In Reply

We thank the authors for their comments. As rightly pointed out, it becomes more and more difficult to cope with both the growth of increasingly complex scientific findings and the need for (and the right to) comprehensible and useful information for patients. It is hoped that the upcoming generation of doctors will be (even better) prepared for these challenges, as the necessary skills have recently been described in the catalogue of National Competency-based Learning Objectives for Undergraduate Medical Education (Nationaler Kompetenzbasierter Lernzielkatalog Medizin, NKLM; see www.nklm.de) and approved by the German Medical Faculty Association (Medizinischer Fakultätentag) (1).

The criticized “low yield” from reading our article may reflect expectations typical for physicians, i.e. primarily “science-focused” and thus not (well) suited for our topic. Similar to research in medical education, it is particularly difficult to generate clear, undistorted and transferable results in clinical research on communication, approaches, and attitudes; however, such results are essential to complement, and perhaps confirm or revise, experience-based practices in everyday clinical settings.

We agree with the view that it is not the primary aim of shared decision making to be “therapeutic”. However, there is strong evidence—as also shown in our review—that shared decision making not “only” meets the needs of patients for participation and enables them to make informed decisions, but also optimizes (or even permits) therapeutic success. The discussion initiated here about the validity (among other issues) of surrogate parameters needs to be continued in a critical manner, which is however beyond the scope of this reply.

In our article, we described how we selected the endpoints that we considered (2). Here, we would like to “defend” only the consideration of adherence. Firstly, adherence has been associated with improved therapeutic success (and even with decreased mortality). Secondly, adherence is significantly affected by physician–patient communication (3– 4).

We have already justified not using the MeSH term “shared decision making”, this can however be seen as a limitation of our study.

We would like to clarify, that studies in which decision aids were examined were only excluded if their use was the only difference between control and intervention groups. It is not called into question the effectiveness of decision aids.

The reactions of readers of Deutsches Ärzteblatt International shows that having a reasonable and scientifically-based patient participation is an issue of concern not only for us. A critical discussion—even beyond “solid evidence”—is an essential part of this process.