Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Part ITI Managing Care in the United Kingdom and Canada
F riority S e1d n g in a Tic e dab use d S y stc m Alan Williams The British beaMb care system is of prickly interest because ~ teas, Tom Us inception iD 1948, sought to reconcUe widespread access to beaRb cage wi1b ~ tight, centraHy controlled budget Before considering bow weH or iD ~ teas performed in that respect ceded key Azures need lo be home . · . ID mlDO: 1. The N~iona1 Heabb Service (NBS) is commuted to providing care according to need, Dot according to wilUngness or acuity to pay. It Cows within ~ strong, but iU-deOned, eg~it~ian ideology. 2. The NBS provides Tout go percent of aH beaMb cage far Bdtisb citizens; the phv~e sector concentrates mainly on providing elective sur- gery for those who wish to avoid NBS mating lists and for those Boo Tab to improve on 1be general level of hospital amenides. Pdv~e treatment is provided in large paw by doctors Boo spend most of tube time in the OHS. 3. The NHS teas a centraHy determined, 1~-ADanced, heed annual budget, ~bicb is disidbuted geogr~bicaMy 10 regional (=d distinct) beabb ~tbobties. Dis~ibudon Chows ~ formula based mosUy on (weigbted) population but to some extent also on geogr~bic visions in depdv~ion and mo~idity/monabty rues. These beabb ~tbobties Be responsible far the provision of ~1 the beabb cage in tube Lea and for coor~naUng the provision of do~cHi~y support services limb local ~tbobties (see below).
so ALAN WILLIAMS 4. The total NHS budget is approximately 6 percent of the gross na- tional product (GNP), of which approximately 60 percent is spent on hospi- tal services. Local authorities, who are formally outside this centrally tax- financed system, provide many health-related services for the domiciliary support of ill and disabled people (services that are very important in the care of the elderly and certain other vulnerable groups). 5. Apart from accidents and emergencies, the patient's first point of contact is always with the general practitioner (GP), who acts as the "gate- keeper" for the system. Every patient is registered with a GP (in principle if not in practice). Under the NHS system, there are small charges for primary care prescriptions, somewhat larger charges for dental and oph- thalmic services, and no charges for hospital care. Thus, billing is totally absent, which significantly reduces both the system's administrative over- head and its knowledge of how much things cost. NHS hospital doctors are salaried employees, and GPs are independent contractors. most of whoop. _ _ 1 _ ~ _ ~ . . ~ · . ~ _ . salaries come from capitation fees paid for the patients registered with them. Fee-for-service charges are rare; they are used only when it is desirable to encourage specific activities (such as home visits or night calls). 6. Political support for this system is strong among the public and among health care providers. The system, however, is widely believed to be underfunded and forced to make painful decisions that it is thought could be avoided if funding could be increased. PROBLEMS OF THE BRITISH HEALTH CARE SYSTEM It will be obvious from the above that the British health care system has no problem with cost containment, if by that is meant controlling total costs. Instead, two other issues predominate: (1) is the right amount being spent on health care? and (2) is good value being received for the money that is spent? The present state of knowledge about the marginal benefits of health care spending precludes an accurate answer to the first question. The second question is more interesting, and most of this paper will be devoted to addressing it and exploring its ramifications. The broad response of most observers to this question is that surprisingly good results are achieved, considering how little is put into the system. Others might argue, this author among them, that the results are good largely because so little is put in. Before considering the way the system works, some general observa- tions are in order about optimization subject to constraints (i.e., about the discipline of economics). The resources devoted to health care must be diverted from other valuable uses; therefore, it is not rational to extend health care provision to the point where all the good that could be done is being done. The system should stop providing health care when the extra health benefit to be gained is of approximately the same value as whatever has been given up to provide the extra resources for health care. This
PRIORITY SETTING IN A NEEDS-BASED SYSTEM 81 principle demands that health care alternatives be ranked in order of cost- effectiveness; in a system with a fixed budget, the choosing of which ser- vices to fund proceeds by starting at the top of the list and working down it until the point in this priority ordering at which the money runs out. Once the system's budget level is set, the optimization problem is one of estab- lishing (and implementing) priorities in cost-effectiveness terms. It is here that this paper's first background point becomes crucial. In a private health care system that seeks profitability (or at least to break even), optimization requires that alternatives be subjected to financial appraisal- that is, to systematic analysis of the revenue and expenditure implications of each alternative. The precise pattern of the rationing that ensues will be the result of (somebody's) willingness and ability to meet (at least) the estimated expenditures. Here, the "somebody" may be an individual, an insurer, a charity or foundation, or a public agency, and they may or may not be in a position to judge the likely effectiveness of treatment when deciding how much to pay. In fact, the multiplicity of actors in this system is likely to lead to great variability in such judgments. But in a public system that has explicitly rejected willingness and ability to pay as a ration- ing device, and that has its revenue fixed according to the population for which it is responsible (rather than according to what it does for them), optimization needs to be informed by economic appraisal. That is, it should be guided by systematic analysis of the benefit and cost implications of alternatives; rationing thus will be guided by (somebody's) assessment of a person's ability to benefit (i.e., by "need" for treatment) in relation to the costs (which will be borne by the citizenry at large). In practice, these economic appraisals have unfortunately limited their consideration of cost implications to the implications for the NHS budget (because that is the constraint uppermost in people's minds), and in practice, the "somebody" who assesses need, at both the clinical and community levels, has for the most part been a doctor. To do this job properly, doctors should know not only about the effectiveness (in outcome terms) of the treatments at their disposal, and how their patients value these various outcomes, but also about the benefits of all the other treatments the system provides for other patients, and how benefits to one patient are to be weighed against benefits to another. Needless to say, no one has this information, but doctors be- lieve that because of their clinical experience they are best placed to fill the void. Consequently, it has been their priorities that have driven the system (with little regard to costs). Concern about costs has been left to the man- agers in the health authorities or in the hospitals, who, because they had little or no influence over (or even knowledge about) the priorities that were being applied by doctors in the use of the system's capacity, found themselves having to make rather arbitrary decisions about increasing or decreasing that capacity, in order to balance their predetermined (fixed) budgets.
82 ALAN WILLIAMS The gatekeeper role mentioned in point number 5 at the beginning of this chapter requires the GP to be the initial "needs-assessor" (once the patient has decided that he or she "needs" to see a doctor, i.e., that he or she might benefit from such an encounter). If the GP refers the patient to a hospital doctor, a second needs assessment takes place, and at this point an important conflict of interest manifests itself. Can the doctor simultaneous- ly act as the patient's agent and the system's agent? Or, to put the point differently, should the doctor be expected to mediate the conflict between the interests of the citizen-as-taxpayer (in keeping costs down) and the interests of the citizen-as-patient (in getting the best possible "free" care)? Can he or she do so evenhandedly, when the citizen-as-patient is sitting there and the citizen-as-taxpayer is vicariously represented only by a busy, somewhat remote manager who is rather detached from the pressures of clinical work and probably located in a different building? Although the majority of British doctors seem comfortable with this role most of the time, every now and again a protest is heard; occasionally, a doctor asserts that it is his role to do everything he can for the patient in front of him, no matter what the cost, and that that is what he proposes to do, because for him to do anything else would be unethical. A brief excursion into medical ethics seems called for here. Put suc- cinctly, medical ethics require a doctor to do no harm, to preserve life, to alleviate suffering, to respect the autonomy of the patient, to tell the truth, and to deal fairly with patients. It is accepted that these principles often conflict and that one of the skills required of doctors is to exercise appropri- ate judgment as to where to strike a compromise between them in any particular clinical situation. The injunction to deal fairly with patients differs from the others in that it requires a comparison to be made between what is done for one person and what is done for another. In a regime of fixed budgets, to offer a treatment to one person, regardless of the cost of that treatment, implies that no account is to be paid to the inescapable consequence that some other patients will be denied treatment, and that those sacrifices by others will be commensurate with the costs that are being disregarded. This approach is not consistent with "dealing fairly" with patients, which, at the very least, requires that the needs of other potential patients be weighed in the balance of treatment decisions. Thus, the argument that it is unethical for doctors to play the gatekeeper role is untenable (which is not to deny that the role might actually be played in an unethical way by some doctors). Let us return to the main theme, however, which is the setting of prior- ities in a needs-driven system. As will be evident from the foregoing, need is defined as the capacity to benefit (people cannot need something that is of no benefit to them). But the capacity to benefit from any health care activity varies (among treatments, among patients, and even among practi ~ _1_. 1 rat, r
PRIORITY SE1TINGINA NEEDS-BASED SYSTEM 83 tioners) and might be so small in relation to its costs as not to be worth providing. In other words, needs must be prioritized, and some will, inevi- tably and quite properly, remain unmet. Thus, if the people on waiting lists are those whose capacity to benefit from treatment is smaller than those who are not kept waiting, this constitutes evidence that the system is work- ing rationally (not that it has failed nor that it needs more money, for even with more money there would still be marginal cases on waiting lists). A more legitimate cause for concern would arise if those on waiting lists would benefit more from early treatment than many of the people who were not being forced to wait. The general point remains, however: the mere existence of waiting lists, or their size, or even the length of the wait, is not important. What is important is the size of the benefits that are being denied to those who are waiting, compared with what other people are receiving that is, a comparison of needs. It is necessary to determine how benefits are to be measured in a need- driven system. The first factor to keep in mind is that what people want is not treatment per se but improved health. Improved health means the ex- tension of one's life expectancy or a better quality of life, or both. Because each is valued, people may be willing to trade off one against the other. Health-related quality of life is judged by examining such characteristics as mobility, self-care, pain, distress, and ability to perform normal, accus- tomed social (including work) roles. Measuring the benefits of treatment requires estimates of the effect of treatment on these various characteristics of health, which, in principle, are encapsulated in the concept of the quali- ty-adjusted life year (or QALY). Yet disappointingly little is known in these terms about the health benefits to be derived from most treatment for most people. So at the very heart of the priority-setting process is a great information void, which renders the system vulnerable to priority setting by less appropriate means. The recent discovery and promotion of outcomes research is a belated response to this information void. THE "REFORMED" BRITISH EIEALTH CARE SYSTEM On All Fools Day 1991, the reforms of the NHS that had been instituted by the Thatcher government went into effect. Their principal feature is the separation of the demand side of the market for health care from the supply side. On the supply side (the providers) are the hospitals, the primary care doctors, and the community services. On the demand side (the purchasers) are the health authorities and those GPs who have been permitted to become budget-holders.) There are some obvious anomalies here, such as the am iThis means that certain large group practices will receive part of the hospital budget, which allows their GPs to purchase certain hospital services on behalf of their patients.
