Lucier, David J.; Frisch, Nicholas B.; Cohen, Brian J. MD, SM; Wagner, Michael MD; Salem, Deeb MD; Fairchild, David G. MD, MPH
Market forces have spurred the development of retainer primary care practices, sometimes called “luxury” or “concierge” practices.1 Patients pay a yearly retainer fee to their primary care physician (PCP), and, in return, the PCP provides nonmedical services not normally extended to other patients. Longer visits, enhanced access to care, and PCP accompaniment to specialist visits exemplify some of the typical services offered to patients in such practices.2 To accommodate these changes in practice, concierge physicians reduce the size of their patient panels to approximately one-fourth the size of a conventional PCP's panel. Provider dissatisfaction about work conditions, such as having a burdensome amount of administrative paperwork, as well as a desire to find an antidote to the “hamster on a treadmill” character of primary care, have driven physician interest in retainer-medicine practices.3–5
Since its inception, patients, payers, and physicians alike have voiced concerns about the ethical foundation of retainer medicine.6–8 These concerns generally fit within one of the three themes outlined by Brennan1: (1) patients who cannot afford or do not wish to pay the retainer fee when their physician transitions to a retainer practice are forced to find a new PCP, (2) the smaller panel size of retainer-medicine physicians effectively reduces access to health care, and (3) it undermines the cross-subsidization “system” where patients with good insurance help defray the cost of care to the uninsured. Further, it is generally held that the profit motive in retainer medicine is a significant enticement for physicians, as those in private retainer-medicine practices can expect to substantially increase their gross revenue.9,10 Not surprising, given these ethical concerns, academic medical institutions have eschewed retainer medicine.
With retainer-medicine practices generally considered “off limits” because of ethical concerns, making primary care financially viable at academic health centers (AHCs) remains a challenge. Although hospital-run primary care practices create indirect positive downstream revenue for the hospital through referrals and ancillary services, many of these primary care practices generate direct negative net margins on each patient provided care. Consequently, many academic primary care practices require financial support. Under this current paradigm, hospital-run primary care practices can only be expanded to the extent that the hospital or the physician practice plan is willing to invest financially. This business model inherently limits the expansion of primary care.
Like most AHCs, Tufts Medical Center emphasizes three main mission objectives: (1) excellence in clinical care, (2) research, and (3) the teaching of residents and medical students. The Division of General Medicine maintains these tenets and further incorporates the following operating objectives: (1) to provide high-value care, (2) to maintain economic viability, and (3) to have a long-term primary care delivery model that is sustainable. With these objectives in mind, a retainer-medicine primary care practice within Tufts Medical Center was created as an entity that could provide additional revenue to help support the mission of the primary care division. The practice was also developed to create an environment within which relationships with potential hospital donors could be cultivated by providing them with a unique atmosphere that conveniently addresses medical needs.
With the goal of developing an innovative retainer-medicine practice that addressed the ethical concerns seemingly inherent in retainer medicine but which retained its profit potential, Tufts Medical Center opened its Academic Retainer Practice (ARP) in 2004 within the Division of General Medicine. This unique business model was developed to respond to several of the ethical issues associated with traditional retainer practices: (1) it does not restrict net access to care, (2) it ensures that participating physicians maintain a socioeconomically diverse practice, and (3) it neutralizes concerns about patient abandonment. To the authors' knowledge, the ARP at Tufts Medical Center is the first and only general internal medicine retainer practice embedded within the structure of an AHC.
The ARP Model at Tufts Medical Center
The ARP at Tufts Medical Center was developed and implemented in collaboration with the for-profit national “concierge medicine” company, MDVIP, Inc. The practice has its own office, waiting room, examination rooms, and administrative staff. A separate clinic area adjacent to the executive administrative suite was renovated and furnished using a $350,000 designated donation by members of the hospital board of trustees. Consistent with private retainer-medicine practices, services offered in exchange for a retainer fee include longer visits than in the Division of General Medicine Teaching Practice (GMP), 24/7 access to an ARP physician, comprehensive wellness and prevention screenings, as well as same- or next-day on-time appointments. Also consistent with many private retainer-medicine practices, the patient's insurance is charged using the same fee schedule and collection of co-payments as in the GMP for regular office visits. A retainer fee of $1,800 per year was determined by a local market analysis. However, unlike some community-based retainer-medicine physicians, ARP physicians do not accompany patients to specialist visits, and there is no guaranteed access to specialty or ancillary services.
