Health Care Expenditures

The Canadian health care system has grown rapidly. Today over $70 billion or 10.1% of Gross Domestic Product (GDP) is spent on health, compared to the 1966 rate of 6.1% when the Medical Care Act was passed by the Federal government. In Ontario, approximately $1.7 billion1 or 10 % is allocated to the delivery of primary medical health care services. Given that approximately 85% of the populace annually accesses primary medical care services, it can be argued that the administration of such services is extremely cost-effective. Notwithstanding this, the surge in proportion to society's resources devoted to health care is equivalent to a 2% average annual increase in expenditures over and above both inflation and the real growth of the economy. This rise in the past has been attributed to a number of factors:

• the increase in the number of health-care professionals per capita;

• increased utilization of institutions because of advances in technology, which is the largest component of health-care spending; and

• population increase and aging.

Declining government revenues, a substantial government debt burden, and a clear election mandate for governments to reduce the size of public sector expenditures, raise significant doubts about whether governments can afford to finance the health care system in a manner to which it has become accustomed.

One of the most critical principles of health care reform is coordination of care. Because of its impact on quality of care and cost to the system, particularly through unnecessary duplication of services, and the potential for compromising the quality of care because of the lack of a designated individual to oversee all aspects of care received by the patient, the need for coordinators of health services in an increasingly complex, technological and yet constricted health care system has never been more important. As outlined in the CFPC "Green Paper," Managing change: the family medicine group practice model," released in September 1995,

"The public assumes that the system of care is coordinated and that health care professionals collaborate. One of the worst problems of the current Medicare system is that it is highly disorganized and fragmented. It is a patchwork of services, providers, facilities and care delivery. There are no formal linkages between functional areas to coordinate care."

In addition, positioning the family practitioner as the point of first contact and involving the physician as the care coordinator enables the physician to act as a patient advocate, enhancing their ability to assist consumers in negotiating the maze of health care services to ensure they receive integrated services in a timely and appropriate manner.

The Physician Advisory Group therefore concurs with the opinions expressed by the CFPC in the "Green Paper" regarding the assessment of the benefits of coordinated care. The beneficial aspects of this approach for patient care are far-reaching in that they ensure:

• improved continuity of care over the patient's lifetime;
• reduced costs as duplication and overlap are eliminated;
• more timely and appropriate care;
• more efficient and appropriate use of the system;
• better preventive care and health promotion;
• better patient mobility through the system;
• less frustration and patient anxiety;
• better outcomes; and
• better understanding by the patient.

Public finance and private provision characterize the Canadian health system. Indeed, while 72 percent of health expenditures are publicly financed, virtually all services are privately provided. The majority of the approximately 10,000 active GP/FPs are generally self-employed independent professionals who rely upon a single funder for the majority of earnings, that is, reimbursed by provincial health plans on a fee-for-service basis.

Hospitals, which are owned and operated on a not-for-profit basis, receive prospective global budgets from provincial governments to finance in- and out-patient services. The terms and conditions under which public finance for health care is translated into private provision are determined through negotiation between providers and government, without consumer input. These arrangements have been criticized for providing:

1. limited opportunities for public funding to "follow patients" (i.e., to be responsive to the needs of patients);
2. restrictions on the availability of supplementary private sector funding for publicly insured services; and for the general lack of economic (i.e., market price and outcome) signals for both system performance appraisal and resource allocation.

1. are consistent with government policy;
2. provide opportunities for greater private sector funding for health care; and
3. that advance the provision of comprehensive, cost-effective, and appropriate health services.

1. capitated primary care funding at a regional level;
2. capitated payment of primary care physicians;
3. the establishment of primary care organizations (PCOs) as the basic unit of primary care in a managed care environment; and
4. the use of non-physician primary care providers.

After reviewing the various models adopted in other jurisdictions, both nationally and internationally, the Physician Advisory Group developed a comprehensive model for primary care, entitled "Reformed Fee-for-Service" (RFFS). Issues of reimbursement alternatives, organizational change, options for public/private funding are discussed. Instead of undertaking a partial analysis of health reform, the committee has conducted a more comprehensive review.

2.0 Current Patterns of Primary Care Practice

It is impossible to comprehensively evaluate current patterns of primary care practice and to exhaustively review all potential reform options in a single report. Given the twin constraints of time and other resources, the committee decided to: outline, somewhat superficially, current patterns of primary care practice; to highlight the principles that may guide the reform process; and to sketch, in broad strokes, the general reform options. Therefore, this chapter does not provide a detailed and exhaustive evaluation of all potential reform options. Rather, the informed implementation of any particular option requires further public dialogue and a detailed strategy for implementation.

Primary care physicians act as agents for an increasingly informed and assertive population of patients. As a result of limits to government health spending, these physicians have been forced to struggle with the conflicting demands of their patients (for increased health service intensity) and their assigned role as "gatekeepers" to the health care system. Canadian medical practice is currently characterized by some rationing of health care services (i.e. acute care bed admissions to hospitals); however, while such constraints have become a characteristic of practice for doctors, they are not particularly significant in the primary care setting where access is quite open for patients.

"Coordinator/Gatekeeper" Role

Usual practice dictates that patients require referral by primary care providers if they are to receive publicly financed diagnostic or specialty services. Without a primary care referral, patients are generally unable to access these publicly financed services. They may, however, receive some services that are not fully publicly financed (e.g. chiropractic) without a primary care referral by financing the cost of these services privately, either through supplementary insurance or direct payments or charges. As a result of patient referral coordination, primary care physicians have the potential to play an effective "gatekeeper" role in restraining the growth of health care expenditures. It should be noted that alternative, or non-physician, health care providers at times refer patients directly to specialty care. However, such referrals tend to occur only in an ad-hoc or non-uniform manner and have the potential to undermine the coordinating role of the family doctor.

Recently, as health care expenditures have grown and the availability of public funds for health care services have been limited, debate has turned to the distinction between "core" and "non-core" health care services. While uniform and precise definitions of "core" services do not currently exist, the health policy literature has highlighted various principles or criteria to distinguish between such "core" and "non-core" services. Indeed, there have been two recent and well-reviewed efforts to determine which health care services should be designated as "core services." In both Oregon and the Netherlands, the primary criterion for determining whether health care services would be publicly insured was based on whether such services were cost-effective. While the Dutch review was based on a broader definition of effectiveness than the Oregon review, and incorporated factors that diminish handicap as well as a traditional emphasis on those factors that alleviate disability, both reviews took an outcomes (or treatment effectiveness) approach to distinguish between "core" and "non-core" services.

To the extent to which primary care reform is concerned with distinctions between public and private funding for health care services, and between "core" and "non-core" services, it is important that "core" services be defined. Indeed, federally or, at a minimum, provincially, "core" services might be defined to satisfy a basic set of principles associated with medical necessity. Specifically, "core" services are health care services that are: clinically effective ("they work"); applied appropriately ("they are needed by the patient"); and that provide for patient choice ("they are wanted by patients"). While Ross and Deber outline these principles for the definition of "core" services, the evidence to support the application of these principles has yet to be detailed. Moreover, even if uniform and precise definitions were developed, and scientific evidence gathered, these principles are silent on the appropriate mix of public/private funding for "core" services.

The Canada Health Act affords the provincial and territorial governments considerable latitude in interpreting the principle of universal coverage on "uniform terms and conditions" for all "medically necessary services." In November 1995, the CMA published an article entitled "Understanding core services" in the series The Language of Health Care which stated that,

"even though medicare is primarily a provincial responsibility, the federal government so far has been able to use its financial clout to maintain the principles stated in the Canada Health Act (CHA) - public administration, comprehensiveness, universality, portability, and accessibility. a national list of core services feasible, practical, or desirable."

"as provinces become more selective about core services, concern has been expressed about how decisions to insure or de-insure are to be made."

The Primary Care Physician Advisory Group recommends that the government of Ontario establish a public commission to examine medical service coverage under the publicly-funded health-care system. The OMA would be pleased to assist the government of Ontario in this endeavour.

3.0 Models of Primary Care Reform

At present there are three models described for consideration in the delivery of primary care in Canada:

1. the fee-for-service model in solo or group settings;
2. the HSO model;
3. capitated/blended payment models.

1. Consistency with the provisions of the Canada Health Act.
2. Consistency with objectives for primary health care reform.
3. Acceptability to policymakers and stakeholders.
4. Potential availability of required resources.
5. Availability of the required technology (or potential to develop it in the short term).
6. Promotion of appropriate accountability.
7. Provision of clinical and practice management tools required for the delivery of effective, efficient care.