84 ALAN WILLIAMS biguous role of primary care doctors and the designation of community services as providers when they are not really part of the system at all but are run by local authorities. (Despite recommendations that the government received to expand and strengthen the community care system, Mrs. Thatch- er prevaricated in regard to this because she could not agree to any strength- ening of the role of local government in Britain.) There are also some less obvious anomalies- for example, certain hospitals (and a few other provid- ers) enjoy a special status as self-governing trusts, which exempts them from some of the restrictions (e.g., on pay and conditions of staff, access to capital, etc.) placed on hospitals that remain within the main NHS system. Generally, however, the idea is that providers will compete with each other for contracts drawn up by the purchasers, who will choose the provider that offers the best deal. Each purchaser's task is to assess the need of the particular population for which it is responsible, set performance criteria for potential providers (which should relate to quality as well as quantity), and evaluate the bids against policy objectives and budget constraints. Certain services (e.g., accident and emergency) must be provided locally, but others (especially elective surgery) might be "bought in" from hospitals outside the geograph- ic boundaries of the purchasing health authority. Clearly, this kind of competition is more viable in large urban areas than in more sparsely populated areas, and it depends on the providers' not forming informal cartels. More crucially, however, it depends on the ability of purchasers to specify the content of contracts in the manner required, which leads one to the information void mentioned earlier. If need means capacity to benefit, purchasers must know what is beneficial and what is not, and just how beneficial the beneficial things are in relation to the costs. They must also monitor performance in outcome (rather than process) terms, a task of no small difficulty, given a situation in which even medical audits directed at technical competence have been voluntary, nonthreatening edu- cational activities pursued only by enthusiasts who set their own local stan- dards and operate in a rather informal (and highly secretive) manner. Even now the official position is that audits are to be strictly professional affairs in which managers are not to be told anything that might enable them to identify individuals who are performing poorly. It is likely to take at least a decade to generate a data base that will enable any of these changes to operate in a manner that is even passably systematic, and for the first few years, even with determination and goodwill, the reforms are unlikely to move beyond the level of ritual. In the new system, the rationing or priority-setting role is vested firmly in the purchasers, but to make their chosen priorities effective they will have to specify the case-mix of each hospital specialty and check on patient selection criteria and quality of performance. It is not clear quite how this
PRIORITY SE~ITING IN A NEEDS-BASED SYSTEM 85 is going to happen if the rhetoric surrounding the protection of clinical freedom and the voluntary and confidential nature of medical audits is to be taken seriously. Making doctors budget-holders may force them to think in terms of priorities and cost-effectiveness, but it will not necessarily lead them to adopt the particular priorities, or notions of effectiveness, held by the health authorities. IMPLICATIONS FOR TECHNOLOGICAL ADVANCE IN MEDICINE In principle, it is clear that what makes a technological innovation a technological advance is that it proves to be beneficial in cost-per-QALY terms. To determine whether an innovation can make such a leap in status, it should be introduced initially in a controlled manner in which it is used only in the context of a carefully designed evaluative trial that is conducted by independent researchers, with full disclosure of data, and financed out- side the main therapeutic budget of the health service. Only when such a test has been passed should the innovation move into the armamentarium of practicing clinicians, and possibly even then only with restrictions on its use. But it is necessary here to distinguish between those innovations that require the use of (expensive) specialized resources, be they drugs, devices, or people, and those that can easily be accommodated within the existing resources for everyday clinical activities. In the latter case, there are few effective sanctions that can be applied at a managerial level to enforce independent evaluation, and the only early indicator that something differ- ent is happening may be a change in patient selection, as an interested clinician seeks more cases on whom to practice the innovation. In the British system, this is most likely to be noticed if the waiting lists for certain conditions lengthen because these cases have been squeezed out to make room for the innovative ones. Health authorities have learned, when new appointments are made, to monitor as closely as possible the likely implications of clinicians' research interests. Even if formal evaluation could be enforced prior to general dissemina- tion of an innovative practice, there are some obvious disadvantages, main- ly to do with delay. It may take quite a while to establish rigorously a clear gain to patients, and even longer to be sure that there are no adverse effects over the longer term. Given the variety and complexity of manufacturers' pricing policies, it may not be easy to extrapolate from the cost data in the trial to the costs that are likely to be faced when the innovation becomes more widely disseminated. Indeed, if it is known that a particular cost- effectiveness ratio has been established by purchasers as the cutoff point between what can be afforded and what cannot, manufacturers may well use
86 ALAN WILLIAMS the results of the trial to price the technology at a level that is as close to that ratio as possible, even though the true cost of producing it is much lower. An additional problem common to all trials is that the results ob- tained by well-disciplined practitioners observing punctiliously the protocol of a scientific trial may not be replicated when the technology is being used in the "real world" of clinical practice. Nevertheless, the system does have to move in the direction of earlier, more formal cost-effectiveness analysis of emerging new technologies. One sign that this has already been accepted is the current burgeoning interest of the drug companies in quality of life measurement and cost-effectiveness analysis. Inevitably, however, their interest is more strongly motivated by marketing considerations than by a desire to improve the efficiency of the health care system or to enhance the methodology of cost-effectiveness analysis, although both of those benefi- cial side effects may follow incidentally. The increasing pressure for earlier and more formal economic appraisal has come from dissatisfaction with the existing situation. Health authorities see themselves confronting collusive activity among manufacturers, practi- tioners, and patients, which prevents or delays systematic evaluation and leads to the weakly controlled dissemination of well-marketed but poorly evaluated new technologies. Manufacturers encourage and exploit the de- sire of practitioners to improve their performance by offering something that helps them resolve some diagnostic, monitoring, or therapeutic difficul- ty of which they are acutely aware, given the current state of the art of medical practice in their particular field. In addition to the desire to help their patients, practitioners are motivated to pursue innovative technologies by the various professional "perks" associated with them, such as subsi- dized attendance at professional conferences in sometimes delightful loca- tions and personal association with published research findings. The pa- tients themselves can usually be induced to go along with the latest "advance" (especially if it has also been promoted in the media as the latest technolog- ical wonder). NHS qualms about the costs of, say, radiological equipment (which nowadays may include qualms about the associated operating ex- penses as well as about capital costs) can be overcome by "giving" the capacity to the service during the experimental phase. Unless it proves to be positively harmful, the technology becomes incorporated into routine practice in the experimental sites, and possibly elsewhere. through the in fluential grapevine of the conference circuit. ~1 ~ALA 1~__1~C _ _,1 ~. 1 ~ . . . . By the time the stage is l~i"~ll~U Wll~ll tI1~ ~leal~n au~norlly itself nas to pICK Up the costs, it will have become politically quite difficult to discontinue it on cost-effectiveness grounds, especially if the media can show a handful of people who have benefited dramatically from the innovation. The constraint of its fixed budget has adc; the DliLlbll bySt~l no~onous~y slow, compared with most countries of equivalent wealth, in diffusing expensive new technologies throughout the system. Yet that very system also offers opportunities, by virtue of its ._^,l~ ~ 1~ D~:~l ~3 ~
PRIORITY SEATING IN A NEEDS-BASED SYSTEM 87 monopsonistic position within the United Kingdom, for guiding this diffu- sion in a purposeful way. Unfortunately, effective strategies by which to exploit these opportunities have yet to be developed. RATIONING HEALTH CARE: IS IT ETHICAL? This much-asked question is always puzzling because it is so oddly formulated. If by rationing is meant deciding who is to get something and who is to go without, then rationing, like death, is unavoidable. No one asks, is death ethical? Instead, practitioners do (and should) ask, when they have some influence over the situation, whether it is ethical to allow (or cause) someone to die in a particular manner at a particular time rather than in some other manner at some other time. So it should be with rationing. The appropriate question is whether it is ethical to impose a particular kind of rationing on a particular community at a particular time, as opposed to imposing some other kind of rationing on that community at that time. Before answering this question, it is necessary to examine the particular ideology that a community wishes to bring to bear on its health care system. Essentially two such ideologies are relevant in this context, the libertarian and the egalitarian. In the libertarian view, access to health care is part of the society's reward system, and as a general rule, people should be able to use their income and wealth to get more or better health care than their fellow citizens, should they so wish. In the egalitarian view, access to health care is every citizen's right (like access to the ballot box or to the courts of justice), and it ought not to be influenced by income or wealth. Each of these broad viewpoints is typically associated with a distinctive configuration of views on personal responsibility, social concern, freedom, and equality, as set out in Table 6-1. The ascendancy of either of these broad viewpoints would generate a distinctive health care system whose characteristics would be very different from those of a system shaped by the other viewpoint. A system shaped by the libertarian ideology would establish willingness and ability to pay as the determinants of access; this kind of access would be best accomplished through a market-oriented "private" system (provided such markets can be kept competitive). In an egalitarian system, equal opportunity of access for those in equal need would be the determining rule, and because this requires the establishment of a social hierarchy of need independent of who is pay- ing for the care, it would be best implemented through a system of public provision of care (provided the system can be kept responsive to social values and changing economic circumstances). Table 6-2 lists the essential characteristics of each kind of idealized system. Note that the success criterion to be applied to the egalitarian system is the level and distribution of health in the community.
