A study published by in the New England Journal of Medicine last week examining the effects of 15 different Medicare care coordination demonstrations received wide coverage by the general media. Unfortunately, much of this focused on the study’s overall finding that these programs didn’t reduce hospitalizations or Medicare spending. For example, the AP story’s headline, “Study finds bid to cut Medicare costs failed,” was used by many papers such as the Washington Times.
However, the actual study had much more complex, important, and useful findings, and the paper’s authors from Mathematica, (which Medicare contracted to do the analysis from this project), deserve a lot of credit for extracting meaningful information from this project.
Complexity of Medicare Care Coordination Demonstration Projects
The study was very complex and presented many analytical challenges, including:
- The “study” involved 15 different and diverse “care coordination” programs: 5 disease management companies, 3 community hospitals, 3 academic medical centers, 1 integrated delivery system, 1 hospice, 1 long-term care facility, and 1 retirement community
- Despite being called “care coordination” programs, the major intervention was nurses communicating with patients outside of the patients’ direct clinical care team
- Most of the programs’ care coordinators didn’t have significant interactions with the physicians treating the individual patients – although reports were provided to the physicians
- The primary outcomes the study looked at were hospitalizations and total Medicare costs
Given that physicians direct most clinical decisions and thus influencing overall healthcare spending, it is not surprising that third party programs were ineffective in changing clinical or cost outcomes. With the patient’s physicians not being directly engaged in the development and implementation of these programs, nor vested – either financially or cognitively – in the success of these care coordination programs, this outcome is not surprising. This is like expecting travel agents to improve the safety and efficiency of air travel, i.e. while they can provide information to individuals they are not engaged in the operations of the airplane or airports. [OK – that’s an overly simplistic example, but I think it helps make the point.]
Positive “Findings” From the Study:
Despite the disparate structure of the programs in this study, the authors found that the more successful programs:
- Had greater in-person contacts with the patients – nearly 1 per month
- Focused their efforts on individuals with higher Medicare costs
- Had greater success educating patients on how to properly take their medicines
- “Worked closely with local hospitals, which provided the programs with timely information on patient hospitalizations and enhanced potential to manage transitions and reduce short-term readmissions”
- “Had frequent opportunities to interact informally with physicians”
Implications for Medical Homes
These findings reinforce the importance of focusing on the role of physicians in controlling and managing clinical care decisions – which overall account for up to 80% of all healthcare spending. This observation is consistent with the conclusions the authors of the Care Coordination study reach in their NEJM article about their findings being able to help support the implementation and success of medical homes: “The successful interventions also may offer more detailed lessons for medical homes about how best to educate and monitor patients, the types of patients for whom they are likely to be most effective, and how to help patients overcome barriers to better self-care.”
This is a very timely and important conclusion since Medicare is getting ready to launch a Medical Home Demonstration Project in 8 states – and possibly more through legislation enacted last summer. Medicare’s Medical Home Demonstration has a greater likelihood of improving quality and lowering costs than the Care Coordination project because the Medical Home Demonstrations will directly put the responsibility for care coordination and improving patient self-management on the patients’ individual physicians – as opposed to providing an outside service for patients. In addition, similar to the two somewhat successful Care Coordination projects, the Medical Home Demonstrations will focus on Medicare patients with chronic conditions.
Implications for Overall Health Reform Efforts
The Care Coordination study’s observation that the successful programs differed from the unsuccessful ones in how they improved patients’ compliance with medications also has implications for broader health reform initiatives. It makes sense that patients with chronic conditions who aren’t receiving adequate medicinal treatments would have increased costs and hospitalizations, and thus improving compliance would reduce both. This finding, along with studies about the costs associated with medication errors, point to the value of focusing on medication compliance and proper usage in initiatives for reforming healthcare delivery.
For example, improving medication compliance might be improved by creating better written instructions for patients. An analogy might be to the checklists that help surgical teams and physicians placing central IV lines reduce errors and infections. Individualized patient “checklists” created in a standard format could help patients with chronic conditions take their medicines correctly and thus improve quality and lower costs. And of course, these “checklists’ should also include ALL the patient’s medicines, reminders about avoiding certain OTC medicines and herbal supplements that might interfere with their prescribed medicines, and to urge them to promptly inform their physicians if they develop certain signs or symptoms related to their medical conditions.
All of this really goes back to the benefits of improving patient-clinician communications, and strengthening their joint engagement with the broader care team that includes pharmacists, nurses, specialized clinical educators, and family members as appropriate.
Strengthening Care Teams Through Better Communications
Such checklists and more standardization of care protocols – for use by clinicians as well as patients – should help foster this communications and team-base management. However, if they are viewed by anyone in the team as cookbook medicine, or impersonal standardization – rather than guidance to protect against overlooking important actions and opportunities – then these protocols will not produce positive clinical and economic outcomes. To encourage positive attitudes and actions about such quality improving changes, clinicians and patient advocacy leaders of all types – and at all levels – will need to be educated about the value of such shifts in clinical operations, and for them to be promoted as champions of these new paradigms for clinical care. (Group visits for patients with conditions like diabetes or congestive heart failure are an example of care delivery innovations that can increase quality and lower costs through greater efficiency and engagement of patients and their care team.)
Changing Reimbursement to Support Care Delivery Reforms
Making these changes to the clinical care delivery so that they are attractive to clinicians – both in private practice and in larger care delivery systems – will likely require altering reimbursement systems to change the financial incentives. The Medicare Medical Home Demonstration project seeks to do this by directly paying physicians for their medical home management activities.
While this Demonstration is being implemented, a more widespread and timely change in Medicare’s physician payment system may happen this year because of the unsustainability of Medicare’s “Sustainable Growth Rate” formula. Under the current SGR formula, Medicare payments for physician services are expected to be reduced by ~22% starting in January 2010. Deliberations to avoid this “broadsword” financial cliff may be a leverage point for making such changes – although it is unlikely that a dramatic, wholesale change will occur immediately. It is more likely that significant changes will be phased in over several years and across geographic regions. A gradual implementation makes sense since it provides opportunities for clinicians and Medicare enrollees to become familiar with the new payment and clinical delivery structures, it enables champions for better care delivery methods to be created, and it allows those who are more cautious to observe the changes while they are implemented for and by others.