There are basically three broad classes of physician reimbursement methods that are commonly used:

• input-based (i.e., salary system);
• throughput-based (i.e., fee-for-service); and
• population based (i.e., capitation).

Numerous articles, books, policy documents, and position papers have focused on the general area of physician reimbursement mechanisms. Some have emphasized issues that are pertinent to a single jurisdiction, while others have attempted to be more general in their application. Indeed, a review of the literature conducted for the OMA's Subcommittee on Health Care Funding concluded that there was no single reimbursement scheme yielding incentives for adequate physician compensation and efficient clinical practice that was also compatible with the provision of comprehensive, cost-effective, and appropriate health care services for patients (and taxpayers). While some aspects of each payment scheme yielded benefits to society, none were without defects.

For the medical profession, the debate about which payment mechanism is best can be traced to the turn of the century, and continues unchanged to this day. In 1935, the Committee on Inter-Relations reported to the OMA Council on this issue:

"The question of the best method of remuneration of the general practitioner will remain ever a debatable point. Nearly unanimous is the opinion that it should not be by salary, which eliminates the element of competition. On the other hand, opinion seems equally divided between the remaining alternatives - the per capita and the per unit of service methods.

The per capita or panel method has much in its favour. By eliminating the reporting on each individual visit, accounting and inspection costs are markedly reduced, leaving more money available for fees. Since the physician's income is not immediately affected by each individual call, he is free to discourage unnecessary calls at irregular hours and to make frequent calls in acute or interesting cases without having his motive misunderstood. The panel (per capita) method also tends to promote more amicable professional relationships since it makes more definite to which doctor a patient belongs and permits the absence of the doctor for holidays and post-graduate study without so great danger of losing his patients or insurance income. By discouraging "shopping around," it promotes greater continuity of the physician/patient relationship. This method also eliminates some of the disadvantages of the rural practitioner for whom distances make for greater hardships but reduced practice per person. Because a fixed basis is provided for actuarial calculations, it is favoured by legislators and actuaries, while an evenly distributed insurance income through the year appeals to many physicians. Incidentally, approval of the per capita method would provide proof of the ethical grounds for the profession's demand that hospitalization and specialized services be by reference from the general practitioner.

While the advantages of the per unit of service method are fewer in number, they are more important. Thus, it eliminates automatically too ready reference to hospital and specialist, an outstanding weakness of the English system. When remunerated in proportion to the actual work done, the medical practitioner is not required to assume the risk of morbidity fluctuation. There is not the same difficulty in providing for mileage rates. Finally, the unit of service method has a distinct tendency to encourage more painstaking effort on the part of the doctor in each individual case and so favours a higher general standard of service. This being the end to which the entire plan (health insurance plan) is devoted, the Committee felt impelled to recommend the per unit of service method of remuneration for all types of professional service."

4.1 Fee-For-Service Reimbursement

Currently, the overwhelming majority of primary care physicians in Ontario are self-employed practitioners who are reimbursed for the provision of health care services on a fee-for-service basis. In Ontario, about three-quarters of GP/FPs receive 90 percent or more of their incomes from fee-for-service payments. There is some consensus within health policy, academic, and government circles that the fee-for-service method of reimbursement provides incentives for service generation to enhance earnings, particularly when patients are less informed about the attributes of the services provided by physicians.

Regardless of the specific payment method adopted, discussions regarding the merits or disadvantages of each model inevitably focus on the element of financial risk of illness each system encompasses. This financial risk of illness is borne by either the provider, patient, or payer according to the reimbursement method adopted. Critics of fee-for-service cite the potential risk to the payer - in the Canadian context, government - of an open-ended reimbursement system, and therefore point to capitation as a mechanism by which the payer can control expenditures. However, the financial risk of illness is transferred from the payer to the provider under a capitation system. As described by the BCMA,

"Risk under capitation may be defined as having to provide services to a patient or patient population at a cost that exceeds the amount received to care for said patient or population, and every physician is potentially at risk for the ‘excessive' health care utilization and costs of the patients for whom they are receiving capitation payments."

At any rate, under a capped budget, service generation is relevant only in respect of the resulting discount of physician payments.

Given the potential for patients to self-select physicians, and given the differences in the characteristics of physicians who are willing to practice under alternatives to a fee-for-service system, it is currently impossible to identify the independent affect of payment schemes on patterns of practice and utilization. However, in the absence of definitive evidence rejecting the service-generation hypothesis, open-ended fee-for-service reimbursement schemes represent a significant financial liability for government. Moreover, even if global payments were capped, this payment schemes may result in inter-specialty concerns regarding income and service intensity inequality, particularly between procedure-intensive specialties and other groups.

Arguments can be developed to support the continued use of the fee-for-service physician reimbursement model for health policy reasons. Since health service provision requires physician time and other inputs, an increase in service provision would eventually be associated with higher "opportunity costs" of clinical practice: as service provision rises, leisure time is lessened, and foregone leisure time is increasingly of value. Therefore, if the fee schedule could be modified such that physician service provision and corresponding payment matched the needs of patients, there would be very little incentive for physicians to over-service patients through the "manipulation of information for self-gain". This, though, is easier said than done.

By anticipating physician responses to modifications in the fee schedule, there are opportunities to redirect patterns of practice in ways that help serve health policy objectives. For example, if the government were committed to enhancing health promotion and disease prevention activities, the introduction of specific fee service codes into the fee schedule may encourage the provision of these services. In addition, modifications to the fee schedule through the use of fees contingent on delivery modality or time-of-day may assist in the achievement of specific health goals. Moreover, if there was the political will to alter fees to address possible earnings inequalities between general practice and other specialty groups or between medical and surgical specialties, the fee schedule may serve as the instrument of change. The elimination of the fee-for-service method of reimbursing physicians would result in the loss of this potential policy lever.

4.2 Capitation Reimbursement

There is a wide variety of capitation-based reimbursement schemes, one example of which is where primary care physicians are reimbursed on the basis of prospectively-determined capitation payments. Such capitation payments reward physicians for the provision of care to roster patients, but may be associated with the under- or over-utilization of diagnostic services and specialty care if the capitation payment includes or excludes the anticipated cost of those services.

As described previously, capitation schemes encourage physicians to attract healthy patients and to discourage those with high service demands (often referred to as "cream skimming" or "cherry-picking"). Ideally, capitation should tailor payments to risk categories accurately enough to prevent this sort of behaviour through the cream skimming of favourable risk patients. While competition for patients may result in the provision of health promotion services, enhance task delegation, and raise patient satisfaction, it is also possible that the reverse may arise relative to the current fee-for-service reimbursement scheme.

The major advantage of a pure capitation payment scheme is that the physician becomes more patient oriented. Since income is a direct function of the number of patients he/she can attract, this should strongly motivate the physician to provide rapid access, courteous staff, good premises, etc. However, there are also certain disadvantages to a capitation system, for example:

1. there is no direct means of using the system to pursue public policy in health care;
2. the patient's standards may be met, but professional standards of health care may not; and
3. the system may encourage some physicians to limit consultation time by issuing more prescriptions or referrals to specialists and hospitals for further investigation.

The contention that competition for roster patients and incentives to reduce health care utilization results in the provision of health promotion services by physicians reimbursed by the capitation method depends largely on a number of components, such as patient expectations, i.e., if patients expect physicians to address their illness needs, then that would be the arena for competition. However, if patients were impressed by the presence and use of sophisticated health care technology, whether or not such technology was appropriate or effective, then competition may be pursued in that dimension. Competition may therefore increase patient satisfaction, but not the actual quality of care. To assume that competition primarily occurs in the educational domain, namely the provision of health promotion and disease prevention services, is rather restrictive and may be at variance with patients' expectations of what constitutes quality health care.

4.3 Salary Reimbursement

Since there is no direct relationship between service provision and the payment of salaried physicians, they do not have financial incentives to attract roster patients, provide services, or offer patients quality care. Such physicians neither face incentives to practice efficiently nor to restrain their propensity to refer patients for diagnostic services or specialty care. Unlike the two physician payment methods discussed previously, salary reimbursement is invariant to client choice. There is no direct relationship between service provision and the payment of salaried physicians.

Salaried physicians do have incentives to engage in non-patient care activities, such as teaching and research, without financial cost; these advantages may account for their adoption in teaching health science centres. There is little support in the literature for the proposition that alternatives to fee-for-service reimbursement improve practice efficiency and increase the quality of patient care. In fact, the primary change attributable to the salary payment method is a lack of incentive to do anything but to show up for work. The method, by itself, does not offer physicians incentives to counsel patients in health promotion activities. Compared to their fee-for-service counterparts, salaried physicians tend to work fewer hours per week, fewer weeks per year, and see fewer patients during the course of a day.

