Primary Care-Oriented Pathways to Support Population-Based Payment Models

Abstract: While clinical pathways have been identified as a key strategy for succeeding under inpatient episode-based payment models, they have not been widely touted as part of primary care population-based arrangements such as accountable care organizations (ACOs). In fact, despite high interest in pathways among primary care physicians, overall utilization remains low. Given limitations of “patient panel” strategies that adopt a population-wide view and encourage focus on improving performance on population-level measures, primary care-oriented patient-level clinical pathways can serve as meaningful complements to drive performance and help achieve payment model goals. Several unique features of primary care delivery will impact pathway creation and implementation, including cadence of care delivery, timeline for expected quality improvements and/or cost reductions, teaming, and breadth of clinical focus. To integrate these features, we propose several considerations for designing primary care-oriented pathways within ACO-type arrangements. In particular, careful attention should be paid to design considerations, eg, starting and ending points, prioritization of condition-based pathways based on breadth of impact, specialist engagement for clinically complex patients, use of noncondition-based pathway definitions, and workflow integration. These are critical elements to ensure that primary care clinical pathways can help organize and improve outcomes and reduce costs in ACOs and other population-based payment models.

Amid the nationwide shift from volume-based to value-based care, providers such as physician groups and hospitals are increasingly engaged in population-based payment models. Compared to volume-based incentives under traditional fee-for-service payment, population-based payment models hold clinicians and organizations financially accountable for quality and cost outcomes across defined patient populations.

Nearly all population-based arrangements emphasize ambulatory care, and primary care in particular. For example, accountable care organizations (ACOs) have been prominently featured and adopted by multiple public and private payers across the country. The predominant ACO quality metrics focus on preventive medicine (eg, immunizations, age-appropriate cancer screenings) and chronic disease management (eg, diabetes, hypertension) as vital aspects of primary care delivery.

Consequently, health care organizations in ACOs often attempt to drive performance through care delivery strategies that center on primary care populations, sometimes called patient panels. For example, common approaches emphasize population-based “panel management,” through which primary care teams (1) stratify populations into high vs nonhigh risk patients; (2) work to routinize processes that improve performance on population-level preventive and chronic disease management measures (eg, proportion of eligible patients screened for breast cancer); and (3) deploy dedicated programs and human resources to supplement routine care among high-risk, costly, or “super-utilizer” patients.

While such interventions may be appropriate, evidence suggests that they also possess limitations.^1,2 Primary care-oriented clinical pathways that focus on individual patients can serve as meaningful complements to drive performance and help achieve population-based payment model goals. However, there is a dearth of evidence about the use of pathways in this context, and our anecdotal evidence suggests that it is often an overlooked intervention, as compared to population- or panel-based interventions. Therefore, in this article, we use the ACO payment model as an emblematic example to articulate this opportunity, highlighting unique features of such pathways through comparison to hospital-based care and corresponding inpatient pathways.

The Opportunity to Increase the Use of Pathways to Support ACOs

While clinical pathways have been identified as a key strategy for succeeding under inpatient episode-based arrangements, such as bundled payments,^3-5 they have not been widely touted as part of a primary care population-based payment model strategy. In fact, despite high interest in pathways among primary care physicians, overall clinical utilization (eg, in the context of clinical decision support within the electronic health record) remains low.⁶ 

Based on our collective experience in value-based payment and primary care transformation, we believe that organizations participating in ACO models have opportunities to increase the utilization of pathways. Used alone, “panel management” strategies emphasize group outcomes (eg, population-level hypertension control rate) over individual outcomes (eg, a given patient’s plan for hypertension management).⁷ Implementing pathways within population-based models improves the ability to provide patient-centered, individual-focused care while promoting population-based outcomes. 

The Need to Design Pathways That Reflect Features of Primary Care Delivery 

Despite the opportunity to implement pathways within ACO arrangements, we also recognize several unique features of primary care delivery that impact pathway creation and implementation.

Cadence of care delivery. Inpatient care is characterized by numerous, high-frequency interactions (eg, daily rounds, ability for continual monitoring) over a relatively short period (eg, an acute hospitalization). In contrast, primary care is characterized by periodic, lower frequency, and non-face-to-face contact over longer periods.