88 ALAN WILLIAMS TABLE 6-1 Attitudes Typically Associated with the Libertarian and Egalitarian Viewpoints Issue Libertarian Viewpoint Egalitarian Viewpoint Personal responsibility Personal responsibility for achievement is very important, and this quality is weakened if people are offered unearned rewards. Moreover, such unearned rewards weaken the motivational force that ensures economic well-being; in so doing, they also undermine moral well-being, because of the intimate connection between moral well-being and personal efforts to achieve. Personal incentives to achieve are desirable, but economic failure is not equated with moral depravity or social worthlessness. Social concern Social Darwinism dictates a Private charitable action is not seemingly cruel indifference rejected but is seen as potentially to the fate of those who dangerous morally (because it is cannot "make the grade." A often demeaning to the recipient less extreme position is that and corrupting to the donor) and charity, expressed and usually inequitable. It seems effected preferably under preferable to establish social private auspices, is the mechanisms that create and proper vehicle; however, sustain self-sufficiency and that charity should be exercised are accessible according to precise under carefully prescribed rules concerning entitlement, conditions, for example, such which are applied equitably and that the potential recipient are explicitly sanctioned by must first mobilize his or her society at large. own resources and, when helped, must not be in as favorable a position as those who are self-supporting (the principle of "lesser eligibility"). Freedom Freedom is to be sought as a Freedom is seen as the presence supreme good in itself. of real opportunities of choice; Compulsion attenuates both although economic constraints are personal responsibility and less openly coercive than political individualistic and voluntary constraints, they are nonetheless expressions of social real and often constitute the concern. Centralized health effective limits on choice. continued
PRIORITY SETTING IN A NEEDS-BASED SYSTEM TABLE 6-1 Continued Issue Libertarian Viewpoint Egalitarian Viewpoint planning and a large governmental role in health lo- . care nnanclng are seen as an unwarranted abridgment of the freedom of patients as well as of health professionals, and private medicine is thereby viewed as a bulwark against totalitarianism. Equality Equality before the law is the key concept, with clear precedence being given to freedom over equality whenever the two conflict. 89 Freedom is not indivisible but may be sacrificed in one respect to obtain greater freedom in some other. Government is not an external threat to individuals in the society but is the means by which individuals achieve greater scope for action (that is, greater real freedom). The only moral justification for using personal achievement as the basis for distributing rewards is the availability of equal opportunities for such achievement for all individuals. Emphasis is then placed on equality of opportunity; where this cannot be ensured, the moral worth of achievement is undermined. Equality is seen as an extension to the many of the freedom actually enjoyed by only the few. SOURCE: Based on A. Williams, "Priority Setting in Public and Private Health Care: A Guide Through the Ideological Jungle," Journal of Health Economics 7:173-183, 1988. Needless to say, in practice neither system fully lives up to its ideals. Most of the problems stem from (1) the peculiar role of doctors in health care systems, (2) market deficiencies on the supply side, and (3) difficulties with information on the demand side. Table 6-3 catalogues the full extent of these problems. In most countries health care is provided through a mixture of systems that have no common ideology, which is probably a reflection of the plural- ized societies of those nations and their attempts to accommodate subgroups with incompatible systems of beliefs. A hypothesis suggested by this anal- ysis is that the structure of the health care system in each country is likely to be systematically related to the equity concerns that have been dominant in the recent past; the health care system is also likely to reflect the ideolo- gy that generated those concerns. An obvious instance is the balance be
9o ALAN WILLIAMS TABLE 6-2 Essential Characteristics of Idealized Health Care Systems Based on Libertarian Views (Private Systems) and Egalitarian Views (Public Systems) System Element Private Systems Public Systems Demand · Individuals are the best · When ill, individuals are judges of their own welfare. frequently imperfect judges of their own welfare. · Priorities are determined by · Priorities are determined people's own willingness and by social judgments about ability to pay. need. · The erratic and potentially · The erratic and potentially catastrophic nature of demand catastrophic nature of is mediated by private demand is made irrelevant by Insurance. the provision of free services. · Matters of equity are dealt · Since the distribution of with elsewhere (e.g., in the tax income and wealth is and social security systems). unlikely to be equitable in relation to the need for health care, the system must be insulated from its influence. Supply · Profit is the proper and most · Professional ethics and effective way to motivate dedication to public service suppliers to respond to the are the appropriate needs of demanders. motivation of suppliers, who should focus on success in curing or caring. · Priorities are determined by · Priorities are determined people's willingness and by identifying the greatest ability to pay and by the costs improvements in caring or of meeting their wishes at the curing that can be effected at margin. the margin. · Suppliers have a strong · Predetermined limits on incentive to adopt least-cost available resources create a methods of service provision. strong incentive for suppliers to adopt least-cost methods r · . . of service provision. Adjustment mechanism(s) · Many competing suppliers ensure that prices are kept low and reflect costs. · Well-informed consumers are able to seek out the most · Central review of activities generates efficiency audits of . . . service provision; manage- ment pressures keep the system cost-effective. · Well-informed clinicians are able to prescribe the most continued
PRIORITY SETTING IN A NEEDS-BASED SYSTEM TABLE 6-2 Continued System Element Private Systems 91 Public Systems Success criteria cost-effective form of treatment for themselves. · If medical practice is profitable at the price that prevails in the market, more people will go into medicine; hence, supply will be demand responsive. If, conversely, medical practice is unremunerative, people will leave it, or stop entering it, until the system returns to equilibrium. · Consumers will judge the system by their ability to get someone to do what they demand, when, where, and how they want it done. · Producers will judge the system by how substantial a living they can make through it. cost-effective form of treatment for each patient. · If there is demand pressure on some facilities or specialties, resources will be directed toward extending them. Facilities or specialties on which demand pressure is slack will be slimmed down to release resources for other uses. · The electorate judges the system by the extent to which it improves the health status of the population at large in relation to the resources allocated to it. · Producers judge the system by its ability to enable them to provide the treatments they believe to be cost-effective. SOURCE: Based on A. Williams, "Pnonty Setting in Public and Private Health Care: A Guide Through the Ideological Jungle," Journal of Health Economics 7:173-183, 1988. tween public and private provision of care, which differs markedly among countries. It was recently observed that Americans regard the British system (of rationing) as unacceptably coercive because it prevents people from getting the care they want and are willing to pay for; the British, on the other hand, consider the American system (of rationing) unacceptably coercive because it denies some people access to services that are available to others with similar needs. Perhaps the answer to the ethical question should be that each system is better than the other if one applies that system's own avowed criterion. Those seeking reform of either system may be reflecting one of the following viewpoints: (1) they want their system to reflect a different ideological position from the one on which it is currently based, and from that new position their own system appears deficient and should therefore be changed; or (2) their system does not perform very well according to its own ideological tenets and should be reformed to embody its ideals more
92 ALAN WILLIAMS TABLE 6-3 Characteristics of Actual Private and Public Health Care Systems System Element Private Systems Demand Supply Public Systems · Doctors act as agents, mediating demand on behalf of consumers. · Priorities are determined by the reimbursement rules of insurance funds. · Doctors act as agents, identifying need on behalf of patients. · Priorities are determined by the doctor's own professional situation, by his or her assessment of the patient's condition, and by the expected trouble-making proclivities of the patient. · Freedom from direct financial contributions at the point of service and the absence of risk rating enable patients to seek treatment for trivial or inappropriate conditions. · Because private insurance coverage is itself a profit seeking activity, some risk rating is inevitable; consequently, coverage is incomplete and uneven, distorting personal willingness and ability to pay. · Attempts to change the · Attempts to correct distribution of income and inequities in the social and wealth independently are economic system by resisted as destroying differential compensatory incentives (one of which is the access to health services lead ability of the rich to buy better to recourse to health care in or more medical care). circumstances in which it is unlikely to be a cost effective solution to the problem. · What is most profitable to suppliers may not be what is most in the interests of consumers; since neither consumers nor suppliers may be very clear about what is in the former's interests, suppliers are allowed a wide range of discretion. · Priorities are determined by the extent to which consumers can be induced to part with their money, and by the costs of satisfying the pattern of ``demand." e Personal professional dedication and public- spirited motivation are likely to be eroded and degenerate into cynicism if others who do not share these feelings are seen as doing well for themselves through blatantly self-seeking behavior. e Priorities are determined by what gives the greatest professional satisfaction. continued
PRIORITY SEWING IN A NEEDS-BASED SYSTEM TABLE 6-3 Continued System Element Private Systems · The profit motive generates a strong incentive toward market segmentation, price discrimination, and tie-in agreements with other professionals. Adjustment mechanism(s) Success criteria · Professional ethical rules are used to make overt competition difficult. · Insured consumers who are denied information about quality and competence may collude with doctors (against the insurance carriers) in inflating costs. · Entry into the profession is made difficult and the number of practitioners restricted to maintain profitability. · If demand for services falls, doctors extend the range of their activities and push out neighboring disciplines. · Consumers will judge the system by their ability to get someone to do what they need 93 Public Systems · Because cost-effectiveness is not accepted as a proper medical responsibility, such pressures merely generate tension between the "professionals" and "managers." · Public systems do not need elaborate cost data for billing purposes and thus do not routinely generate much useful information on costs. · Clinicians know little about costs and have no direct incentive to act on such cost information as they may have; sometimes they may even have disincentives (i.e., cutting costs may make life more difficult or less rewarding for them). · Little is known about the relative cost-effectiveness of different treatments; where some information exists, doctors are wary of acting on it until a general professional consensus emerges. · It is difficult to phase out facilities that have become redundant because such an action often threatens the livelihood of some concentrated, specialized group; in addition, the people dependent on the facility can be identified, whereas the beneficiaries are dispersed and can only be identified as "statistics." · Because life expectancy is the easiest aspect of health status to measure, mortality continued
94 TABLE 6-3 Continued ALAN WILLIAMS System Element Private Systems Public Systems Success criteria cont'd done without making them "medically indigent" or changing their risk rating too adversely. · Producers will judge the system by how substantial a living they can make through it. data and mortality risks . . . predominate in outcomes measurement, to the detriment of assessment of treatments concerned with non -life-threatening situations. · In the absence of accurate data on cost-effectiveness, producers judge the system by the extent to which it enables them to carry out the treatments that they find most exciting and satisfying. SOURCE: Based on A. Williams, "Pnonty Setting in Public and Private Health Care: A Guide Through the Ideological Jungle," Journal of Health Economics 7:173-183, 1988. fully. Those trying to reform the American system are probably in the first position, and those trying to reform the British system are in the second. If health care reforms in Britain are to succeed and succeed without abandoning the system's egalitarian ideology-a way must be found to "mi- cromanage" clinicians more effectively. The best hope for doing this ap- pears to lie in convincing doctors that it is their public and professional duty to carry out the policies that have been established by their health authorities; their acceptance of this responsibility in turn depends on the outcome of a much wider struggle for the hearts and minds of the citizenry. The policy dialogue in the media has been vigorous in the United Kingdom, raising hopes that the public will not only come to accept the need for prioritization (there are signs that they may have accepted it already) but also that they will be able to move on to the much more difficult task of agreeing on the broad principles on which such prioritization (i.e., ration- ing) should be based. The health care system of the future in Britain is likely to be one that in substance will be close to the maximization of equity-weighted quality-adjusted life years (although the public debate is unlikely to be conducted in such abstruse terms). Consensus as to the precise values that go into such a system will probably never be reached- any more than consensus has been achieved in Britain with respect to edu- cational policy, or defense policy, or transport policy. But merely to have
PRIORITY SETTING IN A NEEDS-BASED SYSTEM 95 health care prioritization a topic of continuous dialogue in the public arena, and subject to continuous review in the light of that dialogue, constitutes an important breakthrough. The entire exercise should greatly strengthen both public maturity and political accountability, which are essential to the dem- ocratic management of care and capacity in a national health service in which priorities are based on need.
