High-Intensity Primary Care: Lessons for Physician and Patient Engagement

NIHCR Research Brief No. 9

To prevent costly emergency department visits and hospitalizations, a handful of care-delivery models offer high-intensity primary care to a subset of patients with complex or multiple chronic conditions, such as diabetes, congestive heart failure, obesity and depression. Early assessments of high-intensity primary care programs show promise, but these programs’ success in improving quality of care and lowering costs rests on the engagement of both physicians and patients. A number of factors can foster physician and patient engagement in high-intensity primary care programs, according to a new qualitative study by the Center for Studying Health System Change (HSC). For physicians, key factors include financial commitment and administrative support from health plans and well-designed financial incentives for quality and outcome improvements. In addition, allowing physicians to help identify patients who would benefit from intensive primary care may improve physician comfort and buy in. To encourage patient engagement, a personal invitation from physicians to join a high-intensity primary care program, as well as rapid access to physicians and care coordinators, appear to be highly successful approaches.

High-Intensity Primary Care

In recent years, awareness has grown that fragmented care—especially for complex patients with multiple chronic conditions—contributes to poor quality and high costs. Many believe that better care coordination for complex patients can improve patient outcomes and reduce costs. Approaches to improving care delivery—including patient-centered medical homes—increasingly focus on strengthening primary care to prevent complications leading to costly emergency department visits and hospitalizations. Another goal is to give patients an initial place to consistently seek care, which may reduce use of other, more-expensive care settings and self-referrals to specialists.

By definition, a patient-centered medical home provides team-based care that is integrated and coordinated by primary care physicians (PCPs) for all patients in a practice.1 In an effort to improve care for people with complex chronic conditions, some large purchasers—typically large employers, unions or a combination of the two working with health plans—are using a model of high-intensity primary care. While similar to a patient-centered medical home, high-intensity primary care programs differ in that they focus only on the sickest, highest-cost patients in a given group, providing them with additional care coordination, management and health education far beyond what is offered in traditional primary care practices.

Early assessments indicate high-intensity primary care programs, sometimes called ambulatory intensive care units, or AICUs,2 appear to be effective in the small number of settings where they have been tested to date. The high-intensity primary care model typically involves PCPs supported by care coordinators who help complex patients navigate the health system, adhere to treatment plans and improve self-care of their conditions. Some models also include specialist physicians, nurse practitioners, social workers, patient educators and others.

Care coordinators, who usually work closely with physicians, often are registered nurses or medical assistants with some clinical background and training in health education. These care coordinators are the primary point of contact for patients and coordinate care in a number of ways, including, for example, scheduling primary and specialty care appointments, identifying and connecting patients to social-support services, working with patients to improve medication adherence and other self-care tasks, assisting with follow-up care after hospital discharges, and helping patients set and reach specific health goals.

This Research Brief examines approaches to patient and physician engagement in six high-intensity primary care programs in New Jersey, New York, Washington, Oregon, northern California and multiple sites across the Southwest (see Data Source). The six programs range from a freestanding health center that serves patients in partnership with a local union to a practice-based model that serves patients with several different sources of insurance coverage. While each model has a unique structure, there are common approaches and goals for patients (see box below for more information about the programs).


Examples of High-Intensity Primary Care Programs

Freestanding Programs

AtlantiCare Special Care Center (SCC), southern New Jersey, is a medical clinic owned and staffed by AtlantiCare Regional Medical Center. The SCC was founded in 2007 in partnership with a large multi-employer Taft-Hartley Trust that provides health benefits to union members. The trust approached AtlantiCare for help managing the care of enrollees with poorly controlled chronic conditions. Services at the clinic are paid on a capitated basis, and employed physicians receive a base salary with performance-based bonuses. Only the most chronically ill and highest-cost patients are invited to seek care at the SCC, although patients may apply to participate. In addition to seeing a primary care physician, all SCC patients are assigned a health coach. Employed specialists see patients at the SCC. In addition, the SCC has an on-site pharmacy that monitors patients’ utilization and alerts clinicians if a patient re-fills a prescription too soon or late.