Participating physicians were selected by three criteria: (1) willingness to participate in the model, which included providing care in both the ARP and the GMP, with the understanding that there would be no additional compensation, (2) zip code profiling of each physician's patient panels to identify probable subscribers, and (3) practice style consistent with the in-depth, long-term patient–physician relationship that a retainer-medicine practice requires.
The infrastructure of the ARP differs slightly from that of the GMP. The ARP does not employ physician extenders such as nurse practitioners or physician assistants. This decision, made by the two of us (D.S., M.W.) primarily responsible for the start-up of the ARP, anticipated that the use of physician extenders would not be acceptable to patients paying an annual fee for retainer medicine. However, the ARP does have a full-time registered nurse (RN) who does initial phone triage of urgent complaints and coordinates a number of other aspects of care, including the transition from inpatient to outpatient care and outbound calls to follow-up with sick patients. Having the RN to coordinate home care along with the physician coordination of specialty care allows the ARP to function along the lines of the medical home model. The RN is an essential component in the practice because of the physicians' dual roles within both the ARP and GMP.
The dual role of ARP physicians
The ARP is staffed by four physicians who split their clinical time to provide care for both ARP and GMP patients. One of the ARP physicians is on call at any given time. As such, ARP patients are always guaranteed access to an ARP physician, although not necessarily their own PCP. However if desired or necessary, a patient's PCP can be reached at home by the on-call ARP physician. No teaching occurs in the ARP. Yet, ARP physicians spend half of their clinical time in the GMP seeing patients, teaching medical students, and precepting residents. The practice of teaching medical students and residents in the GMP is an important component of the model, as it allows ARP physicians to maintain alignment with the mission and principles of their AHC. Further, all patients have access to all PCPs, as no one PCP limits his or her practice to the ARP.
ARP physicians have a modified patient panel, reflective of their responsibilities to both the ARP and the GMP. Each physician maintains a maximum ARP panel size of 275 patients—half the number of private PCPs in most retainer-medicine practices.10 ARP physicians also carry a GMP panel of approximately 900 patients—half the number of most full-time PCPs in academic practices.11 Thus, each ARP physician cares for a maximum of 275 ARP patients and 900 GMP patients concurrently, splitting his or her clinical time equally between the ARP and GMP. Despite this altered workload, physician compensation within the Division of General Medicine is independent of practice location; physicians who practice in the ARP are paid at the same rates as those practicing solely in the GMP.
The motivation for PCPs to practice in this dual role setting stems from the satisfaction that results from the in-depth patient interactions. Most PCP visits last between 45 and 60 minutes. Although they do not receive additional financial compensation, ARP physicians find it gratifying to practice medicine without the time pressures of a traditional practice. ARP physicians are motivated to split their time between the two practices in this model because they do not have to abandon their GMP patients, for whom they have been providing care for years, while they are able to derive gratification from the in-depth patient care of the ARP at the same time. Despite having to practice in two different clinic locations, physicians in the ARP are generally satisfied with their dual role, although they do report that the logistics of switching between the two practice settings can be challenging at times.
ARP patient enrollment data
Of 347 patients enrolled in the ARP during the first year of operation (the 2005 calendar year), 202 (58%) were previous GMP patients who converted to the ARP practice setting. Initially, every GMP patient of an ARP physician was sent a letter outlining the benefits and yearly cost of the new practice model. The letter asked the patient if he or she would like to enroll, while making it clear that declining the enrollment offer would not impact his or her care. Those patients who declined would remain in the traditional GMP model. In addition, MDVIP sponsored several information sessions for GMP patients, detailing the services offered and the costs and benefits. Advertisements were placed on cable TV and in the local newspaper, while numerous media sources provided independent coverage of our unique business model. As a result of these campaigns, 144 (41%) patients who enrolled in the ARP during the first year were new patients to the AHC, and 1 (0.3%) was an employee of the AHC.