7 The Meeting of the Twain: Managing Health Care Capital, Capacity, and Costs in Canada Morris L. Barer and Robert G. Evans The Canadian and American health care systems differ in three funda- mental structural respects: entitlement, management, and environment. Fun- damental philosophical differences in the two societies have their outcomes in the different approaches to, and results of, extending entitlement to bene- fits to their respective populations. The most obvious result distinguishing the two systems is that 30 to 40 million Americans have no health insurance coverage; many millions more have inadequate coverage. The managerial differences reveal themselves in the quite different "targets" of manage- ment. In Canada, the targets have largely been at the "macro" level, or systemwide; in the United States, they have been at the "micro" level, in the form of particular clinical interventions, or at the "mini" level, in the form of specific organizational and financial constructs (e.g., health maintenance organizations) intended to manage care patterns for particular subsets of the population. Differences in environment are found in other sectors of the two nations' economies, the interests of which must be balanced against the activities of and benefits from the health care system. In particular, Canada does not have a major industry that develops new technologies with health care applications. Nor is there an informed, articulate private constituency for cost containment, because private employer-based health insurance is vestigial in Canada. Although these differences have resulted in dramatically different cost experiences (Evans, 1986; Evans et al., 1991) and some clear differences in 97
98 MORRIS L. BARER AND ROBERT G. EVANS patterns of health care utilization, this paper argues that the problems of health care system management in both countries are now, and will increas- ingly become, tied to the management of health care capital: human (partic- ularly physicians), physical (facilities and equipment), and technological (know-how). Such management will require considerable political will on both sides of the border. To date, Canada has been somewhat more suc- cessful, largely because of its quite different funding process, but also be- cause those responsible for making major capital decisions are less ham- pered by competing priorities than their American counterparts. This discussion begins by elaborating briefly on the fundamental differ- ences (noted above) between the two systems. It then outlines Canada's approach to health care system management, the relationship of that man- agement to the management of each type of health care capital, and the interconnectedness of the three classes of capital. The paper closes with observations on the likely future direction of strategies for health care sys- tem management in Canada, on the challenges ahead, and on the signifi- cance of some of those challenges for U.S. health care system reform. WHAT MAKES CANADIAN HEALTH CARE DIFFERENT? One of the great ironies in the predicament in which health care in the United States finds itself today is that universal coverage or entitlement to benefits a key to health care system management (and cost control) in Canada and most other Western industrialized countries is seen in the United States as an elusive target because of its alleged cost-expanding implications. Experience elsewhere continues to suggest that overall (as distinct from public) health care cost control in the absence of some form of universality is impossible. Yet universality, while apparently widely sup- ported by Americans (Blendon and Taylor, 1989), remains elusive because of the perception that achieving it will further increase costs in what is already the world's most expensive health care system. Such cost increases, in contrast to the broadening of coverage, find favor with only a very narrow and clearly identified group of Americans- the "vendors" of services. The reconciliation of this apparent contradiction is relatively straight- forward. Universal coverage for medically necessary services is necessary, but not sufficient, for health care cost control. Extending coverage may increase costs in the United States if it is achieved through "fill-in" cover- age, that is, through the addition of more pieces to the jigsaw puzzle of American health care financing. Hospitals operating at relatively low levels of occupancy may be able to raise those rates. But as Wennberg (in this volume) points out, where capacity is relatively fully deployed already (as is presumably the case for physician services), the extension of coverage
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 99 may have minimal effects on overall use.1 It may, however, put significant upward pressure on prices. So those in the United States who are concerned about the cost-expanding implications of extending coverage may well be right. But they will be right only because partisans of "autonomy," "plural- ism," and "taxation anti-bodies"2 have used these approaches with almost religious fervor to suppress more comprehensive proposals for health care system reform and, at least so far, have restricted the policy choices to continued "disjointed incrementalism" (Kinzer, 1990~.3 Within this policy straitjacket, extending coverage to those Americans who are presently uninsured or underinsured would achieve universality but would almost certainly increase health care costs in the United States quite dramatically. Universality is not sufficient to achieve cost control. Where- as those countries with universal coverage have managed to control their health care costs relative to the United States, it has not been the universal- ity per se that has achieved such control. One could quite easily imagine circumstances in Canada that would allow the entire population to be cov- ered for medically necessary hospital and medical care, yet that would pro- duce a cost experience more closely paralleling that of the United States. In fact, during subperiods of the two decades of universal hospital and medical coverage in Canada, some individual provinces have done rather well in mimicking the American cost experience (Barer and Evans, 1986; Hughes, 1991). Yet universality elsewhere has left American health care costs in a league of their own (Abel-Smith, 1985; Schieber and Poullier, 1991~. In what sense is it necessary? In addition to being an objective worth pursuing tin fact, this limited effect was precisely what was found when universal health insurance was introduced in Quebec in 1970 (Enterline et al., 1973). The distribution of beneficiaries changed use by lower income persons increased and use by higher income individuals de- creased but overall use was unchanged. 2Thus, for example, one finds the Health Insurance Association of America going to great lengths to highlight the huge tax burden for Americans implied by the adoption of a Canadian- style system. Malignant neglect of the countervailing side of the ledger, the private out-of- pocket costs and private insurance premiums that would be "saved" (but which in some pro- portion constitute the incomes of members of that association), creates a predictable and highly misleading picture (Neuschler, 1990). 3Never mind that systems structured around universal entitlement, such as that of Canada, have, according to the record to date, provided far more autonomy for physicians than has the increasingly clinically managed American system, or that the Canadian system provides at least as much autonomy for the patient in choosing a practitioner and that most Americans seem dissatisfied with the plurality of health insurance schemes. In the heat of the rhetorical debate, the precise nature of the "autonomy" or "pluralism," apparently so coveted, rarely emerges. Nor is it often made explicit which segments of the American population so value these alleged characteristics of the present U.S. "system."
100 MORRIS L. BARER AND ROBERT G. EVANS in its own right for reasons of equity and altruism, universality is also a means to a management end. It appears to be necessary to the achievement of cost control because it provides the enabling management structure for such control. Cost experiences in different countries, or in different prov- inces, states, or regions, reflect the extent to which those responsible for managing health care systems avail themselves of the management opportu- nities provided by each particular form of universal coverage. And that, in turn, depends on the shifting balance of political influence between vendors and payers. In Canada, the federal legislation that established the hospital and med- ical care insurance programs provided grant and tax point transfers to the provinces that were conditional on universality and the establishment of provincial, nonprofit, publicly administered insurance programs to provide coverage for medical and hospital costs. Responsibility for plan manage- ment rests with each province, but the broad terms and conditions, at least until now, have been largely dictated by federal fiscal powers. Thus, in the Canadian context, it is the manner in which universality evolved that has shaped the form of health care management. Unlike the United Kingdom, Canada chose to establish a collection of social insurance programs while leaving the actual provision of care largely in the hands of private "autonomous" medical practitioners. Although most hospitals are public institutions and are funded globally out of public funds, the clinical management of the patients treated in them is a matter left to private practi- tioners with privileges at each facility. The administration of each institu- tion is handled by an executive staff employed by independent hospital boards. This does not mean, however, that care is not "managed." Health care in Canada is "macromanaged": provinces manage the financing of the system that provides the care, and that financial management affects the volume and mix of care provided. But the management, at least to date, has been one step removed from the bedside or the clinician's office. "Man- aged care," even in a broad sense, has come to be associated with a particu- lar set of activities that have gained favor recently in the United States (chart review, outcomes management, development and application of clini- cal practice guidelines, volume performance standards, second opinion pro- grams, and the like). All of these have the explicit intent of involving more than the attending physician in the micromanagement of patterns of care provided to individual patients. These forms of managed care are explicitly about changing the fundamental nature of the doctor-patient relationship. But they are not the only means available for managing care. The importance of viewing the American and Canadian systems as oc- cupying different positions along a management continuum cannot be over- stated. Too often the debate over system reform in the United States casts
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 101 Canada as the system that "rations,"4 in contrast to the United States, where care is "managed." Such a contrast is, of course, pure rhetorical nonsense. Both (all) systems "ration" (Hadorn and Brook, 1991), but they approach rationing in different ways. Each system chooses to manage the utilization of health care resources differently. The increasingly dominant American approach is to have third parties involved in the physician's microenviron- ment, to have someone attempt to look over the physician's shoulder and push his or her elbow. The Canadian approach has been to attempt to constrain the macroenvironment, the total size of the health care pie, and to assign (implicitly) the responsibility for micromanagement to those eating it. In Canada, at least until quite recently, the micromanagement levers have been left largely untouched. Nothing in principle prevents such activ- ity in the Canadian system, although in practice there are powerful political constraints. Certain approaches, however, such as the development of qual- ity assurance programs and physician peer assessment programs (McAuley et al., 1990) appear to be emerging as a new growth industry in Canadian health care. As suggested later, however, they seem to come with more circumscribed and more realistic objectives than are found in the United States; there, managed care appears increasingly (at least to these authors) to be expected to carry the burden of system cost control along with the more micro objectives of improving the effectiveness and efficiency of clin- ical management of particular health problems. Based on the record to date, its achievements at either level seem underwhelming. The third fundamental structural difference between Canadian and American health care can be found in the environmental contexts in which each sys- tem operates. A significant and economically important share of American productive activity (and capacity) is to be found in technological research, development, sales, and distribution. Much of this activity (e.g., the devel- opment and marketing of new pharmaceutical products, of imaging and laser technologies, of orthotics and prostheses) has the American health care system as one of its major potential markets. More use of its products by the health care system means a more profitable technology sector (not to mention a more prosperous group of health care providers). Thus, the dynamic of growth in the health care sector is intertwined with that of these other "derived-demand" sectors. In the current environment, however, the two sectors create competing political imperatives, and the policy problems 4A leading example of this attitude was the 1989 letter sent over the signature of Alan Nelson, then president of the American Medical Association (AMA), to all AMA members, cautioning that a "Canadian-style health care system could cause rationing of medical services."