Union Health Center, New York, N.Y., is a freestanding clinic that historically has served members of New York City trade unions. First established by the International Ladies’ Garment Workers’ Union in 1914, the center now serves workers and their families and retirees from many unions. In 2006, the clinic established the Special Care Center, offering high-intensity primary care to patients with chronic conditions. In 2008, these services were made available to all patients seeking primary care at Union Health Center. Patients with poorly controlled chronic conditions may be assigned a health coach who closely follows the patients by accompanying them to clinic visits, working with them by phone, as well as teaching self-management techniques. Primary care physicians and non-physician staff are full-time, salaried employees, while specialists are contracted on a part-time basis and work half a day a week in the clinic. Services at the center are paid either on a fee-for-service or capitated basis, depending on the patients’ health plan contracts.

Practice-Based Programs

The Intensive Outpatient Care Program, Seattle, was launched as a pilot in 2007 for Boeing employees and their adult dependents enrolled in the firm’s self-insured plan administered by Regence BlueShield of Washington. In practices participating in the program, existing complex patients were identified through claims-based algorithms and invited to enroll in the high-intensity program. Primary care practices, which received an additional per-member, per-month payment for each high-intensity patient, hired nurse care managers who scheduled and guided patients through appointments, provided health education, and coordinated referrals to appropriate social support services when possible. Patients remained under the care of their regular primary care physician while enrolled in the program. The pilot program ended in 2009, and Regence BlueShield expanded the program in 2010 to include all eligible enrollees regardless of employer and expanded the model to other providers and markets.

The Oregon High Value Patient Centered Care Demonstration, with sites throughout Oregon, is a practice-based program partially administered and funded by the Oregon Health Leadership Council. Participants include 14 medical groups, five health plans and four Oregon purchasing groups. The demonstration began in 2010 and is based largely on the Intensive Outpatient Care Program. Participating health plans employ algorithms to identify patients with complex needs, and primary care physicians invite identified patients to enroll in the high-intensity care program. Nurse care managers, funded by the health plans through a per-member, per-month payment, are assigned to primary care practices to help coordinate patients’ care. Patients remain with their regular primary care physician while enrolled in the high-intensity program.

Priority Care, Humboldt County, Calif., was established in 2011 and is administered by the California Public Employees’ Retirement System (CalPERS), Anthem Blue Cross, the Humboldt-Del Norte Independent Practice Association (IPA) and the Pacific Business Group on Health. Complex patients insured through CalPERS and Pacific Gas and Electric Company are identified through claims-based algorithms and invited to enroll in the high-intensity care program. Patients enrolled in the program continue to receive care from their regular primary care physician. Nurse care managers work in primary care practices to help manage and coordinate care. The IPA and participating physicians share a monthly care management fee. Savings accumulated through the program will be shared among the participating organizations.

Hybrid Program

CareMore, with sites in California, Arizona and Nevada, is a for-profit Medicare Advantage plan established in 1993 that also owns and operates multiple facilities and clinics for specialized chronic disease management. CareMore contracts with outside primary care practices for enrollees’ usual care and employs physicians from various specialties to provide high-intensity services to certain patients at CareMore facilities. CareMore physicians care for patients along the entire continuum of care from outpatient clinics to hospitals to skilled-nursing facilities. Physicians employed at CareMore facilities receive a base salary with bonuses based on patient satisfaction and performance measures. CareMore physicians provide all care to participating patients, and patients do not receive care from their regular physician until their conditions have stabilized. Patients enrolled in the high-intensity program are monitored through phone calls, home visits and, for patients with congestive heart failure, wireless telemonitoring devices.


Identifying Patients

Candidates for high-intensity primary care generally are identified by health plans through algorithms that incorporate claims data to predict which patients without intervention are likely to account for a disproportionate share of costs. Identifying appropriate candidates for participation in a high-intensity primary care program is challenging because the goal is to include patients who will have persistently high medical care utilization by analyzing claims data based on their past utilization. Many patients with high previous utilization, generally because of an acute illness or injury, will return to more typical patterns of use, while other patients with typical previous use will unexpectedly have a period of higher use.