The source of patient enrollment growth in the ARP changed significantly over time. During the period of October 2007 through October 2008, only 9% of new ARP patients enrolled as a result of conversions from the GMP, whereas patient referrals from friends or family who were existing ARP patients accounted for 59% of the growth (see Table 1). This shift in patient enrollment from initial GMP conversion to referral-based growth illustrates an inherent challenge in the ARP model; the initial patient base might be mostly patients moving from the GMP, but the conversion rate from GMP to ARP cannot be relied on to deliver sustainable enrollment.
Enrollment continues to grow and is mostly by “word of mouth” referrals from existing patients and ARP staff. In general, past and new enrollment patients in the ARP are older and require more intense coordination of care, which the ARP practice setting can provide.
The Division of General Medicine of the Department of Medicine has budgetary responsibility for both the ARP and GMP. It collects all retainer fees and office visit professional reimbursements while paying for all costs incurred at the ARP. The division also collects all professional revenue and is responsible for all professional expenses in the GMP. Yet, it is constrained by a physician practice plan budget that must support financial losses from the negative margins on each GMP patient. With this system, the success of the ARP has enabled ARP profit to subsidize the GMP directly. The financial support of the ARP for the GMP allows for several possible scenarios, including
1. Expansion of the GMP's primary care patient population beyond what would normally be covered by the practice plan budget, by offsetting the GMP's negative margins with positive margins from the ARP
2. Allocation by the Division of General Medicine of the extra revenue from the ARP to come in under budget or cushion against the need for contraction during times of financial and economic stress
3. Support, using revenue from the ARP, of the often underfunded mission-related activities within the Division of General Medicine, such as teaching and provision of free care to impoverished patients
4. Bolstering of PCP salaries, again using revenue from the ARP, to aid in the increasingly difficult recruitment and retention of PCPs12
The cross-subsidization of our model addresses the second and third operating goals of primary care—to produce economic viability and to ensure long-term sustainability. In addition, this cross-subsidization has attracted patients to the ARP from other private concierge clinics, because they feel that they are receiving the same quality and type of care but appreciate the fact that their money is being used to support the mission-related goals of the GMP.
Benefits of the ARP Practice Model
Growth and stability
Because each ARP patient contributes a significant amount of profit, and paying PCPs to provide care to each GMP patient costs the Division of General Medicine a small fraction of that amount, one metric of the success of the Tufts Medical Center model is approximately how many GMP patients can be subsidized by each additional ARP patient. Based on the current cost and revenue structure of the ARP, the profit gained from each new ARP patient can subsidize the PCP costs of caring for approximately 50 additional GMP patients.
This calculation is for illustrative purposes only and does not factor in other decisions that might affect the finances of the ARP or the GMP. For example, if additional capital were necessary to support growth in the ARP, the size of the subsidized GMP patient population would decrease. The calculation also does not take into account any additional infrastructure investment in the GMP required to handle the consequent increase in patient population. Irrespective of the many variables in this calculation, however, our experience with the ARP model indicates that each new patient joining the ARP represents a source of support and subsidy for the general medicine division. Thus, growth of the ARP can directly support growth of the GMP without the losses traditionally associated with primary care practices, assuming that the GMP has the infrastructure necessary to support such growth.
PCP salary augmentation
In Massachusetts, 56% of hospitals reported a shortage of internal medicine physicians in their hospital.12 The same study showed dissatisfaction among practicing physicians to be so high that only 51% would choose to practice medicine again as their profession.12 In addition, many PCPs are thinking of leaving the profession altogether.13 Increased payment could help retain those PCPs who are contemplating changing their profession, can attract additional new, high-quality PCPs, and can help reverse physician dissatisfaction. Also, creating additional revenue via the ARP and allocating a proportion of it to PCP salaries may give an AHC a competitive advantage in hiring and retaining quality PCPs.
The revenue from the ARP has the potential to augment PCP salaries but is currently not used for this purpose at Tufts Medical Center. If an AHC were to use the ARP revenue in this way, consideration should be given to providing all ARP and GMP PCPs with the same increase regardless of practice environment. This change avoids creating incentives for PCPs to move to the practice environment with the higher pay, and ensures that each PCP, regardless of practice environment, becomes a stakeholder in the ARP. By sharing in the same revenue pool, all PCPs have an incentive to maximize the patient base of the ARP because doing so directly impacts their earnings.
