UBC Faculty Research and Publications

Accessible, acceptable and affordable financing health care in Canada Evans, R.G. Sep 30, 1990

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-Evans_RG_Accessible_acceptable_affordable.pdf [ 1.7MB ]
JSON: 52383-1.0075922.json
JSON-LD: 52383-1.0075922-ld.json
RDF/XML (Pretty): 52383-1.0075922-rdf.xml
RDF/JSON: 52383-1.0075922-rdf.json
Turtle: 52383-1.0075922-turtle.txt
N-Triples: 52383-1.0075922-rdf-ntriples.txt
Original Record: 52383-1.0075922-source.json
Full Text

Full Text

ACCESSIBLE, ACCEPTABLE AND AFFORDABLE:FINANCING HEALTH CARE IN CANADAR.G. EvansHPRU 90:19D September 1990HEALTH POllCY RESEARCH UNITDivision of Health Services Research and DevelopmentThe University of British ColumbiaVancouver, B.C.V6T lZ6Given as the Rosenthal Lecture, Institute of Medicine, National Academy of Science,Washington, D.C., April 1989Research underlying this paper has been supported by a National Health Scienctist Award,Health and Welfare canada, and by a FelIowship from the canadian Institute for AdvancedResearch.ACCESSIBLE, ACCEPTABLE AND AFFORDABLE:Financing Health Care in CanadaThe Most Popular Public Program in CanadaIn the fall of 1988 there was a federal election in Canada, and theprincipal issue in debate during a very lively campaign was the upcomingfree trade agreement with the United States. This agreement was viewed asmuch more than a commercial treaty with our largest trading partner. It wasfeared as potentially leading to a fundamental change in our whole sense ofnational identity, which has always been powerfully affected by ourrelationship (close but not too close) with the overwhelming presence of theUni ted States. The Progressive Conservative government was committed tosigning the agreement, and was re-elected on that platform, though with wellshort of a majority of the popular vote.The government came very close to defeat, however, during a remarkablethree day period in the middle of the campaign in which the free trade issuebecame entangled with health care funding. The opposition parties began toattack the agreement on the ground that it would lead to the destruction ofthe Canadian health insurance system and its replacement with something moresimilar to the American approach. The public responded to this prospectwith a massive swing against the government, almost overnight, of about 10.0to 15.0 percent in the opinion polls. The opposition Liberals emerged infront. A desperate political damage control exercise by the governmentconvinced enough of the electorate that there was in fact no connectionb e twe e n free trade and health insurance, and the "t.LdaL wave" slowlyreceded.It is hard to think of a more reliable indicator of the extent andintensity of public support for the Canadian system of health care funding.Any government which was widely perceived to be putting that system at risk,would become an ex-government at the next opportunity. No one imagines thatthe system is perfect; it has been surrounded by political controversy sinceits beginnings and is likely to remain so. But that controversy does not2extend to the fundamental principles; after more than 20 years of experienceuniversal Medicare has a broader and firmer base of support than any otherCanadian institution. There is no serious political voice calling forabandonment or major change, and the 1988 election re-emphasized why.But the extraordinary level of interest in the Canadian healthinsurance system which has recently appeared in the United States suggeststhat our concerns may have been misplaced. We feared that Americans wouldregard the Canadian system as an unfair advantage for our firms in theinternational marketplace, and would demand that it be dismantled as acondition of the free trade agreement, to ensure a l'level playing field'!.It did not occur to us that the United States might instead want to tradehealth care systems! (If we were to do so, of course, we would be veryfoolish to trade at par. We ought to charge quite a healthy premium.)The massive popular support wi thin Canada for our form of healthinsurance is a political fact, and its relevance to Americans is simply thatthose who live in the system overwhelmingly approve of it. It meets thetest of public opinion, and political support. That might not necessarilybe a recommendation - Canadians might be wrong, or might simply not knowwhat they are missing. After all, it appears that citizens in mostindustrialized countries, even the poor benighted Brits, are stronglyattached to their particular health care systems, just as they are to theirindividual physicians. The common American rhetoric, that whatever itsproblems, the American health care system is still the finest in the world,presumably implies that Canadians are misinformed.'But it is relevant, I think, to note that Canadians are comparatively1 The somewhat less grandiose claim that "At, its best, Americanmedicine .... etc.", while less self-evidently false, contains a rhetoricalboomerang. This could perfectly well describe a system in which a