102 MORRIS L. BARER AND ROBERT G. EVANS that this fundamental conflict poses are characteristics of the American system that are largely absent in the Canadian.s In this context, it is worth noting that if newly developed technologies were to produce cost savings when applied in the health care sector, part of the political problem of competing priorities (health care cost control versus support for technological research and development) would disappear. But the cost savings on one side of the ledger are still income losses on the other; the incentives for the health care system to pick up such technologies are not obvious. It comes as no surprise, then, that the health care industry has a voracious but selective appetite; cost-enhancing technologies are pre- ferred. This latter point is no less true in Canada than in the United States. The difference is that Canadian politicians do not have to confront these conflicting political objectives to anywhere near the same extent. The de- velopment of new pharmaceuticals, medical devices, and the like in Canada represents a minute industry when set against the health care complex. Political initiatives to control health care costs in Canada do not come up against the kind of need that exists in the United States to support and nurture another major set of economic activities.6 The world of technolog- ical innovation in medicine will carry on (and in all likelihood continue to prosper) regardless of what Canada does in the way of health care manage- ment. Indeed, the future of the technological development sector in Canada is tied more closely to world markets than to the Canadian health care system. This knowledge may raise questions about the extent to which Canada's relative cost control success is simply due to the absence of a domestic industrial complex selling to the health care industry. Here, the experience of other countries may inform the discussion. The United Kingdom falls somewhere between Canada and the United States in this respect. It has a major international pharmaceutical presence, whose interests conflict with SThe military analogy is pervasive here. Not only do some of the technological advances in health care emerge from research that is, in the first instance, militarily motivated, but the military and the health care sectors in the United States share a common current political problem. Both are under intense cost-control pressure, which conflicts with the interests of the complementary technology development sectors for which the military, on the one hand, and the health care sector, on the other, are the major markets. 6Canadian policy, however, may be powerfully influenced by the priorities of the American technology industries. The Canadian federal government's Bill C-22, passed in 1987, signifi- cantly extended the patent protection of new pharmaceuticals and thereby undercut (probably destroyed) nearly two decades of carefully crafted cost-control policy at the federal and pro- vincial levels. The action was a direct response to American pressure applied at the highest political level.
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 103 that of the National Health Service in minimizing the cost of pharmaceuti- cals, at least for Britons. The pharmaceutical manufacturing sector appears to survive and prosper; in addition, health care there represents significantly less of all productive activity than in Canada. Germany is more like the United States than Canada in this respect, having major pharmaceutical, imaging, and other "high-tech" interests; yet its health care cost experience is more like that of Canada. A rational strategy for all of these countries is to promote the export side of domestic technological industries (especially exports to the United States) while attempting to limit their applications in domestic health care markets. In the United States, however, the domestic health care system represents the largest potential market for the new prod- ucts. Furthermore, it is a market eager to adopt new innovations in "half- way" technologies (Evans, 1984~. Thus, the United States continues to promote progress in health care technology, apparently oblivious to the fact that such promotion is at odds with the rhetoric of health care cost control. This paradox was nowhere more obvious than at the 1990 International Summit on Health Care and the Economy, sponsored by the University of Texas Health Science Center just prior to the meeting of the G7 countries in Houston. The keynote speakers, Robert Mosbacher and Denton Cooley, both gave inspirational addresses. Unfortunately, the two fundamental messages could not have been more opposed. U.S. Secretary of Commerce Mosbacher opened the summit by emphasizing the need to extend coverage and improve the efficiency with which health care is provided to Americans. He emphasized that uncon- trolled health care costs were now a serious drag on American prosperity and economic growth. Cooley, on the other hand, celebrated the glories of modern medicine and extolled the economic virtues of the export potential of the products and services developed and offered by the Texas medical complex (the largest such complex in the country) as a major countercycli- cal stabilizer for the Houston economy. Unfortunately, neither speaker heard the other (nor, one suspects, would it have mattered if they had). SO HOW DOES CANADA DO IT? Managing Operations A considerable literature describing the organization and financing of Canadian health care already exists (see, e.g., Evans, 1984, 1988; Iglehart, 1986a,b; Barer et al., 1988; Evans et al., 1989~. Its fundamental character- istics are well known, and any comprehensive attempt at description would detract from the intent of the present paper. In brief outline, each province has its own medical and hospital insurance program, but all adhere to re- quirements set out in federal legislation. A fundamental element of these
104 MORRIS L. BARER AND ROBERT G. EVANS . programs Is that coverage must be offered to the entire population, under uniform terms and conditions. Although a few provinces continue to charge premiums for medical care (in British Columbia, for example, premium revenue represented slightly more than 50 percent of total Medical Services Plan outlays, including the costs of administering the plan, during fiscal 1988-1989), care cannot be denied because of premium payments that are in arrears. The medical and hospital sectors are financed almost entirely from gen- eral revenues (from provincial sources and federal transfers to r~rovin~.~..~N ~. .. ~.. . . _,. _^ V - ~ tA ~ ~ Ad,_, . In ~act, with the enactment in 1984 of the Canada Health Act, "user charges," "extra-billing," and other out-of-pocket medical or hospital costs to patients were largely eliminated because the legislation stipulates dollar-for-dollar reductions in federal cash transfers against any such private charges. In effect, a province that allows such charges asks its population to pay twice, a compelling political deterrent. Not surprisingly, taxation rates are high. On the other hand, private insurance cannot operate in competition with the public medical and hospital programs. With very few exceptions, Canadi- ans do not pay out-of-pocket charges or premiums for these services. Be- cause total health care costs are lower, the savings in private costs outweigh the extra tax payments. Provincial governments are responsible for the allocative decisions within the health care sector in each province and have a key role to play in the "pricing" of services. Here, the discussion is restricted to medical and hospital care. Most physicians in each province are paid fees for service on the basis of a provincewide fee schedule. Overall average changes in the schedule are periodically negotiated between provincial ministries of health and provincial medical associations (there are separate general practitioner and specialist associations in Quebec). The medical associations determine the internal allocation of these increments. Each fee schedule is associated with a set of payment rules that govern the frequency and circumstances under which particular billed items will be reimbursed. As a result of increasing internal pressure, a number of provincial medical associations have recently begun to develop relative value scales that would remove some of the alleged inequities,7 but there is no sign of any movement toward a consistent national relative value scale. Although fee experiences in different provinces and during different periods have varied, sometimes dramatically (Barer and Evans, 1986; Hughes, 1991), overall, the process of 7Much of the internal conflict is over relative incomes, in that procedural specialists have gained some considerable ground over their nonprocedural peers and over general practitio- ners. This conflict tends to be played out on the fee schedule playing field, despite the fact that those at the lower end of the income distribution have, in fact, fared relatively well on the internal fee allocations (Barer et al., 1992).
MANAGING CAPITAL, CAPACITY,AND COSTSIN CANADA 105 bilateral negotiations has held fee increases at, or slightly below, general rates of inflation over the past decade. This process is a key component of managing the Canadian health care system, and it has implications for the management of care (as distinct from costs; Barer et al., 1988~. Physician fees have grown much more rapidly in the United States than in Canada; utilization per capita has grown somewhat more rapidly in Canada. An increasing number of physicians are being paid salaries or are being compensated on a sessional basis, and it appears that most provinces would like to see this trend continue. To date, however, these alternative forms of payment represent a relatively small proportion of total provincial outlays for medical services. Against this payment backdrop, the supply of physicians in Canada has increased at rates well in excess of population growth for almost 40 years. Estimates to 1990 suggest a population/physician ratio of about 450:1 in Canada; this figure is closer to 400:1 in the United States (Evans et al., 1991~. The aggregate income expectations this sustained increase repre- sents, and the growing recognition that fee controls alone do not control costs, have led a number of provinces to introduce so-called macro cost- management techniques (Lomas et al., 1989~. Negotiations in most prov- inces now have utilization "on the table" (over the protests of the medical associations), and some medical associations appear to be willing to give up fee increases to avoid utilization or expenditure caps. There are a rich variety of capping models. A capped reimbursement agreement could provide for quarterly monitoring of global utilization, with fees in subsequent quarters being temporarily rolled back from schedule values so that expenditures remain within the cap. All variants allow utili- zation increases for general population growth; some provide additional utilization room for structural population changes or other factors; some involve the sharing of overages between the profession and the provincial ministry of health (Lomas et al., 19891. Finally, individual general practi- tioner income ceilings have been in place in Quebec for a number of years (Contandriopoulos, 1986; Barer et al., 1988; Lomas et al., 1989), in con- junction with overall expenditure caps. As noted earlier, funding for hospitals also comes largely from ministry of health budgets. (Only one province still retains hospital insurance pre- miums.) Salaries and wages of most hospital workers are negotiated on a provincewide basis, between the unions and associations representing the hospital employers. The hospitals, in turn, negotiate annual operating bud- gets8 with the ministry of health. Historically these negotiations have been 80perating and capital costs are funded differently, although mainly from the same source. Capital costs are addressed in the next section.
106 MORRIS L. BARER AND ROBERT G. EVANS rooted rather firmly in the experience of prior years, with some adjustments for new programs. An emerging trend in some ministries of health is the development of more sophisticated population-based funding formulae for hospitals, which take account of the age, sex, and even ethnic structure of the population. Current consideration by provincial authorities of population-based funding formulae, however, goes beyond the funding of hospitals. There has been a spate of provincial Royal Commission reports in recent years, all of which have recommended some form of regional funding and management (see, e.g., the 1989 report by the Nova Scotia Royal Commission on Health Care and the 1991 report of the British Columbia Royal Commission on Health Care and Costs). But the political pressures against such a policy continue, so far, to thwart any initiatives. One province (Nova Scotia) has already announced that it will not adopt its commission's regionalization recom- mendations; to date, others have not been that explicit. Ontario is commis- sioning work that is intended to develop regional funding formulae (Birch and Chambers, 1990), and recent Quebec proposals would allocate medical budgets regionally (Ministere de la Sante et des Services Sociaux, 1990~. It seems likely that regional management structures will be developed in Ca- nadian provinces over the next half-decade, in part because provincial min- istries of health recognize the inherent logic of population-based funding. Those same ministries, however, may also recognize the advantages of de- flecting the centralized political heat that results from attempts to control costs. Managing Capital The long-run viability of these Canadian approaches to managing health care costs will depend critically on the will and ability of Canadian policy- makers to manage health care capital, including new technology. Health care capital comes in three basic forms: physical (bricks, mortar, machines and equipment), human (health care personnel), and intangible (research and development activities; Barer and Evans, 1990~. All share the charac- teristic that resource commitments at a point in time are intended to gener- ate a future stream of benefits; all correspondingly require the sacrifice of current consumption. Yet the anticipated stream of future benefits is not the only future effect of today's commitments. Health care capital also creates a future stream of pressures for additional, complementary capital and operating commitments. It is this fundamental characteristic of health care capital that poses the major challenge for the macromanagement of the Canadian health care sys- tem and, seen from this side of the border, poses an even greater challenge for the management of health care in the United States.
