Kennedy, David W. MD; Johnston, Elizabeth MBA; Arnold, Ethan MPH
At the University of Pennsylvania (Penn) School of Medicine and Health System, as at many academic medical centers (AMCs), funding for the academic and clinical missions of clinical departments was traditionally based, in significant part, more on historical precedent than on a rationally defined allocation of resources. The funds-flow project, initiated in 2004, restructured the distribution of institutional funds across all the missions of the clinical departments, so as to provide clearly defined support for teaching, research and administrative duties, as well as rationally based clinical support and incentives for strategic clinical and research growth. In this article, the authors present the principles, process, and results of this undertaking. The framework may provide a basis for rationalization of financial support at other integrated academic medical institutions.
Clinical departments at academic institutions have a threefold mission: in addition to conducting research and providing clinical care, they are involved in medical student, resident, and fellow teaching. Although there has been a significant movement towards mission-based budgeting in AMCs, what has been published to date on this topic has been aimed towards developing accountability for the research and teaching missions.1–3 The importance of tracking both teaching effort and outcomes and of providing defined funds for teaching has been well documented in an era when time pressures are significant and clinical reimbursement is tight.4,5 Additionally, external research funding frequently only provides a portion of the financial support required for the overall departmental research mission. Indeed it has been estimated that 12% to 13% of an institution’s total research costs are borne by the institution.6 If the appropriate proportion of these costs are not reimbursed at the departmental level, and incentives are not provided to increase research funding, there is a real risk of decreased research effort in an environment with limited fungible resources.
There has been less published previously with regard to rationalizing funds flow across both clinical and academic missions. However, in many AMCs the departments do not own the facilities or the equipment utilized in their practices, and thus may not receive the technical fees associated with tests and procedures performed in their medical practice. Conversely, in private practice such technical fees have assumed an ever-increasing part of practice revenue, becoming progressively more essential for keeping practices financially sound in an era of increasingly restrained physician professional payments. Spahlinger et al7 reported improved financial performance in the University of Michigan Cancer Center after a realignment of financial incentives, which included departments receiving at least a portion of associated technical charges. The importance of aligning the goals of the institution with the clinical practice plan was identified by Cohen and Fox8 as a factor in the success of a merger between two large AMCs. The process of realigning institutional goals and AMC missions at Penn, described herein, was a broader, rationally based initiative to redistribute income to support the tripartite missions in a more clearly defined fashion.
Clinical departments in AMCs typically receive institutional budgetary support for their unfunded or underfunded missions. Such institutional support may be provided in the context of salary lines, subsidized rent, clinical service line support, faculty recruitment, or faculty retention negotiations. As reported by Mallon,6 such funding may be approached in different ways either as a result of cap-in-hand approach from a department chair or faculty member to the dean or as part of a planned giving model. At Penn, as at many other AMCs, this type of support was traditionally provided both from the dean’s office and from the medical center on the basis of private negotiations. Once negotiated and initiated, it became expected for continued departmental support, even, in some situations, after the original rationale for such support may have changed or may have no longer been applicable. The budgetary support agreements at Penn were not public, and to some extent they were more dependent on the negotiating ability of a particular chair or faculty member than on the basis of a solid strategic or fiscal rationale. In this article we discuss the process, principles, and metrics used to develop a new model of support for the clinical departments at Penn, and the results of this model’s implementation.
The Penn School of Medicine and Health System constitute an integrated AMC which reports to the dean of the school of medicine/executive vice president of the university. The health system includes three Penn-owned academic acute care hospitals, several affiliated hospitals, a primary health care network, a nursing home, hospice, and home care. Eighteen clinical departments within the school of medicine house 1,350 full-time faculty, and the chairs of these departments also serve as clinical chiefs for the overall health system. All full-time clinical faculty belong to the faculty practice plan (Clinical Practices of the University of Pennsylvania, or CPUP) and the CPUP leadership reports both to the dean of the school of medicine and the CEO of the health system. Although the dean’s tax at Penn is low (2% of net clinical revenue), limiting the dean’s ability to directly readjust departmental revenue with school of medicine funds, total institutional overhead for Penn’s clinical departments is benchmarked with other faculty practice plans. As a result, at Penn overall fiscal responsibility for the research, teaching, and clinical missions of the 18 clinical departments is coordinated under CPUP, and departmental financial support is achieved primarily through health system funds. Total revenue for the school of medicine and health system is $3.0 billion (FY2006). The clinical departments’ revenue from clinical operations was $367 million for FY2006, with an additional $300 million in clinical department-sponsored research.