Overall, relative to fee-for-service physicians, there is an expectation that salaried physicians would be less efficient, exhibit greater practice costs, see fewer patients with less intensive forms of care, and have a greater propensity to refer patients for diagnostic services and specialty care.

4.4 Acceptability of Alternatives to Fee-For-Service

Information on the acceptability or attractiveness of a variety of alternative payment methods to physicians was collected by a 1994 Decima survey of Ontario physicians. An analysis of survey findings, disaggregated by type of practice, specifically, by physicians in Family Practice (FP/GPs) and all other physicians (specialists) reveals that the most attractive payment method to FP/GPs is the current fee-for-service model, followed by a composite or blended payment system (salary plus bonus).

• While the fee-for-service option is found to be the most attractive overall, it receives less support from FPs than from non-FPs. Thirty-eight percent of FPs rate the fee-for-service model as "very attractive" compared to 46 percent of non-FPs.
• Remuneration schemes where payment is based on the number of patients to whom a physician is responsible for providing a specified scope of services (i.e., capitation) received very little support from physicians: overall, less than 7 percent find them "very attractive." However, FPs showed substantially less distaste for the capitation option. About 45 percent of non-FPs stated that such an option was "not at all attractive" compared to only 28 percent of FPs.

• The current fee-for-service model:

• Payment based on the number of patients to whom you are responsible for providing a specified scope of services:

• A blended system of fees and fixed compensation, in which a salary is paid for providing designated services to a set number of patients, and any service provided above this is paid through a bonus system:

4.5 Public and Physician Opinion on Patient Rostering

Physician support for patient rostering (or enrollment, or registration), which is usually associated with capitation payment models, appears to be somewhat mixed. In a 1994 Decima survey, over 70 percent of Ontario physicians stated that there should be limits placed on the number of different primary care physicians that a patient can consult each year. However, the same survey found that physicians were quite evenly split in their opinions on whether patients should be "required to register with one physician and restricting access to primary care to this physician or his or her on-call back-up". About one-quarter of general/family practitioners (FPs) were "very supportive" of this approach to utilization control and an almost equal percentage were "not at all supportive".

• Support for requiring patients to register with one physician and restricting access to primary care to this physician or his or her on-call back-up:

Ontario residents are also divided in their views regarding patient rostering. A recent Decima survey (December 1995) of adult Ontario residents found that about one-half (51 percent) of Ontarians approve of the rostering concept, with 21 percent noting "strong" approval and 30 percent approving "somewhat". Notably however, an almost equal proportion disapprove of rostering, with 27 percent indicating "strong" disapproval. The data indicate that to some extent this disapproval stems from a desire to maintain choice and decision-making responsibilities.

• Ontario residents approval of patient rostering:

4.6 Evaluation of Physician Reimbursement Mechanisms

Governments have shown increasing interest in paying General and Family Practitioners in ways other than fee-for-service. This began, in an organized way, in 1991 with the release of the report, Toward Integrated Medical Resource Policies for Canada. The report, authored by Morris Barer and Greg Stoddart, concluded that "the concept of capitated payment would seem to be far more consistent with the objectives of primary care." Their position was that although there does not exist a single best method of paying physicians, the problems associated with the fee-for-service method of payment were such that alternative methods should be chosen unless fee-for-service can be proven to function better in specific settings. Governments took a slightly more tolerant (of fee-for-service) approach, agreeing to replace fee-for-service with alternative methods of payment where fee-for-service is not consistent with policy objectives. However, an additional element of the policy directions cited by the Ministers of Health was the establishment of "predictable medical care expenditures through a combination of global, regional and individual practitioner budgets."

Since that time, the Ministers of Health have commissioned two studies on alternative methods of payment: Paying the Piper and Calling the Tune: Principles and Prospects for Reforming Physician Payment Methods in Canada, by Stephen Birch; and A model for the reorganization of primary care and the introduction of population-based funding, by the Advisory Committee on Health Services, widely referred to as the Kilshaw Report.

Birch released his report in May of 1994, and took a similar stance to that of Barer and Stoddart. While stating that the literature is not helpful in determining a best method of physician payment, Birch argues that fee-for-service is "widely recognized to provide the wrong signals to physicians (i.e., the more services one provides the more income one receives, irrespective of who receives the service and the expected effect of providing that service on the client's health status"), and believes that capitation supplemented with payments to "reward specific achievements" provides the right signals. However, this stance is qualified by the reality that Ontario physicians are considered to be extremely productive and ensure ample opportunity for patient access. It is therefore important to acknowledge the incentives of the fee-for-service system which allow physicians to meet patient demand.

Finally, the Kilshaw Report, released in draft form on July 18, 1995, discussed population-funding models, and while seemingly showing a preference for capitated payments to providers, nevertheless suggested that the agency receiving population-based funding determine the appropriate method of payment of providers.

The above authors show their preference in arguing for capitation-based funding. .Whereas they believe that fee-for-service results in patients receiving medical services they should not receive, little attention is paid to the prospect of under-servicing when the payment mechanism becomes independent of the services provided as it is under capitation or salary. Birch says as much by suggesting supplemental payments to reward physicians for providing certain services or producing specific outcomes prescribed by government. Could not Birch's criticism of fee-for-service, i.e., that it sends the wrong signals by virtue of paying more for more services, be interpreted as providing an incentive to physicians to do all that is possible for their patients? (His subsequent statement that under fee-for-service, payment is divorced from who receives the service and its expected outcome, would apply to more than fee-for-service).

Physicians paid by whatever method should have as their primary consideration the overall health needs of their patient. There is an undeserved presumption that family practitioners paid fee-for-service pay careful attention to each service and its fee, and are prepared to suspend clinical judgement and to practice contrary to the patient's interest for the sake of a few extra dollars. Under fee-for-service, physicians accept that some aspects of needed care, or items of service, may not be well-remunerated; other items, hopefully, will make up for that deficiency. Conversely, it is speculated that capitated or salaried physicians are not prepared to abandon their patients in order to free time for other endeavours.

For equal levels of payment and with all other things held the same, do different methods of payment lead to substantive differences in patient care? By switching to a system other than fee-for-service, will physicians alter their approach to patient care? Is the greater satisfaction of Canadians toward Medicare as compared to the satisfaction of the British public in part due to the predominance in Canada of fee-for-service? The answers to these questions are not known. However, it is the opinion of the Physician Advisory Group that the effects of different methods of payment on a practice are likely to be minor. Different methods of payment may, at the margin, be expected to produce different incentives for, as an example, preventive care, but in the main, it is uncertain whether those differences would be discernible. If properly constructed, each has merit.

5.0 Reformed Fee-For-Service Model

The purpose of this Section is to outline in some detail the features of the Physician Advisory Group's preferred model for the delivery of primary care services, the Reformed Fee-For-Service Model.

5.1 Background

The Social Contract introduced a global ceiling on expenditures for medical services during the three fiscal years beginning with 1993/94. In 1993/94, the cap was set at $93 million below 1992/93 expenditures, and in the 1994/95 and 1995/96 fiscal years, the cap was $138 million below 1992/93 expenditures.

Payments in excess of the globe are ‘clawed back' by reducing each physician's billings by an equal across-the-board percentage. In 1993/94, the clawback was 2.8 percent; in 1994/95, it was 7.5 percent; and for 1995/96, the clawback is projected to be 10.88 percent. Over the term of the Social Contract, physicians of the province will give back to the government more than $800 million in payments they received for medical services rendered to patients.

Billings are exceeding the cap for a variety of reasons. First, the current cap is 3.5 percent below expenditures in 1992/93, and there was an increase to the Benefits (price) of 1 percent effective October 1, 1992. If medical service volume could be held constant, expenditures would still exceed the cap by 4 percent. And yet, volume can be expected to rise, if for no other reason than the growth in population. But population aging, physician redistribution, technological advances, economic conditions, epidemiological trends, hospital downsizing, and other factors, play a part in changing medical service utilization.

Physicians are particularly disturbed by government and other third-party activity that is causing increased billings to OHIP which results in an increase to the amount of clawback. Examples include: government initiatives to expand the cardiac surgery, renal dialysis, mental health and cancer treatment programs without an adjustment to the cap to recognize the increased services that will result; hospitals moving in-patient diagnostic services into out-patient and community settings where they become the responsibility of OHIP; hospitals now requiring pre-admission work-ups so that the services are billable to OHIP, with the resulting increase in service volume.