Improvements in quality of care
Critics of retainer medicine voice the concern that patients in retainer practices may receive a higher standard of care and, further, that this practice environment might lead to unnecessary resource utilization and higher cost burden. The approach at Tufts Medical Center has been to establish one standard of care for both the GMP and ARP. All clinical guidelines and quality goals are the same irrespective of practice setting.
Thus far, we have been unable to detect any differences in the clinical quality of care provided to ARP patients compared with GMP patients treated by the same physicians, as evidenced by standard quality measures such as hemoglobin A1c. Each physician follows the same standardized care algorithms whether in the ARP or GMP. Furthermore, from the onset, we have asked insurance companies to identify differences in costs of care between the two groups, and, to date, none have been noted. We do feel that patients benefit from the enhanced coordination of care that the ARP practice setting allows. Although physicians do not accompany patients to subspecialty appointments, specialists, both within Tufts Medical Center and at other institutions, receive a packet with all the relevant information needed to evaluate each patient. This specialist report and any testing performed is then reviewed and discussed with the patient to make sure that he or she has a full understanding of the findings of the visit and to obtain input on any decisions to be made. This coordination of specialty visits is an important role of a general internist and one that is very difficult to do well with the patient load that most internists have.
When a physician is scheduled to be in the GMP and his or her ARP patient needs to be seen, that patient will be seen by the ARP physician on call. This coverage system is explained up front to the patients and protects the physician from having to prioritize a patient from one clinic over a patient from the other. In the end, though, it is the professionalism of the physicians that ensures the same level of care in both practice settings.
A recent study by Ko et al14 comparing the satisfaction of ARP patients with that of GMP patients who had the same physician within Tufts Medical Center found that ARP patients were more satisfied with care coordination, access to care, and interactions with the office staff (P < .05). However, the quality of interaction between patients and their physicians did not differ between ARP and GMP patients who shared the same PCP (P = .2). This study indicates that patients perceived better service to be the main differentiating factor of the ARP.
Avoiding patient abandonment
In community settings, the transition to a retainer-medicine practice is predicated on the expectation that many patients will choose to pay the retainer fee to stay with their current PCP rather than find a new personal physician. Although this transition strategy does not violate medical ethics guidelines, the practice is considered disruptive and distasteful.15,16
In the transition to Tufts Medical Center's ARP, every patient was given the choice of moving to the ARP setting or staying with his or her physician in the GMP environment. The majority of patients (roughly 95%) elected to remain in the GMP setting, whereas only approximately 5% chose to join the ARP. This highlights another unique aspect of the Tufts Medical Center ARP model—no patient is forced to leave the practice. All prospective GMP patients are given the choice to join the ARP or remain with the GMP. This ability to choose alleviates the potential for patient abandonment found in traditional retainer-medicine practices. It was critical to the adoption of the ARP model in our institution that no patients were given the ultimatum of paying the retainer fee or leaving to find another physician.
Access to quality care for all
The American College of Physicians warns that physicians in retainer practices “risk compromising their professional obligation” to care for patients from all socioeconomic means.17 By practicing in both the GMP and the ARP, the ARP physicians care for patients from a broad socioeconomic spectrum and contribute to the support system required to care for patients who are underinsured. This unique aspect of the ARP model at Tufts Medical Center distinguishes it from other private retainer-medicine practices.
Additionally, the ARP subsidization model allows the GMP to provide access to care for many more patients without incurring additional financial losses. For example, if a GMP physician carries a panel of 1,800 patients, he or she provides access to care for those 1,800 alone. If an ARP physician carries a panel of 275 ARP patients and 900 GMP patients, he or she will directly provide access for 1,175 patients. This results in a direct restriction of 625 patients. However, the profit generated in the ARP by the 275 ARP patients has the potential to subsidize another PCP to care for many more than 625 GMP patients.