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 107 This intertemporal and interclass capital interdependence is nowhere more evident than in the relationship between physical and human capital. "New non-human capital brings with it demands for, or expectations of, new and often quite specialized human capital. Once the human capital is in place, idling physical capital offers the prospect not only of turning off switches on machines, but the redeployment, or costly re-tooling, of the complementary human resources. Not only is the physical capital the raison d'etre for the human resources, but the reverse also becomes true in prac- tice" (Barer and Evans, 1990~. New imaging technologies demanded by hospitals that want to remain up to date create derived demands for physi- cians with the skills and knowledge to manage them. Once the medical care team is in place, a new set of political constituents renders the job of shutting down the capacity (even if it is determined to be obsolete or inef- fective) that much more difficult. But this dynamic works equally in the opposite direction. New, highly specialized physicians create derived de- mands for the complementary "tools of their trade." If fewer cardiac sur- geons are trained, there will be fewer coronary artery bypass graft units. Once the units are in place, however, a demand for, among others, perfu- sionists is created. The interdependence of the less tangible intellectual capital with the other two classes is no less real. New imaging or laser therapeutic tech- niques give rise to a host of new forms of physical capital and create de- rived demands for ever more highly specialized technicians and physicians. The explosion of clinical and technological knowledge makes mastery of any part of it increasingly difficult and creates continuous pressures for subspecialization as a knowledge-control mechanism. In addition, human capital creates demands for itself. New subspecialists covet academic pro- grams through which they can funnel residents to assist with the clinical work. Educational programs tend to be supported not on the basis of whether the products of the programs are required by the population but rather on the basis of the needs of the training institutions and their faculty (Barer and Stoddart, 19911. The expansionary dynamic of health care capital has roots in the explo- sion of the physician supply in North America and Europe over the past three decades (Schroeder, 1984; Viefhues, 1988; Evans et al., 1991~. Not only do new physicians create demands for new complementary treatment space (hospital beds and other facilities) and technology (their diagnostic and therapeutic arsenal), but they play critical roles in the creation of new knowledge and techniques through their roles in research and development. Although Canada's current per-capita supply of physicians is not as "rich" as that in some European countries (Germany, France), and is actually somewhat lower than that in the United States, it is still widely regarded (at least in Canada) as in excess of desirable levels.
108 MORRIS L. BARER AND ROBERT G. EVANS Canada's management of its health care capital has been, like its over- all system management, largely at the macro level. It has not been particu- larly successful in managing its health care human capital because, unlike most physical capital, human capital is mobile. Each province is account- able for health care spending within its jurisdiction and can more or less successfully control overall levels of funding and the proliferation and dif- fusion of physical capital. No individual province, however, has control over the supply of physicians who wish to practice in that province (and who submit claims to the provincial plan). The key to control of physician supply rests with individual medical schools. Yet individual schools will argue (correctly) that reducing their training capacity will have no necessary effect on provincial supply, both because of interprovincial movements of physicians trained in Canada and because of in-migration of foreign medical graduates. There are two para- doxes in these arguments. First, although the logic holds for each school, it does not hold in the aggregate. Yet medical schools in Canada (as in the United States) have shown no inclination to provide collective leadership on this issue. Second, the two dominant problems, to which a continued influx of foreign medical graduates is the solution, are geographic maldis- tribution of physicians and the service requirements of postgraduate train- ing programs. Domestic solutions to both would again require leadership from the medical schools, which has, to date, not been forthcoming (Rarer and Stoddart, 1991~. ~e, ~ ~ _4 The problem of interprovincial mobility is felt most acutely in British Columbia. In the mid-1980s, that province attempted to address its particu- lar problems with physician supply growth by limiting the number of physi- cians who could submit claims for payment to the medical plan. At the end of a rather tortuous legal evolution (Barer, 1988), this policy was over- turned on constitutional grounds. reunite. am r~r`,lht ah^.~+ `',h^~ ~ ~ - ~ ~ _~ ~ 1 · ~ D.~. WOUrl OI Appeals Judgment was consistent with prior and subsequent constitutional decisions (Lepofsky, 1989), no other province has yet tested the legality of this approach. In any case, a policy of limiting the number of physicians billing the system would have a clear effect in the implement- ing province but would make little sense if all provinces did it. First-year enrollments at Canadian medical schools have declined about 6 percent over the past 7 years (although applications per place have not), and there are increasing pressures for further reductions. The pressures come, rather predictably, from provincial ministries of health, which are responsible for meeting, or otherwise dealing with, the financial pressures created by the burgeoning supply. In contrast, despite the fact that the number of graduates peaked in 1985, no reduction has been seen in the overall number of funded post-M.D. training positions or even in the num- ber of such positions funded by provincial ministries of health (Association
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 109 of Canadian Medical Colleges, 1990~. This may reflect in part the length- ening of requirements in some programs. It seems equally likely, however, that the expansion in post-M.D. positions is increasingly driven by the self- generated "need" for students of ever-growing numbers of residency pro- grams (Barer and Stoddart, 1991~. Canada has not, then, to date, adequately addressed the issue of human capital management. Relative to the United States, however, it has better managed the proliferation of subspecialties. The Royal College of Physi- cians and Surgeons of Canada recognizes for certification about one-half the number of specialties recognized in the United States. The ratio of general practitioners to specialists in Canada is about 55:45. This manage- ment of specialty supply is achieved through ministry of health funding of the vast majority of post-M.D. training positions. Each new residency im- plies a requirement of from 4 to 6 new funded positions, because a position is necessary for each year of the training. The financial implications of such requirements are not trivial. Nevertheless, given the downstream income expectations (and the de- mands for complementary capital) associated with each new specialist, as well as the growing divergence between undergraduate training capacity and post-M.D. funded positions, the management record here seems no more worthy of envy than that on overall supply. Some reductions are anticipat- ed in the size of the post-M.D. training establishment in Canada over the next 5 years, despite a move toward a common 2-year post-M.D. training requirement that would be recognized by all provincial licensing authori- ties. This expectation may remain unmet if reductions in training capacity will idle, or force the redeployment of, medical school human and physical capital. The threat of such reductions mobilizes powerful, determined op- position. Such opposition historically has been quite successful, because the distribution of losses is much more concentrated (and identifiable) than is the distribution of benefits from downsizing capacity. As for Canada's management of physical capital, the record is relative- ly good. Institutional capital (new facilities and beds, new capital equip- ment) is funded largely through the same provincial ministries of health (with regional districts picking up most of the rest of the costs), although the specifics of approval and allocation vary across provinces (Deber et al., 1988; Bayne and Walker, 1989~.9 Relatively speaking, this process has resulted in an ample supply of hospital beds (which are, in fact, more fre- quently occupied in Canada than in the United States [Evans, 19901~. But this supply (particularly of acute care beds) has not increased in recent 9The interested reader will find a relatively detailed description of this process in British Columbia in Barer and Evans (1990).
110 MORRIS L. BARER AND ROBERT G. EVANS years to match the growing supply of physicians (Barer and Evans, 1986). Because these two forms of capital are complementary, this asymmetry creates continuous political pressure through physician claims of system underfunding, "shortages," and "waiting lists." The Canadian centralized process of capital approval, however, has Because limited the diffusion of diagnostic and therapeutic technology. physicians cannot generally receive lump-sum or fee-based funding for cap- ital acquisitions, much of Canada's high-tech capacity is restricted to public hospitals. In turn, all hospitals in each province must go through provin- cial, and often regional, approval processes that provide at least the poten- tial for rational planning of the acquisition of such equipment. The hospi- tal- and physician-based pressures to have every conceivable piece of new equipment at every hospital are similar to those in the United States. But the diffusion outcomes, the rates of utilization, and the implications for overall hospital costs are quite different in Canada (Detsky et al., 1983, 1990; Rublee, 1989~. Even when hospitals manage to raise local funds for a CT (computed tomography) scanner, for example, the provincial ministry of health is un- der no obligation to provide the necessary operating funds. In such a case, the hospital must reallocate monies from within its global operating budget or raise the operating funds as well. Taking the former route runs the risk of raising Questions within the mini~trv At how the h^~^it~1 fnllr`A the ~ _ A ~ ~ A ~ ~ ~ ~ ^ ~ ~ = all. t,-~ 1 V ~. Al ~ At 1 ~ necessary --slack In a budget about which it is constantly complaining ---a ~ I- = Although hospitals are raising money for capital expenditures with increas ing frequency, such situations remain the exception. As noted earlier, Canada plays a minor role (on a world scale) in the development of medical technological capital. Most of it simply arrives at the border. Canada has not controlled access to new knowledge in fact, how could it? Instead, in the manner described above, the macro approach to health care management controls the number of "embodiments" of that new knowledge in new machinery. To date, as suggested above, the record is mixed. How successful Canada has been depends on where one sits, and on one's perception of the value of more, relative to lesser amounts, of different types of health care capital. Some American observers find much to envy in the Canadian approach (Marmor et al., 1990~. Others argue that the limitations on the availability of new high tech capital in Canada are a serious drawback to the Canadian system; to support their contention, these observers point to the alleged flow of Canadians in search of high-tech interventions south of the border or to long waiting lists for high-tech inter- ventions. No one suggests that the management process is perfect (Iglehart, 1990~. Every health care system is a dynamic set of solutions to the continu- ing emergence of a series of connected and complex problems. The choice of a health care management approach is a choice among alternative sets. . . ~
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 111 FROM MACRO- TO MICROMANAGEMENT AND BACK AGAIN? Largely absent until recently within this macromanaged system has been micro- or clinical management. Provincial ministries of health generally have been more or less content to manage overall costs and the allocation of funds. Although decisions have been made about the availability and loca- tion of new technology, those decisions have been based more on financial and political factors than on effectiveness or efficiency evidence. There has been virtually nothing that looks or feels like "managed care," as it is understood in the United States. Whether this is better or worse than other alternatives, and whether Canada should (as it appears now to be doing) put more energy into micro- management initiatives (e.g., technology evaluation, continuing competence programs for physicians), depends on the goals of technology and system management. New technologies offer a variety of cost and outcome possi- bilities, but there are few instances in which information on these possibili- ties is known in advance of application. For the rest, policymakers must attempt to acquire it after the technologies are put into use, all the while hoping that they (and the population to which they are accountable) do not get too badly "burned" while the evidence accumulates. (Of course, in many cases the evidence never accumulates, but management decisions must still be made.) Interventions (including any new approach to clinical diagnosis or ther- apy) may have one of several effects: a. reduce health care costs while improving or leaving unaltered the health status of recipient patients; b. increase health care costs but produce substantial and unequivocal improvements in the functional capabilities of recipients; c. increase health care costs and produce small, positive, often diffi- cult-to-measure increments in the health status of some segments of the patient population;l° or d. whatever their costs, produce no or negative effects on health sta tus. Health care managers in any system should welcome all possible occurrences of type (a) interventions. The management of type (b) technologies can be assisted by technology evaluation, but care must be taken in generalizing 0Welch has labeled these the "epsilon effects." 1lThe vendors of services, however, are often less supportive; as emphasized earlier, re- duced costs translate into reduced incomes for some vendors. If these vendors are in a position to insist that absence of harm be proven to their satisfaction before the new technology is introduced, type (a) changes may be slow in coming.