Many of the examples given above, and much of the growth in medical service provision paid by OHIP, comes about through third-parties shifting their financial liability for payment of medical services onto OHIP. This phenomenon--cost-shifting--changes the source of payment to physicians (from third-parties to government) and results in a loss of income to physicians. As third-parties move their liability to OHIP, associated payments cause the expenditure cap to be exceeded: the money is simply clawed back from the profession. The evidence is strong that cost-shifting is on the rise and physicians are looking to government to strengthen regulations governing such activity.

The clawback is having a devastating effect on physician practice.

The amount of the clawback should be viewed in the context of the costs physician incur with respect to providing medical services. Family practice expenses averaged 42 percent of gross revenues in the 1992 tax-year. Current estimates suggest that expenses are between 45 percent and 55 percent. Expenses are comprised of building rent or leases, medical equipment and supplies, staff salaries and benefits, insurance, postage, telephone, accounting fees, and so on. From the roughly 50 percent of revenue remaining, the physician purchases benefits: extended health, dental care, life insurance, and may also make contributions to his or her registered retirement savings plan. Since the clawback is deducted from gross earnings, a 10 percent reduction in gross practice revenue translates into a loss of perhaps 25 percent in taxable income.

The punitive aspects of the clawback have become entirely disruptive and demoralizing to the profession, and consequently physicians may increasingly respond to opportunities to practice outside the province, or may leave the practice of medicine altogether. As stated by Dr. Walter Rosser, Chair of the Department of Family Medicine at the University of Toronto, at a recent news conference convened to discuss the potential impact on medical students of the Ontario Omnibus legislation, Bill 26, eighty per cent of graduates have indicated they plan to leave Ontario and establish practices in other jurisdictions, primarily the US. Physicians who remain will be resentful of the discounts and increased workloads.

Clawbacks have also set up internal divisions within the profession, resulting in further demoralization. Physicians are quick to blame other members of the profession for the increasing amount of the clawback in the belief that colleagues are ‘ramping up' service provision in an attempt to circumvent the negative effect of the clawback. Evidence suggests, however, that although utilization per physician has actually declined, the overall rate of utilization has risen as a result of increased technology and the number of new physicians entering practice.

The current fee-for-service system, characterized by clawbacks, open-endedness, and unreasonable freedom of access by patients, cannot be sustained. At the same time, fee-for-service is the preferred method of payment, preferred in that it is most conducive to fair and reasonable compensation of medical services and to high levels of accessibility to medical services by patients.

If the government can no longer afford to fund an open-ended fee-for-service model, methods must be explored and adopted to curb inappropriate use of the health care system by both patients and providers, to lower utilization of the publicly-funded health care system, and to allocate system resources where they are most needed.

As stated previously, the OMA Physician Advisory Group supports a plurality of models for the delivery of primary care services, i.e., fee-for-service, blended capitation, and the salary model. The committee recommends as an addition to these models the Reformed-Fee-For-Service Model. This model enhances the "gatekeeper" role of the primary care physician, requires patient rostering, and also requires that individual thresholds be placed on physicians according to their roster size. Improving the management responsibilities and capabilities of primary care physicians is a central feature of this model, in accordance with the committee's belief that managed care responsibilities should be gradually introduced.

The essential features of the Reformed Fee-for-Service Model proposed are outlined in the following paragraphs.

5.2 Patient Rostering

The Primary Care Reform Physician Advisory Group recommends the creation of Rostered Family Practices (RFPs) with a requirement for contractual commitment by patients.

Although government has emphasized that utilization of health care services must be controlled in order to comply with the fiscal realties of the province, the focus to date has been on controlling physician services. The profession agrees that it is imperative to make health-care delivery more efficient and effective, and acknowledges the responsibility physicians must bear in ensuring that the system is managed appropriately. However, the profession also recognizes that the system cannot sustain itself if it remains open-ended for patient utilization. It is essential that the public share responsibility for controlling health care utilization, and therefore ultimately expenditures. The OMA believes that patient registration with an individual family practitioner is essential for the coordination of care.

Consequently, the adoption and implementation of patient rostering is central to the primary care reform strategy. Rostering to a general or family practitioner will:

• avoid duplication of services
• create efficient referrals
• constrain certain inappropriate usage by patients (e.g., self-referral, double-doctoring)
• promote continuity of care, which improves efficiency and outcomes
• address physician distribution
• reduce the potential for health care fraud
• provide opportunities for all to obtain a primary care physician

If rostering is accepted as a fundamental component of primary care reform, it is essential that the details for rostering, de-rostering and maintenance of the database necessary to implement this system are clear and applicable in the real world. Because this represents a significant departure from the open-ended system currently enjoyed by the public, this measure will require the enthusiastic support of government including a substantial public education effort. Although the Physician Advisory Group has outlined a number of the details such a rostering system would encompass, it is suggested that further specifics will be developed through pilot project evaluations and consultation with other jurisdictions which have mature rostered systems.

Under the reformed fee-for-service model proposed, the public would assume financial responsibility for non-emergency care if such care is obtained outside the rostered family practice (RFP) but within defined geographic proximity. Practitioners would be required to provide 24-hour response, but could do so by means of a group call practice, i.e., solo practitioners could join together to provide after-hours coverage in a shared-care arrangement. Such coverage would not only reduce the frequency of visits to hospital emergency departments and at times the indiscriminate utilization of walk-in clinics, but would also enhance patient satisfaction by improving access to after-hours care. In addition, by instituting such shared care for after-hours coverage, physicians would not be forced to assume an unreasonable burden.

Although many of the advantages of rostering with commitment have been successfully documented in Europe and American HMOs, this concept has never been tried in the Canadian culture. However, as in other jurisdictions, rostering of patients in Canada should provide a number of advantages for government.

Advantages for government include:

• the avoidance of duplication of services;
• more accurate data collection;
• the creation of efficient referrals;
• the promotion of appropriate use;
• the promotion of continuity of care which improves efficiency and outcomes;
• the provision of improved access to primary care services for patients formerly disenfranchised from the system, e.g. AIDS, Alzheimer, homeless and palliative care patients; and
• most importantly, the issue of physician distribution is addressed.

Re-rostering (changing doctors)

In regard to the issue of re-rostering, a problematic issue is that of how quickly an individual can change to another physician, and how frequently this can be allowed. The following is a description of rostering:

1. patients can change doctors in 6-8 weeks if unhappy with their primary care physicians;

2. the administrative process should be completed by OHIP in 7-10 working days;

3. patients can change doctors twice in a calender year, i.e., they may have a maximum of 3 primary care physicians in one year.

Patient "X"			rosters	Jun 1, 96	with Doctor "A",
			changes	Feb 14, 96	with Doctor "B",
			changes	Apr 1, 96	with Doctor "C",

and cannot change again in this calender year, but may always go outside the rostered system and accept financial responsibility.

The British Experience

Lessons can be learned from the experiences of other health care systems which have attempted to appoint or require physicians to be all things to all people at all times. For example, in the UK, general practitioners ultimately balked at the requirement to provide medical care outside normal office hours because of the extreme liberties taken by their patients in accessing this service. Citing the frequency and triviality of many of the night calls to which they were summoned, British GPs were unmoved by government's attempts to induce physicians via financial incentives to provide these services. Physicians were supportive of such measures to compensate for legitimate health care needs; however, they felt that because the service is paid for by government rather than the individual consumers who utilize such after-hours services, there is no onus on the beneficiaries to change their behaviour. This is exemplified most frequently by those at the lowest end of the income scale: according to inner-London GP Iona Heath, "for the most deprived people (who are devoid of financial muscle) the health service may be the only arena in which they feel able to exert their power as consumers.

Britain's Solution to 24-Hour Response:

The agreement ultimately reached by government with the BMA releases GPs from .their responsibility for 24-hour care if another GP can be appointed to provide this service. Two-thirds of GPs have subsequently indicated their interest in opting-out of this requirement, which would leave the remaining one-third to assume this responsibility. The conclusion to be drawn from this was articulated by Dr. Brian Hurwitz in a recent British Medical Journal editorial,

"It remains to be seen how it can be in patients' best interests to allow a minority of general practitioners to add other doctors' out-of-hours responsibility to their own swollen workload. The need for a responsive primary care service at night, separately organized and separately funded from the contract for general medical services, has never been more urgent."