smallprivileged class received the world's best, and the care of the rest of thepopulation was mediocre or worse. It is thus remarkable that anyone wouldregard this claim as grounds for pride, without further elaboration as towhat proportion of American medicine meets such a standard, how far shortthe rest falls, and how access to "t.he be s t " is determined.3well informed about matters American, for reasons of simple proximity andrelative size. Few Canadians are out of range of American television, mosthave travelled in the United States, and all are immersed in "Northnevertheless likely to be aAmerican" culture. do have a picture of American health care which,if not completeThey(much less completely accurate, who has that?), isgood bit clearer than the typical Americanpicture of Canada. And they know, very firmly, that they do not like whatthey see.Going to the other side of the mirror, the finding that a substantialmajority of a randomly selected poll of Americans expressed a preference fora Canadian-style system - at least as briefly described to them - appearswholly unprecedented in international comparisons (Blendon, 1989). Again,those polled may be wrong, in the sense that if most Americans really hadto live with such a system, they would be much less satisfied. But theineluctable fact - and it appears to be a fact - is that Americans are nothappy with what they have (Taylor, 1990).2 Canadians are.Affordability and Accessibility: Defined How? Judged by Whom?The design of health policy, however, is not judged solely bycomparative popularity polls. Analysts and commentators look for objectivefacts. (Although they neglect at their peril the reality that to theirpolitical masters public opinion is fact.)what contribution different systems make tothey serve. The true tests of a good systemIdeally one would like to knowthe health of the populationswould then be, "Does it work?lt2 As Taylor points out, most Americans do seem to be happy with theirown health care; it is the system as a whole which they report asunsatisfactory. Since a substantial majority of Americans have goodinsurance coverage and ready access to services, this is no paradox. Careis both accessible and affordable for most individual Americans; it is theglobal cost and the large minority who have inadequate or no access whichlead to systemic concerns.4as well as "Do the population like j.t?"3 The test of an innovation wouldbe its potential for improvement on some combination of these measures.Outcome data being notoriously inadequate at the individual level, letalone for entire populations, we fall back upon such intermediate measuresas "affordability" and "acce s s fb i.L'i t.y v. These have been of par t Lcul arconcern to Americans I because the various health care funding systemsoperating in the United States make up a package unique among industrializedcountries, both in the level and rate of escalation of their costs (Schieberand Poullier, 1989), and in the proportion of the American population whichhas either grossly inadequate, or simply no, form of public or privatehealth insurance (Short et al., 1989). Canada, along with all the countriesof Western Europe, has achieved the combination of lower and less rapidlyescalating costs, and broader population coverage, which most Americansappear to regard as proximate but perhaps (for them) unattainable goals.Such goals, it should be noted, are in this context characteristics ofa health care funding system, aggregating the experience of individuals andorganizations. An individual with severe health problems and limitedpersonal resources or insurance coverage might find care inaccessiblebecause it was unaffordable for her. But the affordability of the systemrefers to its overall costs relative to the resources and priorities of thesociety as a whole. Whether or not the American health care system is insome sense "una f fo r dab l.e " for the United States is a separate issue fromwhether needed care is "una f fo r dab l.e " for some Americans, or for theiremployers; either could be true without the other.Of particular importance, because frequently a source of confusion, thecosts of a system do not become more or less affordable by being transferredfrom public to private budgets, or back again. At the end of the day, the3 "Affordability" drops out of consideration as an independentcriterion at this level of generality, because the program consists of itsbenefi ts and its costs. If the program really "works", and if thepopulation is happy with the balance of benefits and costs - then clearlyit is affordable. The population served have chosen to afford it.5people of a country pay for the costs of their own care, and it is the totalthat matters. The total costs of health care do not become less of a burdenon American society, more affordable, simply because a substantially lowerproportion of these costs are funded through public budgets than in othercountries.Similarly the accessibility of a health care system reflects theoverall response of that system to the needs of the