112 MORRIS L. BARER AND ROBERT G. EVANS results from one setting to another, let alone across countries. Furthermore, many evaluations are themselves quite costly. The primary problems of health care system management do not, how- ever, come from type (a) and (b) interventions. Most of the micromanage- ment (and research) efforts are intended to identify and eliminate type (d) technologies indeed, they have no place in any health care system. The great danger in this approach is that a single-minded preoccupation with type (d) interventions may skew the application of management energy out of all proportion to their relative importance. Category (c) may be more important quantitatively than the other three categories combined. More- over, clinical ingenuity and technological progress are likely to ensure a growing stream of such interventions (Wennberg, 1990~. Although category (c) interventions produce small benefits for individ- ual patients, collectively, the high costs of the health gains they offer may swamp the benefits. Heroic measures for the late-stage Alzheimer's disease patient come to mind as an example of a situation in which extremely costly interventions may extend life for a few hours, days, or even weeks (Calla- han, 1987~. Most people, if they were given the choice at earlier stages in their lives, would choose a different process for the final stage of life.12 But a person may not have that choice, because new technologies continue to make more things possible and because they are there they will be used. The problem with many category (c) interventions is that the ethical imperatives within the health care sector malice it exceedingly difficult, if not impossible, to make choices against such "epsilon interventions." Cost- effectiveness evidence is unlikely to be available to the management pro- cess, because many of these everyday interventions are not individually important enough to warrant the use of limited evaluation research resources. The only practical way to reduce the occurrence of category (c) interven- tions is to reduce the capital and capacity that makes them possible. How does Canada manage category (c) and (d) interventions? There seems to be little doubt that Canada lags behind the United States in identi- fying category (d) technologies. As noted earlier, the macromanagement approach has provided few management incentives, and in fact powerful political disincentives, to look over the clinician's shoulder. Some evi- dence suggests that Canada has better addressed the epsilon problem within 12There are, in fact, two conceptually distinct issues here. Life extension per se does not necessarily represent improvement in health status. The person concerned might feel, and genuinely be, "better off dead." An intervention that appears to belong in category (c) when measured only by life expectancy may actually fall into class (d). However, even for "authen- tic" class (c) interventions, the relation of benefit to cost may be such that a representative individual, looking forward in life, might reasonably judge that he or she would prefer to forego the possibility of such interventions.
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 113 the hospital sector than has the United States (Detsky et al., 1983, 1990; Barer and Evans, 1986; Anderson et al., 1989~. But it may be less success- ful in other areas- for example, by providing more physician services for and institutional care of the elderly. As for high-technology diagnostic equipment, Canada's approach of "controlled technological diffusion" has. at least in relation to the United States, controlled technological diffusion. The jury is still out, however, on whether this has made Canadians better or worse off than their better-endowed American neighbors (Evans et al., 1991~. The American managed care approach to category (d) interventions is able to muster political and financial support because it is identifying and promoting the elimination of unequivocally "bad buys." In comparison, the Canadian macromanagement approach may be coming under increasing po- litical pressure as the social consensus on which it rests is threatened by the asymmetry of information dissemination to the public. Much of what Cana- dian patients (like their American counterparts) learn about the possible benefits of interventions comes from their vendors, for whom doing better means doing more. The predictable result is a growing public perception of an underfunded health care system, bled white by continual financial cut- backs. There is no informational counterpart to the provincial financial and managerial roles. Provincial ministries of health are loath to become in- volved in an organized effort to counter the claims of vendors because they fear that they cannot possibly succeed-that they will be perceived as sim- ply projecting a message consistent with their responsibility to control costs, without much regard for outcomes. The research and policy analysis com- munity, which might be expected to assume this role, is too small and, more fundamentally, with few exceptions does not yet see this as a legitimate or appropriate task (Lomas, 1990~. At the same time, the number of "promotional" voices continues to grow far more rapidly than the population, and more rapidly than the real rate of economic growth of the country. Canada, as has been mentioned, has done little to manage its human health capital in a manner consistent with its approach to health care system management. Continued tight con- trol over hospital capacity and over medical care budgets, in the face of a rapidly expanding physician supply, offers very few possible outcomes. That of "loosening the public purse strings" seems unlikely and, on current evidence, unjustifiable. That of forcing physician incomes down, perhaps precipitously, would be politically hazardous, and not necessarily fair to the large majority of the profession. Yet those are the two stark options. They ensure a continuing climate of public conflict. There is, in fact, a third option, favored by many vendors. The constant pressure of the human capital who depend on an ever-expanding health care system for their own survival and prosperity frequently produces renewed
14 MORRIS L. BARER AND ROBERT G. EVANS calls for the introduction of private-sector funding. As Iglehart (1990) noted recently, Canada is alone in the Western world in its "resistance to private funding." User fees, in various forms, are an idea that continues to surface, even in Canada. They are proposed regularly by the medical pro- fession, allegedly as a means of reducing cost pressures. In reality, they are seen by the profession as a means to increase expenditures, while reducing public cost pressures. Nor are vendors the only advocates of greater expenditures through direct access to patients' private resources. It appears that a growing num- ber of relatively well-off Canadians are becoming convinced by the ven- dors' arguments that public funding cannot or will not support ready access to first-class care for themselves and their families. They are realizing that a limited schedule of user fees will give them an advantage: preference for services, and thus first call on the public funds that will always form the backbone of any health care system.~3 In the end, the funds all come from the same source; thus far, Canadian governments have recognized this and stood by the principle of universal access on equal terms and conditions. Yet this, too, may be about to change. Recent federal legislation in Canada (Bill C-69) will dramatically reduce the federal contributions in- tended for provincial health care. The bill froze such contributions for a period of 2 years, and this freeze has recently been extended so that it will now be in place until 1995. The legislation was introduced and passed with surprisingly little fanfare or outcry, either from federal opposition parties or from the provinces, under the cover of a major (and continuing) constitu- tional crisis. Given historical, federal all-party support for the Canadian medical and hospital insurance programs, the relative silence from the op- position parties suggests that they have not yet fully recognized the poten- tial ramifications of the bill. The glue that holds the system together, that ensures adherence to a common set of principles by all provincial plans, is the federal fiscal role. As that erodes, as the contributions from the federal government become less important, provinces are more likely to go their own ways. In the end, it may be the federal government itself, rather than the medical profession, that drives the wedge of private funding into the door. In the process, however, it may destroy the whole system, a possibil- ity clearly recognized by the profession in its public opposition to Bill C- 69. In the view of these authors, this legislation represents a major threat to the Canadian system of financing health care, a threat perpetrated by a federal government increasingly seen by Canadians as slowly disembowel- ing Canada. The implications for the possibility of macromanagement of i3This statement includes the American system, in which the rhetoric of private funding obscures the major public role in subsidizing and regulating the "private" system.
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 115 health care in Canada are not good. Nevertheless, whatever system (or country) emerges over the longer term, the need for such management will not disappear. THE FUTURE OF HEALTH CARE MANAGEMENT IN CANADA Despite these looming dark clouds, macromanagement is likely to con- tinue to dominate the Canadian health care economy for the foreseeable future. But nothing inherent in the Canadian approach guarantees efficien- cy or effectiveness in the use of health care resources. The outcome-of what gets done, to whom, where, by whom, with what complementary re- sources, and with what effects is not necessarily, or even likely to be, the outcome that one might observe if one were able "objectively" to rank all possible interventions and then allocate resources to them up to the current global expenditure ceilings. Macromanagement may be crucial to global cost control, but it is not sufficient to produce the patterns of care sought by micromanagement initiatives. Canada's record with macromanagement, if viewed from the perspec- tive of cost control, is quite good in comparison to the United States, but unimpressive in comparison to any other country. To a large extent, the current condition of the health care economy is a product of medical educa- tion and funding decisions (capital commitments) made in the late 1960s and early 1970s. The demographic projections on which those medical school enrollment decisions were based were made in the early 1960s; it has been known for nearly 20 years that they were grossly in error (too high by about 35 percent by 1991; Barer and Stoddart, 19911. But capital commit- ment is politically far easier than capital contraction. The incentives to encourage adjustment in the face of new demographic information simply were not in place. Macromanagement in Canada over the next decade may begin to look more like that in the United Kingdom, as budgeting and management re- sponsibilities are decentralized. But the challenges for smaller managerial units will be no less daunting than those presently faced by the centralized provincial authorities unless those authorities are willing at the same time to make some hard capital decisions that cannot be made locally. Yet what of micromanagement? It is at best misleading, at worst dis- honest, to promote the notion that micromanagement, if only there was enough of it, would achieve macrocontrol. As Wennberg (1990) has noted, "The inventive nature of the medical mind, the endless possibilities for plausible theories, and the urge all physicians feel to work for and be help- ful to their patients combine to make it impossible for outcomes research to keep up with the flow of new medical ideas" (p. 1204~. Grumbach and Bodenheimer (1990) describe this phenomenon as the "continual attempt
116 MORRIS L. BARER AND ROBERT G. EVANS [by physicians] to extend the borders of the medical pasture" (p. 121). Micromanaged care will continue to change the shape and composition of the health care pie; it is unlikely, however, to have much effect on its size. In this, it seems remarkably (and depressingly) similar to basic medical research. In that sphere, as each "insulting" organism is identified and a clinical assault mounted, three others emerge.l4 Continued uncertainty about the specifics of appropriate care seems likely to hinder the effort for some time to come (Grumbach and Bodenheimer, 1990~. Clinical management in Canada is the object of increasing interest and effort, but resources are not being channeled into this arena because of a belief that it will replace the need for macromanagement. Rather, outcomes research, the development of clinical practice guidelines, clinical compe- tence assurance activities, and the like are seen in the Canadian context as tools for guiding resource allocation and organization within global bud- gets not as replacements for those budgets. No amount of micromanage- ment or outcomes research can tell a society how much of its scarce real resources should be devoted to health care. There will always be more inter- ventions that produce "epsilon effects" than can possibly be evaluated.l5 A commitment to maintaining a system of global budgets still leaves the problem of setting and controlling them. The present Canadian ap- proach of bilateral negotiation and, in the end and if necessary, imposition may not be sustainable politically in the absence of new policy directions for capital management. A reduction in the rate of production of new physicians seems an essential starting point. But even that simple step will require a new, heretofore elusive, national consensus. Physicians are a national resource, budgets a provincial responsibility. The elements push- ing stakeholders toward such a consensus may now be there: the present fiscal climate, a growing understanding of the broader (non-health-care) determinants of population health (Evans and Stoddart, 1990), the increas- ing range of questions raised by research on outcomes and procedural var~a- tions regarding the population benefits of ever-larger allocations of limited public funds to health care (Roos and Roos, 1990), and a common sense of political fatigue from the prospect of having to manage an ever-larger med- ical community. The hope is that these elements will finally come together 140pportunistic infections of AIDS patients offer the clearest example of this phenomenon. As the patient's T-cell count progressively declines, new infections gain a foothold. Increas- ing research efforts find new treatments for each, which are effective only until the count falls sufficiently to bring on the next infection. 15What is transpiring in the United States suggests that micromanagement is being saddled with a far more onerous burden. Outcomes research and managed care appear to be the replacement for the lost promise of, first, more regulation and, then, more competition (which was, in fact, more regulation in a different package) as vehicles of cost control.