In 1999, in the face of serious financial problems, the university began to explore the option of spinning off the health system as a joint venture with a for-profit health care entity. The university eventually backed away from this course, largely because of strong pressure from the medical school faculty, but also in part because of difficulty in defining the most appropriate organizational relationship of CPUP to both the school of medicine and the proposed new health system entity. With a subsequent decision to maintain the health system as a subsidiary of the university, the health system underwent significant restructuring to address the financial problems. Penn’s current dean, Arthur Rubenstein, was recruited in 2001, and was given overriding authority for both the school of medicine and the health system. A new overarching governing board, Penn Medicine, was also created. With a carefully crafted strategic plan and improving operating margins, the school of medicine and the health system together embarked in a major reinvestment in the AMC’s missions, paying down its debt and initiating a major building plan financed almost entirely by operating margin and philanthropy. In FY2006, the health system significantly increased its support to both the school of medicine and the university and posted a margin of 6.6%.
In the late 1990s, in the face of Penn’s impending financial crisis and with the necessity for increased fiscal control, individual departments’ access to their own fund balances had been frozen. Although this improved expenditure control, the lack of access to department fund balances removed a significant incentive for department profitability. By 2004, some departments had grown large fund balances (up to $100 million), whereas other departments had developed negative fund balances (up to −$40 million). Although such variances had developed in part on the basis of differing levels of fiscal management within the departments, they had also developed in significant part as a result of very favorable global billing arrangements and high levels of hospital pass through for services to some of the hospital-based departments, where such payment patterns had continued in lieu of market-based reimbursement practices more typical in other departments and other institutions.
By early 2004, several clinical departments had significantly negative budgets, large accumulated negative fund balances, and little incentive for fiscal improvement. On the other hand, some departments with traditionally strong funding had large faculty and staff components with reduced productivity on a per-faculty basis. Additionally, chairs were both uncertain about the fairness of, or the rationale for, institutional support and were quick to attribute either their negative budgets, or the financial success of other departments, to the variability and unknown nature of such support. Because CPUP is responsible for the fiscal aspects of all of the missions of the clinical departments, after obtaining the agreement of the dean and health system CEO, we undertook an initiative to rationalize all aspects of financial support provided to each of the clinical departments with the goals of making the support logical and open across all departments. Consistent with an aim of prioritizing the academic mission, we made a decision to sequence the funds- flow so that the funding for teaching would be addressed first, followed by that for the research mission, and then clinical support.
Funds-Flow Reallocation Principles
Our founding principles for the funds-flow reallocation initiative were that the methodology be transparent, rational, and mutually accepted by the chairs, and that the overall level of support to the departments should not increase in the absence of either increased clinical volume, increased external research funding, or clinical incentive opportunities which would be beneficial to both the health system and the department. The principle of not increasing support in the absence of one of these criteria was considered an essential prerequisite to the reform process if the possibility of opening the flood gates of additional claims for support was to be avoided. On the other hand, as we explain later, a significant increase in support to the clinical departments did in fact occur by the end of this process, in large part on the basis of increases in case volume, increased case-mix index, and mutually beneficial incentive opportunities, but also in part because some rational additional areas of valid support were identified. We clearly anticipated that the allocation of funds among departments might have the potential to change significantly, with some departments seeing increases in support and others encountering decreased support.
As an initial step in the funds-flow reallocation process, we worked with the clinical practice finance committee to develop an agreement that departments with positive fund balances would perform a one-time transfer of sufficient funds to the departments with negative fund balances to eliminate all fund-balance deficits and, after negotiation and refinement, this proposal was unanimously accepted by the chairs. The fund transfers were based on a sliding scale of fundbalance taxation, with each department retaining 80% of its first $1 million but with the retention decreasing in steps to 17.8% after the first $5 million in fund balance. In return for these contributions, the departments with positive fund balances gained budgeted access of up to 20% of their fund balances in any given year, assuming that they had made budget in the prior year.
In parallel with the fundbalance reallocation, we worked with a committee of chairs and school of medicine leadership to develop appropriate faculty expectations for clinical productivity and to develop expectations for clinical time commitments. The portion of teaching that was performed purely within the standard clinical environment was defined as clinical time, and productivity was benchmarked according to Faculty Practice Solutions Center (FPSC) data for specialty. The FPSC is a joint University HealthSystem Consortium/Association of American Medical Colleges endeavor, which pools data from participating AMCs to provide productivity benchmarks by specialty. The committee called for standardized four-hour clinical sessions and overall productivity at the FPSC 65th percentile for overall departmental clinical time. At the same time, we also worked with Penn’s clinical practice executive committee to develop defined clinic time expectations with variation based on faculty track, external funding salary support, and administrative and teaching commitments. These recommendations were accepted by the chairs.