population it isintended to serve, and may be impeded by a variety of different barriers ofwhich out-of-pocket costs are only the most easily identified.There is, however, a danger that these intermediate criteria may beinterpreted as more "ob j e c t Lve " and/or more readily measureable - more"scientific" than the ultimate goals of health effect and populationsatisfaction. Certainly one can measure such indicators of cost and use asdollars spent, prices, numbers of treatments, persons enrolled, terms ofcoverage, etc., and in principle with great precision. But words likeaffordability or accessibility of care go well beyond measurement, andembody implicit values and choices - judgements - which cannot be derivedfrom the data themselves.What is affordable depends on one's preferences and priorities as wellas on costs; very rarely are wealthy societies constrained in any particularendeavour by absolute shortages of resources. But the setting of socialpriorities is qUintessentially political, not "scientific ll . The "expertl1,medical or economic, has an important task in trying to layout the options,as accurately and honestly as possible. But the actual setting ofpriorities, the making of choices, is the role of the citizen and voter.The expert qua expert is no better equipped than anyone else for this task,and is entitled to one vote. 44 This statement presupposes that a prior decision hAS been made todeal with the allocation question politically, rather t.h an through themarketplace in which people have different numbers of vot0S according totheir wealth. The justification for this approach is both that in the realworld no society, not even the United States, has been willing to let the6Similarly accessibility begs the question of what is to be accessible,to whom, and under what circumstances? Accessibility per se is really ameans to one or more ends, not an end in itself. The end that is soughtthrough health care is health, and the "accessibility" of health care isvalued principally on the belief that such care will contribute to someone'shealth. 5The connection between health care and health is, however, highlyuncertain and contentious; students of the effectiveness of health careemphasize that most of it is at best unevaluated and that even interventionswhich are demonstrably effective in specific circumstances are very widelymisapplied (Banta et al., 1981; Feeny et al., 1986). It follows thataccessibili ty as a normative concept, a proximate objective, cannot beidentified or compared across systems simply on the basis of a set ofmeasurements of utilization. One needs to know what forms of care are beingprovided or denied to persons in particular circumstances, in order todetermine whether differences in access to care correspond in any systematicway to differences in access to health. Better access to useless or harmfulcare is not in general a cause for congratulation.But useless for what? Such a statement presumes an unambiguous andgenerally agreed upon concept of "health H against which interventions canbe evaluated. For some aspects of health this is a reasonableapproximation, but other dimensions are highly debatable and culture-dependent. There will be disagreement among individuals and particularlyacross cultures as to the nature and extent of the "hea l.t.h" whichaccessibility to care may promote. The meaning and the value ofmarketplace govern health care matters, and that, if any societyaggregate issues of "a f fo r dab i Li t.y" and "acce s s Ib Ll.Lt.y"meaningless. Who, other than individual buyers, worriesaccessibility and affordability of Mercedes-Benzes?did so,wouldaboutthebethe5 This is not the only reason; considerations of social solidarity andthe symbolism of caring may justify promoting access to care of dubious orno therapeutic value. But anticipated health benefits are the centralissue.7"accessibility!! to particular states as well as services will then alsovary.Nor is this only an abstract possibility. The individual undergoingregular monitoring of his serum cholesterol level, and on a strict dietaryand drug regimen for life, may be regarded by one person as "heal thy"because his probability of death from heart disease is reduced. But anothermay see the same individual as II sick" , because he is now both physically andpsychologically dependent on care - morbidly concerned with his own health.Should a "good" system promote, or even provide, universal access tocholesterol screening? Hume's Law applies; one cannot derive Il ought II fromIlis ll. And words like affordable or accessible are inherently "ought" words Iladen with normative content. What ought particular people to receive? Andhow much should they or others be willing to pay for this?With this caveat, we shall sketch out some of the basic facts andcentral features of the Canadian health insurance system, notingparticularly the principal similarities with and differences from the formsof funding in the United States. Structural differences then lead intodifferences in performance, although the connection