December 10, 2007
Disease management programs that help guide the care of patients with chronic health problems appear to improve the quality of health care, but there is little evidence that such efforts actually save money, according to a study issued today by the RAND Corporation.
The RAND Health study reviewed all past research on disease management programs, which seek to help patients with conditions like diabetes and congestive heart failure by offering a system of coordinated healthcare interventions. These interventions can range from pre-recorded telephone reminders to home visits by medical professionals.
Researchers selected 29 evaluations, systemic reviews and meta-analyses to focus on, covering 317 unique studies. That review found consistent evidence that these programs can improve health care quality, improve disease control, and, in the case of patients with congestive heart failure, reduce hospital admission rates. But patients with depression who were enrolled in disease management programs were more likely to use outpatient care and prescription drugs, increasing costs. There also is little evidence about whether these programs improve health outcomes over the long term.
“Disease management is viewed as the silver bullet that can fix two problems of the health care system – inadequate quality and high costs,” said Soeren Mattke, lead author of the report and a senior natural scientist at RAND, a nonprofit research organization. “Unfortunately, while there is evidence that disease management programs can indeed improve the quality of care, there is no conclusive evidence that they can actually save money.”
Health insurance plans and employers nationally in 2005 spent about $1.2 billion on disease management programs, with 96 percent of the top 150 U.S. health insurance companies offering some form of disease management service. The topic also has become a key point in the national health care reform debate, as policymakers search for a way to improve health care quality and access, while controlling costs at the same time.
The RAND study analyzed research on various disease management programs and their effect on six chronic conditions: congestive heart failure, coronary artery disease, diabetes, asthma, depression and chronic obstructive pulmonary disease. With the exception of asthma and chronic obstructive pulmonary disease, which showed inconclusive results, researchers found consistent evidence that disease management programs did improve the quality of health care. There also was consistent evidence that patients with congestive heart failure and depression reported improved quality of life.
But evidence of cost-savings was inconclusive for most of the conditions, indicating that further research is needed. It is plausible that disease management programs reduce costs for congestive heart failure patients because many programs reduce hospital admissions for these patients, Mattke said. In contrast, research has shown that patients with depression are commonly under-treated, so a disease management program that actively screens for depression and encourages patients to get treatment will increase costs.
Most of the studies Mattke and his colleagues reviewed followed patients only for about a year, which is not long enough to assess long-term health outcomes. For example, a disease management program may improve a patient's cholesterol levels in the short term, but it can take years to determine whether those interventions — assuming they were sustained for a long period — prevent heart attacks and costly hospitalizations years into the future.
“People take for granted that these programs work and save money, because the concept is very plausible,” Mattke said. “But many things in medicine sound plausible until you do the research and find that promises don't hold true.”
Health insurance plans, employers and policymakers also will have to evaluate whether the benefits of disease management programs are worthwhile, despite the lack of evidence for cost-savings, Mattke said.
Another factor is that “disease management” itself is not well-defined, Mattke said. The types of interventions can vary from congestive heart failure specialists at a local hospital hiring a nurse to educate patients about preventive care to large-scale programs reaching thousands of patients. Some programs may be better than others, but there has not been enough research to properly assess which ones are the most effective or what characteristics are associated with more successful programs.
“Just because one type of intervention can benefit one disease, that doesn't mean it will work the same way for another disease,” Mattke said. “Most of the evidence for disease management comes from small provider-centered programs at academic medical centers or multi-specialty groups.”
The study, “Evidence for the Effect of Disease Management: Is $1 Billion a Year a Good Investment?” will appear in the December issue of The American Journal of Managed Care. The other authors of the study are Michael Seid of Cincinnati Children's Hospital, and Sai Ma of Johns Hopkins Bloomberg School of Public Health.
RAND Health, a division of RAND, is the nation's largest independent health policy research program, with a broad research portfolio that focuses on health care quality, costs and delivery, among other topics.