Funds-Flow Reallocation Process
To lead the process, we developed a committee consisting of five clinical chairs, the vice dean for professional services, the executive director for CPUP, the vice dean for administration, and representatives from the hospitals and health system finance. The committee met one or two times per month for a period of approximately 12 months. Because of the sensitivity of issues involved, it was agreed that the deliberations would remain confidential, but that effective, clear communication of the progress was essential. The evolving recommendations were reported at regular intervals to the clinical practice executive committee and to all the chairs, as well as to school of medicine and health system leadership. Principles for the funds-flow reallocation were further refined and endorsed by all clinical chairs as an initial step in the overall process (Table 1).
On the basis of the agreed-on founding principles, the committee recommended support should be provided to the departments for didactic and small-group medical student teaching, for faculty time spent supervising residents and medical students, and for departmental administrative support for teaching programs. Medical student teaching had been previously tracked on a database allowing allocation of “teaching RVUs” according to actual hours of teaching. Teaching support was also provided for course directors and for other teaching administrative duties. Funds for resident supervisors were allocated on the basis of the number of residents within a department and the average faculty salaries for the department, using a ratio of resident supervisors to residents of 1:6 in the cognitive specialties and 1:10 in the procedural specialties. Program director support was based on the number of residents within a program and adjusted on specialty-specific Accreditation Council for Graduate Medical Education requirements. In the absence of specialty-specific requirements, programs with 0 to 15 residents received 0.125 FTE of GME program director support, and this increased in gradations up to 1.0 FTE for programs with more than 75.
The committee recommended that three categories of research support be provided to the departments: academic development funds, indirect cost sharing, and salary coverage incentive payments. As previously, academic development funds would be provided by the dean’s office for specific academic development projects within departments, such as strategic research start-up packages for research scientist recruitments. The committee recommended that all such support be time-limited (typical maximum support is three years). They also recommended a new program of sharing indirect cost recovery with the department. Under this proposal, funds equivalent to 16.5% of indirect cost recovery dollars would be provided to the departments. The committee also called for financial support for the departments in order to provide compensation for some unfunded academic time. Departmental support was provided at a level equivalent to 12.5% of salary for tenure-track faculty, and 7.5% of clinician educator faculty up to the National Institutes of Health (NIH) cap. For departments with salaries above the NIH cap, additional support could be provided up to 30% of total funded departmental research. As with teaching support, the committee recommended the funds developed for research be provided at the departmental level rather than directed to an individual faculty member, so as to provide a departmental incentive to increase peer-reviewed funding, while at the same time allowing chair discretion as to how the funds are expended within a department.
Financial support of the clinical mission was defined in several categories: new program start-up/new recruit, purchased services, programmatic support and incentive payments, and pass-through payer contracts (Table 2).
New program/new recruit funding is a cash guarantee agreed to by the hospitals on the basis of profit and loss projections for the program or recruit; it is provided to the department rather than the individual and is typically limited to a maximum of three years. Templates were developed for new-hire support; they included defined time periods for clinical practice ramp-up and identified appropriate costs, while capping departmental overhead costs by specialty.
Twenty-five percent of each chair’s effort, as well as other faculty time for health system or school of medicine administrative duties, directorships, or regulatory positions, are purchased services. Purchased services include salary, benefits, and malpractice.
Clinical programmatic support was defined in conjunction with the hospitals. Such support was targeted according to the overall importance of clinical programs, which, even when performing at the expected clinical productivity levels (65% FPSC) and with mutually agreed overhead/infrastructure rates, were not expected to break even. The committee recommended that appropriate mutually agreed-on expectations be established between the hospitals and practices and that these arrangements be reviewed in relation to market conditions every other year.
In certain areas, incentive payments for departmental, divisional, or programmatic growth were developed. Such incentives involved revenue-sharing agreements between the hospitals and the clinical departments based on new operating margins.
Some third-party contracts had been negotiated as a global fee or with a global health system reimbursement perspective. In these cases, an appropriate fund pass-through would be continued on the basis of market equitability and clinical volume.
Communication and implementation
All chairs accepted the preliminary principles of the funds-flow reallocation, and the chairs were kept informed of the process through formal communication channels as the more-detailed definitions of support were defined for each of the clinical department missions. The foundations for appropriate support were discussed with the chairs and modified where appropriate on the basis of consensus. Although the potential impact of proposed changes was modeled during the process, these figures were not discussed until the process was completed. Following identification of the different clinical funds-flow categories, there was careful discussion between the hospitals and each department to develop a mutual understanding regarding the clinical funds flow provided in each category as well as the expectations including short-term and strategic goals.
After the chairs and the health system leadership defined and accepted the final principles of the entire funds-flow reallocation, one-on-one meetings were held with the leadership of each department to review the individual departmental impact of the proposed changes. In situations where the reallocation resulted in a significant negative impact to a department, leadership attempted to identify potential areas where incentives which would be beneficial to both the department and the health system could be developed. Where incentives could not be implemented, transitional payments were negotiated so as to allow the departments with decreased funding to take appropriate measures to readjust to the new level of funding. At the end of the process, departments were given one year to model the changes into their respective budgets, and a transition support plan was developed in which significant reductions in support were planned.
