Tuesday, December 7, 2010
Disease Management Plus Chronic Care Model Is Better Than the Chronic Care Model Alone: Two Reports
Fans of “disease management” may want to be aware of two just released reports.
The first is a hot-off-the-presses Health Affairs report "German diabetes management programs improve quality of care and curb costs" by Stepahie Stick and colleagues at the University of Cologne. It describes the outcomes associated with a German disease management program. While the lessons from Europe’s health care system may not always apply the United States, the Disease Management Care Blog thinks this particular manuscript may have an important lesson for the United States.
First, a quick description of Germany’s system. As the DMCB understands it, the German "Statutory Health Insurance" system consists of 115 not-for-profit “sickness funds” that are financed with a combination of employer contributions and government funding. It covers 90% of the German population. A “Risk Compensation Scheme” actuarially redistributes funding on a risk-adjusted basis between each of the funds. That’s important because insurance enrollees with a higher burden of illness “attract” extra funding. As described below, that extra funding was used to help bankroll disease management.
In 2002, Germany used the Chronic Care Model (CCM) to institute nationwide "primary care-based programs" for diabetes, breast cancer, asthma, COPD and coronary heart disease. They promoted guidelines, quality assurance programs, access to trained facilitators, benchmarking, reminders, physician education and peer feedback with risk adjusted measures. Physicians were expected to provide education to their patients, coordinate care and document treatment plans.
Doctors and their patients were also given access to vendor-run disease management programs. According to the Health Affairs article, while physician and patient participation DM was voluntary, patients who signed up got breaks with their office co-payments and “better service such as appointments within a certain time frame." This was partly possible because in 2007, the sickness funds received a payment amounting to $1,916 from the Risk Compensation Scheme for each patient who was enrolled in a disease management program. That figure was later adjusted using a comprehensive “morbidity oriented risk compensation” methodology.
So with that as background, the authors compared the patients who signed up for disease management to those who didn't in the BARMER Ersatzkasse” sickness fund. The period of study was for four years. On the basis of insurance claims, the authors found approximately 234,000 persons who appeared to have diabetes. The authors then further limited the population to persons who had filled three prescriptions for a diabetes medication during the 2002 baseline year. This yielded a population of 91,696 persons. Propensity matching, using age, gender, zip code, insurance status, baseline costs, hospital diagnoses and drug prescriptions was then used to find otherwise identical patients for study. This yielded 19,882 one-to-one matches.
There were big differences in outcomes between the two groups. By 2007, death rates were 2.3% for the disease management group vs. 4.7% for the matched group. There were lower rates of myocardial infarction, stroke, chronic renal insufficiency and amputations in the disease management groups. Mean cost differences (euros had to be converted to dollars) from the 2002 baseline to 2007 for the disease management group vs. the control group totaled $1443 vs. $1890, respectively, which was statistically significant. It was mostly accounted for by lower hospitalization rates. When the per patient difference of $447 was compared to the disease management program cost $238, the net savings was $209
While there are limitations from any retrospective observational study, this publication is important because it shows that disease management layered on top of the chronic care model can yield additional savings. While it is possible that disease management "bested" the Chronic Care Model, the DMCB thinks it's more likely that disease management's impressive impacts on complications and costs was made possible by the synergies of being associated with the Chronic Care Model. The DMCB speculates that the Model prompted awareness of population-based care among the physicians and helped them and their patients take advantage of the vendors' offerings.
There may be a lesson for the United States. The authors point out that Germany's Nike "just do it" adoption of disease management contrasts with the cautious “piloting” of population health in the United States. Based on what we know about the Chronic Care Model, Patient Centered Medical Home and disease management on our shores, there is little reason to doubt that a similar approach with a heavy emphasis on the primary care physician as the program manager would succeed. Hopefully the upcoming Accountable Care Organization (ACO) pilots will take note and include this in their planning.
There's an important second report that has been ably reviewed by Vince Kuraitis over at his e-CareManagement blog. If the wording includes the term "disease/care management" and Vince calls it a Home Run, then you know it's something special. This involved Massachsetts General Hospital and its physicians in an initiative that was similar to the German experience outlined above: physicians were in an "integrated" care setting (sounds like Chronic Care Model to the DMCB) and worked with case managers who provided education, support and coordination. There was a return on investment that, depending on the methodology, ranged from 2.65 to 3.35
Image from Wikipedia. That's the magnificent Cologne Cathedral in Germany, the home town of the Health Affairs authors. The DMCB climbed one of those towers not too long ago.