provides fertile groundfor interpretation and disagreement over precisely why things have evolveddifferently on each side of the border - or for that matter how differentthey really are.Differences in performance are then matters for evaluation, and we willconsider some of the problems of interpretation which arise in moving from"c o s t;" to "a f fo r dab L'Li t.y ?, and from "cove r age " to "acce s s Lb i Lfuy ". Theseproblems are not insurmountable, but they do involve certain unavoidablevalue judgements which provide a context for the choices inherent in anyprocess of health care funding. Subject to this qualification, however, itdoes appear that the Canadian system of health care finance is both moreaccessible and more affordable than that of the United States, and that itsadvantage is growing over time. The two populations are not wrong in theirrespective evaluations of their systems.8Health Care Funding in Canada and in BriefThe "stylized facts" of health care funding in Canada, stripped of amultitude of fascinating but inessential footnotes, are as follows. Canadadoes not have "socialized medicine ll , but it does have socialized insurance,for hospital care and physicians' services. Each of the 10 provincesoperates a payment system which reimburses private, fee-for-servicephysicians for the care they provide to their patients, according to auniform fee schedule negotiated at periodic intervals between the provincialmedical association and the provincial government. The schedules differacross provinces. Physicians have admitting privileges in hospitals run bycommunity or municipal boards; these hospitals derive their operatingfunding from annual global budgets negotiated with the provincial ministriesof health.The costs of this system are met by each province out of its generaltax revenue. But the federal government also makes a substantialcontribution to the provinces, currently about 40 percent on average ofprogram costs, in the form of a block grant rather than as a share ofaudited costs. The federal government requires that the provincial plansmeet certain conditions to be eligible for these funds, hence the closesimilarity among provincial plans despite their technical independence.In particular, the provincial plans must cover 100 percent of theirpopulations, for all ltmedically necessary" services. This is significantin those two provinces which still require their residents to pay premiumsfor health care. (The revenue from these premiums is not "ea rrnarked"specifically for health care, but is in effect pooled with generalprovincial revenues.) One cannot be denied services for failure to paypremiums; such "prerniumsll, which are also unrelated to risk status, are infact a form of poll tax. (Most people do not, however, know that theycannot be denied care, and provincial governments do not try to disseminatethe information.)9Furthermore I while the federal conditions do not ban charges topatients, they do provide that a province's grant must be reduced by anyamount which the province charges, or permits to be charged, to patients forinsured services. In response to this I provinces have in various waysdiscouraged physicians from extra-billing patients in amounts above theprovincial fee schedule, and do not impose charges for hospital services.(Patients in long~terrn care institutions, however, are charged a daily ratecalculated to recoup most of the public minimum pension. And patients inacute care, who in the judgement of their physicians do not require semi­private or private room care, may nevertheless choose such care on paymentof a "preferred accommodation differential II • If medically required I ofcourse, such care is free.)Accordingly, all residents of Canada are fully insured for allI1 medically necessaryl1 hospital and medical services. s Access is universal,and complete, in the sense that there are no financial barriers to care.,,'hile it is clear that this does not exhaust the possible content of"acces s Lb i Lty'", it does mean that the phenomena of medical indigence andbankruptcy, uncompensated care, patient dumping and other forms of financialdiscrimination simply do not exist. The anxiety and distress suffered byso many individual Americans as they contemplate the potential or actualimpact of ill-health on their economic situation, has no counterpart inCanada, while those responsible for managing or paying for the system do nothave to cope with the problems and costs raised by the multiplicity ofmanoeuvres to pass costs on to someone else.Universality is Cheaper: A Paradox, But a Small OneNor is it the case, as so often claimed in the United States, thatuniversality implies national bankruptcy. or at least "unaffordability".6 The significance of that qualifying phrase has never been explored.It excludes elective cosmetic surgery, though obviously repair of traumaticdamage or congenital defects is covered. Conceivably the legislative phrasecould serve as a basis for "de LnsurLng " services evaluated by expert opinionas having no actual or anticipated