The end result is that restricting the patient panels for ARP physicians can actually result in a net expansion of access to care if ARP profits are used to subsidize the GMP, which has been done thus far at Tufts Medical Center. This potential for a net improvement in access to primary care is a strong contrast to community retainer-medicine practices where the transition from traditional primary care to retainer medicine creates a net reduction in patient access to primary care.
How it all fits together
The ARP business model enables an AHC to obtain additional revenue from a niche market patient seeking supplementary services. Patients are informed of the services they can expect and understand that the quality of care is the same across all hospital divisions. Receiving retainer fee revenue provides for a more ethical environment in which it is clear that the financial exchange is purely for additional services and not higher quality of care. This revenue provides flexibility for a hospital's general medicine division. Uses of the revenue could facilitate expansion of primary care services while remaining within a predefined budget, reduce the external financial contribution required to support primary care, or support mission-related activities such as teaching medical students and residents and providing care for those patients who are unable to pay. The revenue can also be used to augment PCP salaries, thus improving physician satisfaction and retention. Furthermore, the model addresses some of the ethical concerns with retainer medicine, particularly regarding patient access to primary care and the potential for patient abandonment. Now in its sixth year, the ARP practice at Tufts Medical Center has generated profits supporting the mission of an AHC while allowing physicians to remain aligned with the ethics of an academic medical institution.
The Tufts Medical Center ARP Model: Why It Works
The growth of retainer medicine is emblematic of the disquieting trend away from traditional primary care.18–21 Fewer medical school graduates are choosing primary care careers, leading to a critical lack of primary care providers.22–24 The current market environment of physician dissatisfaction due to low reimbursement for primary care, coupled with patient dissatisfaction due to inadequate customer service from overworked physicians, provides incentive to patients and physicians alike to move toward retainer-medicine practices.25–27 Yet, AHCs have hesitated to embrace retainer medicine, likely because the traditional retainer-medicine business model is not aligned with the culture and ethics of an academic medical institution.
In this paper, we have presented an innovative business model for a retainer-medicine general medicine practice that we believe is aligned with the mission and ethics of an AHC. One of the challenges during the start-up of the ARP was recruitment of patients. As a result of slow patient recruitment, the ARP required greater financial subsidy than was originally anticipated. This model may therefore only be viable within a large AHC that is capable of providing external financial support. However, our experience also suggests that after the initial ramp-up, an academic-based retainer-medicine practice will have a better financial performance than the traditional academic primary care teaching practice and can lead to subsidized expansion of the traditional GMP. After two years, the annual investment (subsidy) required per physician FTE in the ARP fell below the level required for physicians in the GMP, and the ARP became profitable by the end of the third year. Although it is generally accepted that primary care practices generate indirect downstream revenue, improving the bottom line of primary care clinical operations holds significant benefit given financial pressures on general medicine divisions.28–31
The ARP business model also addresses some of the ethical concerns voiced by the medical community regarding retainer medicine, including the potential for patient abandonment and reduction in access to care. Patients can choose their PCP and the practice type because all ARP physicians carry a half-panel of traditional GMP patients and a panel of ARP patients, splitting their clinical time between the two practices. Currently, ARP physicians do not receive any additional payment over GMP physicians. Compensation follows the same formula for all physicians within the general medicine division whether they practice in the ARP or not.
Further study is required to determine whether there are in fact differences in resource utilization and quality between the ARP and the GMP setting. Although critics are concerned that resource utilization may go up in the ARP setting, it is also possible that utilization of resources may actually be lower in a retainer practice because physicians spend more time with patients and can appropriately reassure them when tests are not necessary. Hurried physicians can order more tests because of the typical time constraints forced on them by the current primary care system.32
Profits from the ARP subsidize losses associated with the GMP practice within the Department of Medicine, effectively allowing an AHC to increase its primary care patient base without incurring the additional losses normally associated with expanding these services in the hospital setting. ARP revenue can also serve to increase PCP salaries, providing an AHC with a competitive advantage for hiring and retaining quality PCPs. Although further research will be required to determine whether this model is generalizable to other AHCs, our experience suggests that the ARP business model has the potential to support the AHC mission while improving overall access to primary care services and providing financial elasticity for the Division of General Medicine in times of economic stress.
