We thank our correspondents for their interest in and critical comments on our study. Behrens and coauthors mostly confirm our results in their own study of guideline knowledge and implementation in secondary coronary prevention. In contrast to our data, however, they found that knowledge and acceptance of guidelines were better in younger physicians, specialists in internal medicine, and doctors who participated in quality circles. Although we suspected similar barriers to guideline implementation as Behrens et al, we were not able to investigate this question explicitly in our current project owing to scarce funding. We unreservedly agree with the comment that implementation of guidelines needs to happen at multiple levels and should acknowledge the known barriers in the sense of a complex intervention. In this sense we also understand the comment by Unverzagt and Klement— that cognitive communication of guideline recommendations is only one building block in the implementation chain and that the patient’s perspective and the influence of the specialist should be considered in general practice. This is in principle identical to our own opinion. However, since in our study we attempted to prove the benefits neither of a guideline or therapeutic recommendation nor of a particular implementation strategy, we did not conduct a randomized controlled study with hard fixed end points.
Rümenapf names communication problems because of incomprehensible specialist jargon as a possible barrier to implementation. We agree in the sense that guideline authors should use language that reflects clinical practice and is concrete.
Beller and Hector mention the multimorbidity of elderly patients and the often associated limitations regarding the extent to which therapeutic recommendations can be implemented by doctors, but also by patients, as a crucial reason for why the treatment did not meet the expected guideline recommendations. Their criticism is that neither the guidelines nor our study paid enough attention to these patient related aspects. However, our study did not focus on implementation barriers from the patients’ perspective. We did, however, take this aspect into consideration in that we did not include treatment data from patients older than 79 in our evaluations. Fundamentally we agree with Beller and Hector that the topic of multimorbidity has not received enough attention in guidelines—because of the logic inherent in guideline development, which relies on high-level evidence from studies and therefore does not sufficiently include the general practice clientele. This problem is currently the subject of intense interdisciplinary debate among guideline authors—for example, in developing the National Disease Management Guidelines Program (“German DM-CPG Program”) or even in general practice guidelines (1). Whether multimorbidity is the crucial reason for deviating from guideline recommendations, however, remains to be investigated.
Our considerations on implementing guideline recommendations as in-process control variables in standardized software into to support doctors’ decision making triggered fears of increasing bureaucratization of doctors’ everyday clinical practice. Lischka furthermore questions the motives of our research team and suspects that the primary reasons for our study was to establish reasons for implementing “software corridors”. This was not our study objective but the conclusion we drew from our observation that both groups of doctors saw room for improvement in implementing the recommendations selected for our study. In any case, as Unverzagt and Klement say, retrievable guideline knowledge in itself cannot be used as a surrogate parameter for guideline-conform clinical practice. How to implement evidence based knowledge is a major challenge facing us all—as evidenced by the discussions at the 15th annual meeting of the German Agency for Quality in Medicine (Ärztliches Zentrum für Qualität, AQuMed).
Dr. rer. pol. Ute Karbach
Institut für Medizinsoziologie, Versorgungsforschung
und Rehabilitationswissenschaft der Humanwissenschaftlichen
Fakultät und der Medizinischen Fakultät der Universität zu Köln
Conflict of interest statement
The author declares that no conflict of interest exists.
|1.||Hausärztliche Leitliniengruppe Hessen. Leitlinie Geriatrie, Teil I Version 1.00 vom 17.11.2008 und Teil II. Version 1.02 vom 14.09.2009|
|2.||Karbach U, Schubert I, Hagemeister J, Ernstmann N, Pfaff H, Höpp HW: Physicians’ knowledge of and compliance with guidelines: an exploratory study in cardiovascular diseases. Dtsch Arztebl Int 2011; 108(5): 61–9. VOLLTEXT|