High rates of hospital readmissions are a persistent problem in the USA. Widespread efforts to implement hospital-based care transitions programs, such as RED (Re-engineered Discharge) and others, have garnered considerable attention and have yielded favorable results over the past several years.1,2,3,4 While hospitals continue to refine their readmission reduction efforts, their ability to control what happens after discharge and outside of their walls remains limited. With recent estimates suggesting that approximately a quarter of hospital readmissions may be avoidable, it is essential for us to pursue new approaches to address gaps in the care transitions process.5

One area ripe for exploration is primary care–based initiatives, which to date have received less emphasis. Many successful hospital-based programs involve the coordination of post-discharge ambulatory services, but outpatient clinics may play only a passive and limited role in the process. Given that the core components of primary care, as described by Barbara Starfield in her seminal book, are contact accessibility, coordination, comprehensiveness, and continuity, it follows that primary care ought to serve a more prominent role in care transition efforts.6 Yet there are few concerted efforts underway to develop, implement, and assess primary care–based transitions programs.

In a recently published environmental scan commissioned by the Agency for Healthcare Research and Quality (AHRQ), we assessed the current state of the literature on primary care–based care transitions programs.7 We identified numerous peer-reviewed and non-peer-reviewed studies ranging from narrow interventions to more comprehensive multi-component initiatives. While some of the included studies had methodological limitations (there were few rigorous controlled analyses), collectively, they shed light on the types of interventions that are likely to be successful, revealing practical considerations and policy changes needed for making these initiatives sustainable. In this piece, we summarize insights from the environmental scan and make a case for an enhanced and expanded role for primary care in the care transitions process.

BUNDLED VERSUS SINGLE-COMPONENT INTERVENTIONS ARE MORE LIKELY TO SUCCEED

First, we found that multi-component, or “bundled,” programs addressing multiple care transitions challenges that patients and providers face were some of the most successful. These programs often utilize clinic-based care managers and include several distinct processes such as close follow-up with a provider, post-discharge phone calls, medication reconciliation, addressing transportation barriers, and scheduling follow-up with social workers and other critical primary care team members.8,9,10,11,12,13,14 Conversely, studies examining the impact of more narrowly focused programs, such as implementing automated admission notifications, medication reconciliations, or post-discharge phone calls alone, were less encouraging.

Second, primary care practices with bundled transitions programs tended to be initiated early in the patient’s hospital course. While primary care notifications for patient hospitalizations alone were not enough to reduce readmissions, hospitalization alerts were frequently an important component of successful interventions. These bundled programs often “reached in” to patients in the hospital prior to discharge—either to assess the patient’s biopsychosocial needs, obtain accurate patient or caregiver contact information, set up follow-up appointments, communicate with the inpatient care team, or assist with developing discharge instructions.10,11,12,13,14,15

Third, the vast majority of bundled primary care–based interventions described in the literature occurred either in primary care practices that were affiliated with hospitals or in settings in which a third-party payer initiated the program. Evidence is lacking as to whether small, independent practices have the capacity to implement multifaceted interventions without support from an affiliated hospital or third-party payer.

Lastly, some of the most promising transitional care interventions occurred in the setting of larger scale practice transformation, such as those guided by the Primary Care Medical Home framework. The success of transitions programs in a primary care medical home setting is perhaps not surprising given that care coordination is a key building block in the medical home models described by Bodenheimer and Wagner.16, 17 We also suspect that these broad clinic “redesign” efforts may have had a synergistic effect with the transitional care interventions, leading to a more pronounced effect on readmission rates.

INNOVATION IN PRIMARY CARE REQUIRES SYSTEM-WIDE REFORMS

Our findings suggest a number of changes in clinical practice and policy that could help promote engagement by primary care practices in the care transitions process.