If a rostered patient is refused needed services from the RFP and goes elsewhere, the patient, if charged for the service, would be able to recover costs from OHIP. The RFP under which the patient is rostered, would have its cap reduced by the amount of any primary fee-for-service costs incurred as a result of its not providing the necessary coverage. Under this circumstance, there would exist a strong incentive for RFPs to develop ‘call cooperatives' to deal with after-hours work. Alternatively, the RFPs may contract with other physicians, or leave it to other physicians, to provide the services and bill fee-for-service, recognizing that the earnings cap of the RFP to which the patient is rostered would be reduced by the amount of fee-for-service billings for those services. In any event, total fee-for-service billings for primary care services would not exceed the prescribed limit.

New Fee: "Rostering Interview"

The Primary Care Reform Physician Advisory Group recommends that primary care physicians be permitted to bill, on a FFS basis, for the initial patient "rostering interview"for transferring patients or re-rostering. There will be no new fee for establishing initial rosters.

Rostering is an essential feature of the Reformed Fee-for-Service model which offers a number of advantages over the status-quo. As mentioned earlier, it avoids duplication of services, better controls referrals, constrains certain patient abuses and promotes continuity of care which will itself improve efficiency and outcomes. But it may also ameliorate problems of patients finding family practitioners, insofar as the incentive to allow patients to roster within a busy practice is greater.

Payment of Physician for After-Hours Work in Defined Geographic Proximity

BMT: all office-based services costed to rostering physician.

After-hours call:

Patient "X" goes to after-hours clinic of rostered group call of his/her RFP. Patient is rostered to Doctor "A".

1. Doctor "A" foregoes potential income.
2. Doctor "B" in after hours clinic bills OHIP and takes additional funds to be added to Doctor "Bs" total income.
   Doctor "Bs" Bench Mark Threshold is not affected.
3. Doctor "A" covers Doctor "B" reciprocal arrangement; Doctor "C" covers Doctor "A" and Doctor "A" accepts lose of potential income (no income loss).

The contractual" provisions of this rostering scheme are essential to control the costs within the publicly financed single payer system and to minimize the financial exposure of the RFP. If patients are not contractually committed, bears the cost of patients choosing to go outside the RFP. If government absorbs the cost, overall costs may exceed the current fee-for-service arrangement. If the RFPs assume the cost, the resulting recoveries could bankrupt the practice or, at the very least, make the model unattractive to physicians. As well, the burden of administrating negations for 11 million patients covering 10,000 family practitioners would be enormous.

The requirement to have patients assume financial responsibility for the cost of discretionary care accessed outside their RFP would to some extent have the effect of making patients accountable for responsible use of the health care system, but would still allow for reasonable access as per the CHA. Although limiting access to primary care services through a defined RFP will not be welcomed by all, the Physician Advisory Group does not believe that this will be perceived as a burden by the majority of patients. The committee believes the public is prepared to take this important step in helping to control health care expenditures once they are fully informed of the details, and will not find it an onerous commitment.

5.3 Utilization and Expenditure Control

Upper Limit Roster Size: determined by current threshold. Payment set at $404,000 for all physicians (OHIP earnings).

Upper Limit Expectation: approximately 3000 patients per physician.

NB 1.	average panel size for the Netherlands is 2,350 at present.
NB 2.	threshold lower than 404,000 will aggravate underserviced regions that may have 1 physician with the equivalent of 2 rostered practices in underserviced areas, e.g. Doctor "A" works in city with a population of 500,000 people from Monday to Thursday, and Doctor "B" works in a town with a population of 5,000 people on Fridays.

The advantage of this approach is that it keeps the system manageable through simplicity of application.

The Physician Advisory Group recognizes that the Reformed Fee-For-Service Model must fit within the existing budgetary constraints imposed by the Government of Ontario. Two methods of achieving budgetary targets are in place under the current fee-for-service regimen. The first is payment discounting, discussed earlier, where payments to physicians are discounted by the percentage by which billings exceed the expenditure cap. The second method involves threshold payment adjustments, where discounts apply to billings of individual physicians once those billings go beyond a prescribed threshold.

The threshold payment is currently set at $404,000. Billings beyond that amount, but less than $454,500, are discounted one-third. Billings equal to or greater than $454,500 are discounted two-thirds. This form of income limitation suffers two problems. First, the one-third discount is not sufficiently high to give the degree of budgetary predictability sought. Physicians are prepared to continue practice beyond the first stage of discount. Second, the thresholds can exacerbate existing supply deficiencies. In regions where the patient to physician ratio is high, the value of services that can be allocated to each patient is lower than in areas where the patient to physician ratio is low. This is contrary to considerations of equitable access. To resolve these problems, the Primary Care Reform Physician Advisory Group recommends that each family physician have his or her fee-for-service earnings capped at an amount equal to the roster size multiplied by the provincial average capitation rate, adjusted for the age and sex composition of the roster. Let us refer to this figure as the "Bench Mark Threshold".

Bench Mark Threshold = Size of Patient Roster X Average Capitation Rate

This approach produces the budgetary certainty associated with capitation, promotes equity of access to primary care services, and maintains the positive incentives for servicing associated with fee-for-service. The approach would also have a powerful influence on the distribution of family practitioners.

The Physician Advisory Group recommends that the Bench Mark Threshold (BMT) be adjusted quarterly to ensure access of provision.

5.4 Exemptions from the Earnings Bench Mark Threshold

(total physicians income (maximum $404,000) = BMT + exemptions)

• obstetrical deliveries;
• emergency room work;
• anaesthetic services;
• surgical assists; and
• work in nursing homes and long-term care institutions.

Special Consideration for AIDS, Alzheimer and Homeless Patients

Recognizing that fee-for-service is not ideal in all circumstances, the Group recommends that family practitioners be paid a monthly fee per patient for care provided in nursing homes and long-term care institutions, for care provided to AIDS patients, to palliative care patients, and to non-institutionalized patients suffering from Alzheimer's disease. These monthly fees would also be exempt from the RFP Bench Mark Threshold, and as with the service exemptions, these exemptions would create an incentive for family practitioners to undertake this very difficult work. As with exempted medical services, the value of the monthly fees would be ‘backed out' of the capitation rates.

Some GP/FPs whose practices are limited to defined services and therefore unable to roster, e.g. GP psychotherapist, sports medicine clinics, should remain for now on fee-for-service. They are considered valuable services and they should be encouraged to align themselves with RFPs and be considered as referral practices. Most of the literature suggests that this area could be addressed through sessional fees, salaries, or reasonable FFS compensation. The Physician Advisory Group recommends consideration of this specialty subgroup in the evolution of primary care providers.

5.5 Physician Resource Planning

Under open-ended fee-for-service, it is still possible for a family practitioners with relatively small patient rosters to make an acceptable income. Under the Reformed Fee-For-Service Model, with rostering and an income BMT based on roster size, it would be difficult for physicians to locate in the larger urban areas, unless they were able to garner a reasonable amount of exempted work. This would apply to many established family practitioners in physician-dense areas, who would find it necessary to relocate to areas where the roster could be increased to an adequate size, or alternatively to provide more exempted services or locum tenens.

Physician distribution is addressed through the creation of rostered family practices as there is a finite number of patients to roster in a geographic location. The attractiveness of less physician-dense areas would be enhanced, and would provide viable opportunities to engage in fulltime practice, unless, or until, the earnings limits are raised. Physicians practising in rural or remote areas may, because of their large roster size, have an earnings BMT that they would be unlikely to achieve through their fee-for-service billings. This may be an inducement to these physicians to seek out other physicians, either on a contractual basis or to join their practice; or it will attract physicians to the area unsolicited by established physicians in the area. It is important that information be available to all physicians regarding practice opportunities throughout the province, which could be achieved through an enhancement of the role of the OMA's Physician Placement Service.

5.6 Integrated Services Funding

The Primary Care Reform Physician Advisory Group does not have comprehensive statistics on the amount of public funding for primary care services provided by other than general/family practitioners. It is known that $225 million was spent by the government of Ontario in 1993/94 on dental, optometric, chiropractic, osteopathic, and physiotherapy services. However, these figures do not include payments for nursing, midwifery, psychology, and other services.

Except in institutional settings, each of the above groups work independently of each other, and with little coordination or integration in relation to the overall primary care health needs of patients. This deficiency could be remedied by allowing RFPs. Integrated Services Funding (ISF) would be obtained through a reallocation of "government silos", and from savings resulting from health care system restructuring (i.e., hospital closures/amalgamations).

Integrated Services Funding would be used to contract with alternative health care personnel (i.e., nurses, dietitians, social workers, etc.) to work within the RFP. Currently, family practitioners may employ allied health care personnel in their practices to assist with the provision of medical care to patients. These staff are paid through practice revenues generated through billings to OHIP.