Along with using claims data to identify patients, in some programs, physicians can recommend candidates for high-intensity care based on their clinical knowledge and experience with the patient, helping to ensure suitable patients are included. Patients included in high-intensity primary care programs generally have multiple, poorly controlled chronic conditions, including diabetes, hypertension, congestive heart failure, asthma, depression and chronic pain syndromes, often complicated by obesity.

Under a high-intensity primary care model, certain physician responsibilities are delegated to a care coordinator, making some physicians and patients reluctant to participate and engage. Previous research has identified a range of barriers to physician engagement in other care-management approaches, including changing the structure and work flow of physician practices, insufficient information technology infrastructure to carry out program requirements, physician concerns about their autonomy, and insufficient financial incentives to change physician behavior.3

Likewise, engaging patients to take part in high-intensity primary care models also poses challenges. These patients frequently have complex medical conditions and are more likely to have mental health conditions and socioeconomic challenges that can hinder the intensive interaction with care coordinators and providers expected in the model.4

Currently, little is known about effective ways to engage physicians and patients, in part because the high-intensity primary care approach is relatively new, and few studies have examined physician- and patient-engagement strategies.5 With its resource-intense, “high-touch” approach to care, much of the value of high-intensity care depends on sufficient engagement from primary care providers and patients. Indeed, studies of similar models show that improvements in health outcomes and cost savings are linked to physician and patient engagement.6


High-Intensity Primary Care Models

There are three general approaches to high-intensity primary care: freestanding, practice-based and hybrid models. Regardless of approach, the goal is the same: Targeting high-intensity care to the right patients and motivating physicians to provide that care effectively by working with and supporting the efforts of care coordinators.

In the freestanding model, patients are recruited to receive care at a dedicated clinic or facility that exclusively or chiefly provides high-intensity primary care to a select group of patients. Once enrolled in the high-intensity program, patients no longer receive ongoing care from their regular primary care physician or that physician’s referral network. Physicians in freestanding models generally are salaried employees and sometimes receive performance incentives.

In the practice-based model, patients receive all care from their regular primary care physician but are recruited by physicians or care coordinators during visits or by phone to join a high-intensity primary care program. Physicians who treat these patients maintain their existing organizational structure and their patient panels. Additional high-intensity services, often managed by a care coordinator, are offered exclusively to the practice’s high-intensity patients. Physicians continue to provide traditional primary care for all other patients in their practice. Payment for high-intensity care in the practice-based model is typically an additional per-member, per-month fee beyond regular capitated payments or fee-for-service rates. The additional payment is for the costs of the care coordinators, extra time communicating with patients, and communication and coordination across providers sharing care for the same patient.

In a hybrid model, a health plan contracts with primary care practices to provide care for most enrollees but shifts the most complex and chronically ill enrollees to a dedicated high-intensity care clinic or facility operated by the health plan. The clinics are staffed by employed physicians, nurse practitioners, case managers and others who temporarily assume all care and coordination responsibilities from the patient’s regular primary care physician. Patients may return to their regular primary care physician once their conditions stabilize.


The Role of Purchasers and Payers

Committed purchasers/payers are critical to developing high-intensity primary care programs, according to respondents. “You’ve got to find a payer that is willing to pay you differently,” said a consultant involved in creating high-intensity programs. “That willingness from the payer side is a precondition for setting up a successful clinic.” When a single payer—sometimes a self-insured employer—accounts for a significant proportion of a physician’s patient panel, that payer has both the leverage to negotiate successfully for changes in care delivery and the patient volume to enable practices to make changes more efficiently.

If multiple payers are participating, standardizing elements of the program proved useful, according to respondents. For example, payers may decide to use the same algorithm to identify candidates, track the same quality or cost measures, implement the same per-member, per-month payment structure and adopt the same shared-savings model.

In cases where a high-intensity primary care program is not part of an integrated delivery system, health plan patient data may be much more complete than the primary care provider’s record and may be the only way a provider learns of, for example, emergency department visits or hospitalizations.