First, poor communication between hospitals and primary care clinicians may impede the development of primary care–based care transitions programs. Many of the articles on care transitions that we reviewed highlighted challenges with care coordination.18,19,20 Primary care–based efforts might be more effective if they align with hospital-based interventions, with well-defined roles and accountability throughout the care transitions process.8, 10, 11 The existence of integrated or compatible electronic health records for all clinics in the system may facilitate collaboration but is insufficient in isolation.21

Second, providing adequate support, resources, and compensation for primary care clinics to lead care transitions efforts is imperative. Several of the successful primary care–based transitions programs we reviewed required significant investment (in many cases from health plans) in practice innovations such as home visits and the hiring of care coordinators and social workers (who typically cannot bill for their services).22,23,24 Though many practices lack the resources to invest in such team members, these groups might be able to hire necessary personnel by pooling their resources. A couple of years ago, the Centers for Medicare and Medicaid Services launched fee-for-service programs that reimburse primary care providers for transitional care and chronic care management.25, 26 While these programs are a step in the right direction, they have limitations. The recently evaluated Comprehensive Primary Care Initiative, which focused on clinics serving fee-for-service Medicare beneficiaries, resulted in improved coordination of care transitions and reduced emergency department visits, but failed to reduce readmissions or total cost of care.27 Fee-for-service payments alone may be insufficient to support the comprehensive, clinic-level transformation that is often required to reduce readmissions.28

Many experts have proposed alternative payment strategies, particularly value-based payments, to improve outcomes among primary care providers. Pay-for-performance programs and capitated risk-bearing payment models might better incentivize primary care practices to invest in care transitions activities, enabling primary care providers to accept increasing risk and accountability for the populations they serve. To date, evidence on value-based payment models in primary care shows modest benefits with respect to some process measures, while demonstrating a benefit with respect to health outcomes has thus far proven difficult.29 Some cost savings, however, have been suggested in physician-led models.30 Furthermore primary care providers may be reluctant to accept risk for variables outside of their control—notably care that patients receive in the hospital as well as social and environmental health determinants (factors that are generally not addressed by the health care system, at least not in the USA). Next year, the Centers for Medicare and Medicaid Services will be launching Primary Care First, a program designed to encourage primary care to reduce health care utilization and increase care quality through flexibility, enhanced revenue and value-based reimbursement.31 Examining the implementation and impact of this program could enhance our understanding of the effectiveness of value-based payment programs in enhancing care coordination and reducing avoidable readmissions.

Finally, our review for AHRQ focused on traditional primary care approaches for reducing readmissions. However, it is important to acknowledge a growing consensus among policy experts that social determinants of health play a significant role in overall health outcomes including hospital readmission.32 Primary care clinicians have long understood the value of working with organizations that utilize community health workers and other social service agencies to help patients facing poor housing, food insecurity, and social isolation.33, 34 The review found only limited research on primary care collaboration with community health organizations during the care transitions process. Understanding how primary care could effectively partner with programs that address socioeconomic challenges offers an intuitive approach for preventing readmissions and improving quality of care.

PRIORITIZING PRIMARY CARE INNOVATION IS ESSENTIAL FOR REDUCING READMISSIONS

While hospital-based efforts have modestly lowered overall readmission rates in the USA, primary care clinics must take on a central role in the care transitions process to achieve greater impact.35 In fact, we believe that primary care clinicians and teams should “own” this process. Our review reveals several ways in which primary care can take on this central role. First, care transitions efforts should include bundled or multifaceted interventions that address gaps on several fronts, ranging from medication management and discharge alerts to patient education and home care, depending on patient needs. These efforts may be most effective when implemented early in the hospital course and within the context of large-scale primary care transformation. Second, such efforts should be aligned with hospital-based efforts. This means primary care practices should consider developing relationships with hospitals frequented by their patients, perhaps through a formalized “community contract”, with a mutual goal of sharing data, identifying high-risk patients, communicating during a patient’s hospitalization, and coordinating care transitions processes.36 Third, and most importantly, to study, implement, and expand care transitions programs, health systems leaders and policymakers must invest in primary care staff and infrastructure that will enable primary care clinics to take the lead.

Our review of the body of evidence leads us to believe that primary care represents the “next frontier” in efforts to reduce readmissions. It is time for policymakers and other key stakeholders to provide the resources, incentives, and authority necessary for primary care clinics to take charge of the care transitions process.