The staff employed by the RFP under ISF may perform tasks different from those performed by current staff who provide components, or ‘constituent elements', of medical services billed to OHIP. In any event, the ISF would not be used to subsidize the expenses of family physicians in the provision of medical services billed to OHIP. ISF personnel associated with RFP would not provide services leading to the billing to a fee to OHIP. Alternative health care personnel would care for patients in a collaborative manner within the integrated services model as part of the care team.

The Group recommends that the alternative health care personnel contract with RFPs under an established compensation grid. That is, RFPs should not be directly involved in the negotiation of salaries and benefits of ISF staff. Integrated Services Funding must be available to defray the increased practice costs--for facilities, supplies, equipment, etc.--associated with the ISF personnel. In addition, the RFP would receive compensation for the additional time involved in directing and coordinating the non-medical primary care being given to rostered patients. However, it is not possible to predetermine the increased practice costs, or the increased management or administrative burden associated with integrated services. This will depend upon the services the RFP would like to make available to patients, the availability of alternative health care personnel in the community, whether those staff are contracted fulltime or part-time to the RFP, and so on. For this reason, the Group is not prepared to suggest an appropriate capitation, or a monthly fee, for Integrated Services Funding. Instead, the Group recommends that each RFP apply to government for Integrated Services Funding, specifying the nature of the integrated services programs it would like to provide to its patients, the number of staff by type that are needed to carry out the programs, and the amount of funding required to defray costs.

The advantage of having alternative providers work in the RFP is that the primary care physician could presumably offer a more comprehensive range of services in their practice Busy physicians who are unable to visit nursing home patients as often as they would like may, for example, ask a nurse to attend to the patients between physician visits. The same may apply for some home visits. Alternative health care personnel may be used selectively for crisis counselling, screening follow-up, weight counselling, telephone advice, or foot care.

Integrated Services Funding will be a boon to physicians practising in rural and remote areas where doctor shortages mean that patients must either travel great distances or do without. Although allied health care personnel cannot be expected to substitute for medical practitioners, their services can be used to complement needed medical services.

The Primary Care Reform Physician Advisory Group is convinced that this approach will not only improve overall quality of care but will also promote efficient delivery of primary care services across a broad spectrum of providers, and in the longer term, overall costs should be reduced.

RFPs should receive funds for coordinating social service program, such as home care. In addition, alternative providers working in an office setting may pay to the RFP a facility fee determined by application by government and paid for by government.

Primary care services are provided by specialty groups other than family practice. The Reformed Fee-For-Service Model must guard against creating an incentive to RFPs to refer patients to specialists for consultation who, on follow up, may take over a large portion of primary care that should be undertaken by the RFP. This can be accomplished by making family practitioners more responsible for the coordination of medical services.

The Primary Care Reform Physician Advisory Group recommends that patients be required to receive a referral from their RFP physician before seeing a specialist for consultation and that referral be documented. This would mean that specialists who receive a referral would need to refer patients back to the RFP physician after completion of the consultation or once a treatment regimen has been established before a new unrelated treatment program could take place.

This differs from the traditional "gatekeeper" role which, once the gate to secondary care is opened, gives little responsibility to the family practitioner for the coordination of specialist care. The Primary Care Reform Physician Advisory Group is concerned that this coordination role not become overly bureaucratic, and that it not be an impediment to timely access to secondary and tertiary care.

For a physician to claim a consultation fee, he or she must submit his or her findings, opinions and recommendations in writing to the referring physician. The Group believes strongly, that mechanisms must be in place to ensure that the RFP receives all consultants reports pertaining to its rostered patients.

5.8 Telephone Triage

Given that many services presented to family practitioners are self-limiting, information can play an important role in reducing physician encounters. The Primary Care Reform Physician Advisory Group recommends that government introduce a province-wide toll-free telephone line to provide advice to patients to allow them to make informed decisions about the need to see their family physician. It may also reduce unnecessary use of visits to a walk-in clinic, from a housecall service, or to an hospital emergency department.

Alternative care providers would be available to give advice on medical problems and the proper use of the health care system in particular circumstances through the toll-free telephone service. The key elements of the system would consist of the following:

1. Pre-recorded Telephone Information

• A series of pre-recorded messages with up to date information on current medical topics and crises would serve as the initial point of contact with patients. An example of this information would include recent information on meningitis epidemics, recent medical information on cholesterol/PSA testing, etc. The patient would access the relevant information before moving to the next stage.

2. Alternative Health Provider as the Second Line of Contact

• A nurse would be available to answer the calls if patients wish to consult a health care professional about their problem. The Group believes that this professional should have a well thought out medical algorithm for what appear to be routine problems. The nurse would likely be able to answer 30-40 percent of the incoming calls without referring them on to the next stage. Dangerous/suspicious symptoms would be referred onto the physician on the system.

Telephone Triage Reduces Costly Emergency Room Visits

With hospital restructuring evident for appropriate use of emergency rooms in a hospital setting has become a critical issue.

The Hospital For Sick Children, in preparation for the traditional overuse of its emergency facilities during the Christmas rush, reacted proactively in 1995 by sending letters asking primary care physicians to provide alternative coverage for their patients while they were on vacation to avoid unnecessary visits to the emergency department.

Some studies suggest that up to 50% of visits to emergency rooms are unnecessary. According to Jeremy Noble of the Harvard Medical School Department of Health Policy and Management, $17 billion is wasted annually in the United States. The unnecessary visits contribute to delays and the duplication of service provision that is obviously not cost-effective.

One potential solution which is attracting growing attention is that of a 1-800 telephone access or telephone triage that patients may call prior to leaving for the hospital or after-hours clinic. Contacting this information service would not be a requirement prior to visiting an emergency department, but rather an additional service. On most occasions there would be telephone menus listed offering a variety of recorded messages which would serve to inform patients about how to handle less serious problems, such as colds, or to advise on the correct medication dosage for fever. After consulting with a Canadian agency that operates a 1-900 telephone line, it is apparent that at the present time it is not feasible to recommend the implementation of a 1-800 service to provide either recorded medical information or on-site physician advice or diagnosis. A number of issues must first be considered, for example, ‘live' physician telephone advice is extremely costly. The potential exists for considerable physician liability when diagnosis and subsequent advice is given over the telephone. Secondly, if the patient perceives the situation to be an emergency requiring the attention of a physician, it is not the intent of this telephone access line to impede appropriate consultation when needed.

Alternative care providers on the other hand may be qualified to teach patients and provide basic information in response to the most commonly-asked questions.

Telephone triage has been addressed by a number of companies in the US including Access Health Inc., a Sacramento, California-based company. The company employs 172 Registered Nurses who, in recognition of the potential for malpractice claims, are under strict instructions not to diagnose or suggest specific treatments. To date, no lawsuit has been filed against the company. The Physician Advisory Group suggests that a 1-800 information line should be explored in partnership with the OMA, government and the private sector to provide basic information in an effort to reduce costly emergency visits and enhance quality of care.

The savings of such a program, if properly run and with the necessary safeguards in place, would more than offset the associated operating costs. The telephone triage system would provide a valuable adjunct to the 24-hour response responsibilities of RFPs.

5.9 Managed Care

The Physician Advisory Group believes that family practitioners should play a greater role in the management of health care services. Providing financial incentives to physicians for more efficient use of health care resources--managed care options--can do just that. The Group proposes that RFPs enter into managed care of Ontario Drug Benefit Program (ODB) on a pilot basis. Expenditures below a prescribed target would be pooled, but because the number of rostered patients of individual RFPs affected by ODB may be relatively small, those savings should be pooled across the entire family practice population. Savings would be shared equally between family practitioners and government, with family practitioners receiving theirs on a per capita basis depending upon rostered population aged 65 and older (and covered dependents).

In order to avoid the potential to compromise quality of care, acceptable guidelines for pharmacotherapy must be pursued. As with referrals, this option would require monitoring; in this case to ensure that the guidelines are being followed.

5.10 Schedule Reform

Significant changes will be required to accommodate the change in practice that will be occasioned by the introduction of the Reformed Fee-For-Service Model. The Ontario Medical Association has invested heavily, since 1991, to have the Committee on Economics, in consultation with the clinical sections, do a systematic review of fee relativity in the Schedule of Fees. The intent is to produce a resource-based relative value schedule (RVS), and hopes to complete that task in the near future. The Primary Care Reform Physician Advisory Group fully supports this endeavour, and hopes that the work of the Committee on Economics can be expedited. Once this work is complete, apart from introducing other necessary changes as mentioned in the previous paragraph, the Physician Advisory Group expects that the earnings BMTs of RFPs will be adjusted accordingly.