Providers and administrators in high-intensity programs reported that they struggled to design and troubleshoot interventions when they did not receive utilization data that could help them to determine whether their interventions were helping patients to stay out of the hospital until months after the fact. “That is something that needs to be fixed and made as close to real time as possible,” said a PCP in a high-intensity care program. “Even if it is quarterly with retrograde adjustments, it would give you a near-term target that you could really look at and say, ‘Okay, I think we did a good job this quarter. Let’s see what they tell us in our shared savings and try to make improvements.’”


Physician Engagement

Physician engagement requires significant time and resources. Programs relying on existing independent primary care physician practices invested heavily in recruiting PCPs to participate. Leaders of a high-intensity primary care program working with independent practices reported the physician recruitment process took three to six months and included cultivating relationships not only with physicians but also other clinical and front-office staff.

“[Physicians] were worried and possibly threatened that we would be challenging what they are doing with their patients,” said a leader from a practice-based model of their recruitment process. “We had to answer their questions and eliminate any concerns that they had. Sometimes it consisted of buying lunch, buying time, paying for appointment time with providers. We utilized a lot of time just to get the information about this new model out there.”

Respondents involved in high-intensity primary care program development described recruitment as a multistage process, where physician champions—early adopters—through their enthusiasm for the program and success stories with complex patients persuaded other physicians to participate.

Without such concrete examples of success, respondents noted it can be difficult to recruit physicians who are willing to work with the inherent ambiguity and uncertainty of developing an innovative program. Some programs found that a physician-led introductory meeting facilitated engagement and buy in, and in one practice, a second kick-off meeting was required to bolster physician engagement. Regular meetings in which physicians participated in ongoing decision making about the program’s structure and goals helped maintain physician engagement once it was established.

Identifying the right patients can improve physician engagement. Primary care physicians reported frustration with high-intensity programs when health plan algorithms identified patients who the physicians themselves did not identify as high need. “There were a number of patients that I really frankly was surprised [to see included],” said a physician in a practice-based program. When these algorithms were seen as error filled, physicians lost trust in the program and worried they would be caring for their sickest patients without additional support while devoting additional paperwork or care-coordination efforts to other patients who did not need these services.

According to several providers, engagement would be enhanced if PCPs had the ability to nominate potential patients who could benefit from the program and to reject potential patients who experienced acute high-cost medical events but did not, in their opinion, require ongoing high-intensity care. “With the selection of patients, it should be people that the clinician really believes they need help with,” said a physician in another practice-based program, noting that even patients with relatively minor diagnoses may still consume a physician’s time and may benefit from having a care coordinator. “It may just be someone with anxiety, but there’s a burden of care.”

Related to identifying the right patients was the observation that provider engagement is enhanced when enough patients are enrolled in high-intensity primary care. Physicians reportedly are more willing to invest time and resources into high-intensity primary care programs when the benefits extend to a critical mass of patients rather than to just one or two.

Providers also recommended that algorithms pre-screen and reject patients who are unlikely to remain in the program because of expected changes in health coverage, including those enrolling in Medicare. “Now we are left with this moral obligation,” said a provider in a practice-based program. “We’ve had a relationship and made a promise, and their employer changed their insurance on them, and they are no longer getting the services. That is a substantial number of patients.” Some providers reported screening patients themselves to avoid the inefficiencies of recruiting patients who would be unlikely to remain in the program for more than a few months given projected changes in their coverage.

Overcoming physician reluctance to delegate tasks important to engagement. Primary care physicians in some sites initially were reluctant to delegate tasks to care coordinators. “The engagement was brutal when I started,” said a care coordinator working in a practice-based program. “[Doctors were asking] who I was, what my role was, how they could use me, why was I here, was I going to add work for them.”

