Open Access Highly Accessed Open Badges Study protocol

The Chronic CARe for diAbeTes study (CARAT): a cluster randomized controlled trial

Anja Frei12*, Corinne Chmiel13, Hansueli Schläpfer4, Beatrice Birnbaum5, Ulrike Held2, Johann Steurer2 and Thomas Rosemann1

Author Affiliations

1 Institute of General Practice and Health Services Research, University of Zurich, Zurich, Switzerland

2 Horten Centre for Patient-oriented Research, University of Zurich, Zurich, Switzerland

3 Division of Internal Medicine, University Hospital of Zurich, Switzerland

4 Ärztenetz säntiMed, Herisau, Switzerland

5 Schweizerischer Verband Medizinischer PraxisAssistentinnen, Bern, Switzerland

For all author emails, please log on.

Cardiovascular Diabetology 2010, 9:23  doi:10.1186/1475-2840-9-23

Published: 15 June 2010



Diabetes is a major challenge for the health care system and especially for the primary care provider. The Chronic Care Model represents an evidence-based framework for the care for chronically ill. An increasing number of studies showed that implementing elements of the Chronic Care Model improves patient relevant outcomes and process parameters. However, most of these findings have been performed in settings different from the Swiss health care system which is dominated by single handed practices.


CARAT is a cluster randomized controlled trial with general practitioners as the unit of randomization (trial registration: ISRCTN05947538). The study challenges the hypothesis that implementing several elements of the Chronic Care Model via a specially trained practice nurse improves the HbA1c level of diabetes type II patients significantly after one year (primary outcome). Furthermore, we assume that the intervention increases the proportion of patients who achieve the recommended targets regarding blood pressure (<130/80), HbA1c (=<6.5%) and low-density lipoprotein-cholesterol (<2.6 mmol/l), increases patients' quality of life (SF-36) and several evidence-based quality indicators for diabetes care. These improvements in care will be experienced by the patients (PACIC-5A) as well as by the practice team (ACIC). According to the power calculation, 28 general practitioners will be randomized either to the intervention group or to the control group. Each general practitioner will include 12 patients suffering from diabetes type II. In the intervention group the general practitioner as well as the practice nurse will be trained to perform care for diabetes patients according to the Chronic Care Model in teamwork. In the control group no intervention will be applied at all and patients will be treated as usual. Measurements (pre-data-collection) will take place in months II-IV, starting in February 2010. Follow-up data will be collected after 1 year.


This study challenges the hypothesis that the Chronic Care Model can be easily implemented by a practice nurse focused approach. If our results will confirm this hypothesis the suggestion arises whether this approach should be implemented in other chronic diseases and multimorbid patients and how to redesign care in Switzerland.