The article in this issue “Community Care of North Carolina: Improving Care Through Community Health Networks” describes a decade-long program of quality improvement and cost reduction created by North Carolina Medicaid. The article provides lessons of value to all health professionals and leaders concerned with the cost and quality of care.1
LESSON 1: SIZE MATTERS
During the past 10 years, improvement work has flourished in a number of developed nations, including the United States. Most of this work, however, takes place in one institution—a hospital safety project or a primary care diabetes program—and is small. The Community Care of North Carolina innovation encompasses many institutions—1,200 primary care practices—and is large, 750,000 patients. How long will it take for improvement work involving one disease in one practice to transform health care? Perhaps as long as reversing global warming one gas-guzzling SUV at a time. Health care needs large projects, and Community Care of North Carolina should grab many leaders’ attention for its size alone.
LESSON 2: IMPROVEMENT TAKES TIME
In addition to being small, much improvement work is short-lived, because it is often based on a regional collaborative or a 3-year project grant. Without being permanently sustainable, such work is of little value and, in fact, creates discouragement among reformers. Community Care of North Carolina began as a small pilot in 1988 and was launched in earnest in 1998. Had it not had able leadership, the small 1988 pilot could easily have died, joining the cemetery of so many other successful pilots that failed to become scalable and sustainable. Moreover, many project leaders expect their metrics to improve in a year or 2, failing to recognize that improvement takes time. By 2006, Community Care of North Carolina was showing cost savings of $161 million per year, an amount, when compared with baseline data from 2000, 2001, and 2002, that is more impressive than the findings of only 1 or 2 years of cost tracking would have shown.
LESSON 3: COLLABORATION IS THE KEY
Payers and clinicians have a long tradition of hostility. Payers blame clinicians for uneven quality and lack of prompt patient access to appointments. Clinicians complain about low pay, mindless paperwork, and ruthless denials of care. Much of this hostility has a factual basis, and sometimes the battles produce needed change. But imagine the potential for improvement if practices and payers worked together. Collaboration may be the central lesson taught by Community Care of North Carolina, which is a compact between payer and practices. North Carolina Medicaid agrees to pay additional money (a management fee to primary care practices) and provide additional services (care managers who may work with several practices); in return, practices agree to provide 24-hour on-call coverage, care coordination with other entities, and a willingness to engage in practice improvement work initiated by Medicaid or by the regional networks that have strong local clinician participation.
Many primary care advocates feel that payers should first provide the funds, and only then will they initiate practice improvement. Many payers feel that primary care practices need to show improvement to deserve new revenues. Pay-for-performance provides some new money only after improvement has been demonstrated. Under a different model, payers would invest in primary care improvement before the improvements have taken place.
Community Care of North Carolina lies closer to the investment model, but the investment is not an open-ended “we will pay you extra and we hope you will improve.” Rather, the investment is a compact: “we will pay you extra and you are expected to improve.” Furthermore, structures will be developed—in particular the regional networks—to guarantee ongoing conversation between payer and clinicians and to monitor and further develop the entire process.
LESSON 4: LINK PATIENTS TO A MEDICAL HOME
The medical home seems to be the current panacea that will cure all that ails US medical care. Community Care of North Carolina highlights a key feature of the medical home (sometimes lost in the myriad of requirements being proposed to define medical homeness): each patient is linked to a medical home. Many developed nations base their health care systems on the principle that each person in society should sign up with a primary care practice and that each primary care practice should assume full responsibility for the health care of patients in their panel. The United States is a long way from achieving even that simple characteristic of the medical home and should perhaps, as did Community Care of North Carolina, begin at the beginning in implementing the medical home concept. After all, is not medical home just a modern appellation for primary care with its 4 pillars of first-contact care, longitudinal care, comprehensive care, and coordination of care?
LESSON 5: FEE-FOR-SERVICE DOES NOT MIX WITH CHRONIC CARE
Acute care and the performance of procedures generally involve a clinician (physician, nurse-practitioner, or physician assistant) providing a limited number of distinct services for a patient. Although fee-for-service payment is by no means the only way to pay for those episodic services, it is a reasonable option. Chronic care, in contrast, involves long-term continuous management of patients, often by care teams with nonclinician caregivers, often relying on telephone, e-mail, Web-based, or group encounters that are not traditional visit-based services. Moreover, because many patients with chronic conditions have complex medical and psychosocial problems that require more services than the average primary care practice can offer, time-consuming care coordination is required, generally taking place between visits. Thus much of the care appropriate for the management of both uncomplicated and complex chronic conditions is not reimbursed and a poor fit for the fee-for-service payment mode. Proposals have been made for both a blended payment model (fee-for-service with additional payment for nonvisit care)2 or a comprehensive, risk-adjusted per-patient per-month model that banishes fee-for-service from primary care.3
Community Care of North Carolina has opted for the blended payment model: a $2.50 per-member per-month management fee to primary care practices on top of fee-for-service visit-based payments, plus chronic care team support provided by Medicaid-paid case managers (often called care managers by other organizations) who assist clinicians caring for patients with highly complex clinical conditions. For larger practices, the management fee is sufficient to pay for 1 or 2 new staff members, allowing practices to forge care teams essential to improving care in an environment of rushed physicians struggling through the 15-minute-visit–studded day.
LESSON 6: SMALL PRACTICES NEED SUPPORT
A dilemma of US health care is the future of small primary care practices; 36% of primary care physicians continue to work in practices of 1 or 2 physicians.4 Small practices have failed to match the quality metrics of larger integrated systems.5 They are financially stressed and in some areas may be on the way out. Yet patients like small practices, as do some clinicians. What to do?
If small practices are to survive and provide high-quality care, they must become part of an aggregating organization that links them and supports them with personnel (health educators, care managers, clinical pharmacists), billing services, and data. Such aggregating organizations could be hospitals (unlikely because primary care physicians are increasingly divorced from hospitals as a result of the hospitalist movement), independent practice associations (which have taken on this role to some extent in California), or payers. Community Care of North Carolina, through its regional networks, has taken on the role of an aggregating organization for small North Carolina practices.
LESSON 7: PUBLISH INNOVATIONS
The final lesson addresses the contribution of the editors of Annals of Family Medicine. Often descriptions of practice and payment innovations are excluded from journal pages because they lack perfect research designs, fail to compare an intervention with a control group, and present unorthodox end points with non-robust statistical analyses. Yet the perfect randomized controlled trial may be of little help in designing a real-world improvement, whereas a real-world improvement description without a careful research design may prove enormously useful. Hats off to the Annals for publishing this article. I hope all journal editors are reevaluating their criteria for publishing health care improvement manuscripts.
“Community Care of North Carolina: Improving Care Through Community Health Networks” is step 1 in informing the medical community of an important innovative model. I hope there will be a step 2: information (both quantitative data and qualitative interview material) from a sampling of the 1,200 practices to describe what improvements the practices have made, how widespread these improvements are, and how the clinicians in these practices view the endeavor. In the meantime, it is prudent to heed the authors’ conclusion that Community Care of North Carolina “is a model of care that has moved beyond theory and could be implemented across the country.”
Footnotes
-
Conflicts of interest: none reported
- Received for publication April 8, 2008.
- Revision received May 2, 2008.
- Accepted for publication May 16, 2008.
- © 2008 Annals of Family Medicine, Inc.