6.0 Composite Funding Model

Under composite funding, physicians receive a combination of fee-for-service and non-fee-for-service earnings. The blended funding proposal of the College of Family Practice of Canada (CFPC), and the recommendation by Birch for blended funding, have garnered significant support from family practitioners in Ontario. The attraction of the approach is having the security of a base payment, while rewarding physicians through fee-for-service payment for either higher volumes of service to patients, or for providing specific medical services to their patients.

The Primary Care Reform Physician Advisory Group looked closely at the CFPC model, and attached numbers to each of the elements contained therein. The OMA model differs from the CFPC model in that the base salary is not fixed across family practitioners. The base salary must in some way be linked to medical service volume, so as not to over-compensate part-time practitioners, or under-compensate physicians who have large practices.

The Group's Composite Funding Model (CFM), which is suggested for consideration as a possible option for some physicians, splits the family practice funding pool equally into capitated and fee-for-service payment.

• As with the Reformed Fee-For-Service Model described previously, the CFM would include patient rostering with patient commitment provisions.
• The base salary plus benefits would equal the number of patients rostered in the practice multiplied by a capitation rate (after adjusting for age and sex of the patient), that equals one-half of fee-for-service payments to family practitioners divided by the patient population.

CFPC Blended Funding Model

As mentioned earlier, the Composite Funding Model (CFM) differs from the CFPC Blended-Funding Model with respect to base earnings. Under the CFPC model, base earnings are a fixed salary and in our analysis we assume that physicians whose fee-for-service earnings are less than that amount do not qualify for blended-funding and are left on straight fee-for-service. With CFM, base earnings are available to all physicians, but are dependent upon roster size. The remaining elements need not differ except in quantum.

The following presents an analysis of the CFPC model which, without loss of generality and recognizing the differences in the derivation of base earnings, can be extended to the CFM.

The model incorporates a combination of volume-driven payment systems and incentives aimed at providing high-quality care. This is done through the inclusion of four basic elements: a base payment, an occupational costs portion, non-volume modifiers, and volume modifiers.

The base payment is for a minimum of 35 hours office time and a minimum of resource-based relative value units (RBRV). It includes allowances for holidays, CME, pension contributions and personal comprehensive insurance costs. Occupational costs are meant to cover overhead costs associated with running a practice. Non-volume modifiers are meant to act as incentives in the form of: isolation allowances, obstetrics, maintenance of CFPC certification, teaching, on-call, nursing home/residential care, hospital practice, hospital committees, maintenance of special skills and other miscellaneous items. These would be monetary bonuses except in the case of on-call where they would be in RBRV units. Categories of RBRV units would be established and a payment per unit established for each category. Physicians would be paid according to the category they achieve.

Salary

A "salary" of $163,000 (excluding any volume or non-volume modifiers) per GP in Ontario would be optimal. This would represent an average 35 hour work week comprised of a set number of relative value units. This salary was derived by first determining the mean income of GPs, excluding those earning less than $30,000. The mean income was $180,000 and the mean number of hours worked was 39 hours per week with 4 weeks vacation per year. Adjusting this mean income to reflect a 35 hour work week resulted in a salary of $163,000.

This salary can then be broken down into its component parts: (a) base; (b) occupational costs; (c) non-volume modifiers; and (d) volume modifiers.

These four components could be weighted at different levels depending on the desires of those involved, and could be negotiated with government in a fashion similar to what is in place now for a fee-for-service schedule.

Base

The base is to be for a minimum of 35 hours office time plus a minimum number of resource-based relative value units (RBRV). This would include:

• Holidays: statutory holidays and paid vacation: initially 4 weeks; 5 weeks after five years and 6 weeks after 10 years.

• CME: 2 weeks paid CME leave for approved courses; $1,000/week allowance for travel and expenses, $250 allowance for journals, tapes, etc., and $250 allowance for texts, computer library, etc., as well as a programme allowing accumulation of leave for CME where these weeks would be deducted from paid vacation leave as per a set schedule.

• Pension

• Insurance: life, disability, extended health and drug, medical/dental and vision.

Occupational costs would be comprised of: rental space, equipment, legal/audit and staffing (nursing and ancillary) costs.

Non-volume Modifiers

Non-volume modifiers are meant to act as an incentive to encourage/reward location choice, the undertaking of "extracurricular activities," or the maintenance/improvement of skills. When any of the conditions are satisfied, the incentives would be distributed in the form of monetary bonuses, bonus RBRV units, or extra vacation/CME leave time. The modifiers are intended to specifically encourage: location in isolated areas, obstetrics, maintenance of certification (CCFP), teaching, on-call, miscellaneous special needs and services, nursing home/residential care, hospital practice, hospital committees or maintenance of special skills.

Volume Modifiers

Payment is based on a Resource Based Relative Value (RBRV) fee schedule, but RBRV units are not assigned a monetary value. Instead of having a direct link between service and fee amounts, doctors would be rewarded increasing amounts depending on which RBRV category they achieved. As volume increased beyond what is considered possible for quality care, the incremental increase from category to category would decrease or alternatively no increase would be given above a certain number of RBRV units.

For example:

0-1,000 RBRV units:	Category I	payment is $X
1,000 - 2,000 RBRV units:	Category II	payment is $Y

It should be noted that some provinces are without an RBRV schedule, yet the current fee schedules should have an inherent relativity developed over time. Thus, a numerical value could be placed on individual fees instead of a dollar value.

Analysis

The following is an analysis of how such a model may affect the current income distribution of GPs in Ontario. The analysis is limited for several reasons: first, Ontario does not currently have a schedule of RBRV units. The second reason is due to data constraints. Rather than having an aggregate data set, income data for physicians is limited to a summarized income distribution according to income categories.

This leads to the third limitation involving non-volume modifiers: given the lack of a direct link between the current income data and physician practice characteristics, it cannot be determined which physicians would benefit from such a modifier. The following analysis is thus approximated on income categories for general practitioners, and uses dollar values for volume modifiers. The non-volume modifier is assumed to be equal to one.

Three scenarios were created to examine the effect of composite funding on the existing income distribution. These simulations were produced with "salaries" ranging from $120,000 to $180,000, and various volume modifiers applied. Any discount schemes were only applied to that portion of fee-for-service over and above the work required to satisfy the salary portion requirements. Each simulation was then run to fulfil the constraint that total payments to General Practitioners equal $1.62 billion (total payments to G.Ps. in the 1993/94 fiscal year).

Simulations

Income distribution graphs and cumulative income distribution graphs are provided in Appendix 1. These graphs highlight those areas where a simulation will cause a deviation from the current distribution. The current cumulative income distribution of general practitioners is also presented.

7.0 Extensions to the Models

As part of its examination of capitation based models, such as the Composite Funding Model, the Primary Care Reform Physician Advisory Group, in parternship with the British Columbia Medical Association and the Alberta Medical Association, commissioned a study by Dr. Dale Rublee entitled "European Models of Physician Capitation." The complete paper is provided in Appendix A.

The paper provides an interesting background to the workings of European capitation systems, and concludes that capitated payment alone tends to be unresponsive to health policy initiatives and provides little in the way of useful planning information. Combined models of payment, such as the Composite Funding Model, in which physicians receive a base payment along with incentives or fee-for-service for select services, show some promise. The Rublee paper highlights the difficulties of the Composite Funding Model that should be addressed before considering the implementation of a capitation based model.

The Primary Care Reform Physician Advisory Group also paid close attention to the experience of the United States with managed care, and the results of the United Kingdom experimentation with GP Fundholding. Although the Group has restricted its recommendations to a managed care structure involving only ODB, it recognizes that the potential is far greater.

• The use of diagnostic testing within a family practice exemplifies the potential for a "managed care" structure. The cost of diagnostic services referred by family practitioners amounted to $759 million in 1993/94, or an average of $72,032 per family practitioner. [ Note: these services are not performed by family practitioners, and a fee is not received by them.]

The Primary Care Reform Physician Advisory Group considered a number of different models, and from them have offered two; (1) the reformed fee-for-service model, and (2) the composite funding model. The reformed fee-for-service model has the most promise in terms of the efficiency and quality it brings to family practice within the existing political and economic environment. The composite funding model may be an option for some physicians in the future.

8.0 Topics for Further Discussion No Recommendations at Present

In this section, two additional issues for discussion are identified: (1) health utilization as a taxable benefit and (2) health care delivery models. The Physician Advisory Group has no specific recommendations with respect to these issues; however, they warrant further discussion and consideration. The options require public debate.