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 117 in Canada to produce a new contractionist era in medical resource policy. It will not have come a year too soon. REFERENCES Abel-Smith, B. 1985. Who is the odd man out: The experience of Western Europe in containing the costs of health care. Milbank Memorial Fund Quarterly 63:1-17. Anderson, G. M., Newhouse, J. P., and Roos, L. L. 1989. Hospital care for elderly patients with diseases of the circulatory system: A comparison of hospital use in the United States and Canada. New England Journal of Medicine 321:1443-1448. Association of Canadian Medical Colleges. 1990. Canadian Medical Education Statistics 1990. Ottawa: The Association. Barer, M. L. 1988. Regulating physician supply: The evolution of British Columbia's Bill 41. Journal of Health Politics, Policy and Law 13:1-25. Barer, M. L., and Evans, R. G. 1986. Riding north on a south-bound horse? Expenditures, prices, utilization and incomes in the Canadian health care system. In: R. G. Evans and G. L. Stoddart, eds. Medicare at Maturity: Achievements, Lessons and Challenges. Calgary: University of Calgary Press, pp. 53-163. Barer, M. L., and Evans, R. G. 1990. Reflections on the Financing of Hospital Capital: A Canadian Perspective. HPRU Paper No. 90: 17D. Vancouver: University of British Columbia, Health Policy Research Unit, Division of Health Services Research and Devel- opment. Barer, M. L., and Stoddart, G. L. 1991. Toward Integrated Medical Resource Policies for Canada: Background Document. Discussion Paper 91:6D, Centre for Health Services and Policy Research, University of British Columbia; and Working Paper 91-7, Centre for Health Economics and Policy Analysis, McMaster University. Barer, M. L., Evans, R. G., and Labelle, R. 1988. Fee controls as cost control: Tales from the frozen north. Milbank Quarterly 66:1-64. Barer, M. L., Evans, R. G., and Haazen, D. S. 1992. The effects of medical care policy in British Columbia: Utilization trends in the 1980s. In: R. Deber and G. Thompson, eds. Restructuring Canada's Health Services System: How Do We Get There From Here? Toronto: University of Toronto Press, pp. 13-17. Bayne, L., and Walker, M. 1989. Capital Equipment Acquisition: A Discussion Paper. Vancouver: Stevenson Kellogg Ernst and Whinney. Birch, S., and Chambers, S. 1990. Development and application of a needs-based methodolo- gy for allocating health-care resources among populations at the county level. Unpub- lished mimeo. Hamilton: McMaster University, Centre for Health Economics and Policy Analysis. Blendon, R. J., and Taylor, H. 1989. Views on health care: Public opinion in three nations. Health Affairs 8:149-157. British Columbia, Royal Commission on Health Care and Costs. 1991. Closer to Home. Vol. 2, Report. Victoria, B.C.: Crown Publishers. Callahan, D. 1987. Setting Limits: Medical Goals in an Aging Society. New York: Simon and Schuster. Contandriopoulos, A.-P. 1986. Cost containment through payment mechanisms: The Quebec experience. Journal of Public Health Policy 72:224-238. Deber, R. B., Thompson, G. G., and Leatt, P. 1988. Technology acquisition in Canada: Control in a regulated market. International Journal of Technology Assessment in Health Care 4:185-206. Detsky, A. S., Stacey, S. R., and Bombardier, C. 1983. The effectiveness of a regulatory
118 MORRIS L. BARER AND ROBERT G. EVANS strategy in containing hospital costs: The Ontario experience, 1967-1981. New England Journal of Medicine 309: 151-159. Detsky, A. S., O'Rourke, K., Naylor, C. D., Stacey, S. R., and Kitchens, J. M. 1990. Contain- ing Ontario's hospital costs under universal insurance in the 1980s: What was the record? Canadian Medical Association Journal 142:565-572. Enterline, P. E., Salter, V., McDonald, A. D., McDonald, J. C. 1973. The distribution of medical services before and after"free" medical care the Quebec experience. New England Journal of Medicine 289:1174-1178. Evans, R. G. 1984. Strained Mercy: The Economics of Canadian Health Care. Toronto: Butterworths. Evans, R. G. 1986. Finding the levers, finding the courage: Lessons from cost containment in North America. Journal of Health Politics, Policy and Law 11:585-616. Evans, R. G. 1988. "We'll take care of it for you": Health care in the Canadian community. Daedalus 117:155-189. Evans, R. G. 1990. Accessible, acceptable, and affordable: Financing health care in Canada. In: Improving Access to Affordable Health Care. The Richard and Hinda Rosenthal Lectures. Washington, D.C.: Institute of Medicine, pp. 7-47. Evans, R. G., and Stoddart, G. L. 1990. Producing health, consuming health care. Social Science and Medicine 31(12):1347-1363. Evans, R. G., Lomas, J., Barer, M. L., Labelle, R. J., Fooks, C., Stoddart, G. L., et al. 1989. Controlling health expenditures the Canadian reality. New England Journal of Medi- cine 320:571-577. Evans, R. G., Barer, M. L., and Hertzman, C. 1991. The twenty year experiment: Accounting for, explaining, and evaluating health care cost containment in Canada and the United States. In: G. S. Omenn, J. E. Fielding, and L. B. Lave, eds. Annual Review of Public Health, Vol. 12. Palo Alto, Calif.: Annual Reviews, Inc., pp. 481-518. Grumbach, K., and Bodenheimer, T. 1990. Reins or fences: A physician's view of cost containment. Health Affairs 9(4):120- 126. Hadorn, D. C., and Brook, R. H. 1991. The health care resource allocation debate: Defining our terms. Paper presented at the conference, "Creating a Fair and Reasonable Basic Benefit Plan Using Clinical Guidelines," sponsored by the California Public Employees' Retirement System's Health Benefits Advisory Council, Sacramento, California, April 24-26. Hughes, J. S. 1991. How well has Canada contained the costs of doctoring? Journal of the American Medical Association 265:2347-2351. Iglehart, J. K. 1986a. Canada's health care system (Part 1). New England Journal of Medi- cine 315:202-208. Iglehart, J. K. 1986b. Canada's health care system (Part 2). New England Journal of Medicine 315:778-784. Iglehart, J. K. 1990. Canada's health care system faces its problems. New England Journal of Medicine 322:562-568. Kinzer, D. M. 1990. Universal entitlement to health care: Can we get there from here? New England Journal of Medicine 322:467-470. Lepofsky, M. D. 1989. A problematic judicial foray into legislative policy-making: Wilson v. B.C. Medical Services Commission. Canadian Bar Review 68:614-629. Lomas, J. 1990. Finding audiences, changing beliefs: The structure of research use in Canadian health policy. Journal of Health Politics, Policy and Law 15:525-542. Lomas, J., Fooks, C., Rice, T., and Labelle, R. J. 1989. Paying physicians in Canada: Minding our Ps and Qs. Health Affairs 8(1):80-102. Marmor, T. R., Mashaw, J. L., and Harvey, P. L. 1990. America's Misunderstood Welfare State: Persistent Myths, Enduring Realities. New York: Basic Books. McAuley, R. G., Paul, W. M., Morrison, G. H., Beckett, R. F., and Goldsmith, C. H. 1990.
MANAGING CAPITAL, CAPACITY, AND COSTS IN CANADA 119 Five-year results of the peer assessment program of the College of Physicians and Sur- geons of Ontario. Canadian Medical Association Journal 143:1193-1199. Ministere de la Sante et des Services Sociaux. 1990. Une Reforme Axee Sur Le Citoyen. Quebec City. Neuschler, E. 1990. Canadian Health Care: The Implications of Public Health Insurance. Research Bulletin. Washington, D.C.: Health Insurance Association of America. Nova Scotia, Royal Commission on Health Care. 1989. Towards a New Strategy (report). Halifax, N.S.: The Queen's Printer for Nova Scotia. Roos, N. P., and Roos, L. L. 1990. Limiting Medicine. Document No. 17B. Toronto: Program in Population Health, Canadian Institute for Advanced Research. Rublee, D. A. 1989. Medical technology in Canada, Germany, and the United States. Health Affairs 8(3):178-181. Schieber, G. J., and Poullier, J.-P. 1991. International health spending: Issues and trends. Health Affairs 10(Spring):106-116. Schroeder, S. 1984. Western European responses to physician oversupply. Journal of the American Medical Association 252:373-384. Viefhues, H., ed. 1988. Medical Manpower in the European Community. New York: Spring- er-Verlag. Wennberg, J. E. 1990. Outcomes research, cost containment, and the fear of health care rationing. New England Journal of Medicine 323:1202-1204.