Over time, the presence of on-site care coordinators promoted PCP engagement, since care coordinators were able to develop rapport with patients and take over certain tasks, creating an immediate and concrete benefit to PCPs. The same care coordinator said that she was welcomed on subsequent visits: “It was an opportunity where they [physicians] didn’t have to go into their computer or to do a long, involved phone note that went through six different tracks. They could just say, ‘Oh…by the way, Carol needs x, y and z,’ and they know that I will take care of it.” High-intensity primary care program leaders also noted that frequent reinforcement, through feedback of utilization data, quality data and financial incentives (if any) were important in encouraging provider retention.

Financial impact on practice affects willingness to engage. Freestanding high- intensity primary care health centers must maintain relationships with community-based PCPs, who may serve as important referral sources. A leader of a freestanding center reported that the best approach is to present the high-intensity care program as a benefit, since freestanding centers could take over the care of a PCPs’ most time-consuming and difficult patients, leaving them with time to see more patients.

Respondents noted that community PCPs are sometimes skeptical about this argument, particularly when the patients targeted for participation in high-intensity care represent a large share of a practice’s insured patient base and the providers are not confident they can make up the lost revenue. Among PCPs participating in practice-based models of high-intensity primary care, concerns about lost revenue if patients used fewer services were rare but real.

Not all physicians can successfully transition to high-intensity primary care. Freestanding programs with a dedicated clinic avoided some of the challenges associated with physician engagement by recruiting and employing highly motivated clinicians and other professionals who frequently were younger and “still fairly malleable,” in the words of one respondent. “Clearly we don’t train doctors [to have the attitude we seek]. They come to us,” said a leader of a freestanding program. “We’ve gotten better in identifying physicians and health coaches with the right attitude through the interview process.”

In two freestanding centers that evolved from traditional practices, some physicians—about half at one site—were asked to leave or quit when they were unwilling to change their practice style and delegate to others. “They didn’t really like the model,” said a leader from a freestanding program. “I think one of the elements of traditional medicine is that the doctor takes care of everything. I think doctors were wary of the expansive roles of non-licensed professionals like the medical assistants in our model.” Encouraging physicians to participate in the training and certification of medical assistants taking on the care coordinator role helped reassure some physicians about delegating important tasks.


Patient Engagement

Transfer of trust key to patient engagement. Because patients have little experience with high-intensity care, they may have difficulty appreciating the benefits, such as having a health coach or learning more about their chronic conditions, until they are in the program. Nearly all respondents reported that patients are much more likely to enroll if their own physicians invite them, either during a scheduled appointment or in a separate interaction, and explain the advantages of high-intensity primary care.

“Initially, we thought providers would call their patients, and the patients would gleefully say, ‘Oh, yes, I’d love to participate,’ and sign their enrollment forms,” said a leader of a practice-based program. “But, the reality is that something new makes people skeptical. Getting a call from your doctor can be frightening, and getting a call from your insurance company can be even worse. So we had to develop letters, and scripting, and a format where first patients would get a call from the office, either the providers themselves, or the office staff, [and then we would] send out an informational letter.”

Some patients were reluctant to enroll if they perceived that the high-intensity program was run by their insurance company or their employer, so providers were more successful with enrollment when they emphasized that the program is managed by the providers themselves. Providers and administrators believed this approach fostered a greater sense of trust from patients. “Lesson one is that people trust their doctors, and if it’s anyone else that the patient is not familiar with, the outreach does not work,” said a leader from a practice-based program. “And certainly health plan outreach does not work as well. It can have an adverse effect, actually.” Sometimes, however, respondents reported that all attempts at recruitment failed when patients simply did not believe they needed high-intensity care.

Patients value rapid access to care providers. The strongest selling point for patients is the prospect of direct access to their care coordinator via phone or email, which allows them to schedule appointments, follow up on test results, and speak to their physicians in a timely manner. Care coordinators often were able to more efficiently shepherd a patient through their care, including obtaining appointments with specialists, aiding transitions home from the hospital, and connecting patients with community resources for social support.

Providers reported that some patients initially were skeptical of working with care coordinators but found that team visits involving both the PCP and care coordinator early in the relationship promoted transparency and trust in the care coordinator. Later, patients spent more one-on-one time with care coordinators. Care coordinators who were linguistically and culturally matched to their patients were reported to be most successful in building trust.