8.1 Health Utilization as a Taxable Benefit: Sickness Tax or Wellness Rebate?

Payment methods of the type the Group has proposed expose physicians to the potentially limitless demands of an increasingly informed and assertive population of patients. While demands on the publicly financed "core" medically necessary services may be alleviated through the adoption of direct charges, a vast literature suggests that such fees on utilization are administratively burdensome, are inequitable, and primarily reduce utilization for those of limited financial means. In contrast, there appears to be a growing realization that the income tax system might be used to enhance the provision of comprehensive, cost-effective and appropriate health care services.

Specifically, a synergy of incentives, which enhances the cost-effective provision of publicly financed health care services, may be achieved by reimbursing primary care physicians through the reformed fee-for-service (or composite funding) models and by designating the utilization of publicly financed health care services a taxable benefit.

With this approach, health service utilization would be a taxable benefit, and depending on the utilization and income thresholds employed, would result in differential income tax rebates that vary with utilization. Individuals with relatively low age-sex adjusted utilization would receive greater rebates than those with relatively high utilization.

Use of an income tax rebate scheme to encourage appropriate utilization of publicly financed health care services has several advantages to the use of direct charges. First, it builds on Revenue Canada's current revenue collection infrastructure and is therefore administratively less burdensome than direct charges. Second, by using the progressive income tax system, the scheme is more equitable that the use of direct charges that are regressive. Finally, as Zweifel has shown, health utilization was reduced when German private health insurers offered consumers premium rebates for "low-utilization". Thus, by making publicly financed health utilization a taxable benefit, utilization and the overall cost of such services will be expected to decline.

Given the current capability of provincial health service utilization and cost information systems, it is possible to compute, at the level of each patient/taxpayer, the cost of publicly financed health care services acquired over a given period. This cost may be considered as a taxable benefit and used to implement the income tax rebate scheme. This fiscal policy complements the payment models recommended by the Group for primary care physicians, is congruent with the provision of comprehensive and cost-effective health care services, and provides useful information to patients on their patterns and cost of health service utilization. However, the more encounters an individual has with the system, the lower the tax rebate: this is a benefit to those who remain well. It is of note that health status is inversely proportional to income: the poor are sicker.

8.2 Health Care Delivery Models

Modifications to the method by which primary care physicians are reimbursed for the provision of health care services will significantly affect the manner by which health care is organized and delivered. Indeed, the adoption of the committee's recommended models encourages the vertical and horizontal integration of health care organizations so that the financial risks to primary care physicians from service provision are minimized. Since this form of organizational arrangement enhances opportunities to monitor the utilization of health care services, it will strengthen the continuity of care and service co-ordination. Such health system restructuring may yield improvements in quality service provision, health care costs, and patient satisfaction.

To the extent to which integrated health care organizations are able to take advantages of economies of scale and scope in the provision of specialized health services, such organizations may provide these specialized services themselves.

In-house service provision enables primary care physicians with opportunities to contract with and monitor alternative health care providers with skills, thereby enhancing team work and task delegation, with patients serviced by the most appropriate quality provider. Furthermore, as the number of roster patients grow, primary care physicians may be confronted with opportunities to engage in selective (or preferred) contracting with specialized suppliers. Such contractual arrangements may enhance health system competition, and moreover, yield price and outcome signals for both performance appraisal and resource allocation. Furthermore, since roster patients may switch between RFP's, selective contracting is expected to ultimately be responsive to patient preferences for quality services.

GP Fundholding

The recommended payment models accentuate the "gatekeeper" role for primary care organizations (Rostered Family Practices). This payment scheme provides such organizations with opportunities to pursue contractual arrangements with alternative health care providers either through in-house service provision or through selective contracting. Mechanisms should be designed to facilitate the development of these contractual arrangements as the public interest would be served through increased health system competition and the resulting provision of price and outcome signals. Such signals may be used for both performance appraisal and resource allocation, thereby enhancing the provision of comprehensive and cost-effective services that are responsive to patient choice.

Conclusions

The Primary Care Reform Physician Advisory Group has considered a number of different physician reimbursement approaches and has recommended two additional models. The first, a fee-for-service model, attempts to add refinements to the current system to better meet system objectives. The second consists of a composite funding method (i.e., a combination of capitation and fee-for-service payment). The Group believes that the reformed fee-for-serviced model has the most promise in terms of the efficiency and quality they bring to family practice within the existing political and economic environment. The composite funding model may be an option for some physicians.

The committee advocates physician choice in the method of payment for primary care services and the continuation of a reformed fee-for-service model, but also supports the composite payment scheme for primary care as an option to Gps/FPs. These recommended payment methods provide for patient choice, allows public funds to "follow patients", offers incentives for task delegation, and enhances health system competition, and contains the financial liability for government. Indeed, such payment models, by emphasizing patient choice, and the associated competition for patients, motivates primary care physicians to be assertive in their role as agents for patients in the provision of quality health care services.

The health system in all countries is undergoing change. The fiscal and political reality raise doubt about whether governments can afford to finance the health care system in a manner to which it has become accustomed. These pressures are the catalyst to reform the health system. The proposals address concerns over the provision of comprehensive, cost-effective and appropriate health care services that would be responsive to patient choice. The Physician Advisory Group recommends maintaining physician choice over the method of reimbursement and also recommends that patients be granted choice in their opportunity to select physicians.

The recommended payment models provide physicians (as agents for patients) with opportunities to negotiate service contracts with cost-effective health providers. Such arrangements enhance health system competition, and moreover, yield price and outcome signals for both performance appraisal and resource allocation. Furthermore, private funding for health care, either through direct payments or health insurance, enhances patient choice when health care services are rationed. These payment models of public/private finance provides for client choice, it enhances competition between providers, and it allows governments to contain their share of health expenditures.

The recommendations proposed in this report represent a comprehensive framework for primary care reform, instead of previous reports that have focused on only one or two important issues. Rather than competing with previous proposals, the Physician Advisory Group has endeavoured to build on the work of others.

There is a recognized need for primary health care reform. While the proposals presented by the OMA Primary Care Reform Physician Advisory Group may not satisfy all interest groups, the committee has endeavoured to build consensus.

Return to Primary Care Main Menu

Referrals

(...resource allocation)Supplementary services are those that do not generally, in and of themselves, improve the service that was provided under the public-insurance scheme, but which may be viewed as providing, overall, a superior service. An example would be providing coverage for service beyond a prescribed limit. Whereas, complementary services are those that complement or "complete" total care. Drug insurance and dental insurance are examples. Generally, there are barriers to private companies providing supplementary health insurance while there is no prohibition on complementary medical coverage.
Return to reference point

(...by the BCMA)Capitation: a wolf in sheep's clothing?, BCMA, November 1995.
Return to reference point

(The Language of Health Care)CMA News, November 1995, p.6-7.
Return to reference point

(...number of models) Hutchison B, Abelson J, Models of primary health care delivery. Building excellence through planned diversity and continuous evaluation, Prepared for The Task Force on the Funding and Delivery of Medical Care in Ontario, November 10, 1995.
Return to reference point

(...capitation payments)Capitation: a wolf in sheep's clothing?, BCMA Project Group on Capitation Policy, Council on Health Economics and Policy Analysis, November 1995.
Return to reference point

(Reformed Fee-For-Service Model)RFP defined as solo providers or functionally grouped providers.
Return to reference point

(...primarily the US)The Globe and Mail, December 1995.
Return to reference point

(...increased workloads)The productivity of Canadian physicians is recognized internationally.
Return to reference point

(...six weeks)Payment models that encompass patient rostering (i.e., capitation in Europe) usually provide for a longer period of patient "commitment" - 6 or 12 months. However, if patients have the opportunity to change physicians over a relatively shorter interval of time as we suggest, then it may result in a greater acceptance of the concept by patients. The Decima survey revealed that, in large part, patient disapproval of rostering stemmed from the accompanying constraints on their choice or freedom to switch doctors.
Return to reference point

(...consumers)Richmond C, Triviality and frequency of night calls irk British Gps, Can Med Assoc J, Jan 1, 1996; 154(1).
Return to reference point

(...referral be documented)This requirement would be waived in situations requiring emergent/urgent care.
Return to reference point

(...against the company)Wall Street Journal, Tuesday October 4, 1995, B9.
Return to reference point

(Such contractual arrangements)The range of terms and conditions under which selective contracting may be pursued are almost limitless. While individual provinces, professional associations or institutional associations may establish recommended fee guidelines, market determined prices might be the preferred solution. Of course, these contractual arrangements would be subject to regulatory review to ensure competition and to prohibit predatory pricing.
Return to reference point