Timeline for expected quality improvements and/or cost reductions. Inpatient interventions can be expected to generate quality gains and/or cost reductions quickly (eg, during or shortly following a given hospitalization), whereas the expected effects of primary care optimization emerge further downstream in the care continuum (eg, better prevention, decreased total costs of care and/or reduced emergency room and hospital utilization). 

Teaming. Differences in clinical cadence and timeline for expected benefit also create potential distinctions in how to optimally assemble multidisciplinary teams to deliver inpatient vs primary care. Inpatient teams are optimized by co-location and ability to provide high-intensity input (eg, same day clinical consultations or procedures), whereas outpatient teams are generally more dispersed over time, location, and even organizations.

Breadth of clinical focus. In most cases, inpatient care focuses on several major acute conditions, while primary care is often delivered while considering the interplay of multiple chronic conditions,⁸ none of which may be considerably more acute or higher priority than others. For example, managing acute heart failure exacerbation in the hospital may involve setting aside lower acuity, chronic conditions, such as depression, whereas both conditions may possess equal weight and focus in the context of primary care.

Design Considerations

To integrate these features of primary care delivery, we propose several considerations relevant for designing primary care-oriented pathways within ACO-type arrangements.

Starting and Ending Points

In inpatient settings, there is clear rationale for designing clinical pathways that begin with admission and end with either discharge or after a discrete period of post-acute care. In contrast, primary care and the management of chronic conditions often extends longitudinally without absolute end points (eg, ongoing management
of essential hypertension) and traverses both in-person and non-face-to-face encounters (eg, management of chronic heart failure through a combination of in-person
visits and intervening phone calls). In turn, one understandable but still potential problematic risk is implementing primary care-oriented pathways that have ambiguous or no starting and ending points. A lack of specific timepoints can complicate efforts to track patients across different pathways for a given condition and to know when to adjust pathway design vs address patient factors, or both. 

Instead, potential solutions include defining pathway starting and ending points based on changes in health status either in a negative (eg, patients with diabetes may trigger a pathway for intensive disease management when they develop evidence of new microvascular complications) or positive (eg, patients may finish or come off of a diabetes disease management pathway after a period with stable hemoglobin A1c or without worsening or new microvascular complications) direction. By creating separate pathways for the same condition and implementing them in sequence as appropriate for specific patients, this type of approach can still align with a longitudinal perspective on primary care and chronic disease management (eg, a diabetic with well-controlled, stable hemoglobin A1c would graduate from the intensive glucose management diabetes pathway to a lower-touch diabetes monitoring pathway). Pathway end points should accommodate non-face-to-face care elements that are relevant to many aspects of primary care (eg, telephone encounters, secure e-messages).

Prioritization of Condition-Based Pathways Based on Breadth of Impact

On one hand, it is infeasible to deploy multiple longitudinal primary care pathways, each with a specific set of actions and interventions, for patients who have multiple chronic conditions simultaneously. For example, if a primary care practice implemented clinical pathways for history of stroke, diabetes, hypercholesterolemia, and depression, patients with these concurrent conditions would trigger multiple pathways that could create information overload or confusion for patients and alert fatigue for clinicians. On the other hand, multiple conditions can possess equal weight and focus in the context of primary care (eg, heart failure and depression, as noted above).

In contrast to inpatient pathways, where higher urgency issues take clear priority, primary care clinical pathways could focus on (a) areas with the broadest potential clinical impact over a longer time horizon (eg, prioritizing blood pressure control given its role in preventing outcomes for multiple diseases, such as diabetes, chronic kidney disease, stroke and coronary artery disease) and (b) shared decision-making (eg, prioritizing low back pain management based on patient preferences and values). Specific prioritization strategies will likely differ by local organizations and patient population factors.

Specialist Engagement for Clinically Complex Patients

In the hospital, patients with clinical complexity are able to access the appropriate specialists either as their primary provider (eg, the patient is admitted to a given specialty’s “inpatient service”) or through close inpatient consultation. Therefore, inpatient pathways do not always need to be designed to include detailed processes for specialist engagement. In contrast, under ACO cost containment incentives, primary care-oriented pathways should increasingly incorporate best practices for specialty referral.⁹ For example, when available, pathways could incorporate e-referral processes through which primary care providers pose questions virtually and defer decisions about whether patients need in-person referral to specialists.

