; ; ; ;
During the introduction period of the DMP, an opportunity was missed to generate the best possible evidence by means of a prospective randomized controlled trial (RCT) before the DMP was widely implemented. It therefore remains an ongoing challenge to support the best possible evidence by means of studies. In spite of all methodological hurdles it would now still be possible to set up an RCT. As a complementary measure, additional data sources—such as data from the sickness funds—could be used for the purposes of evaluation. Using the available funds efficiently in the statutory sickness funds requires comprehensive and valid evaluation of the DMP, in order to modify these programs in such a way that the optimum cost-benefit relation can be achieved. It is therefore essential to demand that the funding used for the DMPs, of more than 1.1 billion Euros per year, have a tangible effect for those affected.
Drabik et al mention the 1:1 matching used in the study published by Stock et al (1). According to our own benchmarking, the selection of the matching method has a negligible effect in the present study. Excluding insurance members participating in more than one DMPs served the purpose of studying a population that can be defined as clearly as possible. The relevant selection effects explained by Drabik et al are much more pronounced in the study reported by Stock et al, which included only those diabetes patients with at least three prescriptions of antidiabetes drugs in 2002. Because—as Stock et al themselves explain—25–30% (according to our own investigations: 39%) of diabetes patients do not take pharmacotherapy, mildly ill patients are not considered in the study and the effect of the DMP is therefore overestimated. A similar effect results from excluding patients younger than 40 and those who changed sickness funds, who were as a rule less severely ill. The result is a subgroup with more than 40% of diabetes patients, who are mostly severely ill. This is a non-permissible subgroup formation when one considers whether the currently practiced watering can principle in DMP registration makes any sense at all. It also explains why—as Altenhofen et al say in their letter—different sickness funds with comparable methodologies reach diametrically opposite conclusions. However, we share the insight that especially a subgroup of severely ill diabetes patients benefits from the DMP, which is also the result of the subgroup analysis conducted by the WINEG.
Szecsenyi calls for taking further comorbidities into consideration, Chantelau for including leg amputations, but objective selection criteria are lacking. Including further parameters can risk the matching in as far as it relativizes the influence of undoubtedly important influential variables. It was not possible for us to consider the study reported by Stock et al that Drabik et al mention in their letter as it was published only after our own study had been submitted for publication. Heinsch in his letter points out the noticeable, growing importance of the DMP in North-Rhine (2) on the basis of documentation sheets. This does not contradict the WINEG study, because the general quality of medical care for diabetes patients has improved simultaneously (3). The quality reports can thus not provide proof of causality. Heinsch rejects the evaluation of surrogate parameters and asks for hard end points instead. This is exactly the approach taken in the WINEG study. Mortality as the hardest end point was not investigated in the WINEG study. The ELSID study (4) concludes that there is no causal association between mortality and DMP registration.
In sum, all evaluations of routine data from the statutory health insurers (which are meant for accounting purposes) are subject to certain limitations. The method suggested by Sawicki to minimize “sponsor bias” is very interesting. At the moment, studies reflect an inconsistent picture. Further studies give rise to the assumption that structured treatment programs do not necessarily lead to cost savings compared with standard treatments (5). It has not been satisfactorily explained either whether the additional costs associated with the DMP are in proportion to their additional effects (6). The healthcare system is challenged to generate valid and reliable data on the cost effectiveness of the DMP.
Prof. Dr. med. Roland Linder
Dr. rer. medic. Susanne Ahrens
Dr. rer. nat. Frank Verheyen
WINEG – Wissenschaftliches Institut der TK für Nutzen und Effizienz im Gesundheitswesen, Hamburg, email@example.com
Dagmar Köppel, Thomas Heilmann
Fachreferat Disease-Management-Programme der Techniker Krankenkasse, Hamburg
Conflict of interest statement
The Wissenschaftliches Institut der TK für Nutzen und Effizienz im Gesundheitswesen (Scientific Institute of TK for Benefit and Efficiency in Health Care, WINEG) is tasked to critically appraise the impact of innovations and new programmatic approaches within the statutory health insurers. The authors declare that because of their membership in the TK, a potential conflict of interests exists.
|1.||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|
|2.||Qualitätssicherungsbericht 2009: Disease-Management-Programme in Nordrhein. (Last accessed on15.04.2011), 2010.|
|3.||www.faz.net/s/Rub7F74ED2FDF2B439794CC2D664921E7FF/Doc~E7EDF5B60C426439AA65CCFC2BE643687~ATpl~Ecommon~Scontent.html (Last accessed on 07.10.2010) and in personal communication with Prof. Berthold (Charité).|
|4.||Miksch A, Laux G, Ose D, et al.: Is there a survival benefit within a German primary care-based disease management program? Am J Manag Care 2010; 16: 49–54. MEDLINE|
|5.||Busse R: Bekämpfung chronischer Krankheiten und Versorgung chronisch Kranker – international. Die BKK 3/2011; 142–5.|
|6.||Nolting H-D: Disease Management Programme im Spiegel der Versorgungsforschung. Die BKK 3/2011; 146–9.|
|7.||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.|