Respondents described other approaches as helpful. Financial incentives, such as waiving copayments for the initial program intake visit and complimentary parking, also may encourage participation by removing logistical barriers. For patients skeptical about the benefits, patients who are already in the program can be good ambassadors. “When they first invited me, I said no because I didn’t want my own employer to be so involved in my care,” said a patient in a freestanding program. “But then another…employee who goes there eased my anxieties about that. So I gave it a chance, and after the first visit, I really believed that this was what I needed.”

Financial incentives help with initial patient interest but do not guarantee sustained engagement. In contrast with practice-based models, freestanding models usually require patients to leave their regular primary care physicians to obtain care at a facility that focuses on high-intensity care. Some freestanding programs offer incentives, such as discounts on chronic disease medications or personal fitness training at no extra cost, for patients who agree to receive high-intensity care.

While incentives may help spur patient participation, consistent encouragement from physician and care coordinators was critical to maintaining patient engagement. As one physician noted, “The financial incentives don’t necessarily encourage patients to follow everything that the health coaches ask them to do like checking their finger sticks…but they do get them in the door.”


Notes

1. Hibbard, Judith H., Jessica Greene and Martin Tusler, “Improving the Outcomes of Disease Management by Tailoring Care to the Patient’s Level of Activation,” American Journal of Managed Care, Vol. 15, No. 6 (June 2009); and Coulter, Angela, and Jo Ellins, “Effectiveness of Strategies for Informing, Educating, and Involving Patients,” British Medical Journal, Vol. 335 (July 7, 2007).

2. Milstein, Arnold, and Elizabeth Gilbertson, “American Medical Home Runs,” Health Affairs, Vol. 28, No. 5 (September/October 2009).

3. Kuraitis, Vincent, “Disease Management and the Medical Home Model: Competing or Complementary?” Disease Management & Health Outcomes, Vol. 15, No. 3 (2007); Leider, Harry L., “Gaining Physician Buy-In for Disease Management Initiatives,” Disease Management & Health Outcomes, Vol. 6, No. 6 (December 1999); and Berenson, Robert A., Challenging the Status Quo in Chronic Disease Care: Seven Case Studies, California HealthCare Foundation, Oakland, Calif. (September 2006).

4. Fleishman, John A., and Joel W. Cohen, “Using Information on Clinical Conditions to Predict High Cost Patients,” Health Services Research, Vol. 45, No. 2 (April 2010); Bodenheimer, Thomas, and Rachel Berry-Millet, “Follow the Money—Controlling Expenditures by Improving Health Care for Patients Needing Costly Services,” New England Journal of Medicine, Vol. 361, No. 16 (Oct. 15, 2009); Unutzer, Jurgen, et al., “Healthcare Costs Associated with Depression in Medically Ill Fee-for-Service Medicare Participants,” Journal of the American Geriatrics Society, Vol. 57, No. 3 (March 2009); and American Hospital Association 2011 Committee on Research, Caring for Vulnerable Populations, American Hospital Association, Chicago, Ill. (2011).

5. Scholle, Susan H., et al., Engaging Patients and Families in the Medical Home, Agency for Healthcare Research and Quality, Rockville, Md. (June 2010); Springrose, James V., et al., “Engaging Physicians in Risk Factor Reduction,” Population Health Management, Vol. 13, No. 5 (October 2010); and Leider (December 1999).

6. Hibbard, Greene and Tusler (June 2009); Springrose (October 2010); and Congressional Budget Office, Lessons from Medicare’s Demonstration Programs on Disease Management, Care Coordination and Value-based Payment, Issue Brief, Washington, D.C. (January 2012).


Data Source

In addition to performing a literature review, Center for Studying Health System Change (HSC) researchers conducted 26 telephone interviews with representatives of benefits consulting firms, as well as program directors, primary care physicians, specialists, non-clinician staff and patient representatives from six high-intensity primary care models across the United States. Interviews were conducted by two-person research teams between February 2012 and June 2012.