DÄ internationalArchive42/2011Matching Should Take Place Before the Start of the Study
LNSLNS

Linder et al criticize the evaluation of DMPs in Germany and present their own study (1). Their study has methodological limitations that put in question the validity of their results. Firstly, we wish to point out the incomprehensible formation of groups. Matching has to be done before starting an intervention. In the Linder et al study, patients registered before 2007 were not excluded, however, and paired in 2006, which already partly eliminates the effect of the intervention. What is not clear either is why the researchers did not pair each DMP participant with a control from the sufficiently large pool. Patients who participated in more than one DMP were also excluded. This means that seriously ill people were excluded who would have benefited particularly from an improved healthcare structure. Patients who were registered very recently were included, in whom no effect has had time to manifest. Furthermore we cannot follow how subjects can be paired according to the variable “Education”, if pertinent data were lacking in almost 70% of identified diabetes patients. It is not permissible to use “missing” as a valid pairing variable. The authors have therefore missed their own objective, namely that of considering more carefully selection effects in evaluating the DMPs.

A study published in December 2010 that showed a positive trend in the quality of care and efficiency in participants in the DMP diabetes compared with a group of non-registered diabetes patients was not discussed (2). This is surprising as this was the first study in Germany that evaluated the T2DM by means of propensity score matching.

DOI: 10.3238/arztebl.2011.0723a

Dipl. Math. Anna Drabik

PD Dr. rer. pol. Markus Lüngen

PD Dr. med. Stephanie Stock

Institut für Gesundheitsökonomie und klinische Epidemiologie
der Universität zu Köln

Dr. rer. soc. Christian Graf

BARMER GEK, Wuppertal, christian.graf@barmer-gek.de

Conflict of interest statement

The authors declare that no conflict of interest exists.

1.
Linder R, Ahrens S, Köppel D, Heilmann T, Verheyen F: The benefit and efficiency of the disease management program for type 2 diabetes. Dtsch Arztebl Int 2011; 108(10): 155–62. VOLLTEXT
2.
Stock S, Drabik A, Büscher G, et al.: German diabetes management programs improve quality of care and curb costs. Health Aff (Millwood); 2010; 29: 2197–205. CrossRef MEDLINE
1.Linder R, Ahrens S, Köppel D, Heilmann T, Verheyen F: The benefit and efficiency of the disease management program for type 2 diabetes. Dtsch Arztebl Int 2011; 108(10): 155–62. VOLLTEXT
2.Stock S, Drabik A, Büscher G, et al.: German diabetes management programs improve quality of care and curb costs. Health Aff (Millwood); 2010; 29: 2197–205. CrossRef MEDLINE