Noncondition-Based Pathway Definitions 

There are multiple types of primary care visits, encompassing initial and follow-up encounters for clinical conditions as well as encounters that are not necessarily based on specific diseases or conditions. For example, the Medicare Transitional Care Management (TCM)¹⁰ program is designed to encourage primary care clinicians to coordinate care for patients transitioning from inpatient hospitalization back to ambulatory care. Clinical pathways could help streamline the process of providing TCM services, which are associated with clinical benefit and cost reductions¹¹ but require significant coordination of care by the primary care team (eg, requires a post-discharge phone call within two business days of discharge and primary care visit within 14 days of discharge). 

By helping to coordinate care among multiple team members within specific timeframes, pathways can also support primary care encounters focused on other transitions (eg, between primary care and emergency room visits). Other focuses for noncondition-based pathways could include social determinants of health (eg, a homeless health pathway with integrated social work services) and patient self-management (eg, connection to health coaching and community resources). One critical factor that supports such efforts and avoids gaps between hospital and ambulatory care is provider accountability across the care continuum. In turn, ACOs and other population-based models designed explicitly to create this type of broad accountability are particularly well suited for noncondition-based pathways.

Workflow Integration

While integration is important for all types of pathways, it is particularly challenging and critical to emphasize for those deployed in primary care. This is primarily because, in contrast to inpatient or ambulatory specialty care, primary care is already marked by breadth of scope and geographically dispersed care team members.

For example, potential tasks involved in primary care delivery can range from urgent administration of medications and referral to the emergency room to timely counseling about new diagnoses and completion of disability paperwork. Given the range of existing tasks and workflows, primary care-oriented pathways should be effectively integrated into clinical processes and electronic health record platforms.¹²

Conclusion

As a care redesign strategy, primary care clinical pathways can help organize and improve outcomes and reduce costs in ACOs and other population-based payment models. Critical to this success are effective design, prioritization, implementation, and workflow integration.

References

1. Edwards ST, Peterson K, Chan B, Anderson J, Helfand M. Effectiveness of intensive primary care interventions: a systematic review. J Gen Intern Med. 2017;32(12):1377-1386. doi:10.1007/s11606-017-4174-z

2. Zulman DM, Pal Chee C, Ezeji-Okoye SC, et al. Effect of an intensive outpatient program to augment primary care for high-need Veterans Affairs patients: a randomized clinical trial. JAMA Intern Med. 2017;177(2):166-175. doi:10.1001/jamainternmed.2016.8021

3. Liao JM, Holdofski A, Whittington GL, et al. Baptist Health System: succeeding in bundled payments through behavioral principles. Healthc (Amst). 2017;5(3):136-140. doi:10.1016/j.hjdsi.2016.04.008

4. Liao JM. Bundled payments: a rising tide for clinical pathways. J Clin Pathways. 2018;4(10):58-60. doi:10.25270/jcp.2018.12.00051

5. Liao JM, Chen C. Getting “bundle ready”: how provider organizations can design clinical pathways in the context of bundled payments. J Clin Pathways. 2019;5(1):49-51. doi:10.25270/jcp.2019.02.00056 

6. Toy JM, Drechsler A, Waters RC. Clinical pathways for primary care: current use, interest and perceived usability. J Am Med Inform Assoc. 2018;25(7):901-906. doi:10.1093/jamia/ocy010

7. Porter ME, Kaplan RS. How to pay for health care. Harv Bus Rev. 2016;94:88-100.

8. Grimsmo A, Løhre A, Røsstad T, Gjerde I, Heiberg I, Steinsbekk A. Disease-specific clinical pathways – are they feasible in primary care? A mixed-methods study. Scand J Prim Health Care. 2018;36(2):152-160. doi:10.1080/02813432.2018.1459167

9. Song Z, Sequist TD, Barnett ML. Patient referrals – a linchpin for accountable care. JAMA. 2014;312(6):597-598.

10. Bindman AB, Blum JD, Kronick R. Medicare’s transitional care payment--a step toward the medical home. N Engl J Med. 2013;368(8):692-694. doi:10.1056/NEJMp1214122

11. Bindman AB, Cox DF. Changes in health care costs and mortality associated with transitional care management services after a discharge among Medicare beneficiaries. JAMA Intern Med. 2018;178(9):1165-1171. doi:10.1001/jamainternmed.2018.2572

12. O’Malley AS, Draper K, Gourevitch R, Cross DA, Scholle SH. Electronic health records and support for primary care teamwork. J Am Med Inform Assoc. 2015;22(2):426-434. doi:10.1093/jamia/ocu029=