidenzgrade