About the Discussion

@NEJM Ask the Experts & Authors: Bridging the Hospitalist–Primary Care Divide through Collaborative Care

Perspective

Bridging the Hospitalist–Primary Care Divide through Collaborative Care

Article

One of the most notable changes in the delivery of medical care in the United States in the past quarter-century has been the near disappearance of primary care physicians (PCPs) from general medical inpatient care, replaced by a new breed of generalist: the hospitalist.1 There are many reasons for this shift, ranging from revenue-driven pressures to increase office visits and minimize lengths of hospital stays to physicians' desires for more manageable lifestyles as both outpatient and inpatient care become more intensive and demanding. Many young physicians who aspire to be generalists — especially graduates of general internal medicine residencies — now choose hospitalist positions over office-based primary care opportunities.2

Hospitalist care is more efficient than traditional models of inpatient care and, on average, appears to be of similar quality.3 Good communication among hospitalist team members and between hospitalists and PCPs can lead to seamless, efficient, well-coordinated care; however, shift-work schedules and suboptimal communication and pass-offs can result in fragmented, impersonal care and excessive testing.4 Inpatients often ask, “Who is the doctor here?” and wonder, at moments of acute need, “Where is my own doctor?” Hospitalists can become overwhelmed with the care of patients with complex conditions whom they've only just met; especially difficult are personalized decisions about the appropriate intensity of care, especially near the end of life. PCPs may feel isolated from their patients and disconnected from decision making, which can compromise postdischarge management.

These shortcomings suggest that the current dichotomous division of labor between hospitalists and PCPs warrants reconsideration, though any alternative approach needs to respect the achievements of the current system, which have included reduced lengths of stay, enhanced standards for inpatient care, the development of a new cadre of generalist clinician-educators in teaching hospitals, and the transformation of primary care practices into medical homes. Past solutions to address issues of continuity of care and information transfer have included designating members of a primary care practice or group to serve, on a rotating basis, as the attending physicians for their patients. A related option is to assign a small number of hospitalists to a primary care group or to embed them within it. Logistic and professional barriers such as equity in scheduling and payment, cross-coverage needs, and the challenge of interfacing with unfamiliar and complex medical and system environments have limited the adoption of these models. Mindful of the forces that maintain the current division of labor, but taking into consideration emerging reforms in primary care payment and practice, we have been exploring alternative strategies for bridging the chasm between inpatient and primary care. The endeavor is timely in an era of multidisciplinary teams, patient-centered care, and enhanced payment for care management and coordination.

In 2002, Wachter and Pantilat suggested that PCPs make a “continuity visit” when their patients were hospitalized, to maintain involvement and help coordinate their care.5They envisioned a visit once or twice during a patient's hospital stay that would involve a brief discussion with the patient, a focused exam, and documentation to assist the hospitalist team, for which physicians would receive a modest payment. Although PCPs do sometimes make such visits, our experience suggests that this practice never became widespread because of persisting financial and work pressures. With the advent of primary care practice transformation and increasing interest in payment reform, we believe it might be time to reconsider and revitalize this model.

We propose a collaborative inpatient care model that incorporates the PCP into the hospitalist team as a consultant. Under this voluntary system, PCPs would visit their hospitalized patients within 12 to 18 hours after admission to provide support and counseling to them and their families and consultation to the hospitalist team. The consultation would focus on the direction and scope of the patient's workup and care. The PCP would write a succinct consultation note, highlighting key elements of the patient's history (including pertinent family and psychosocial components), physical exam, and recent testing, and conclude with a prioritized differential diagnosis and recommendations for personalized inpatient evaluation and management. The hospitalist team would still retain full attending-physician responsibilities.

The initial consultation — contributing insights from an established doctor–patient relationship — would be designed to complement and help inform the hospitalist's admission workup and care plan, aiming to reduce hospitalist workload while increasing personalization of care. Subsequent to the admission consultative visit, the PCP would be available to meet with the patient, family, and hospitalist team on an as-needed basis, returning just before discharge to consult on the design of a coordinated posthospital program.

A key challenge to implementing collaborative inpatient care will be ensuring that PCPs have time to fulfill this role, which can be especially daunting for those in solo or small-group practices that are still reimbursed under traditional fee-for-service arrangements. Many such practices cannot afford the often-recommended multidisciplinary team structure that can shift PCP workload to more value-added activities such as care coordination and management. Having a well-functioning primary care team is probably essential to PCP participation in collaborative inpatient care. Emerging electronic technologies could play an important facilitative role, making possible “virtual visits” when a live visit is not deemed necessary, freeing up time for essential live visits.

For collaborative care to succeed, the hospitalist workflow also needs adjustment to make best use of the PCP's input. The inpatient team's organization and communication protocols will need to be revised. Enhanced information flow (made possible by an interoperable electronic medical record) and timely collaboration and consultative input from the PCP (some of which might be virtual) should allow hospitalists to care for inpatients more efficiently and cost-effectively.

Implementing collaborative inpatient care would also require payment reform that removes incentives for maximizing the volume of office visits. Risk-adjusted bundled and global payment models are one approach to payment reform; capitated care-management fees that supplement fee-for-service contracts are another. Even a traditional fee-for-service payment system could support a collaborative inpatient care model if it were complemented by a new set of evaluation and management codes for collaborative care, with appropriate scores reflecting the total amount of time and effort that PCPs spend on inpatient consultations and taking into account the complexity of the patient's conditions and the value created by these visits.

Budget-conscious observers may raise concern that payment to PCPs for inpatient work could increase total costs. Savings should accrue, however, from improved diagnostic efficiency and accuracy; from reductions in lengths of stay, unnecessary testing, preventable readmissions, and inappropriate discharges; and from enhanced compliance, follow-up, and patient satisfaction — measurable parameters that can be used to evaluate the efficacy of the proposed reforms. We would expect savings to be realized in both inpatient and outpatient settings. Pilot studies of this proposed model could focus on overall cost, clinical outcomes, operational metrics (e.g., length of stay, prenoon-discharge rates, 30-day readmissions), and patient satisfaction.

The current hospitalist–ambulist division of general medical care has made important contributions to patient care, but it leaves much to be desired, especially with regard to personalization and continuity of care. A collaborative inpatient care model that incorporates the PCP into the hospitalist team as a consultant has the potential to cost-effectively improve the patient's care experience and other important outcomes, as well as enhance professional satisfaction at this critical time for both hospitalists and PCPs.