Outcomes of the Funds-Flow Reallocation Process
Overall, whereas annual total funding for the clinical departments from the health system and school of medicine increased 30.8% from $121 million to $158.6 million between FY2005 and FY2007, school of medicine funding to the clinical departments decreased 25% from $12 million to $9 million. The largest component of the increase in total funding was associated with new hires and additional programmatic support for clinical program strategic growth; however, other factors in the increase were inflationary increases on the teaching, research, and clinical purchased services programs, new academic development funds, and increased third-party pass through associated with additional volume on global contracts.
In terms of the mission-specific funding, support designated for teaching increased from $7.6 million to $23.9 million, and funds designated for research support increased from $11.1 million to $33.0 million, on annual funded research-budgeted expenditures within the clinical departments of $300 million. On the other hand, funds designated for clinical support remained essentially stable (Figure 1).
At the departmental level, there was significant variation in the impact of the introduction of this mission-based support methodology (Figure 2). Whereas in some departments there was little change in funds-flow support, there were significant financial gains in some departments; the largest gains were in Medicine ($12.9 million) and radiation oncology ($6.6 million). The largest funds-flow reductions occurred in emergency medicine ($2.1 million) and pathology ($2.0 million).
Overall, there has been progressive and sustained improvement in the financial position of the clinical departments during the past three years, with significant additional improvement in the fiscal year that is currently closing (FY2007). In FY2004, the clinical departments had a decrease in net assets of $34 million, and for FY2007 a gain in net assets of $1.3 million is budgeted. This financial improvement is significantly attributable to the increased support provided under the funds-flow process. However, the development of clearly identified lines of support with defined expectations was an important step in the process, as were the abilities to restore confidence in the equitability of the financial support process, to align incentives with the strategic goals of the health system and school of medicine, to define financial support for the different missions, as well as to restore rewards for profitability at the departmental level.
The funds-flow process undertaken at Penn over the past few years has rationalized the flow of institutional funds to the clinical departments on the basis of mission and alignment with the overall strategic plan of the institution. The ability to provide fair and transparent institutional support for each of the missions on the basis of well-defined and broadly accepted principles aligned with the overall strategic plan has aided in providing appropriate incentives for the varied missions and in improving the overall financial performance of the clinical practices. Except in the case of faculty with major administrative teaching roles, the funds are provided to the department and are not earmarked at the faculty level. It thus remains the chair’s role to identify the best use for these funds within the departmental setting. The research incentives provided through the funds-flow process should provide additional incentives to encourage pursuit of external funding opportunities. In the future, the vice dean for education will participate in the departmental budget process so that clearly-defined objectives are linked to the funds provided for teaching. At this point in time, teaching outcome measures have not been directly linked to the funds flow for teaching, although as identified by others, this does remain a possibility at a future point in time.9,10 The ratio of funds provided for teaching ($24 million) compared with those provided for research ($33 million) is different from that reported previously from other institutions.2 However, this may be distorted by the research-intensive nature of Penn, or by the fact that some other institutions already pass a portion of the F&A dollars back to the departments and consider this outside the discretionary funds-flow process.
The funds provided for the support of clinical care delivery are now clearly categorized and identified. Funds are provided for new program start-up and for new recruits, and a template devised for cost allocation with a preagreed departmental overhead should frame and ease future negotiations. Purchased services are also clearly defined and should be identified with job descriptions. Such proportional FTE funding also implies input into the annual evaluation for that faculty member. Having established productivity expectations for clinical time and expected clinical time commitments for different faculty tracks and differing levels of external funding eases the identification of where programmatic support is required. This, combined with incentive payments, should help to ensure that areas of strategic importance continue to be developed.
The integrated nature of our health system and school enabled a rebalancing of the problem of the increasing gap between professional and technical/hospital reimbursement, consistent with the academic missions of teaching, research, and clinical care. The connection also made it possible to align the clinical goals with those for the health system. Defining the funds-flow clearly to support the different missions of the clinical departments and making the reallocation process open and based on principles that are both accepted and fair has significant advantages in terms of realigning strategic goals. On the basis of our experience, this process seems to be a significant factor in overall improvements, both financially and in terms of productivity, within the clinical practices. In order to maintain chair flexibility and autonomy, the majority of funds are allocated at the department level rather than being earmarked for specific faculty. Since funds-flow may be based more on historical precedent and individual negotiations at academic medical institutions, it seems that a broad reevaluation such as the one performed at Penn might be helpful at other AMCs.