1Brennan TA. Luxury primary care—Market innovation or threat to access? N Engl J Med. 2002;346:1165–1168.
2Alexander GC, Kurlander J, Wynia MK. Physicians in retainer (“concierge”) practice. A national survey of physician, patient, and practice characteristics. J Gen Intern Med. 2005;20:1079–1083.
3Morrison L, Smith R. Hamster health care: Time to stop running faster and redesign health care. BMJ. 2000;321:1541–1542.
4Sandy LG. Homeostasis without reserve—The risk of health system collapse. N Engl J Med. 2002;347:1971–1975.
5Grumbach K, Bodenheimer T. A primary care home for Americans: Putting the house in order. JAMA. 2002;288:889–893.
6Donohoe M. Luxury primary care, academic medical centers, and the erosion of science and professional ethics. J Gen Intern Med. 2004;19:90–94.
7Charatan F. US “boutique medicine” could threaten care for the majority. BMJ. 2002;324:187.
8Newman L. Doctors' groups must rally to preserve the US public's health. BMJ. 2002;324:1335.
9Reinhart E. Doctors are more interested in having high incomes than providing better health care. BMJ. 2002;324:1335.
10Zuger A. For a retainer, lavish care by ‘boutique doctors.’ New York Times. October 30, 2005.
11Sussman AJ, Fairchild DG, Coblyn J, Brennan TA. Primary care compensation at an academic medical center: A model for the mixed-payer environment. Acad Med. 2001;76:693–699.
13Solomon J, Landon BE, Reschovsky JD, Pham HH, Blumenthal D. How strategies for managing patient visit time affect physician job satisfaction: A qualitative analysis. J Gen Intern Med. 2008;23:775–780.
14Ko JM, Rodriguez HP, Fairchild DG, Rodday AMC, Safran DG. Paying for enhanced service: Comparing patients' experiences in a concierge and general medicine practice. Patient. 2009;2:95–103.
16Galewitz P. The doctor is in—But at what price? Palm Beach Post. March 13, 2006.
17Snyder L, Leffler C; Ethics and Human Rights Committee, American College of Physicians. Ethics manual: Fifth edition. Ann Intern Med. 2005;142:560–582.
18Brennan TA. Concierge care and the future of general internal medicine. J Gen Intern Med. 2005;20:1190.
19Woo B. Primary care—The best job in medicine? N Engl J Med. 2006;355:864–866.
20Blue Ribbon Panel of the Society of General Internal Medicine. Redesigning the practice model for general internal medicine: A proposal for coordinated care. J Gen Intern Med. 2007;22:400–409.
21Reuben DB. Saving primary care. Am J Med. 2007;120:99–102.
22Moore G, Showstack J. Primary care medicine in crisis: Toward reconstruction and renewal. Ann Intern Med. 2003;138:244–247.
23Sox HC. Leaving (internal) medicine. Ann Intern Med. 2006;144:57–58.
24Bodenheimer T. Primary care—Will it survive? N Engl J Med. 2006;355:861–864.
25Safran DG. Defining the future of primary care: What can we learn from patients? Ann Intern Med. 2003;138:248–255.
26Fairchild DG, Wilcox CM. Comprehending coordinated comprehensive care: The devil is in the dollars. J Gen Intern Med. 2007;22:224–225.
27Goroll AH, Berenson RA, Schoenbaum SC, Gardner LB. Fundamental reform of payment for adult primary care: Comprehensive payment for comprehensive care. J Gen Intern Med. 2007;22:410–415.
28Woodcock EW. Assessing primary care's contribution to academic health centers. Med Group Manage J. March/April 1999;46:14–18, 20–22.
29Fahey P, Cruz-Huffmaster D, Blincoe T, Welter C, Welker M. Analysis of downstream revenue to an academic medical center from a primary care network. Acad Med. 2006;81:702–707.
30Centor RM, Huddle TS. Academic general internal medicine: Past, present, future. Am J Med. 2006;119:172–175.
31Walker DL. Physician compensation: Rewarding productivity of the knowledge worker. J Ambul Care Manage. 2000;23:48–59.
32Kuttner R. Market-based failure—A second opinion on U.S. health care costs. N Engl J Med. 2008;358:549–551.