benefit, but this has not been tried.10The assumption that there is an inevitable trade-off between accessibilityand affordability is one of the more deceptive and disabling fallaciesinjected into public debate, often by economists suffering from a bad caseof a priorism and a low level of comparative information.The fact is that Canadians spend substantially less of their nationalincome on health care than do Americans, about one dollar in twelve comparedwith nearly one dollar in eight south of the border. In proportionateterms, this amounts to a saving of about one - quarter. And all of thisdifference is in the total costs of hospital and medical care services ­those components of national health expenditure covered under the universalpublic Medicare program and in the overhead costs of the insuranceprograms themselves. Such items as dentistry, out-af-hospital drugs, andpublic health are not covered by that program, and their costs do not intotal differ very much from South to North (Barer and Evans, 1986; Evans,1986).The Canada-United States divergence, which now amounts to twopercentage points of Gross National Product (GNP), or in American termsabout 100 billion ?ollars, has emerged in the two decades since the Canadiansystem was fully established. The last province entered Medicare on January1, 1971; in that year both Canada and the United States spent roughly equalshares of their national income on health care. Furthermore, the patternof cost escalation in the two countries had been virtually identical overthe previous 20 years. Between 1971 and 1987 the health spending share inthe United States rose further, from 7.6 percent of GNP to 11.1 percent,while the corresponding Canadian increase was from 7.4 percent to 9.0percent. And virtually all the Canadian increase occurred in one year ofdeep general recession 1982 when real national income fell sharply(Canada, 1987; Levit et al., 1989; unpublished data from Health and WelfareCanada, 1989).The Canadian experience thus demonstrates that, far from being inconflict, affordability and accessibility are complementary goals. It is11the universal system, channelling all reimbursement through a single payer,which has made both possible. More detailed analysis of the functioning ofthe health care systems on both sides of the border confirms this view.As further evidence! most countries in Western Europe have since 1980stabilized the growth of their health care sectors to a roughly constantshare of national income. All have universal, public or quasi-public healthinsurance programs. Sweden and Denmark have actually significantly reducedthe share of health spending, from 9.5 percent and 6.8 percent of GrossDomestic Product (GDP) respectively in 1980, to 9.0 percent and 6.0 percentin 1987. Sweden began the decade with the highest share reported among thenations of the Organization for Economic Cooperation and Development (OECD);but Denmark in 1980 was already below average. For the OECD as a whole, theaverage share of national income spent on health has moved from 7.0 percentof GDP in 1980 to 7.3 percent in 1987 - but this average includes the UnitedStates (Schieber and Poullier, 1989). Canada is no longer unique, althoughwe do have the longest record of cost control.But the total costs of health care in any country are also bydefinition the total incomes earned from the provision of health care. Thiselementary mathematical identity is extremely important to the understandingof the air of continuous controversy surrounding the system which I haveportrayed as affordable, accessible, and overwhelmingly popular with thecitizenry of Canada. That controversy, which is real, long-term, and likelyto continue indefinitely, may mislead some external observers (and even someinternal observers) into wondering if the system is collapsing.Hospital and medical care is "free ll to the -user, but of course not tothe society as a whole. And while the overall cost is much lower than inthe United States, the fact that provincial treasuries bear all of that costplaces them in continuing conflict with the physicians, nurses, hospitals,and other providers of health care for whom no amount of spending is everqui te enough. Funding health care is the largest and most pol i tica11yvolatile respons ibili ty of any provincial government, with the greatest12political dangers. Precisely because the controls on spending work, thepayment systems are a lightning rod for professional dissatisfaction. Asa group, providers have learned to live with cost control; but they havenever accepted it in principle, and it would be naive to imagine that theyever will.But this inherent conflict of interest between payers and providers iscommon to all financing systems .. And the fact that a sense of financialII crisis'! is observed in so many national systems I at very different levelsof funding in both absolute and relative terms, suggests that controversyis the result, not of spending levels per se, but of any attempts to containcost growth, regardless of the level of spending. Controversy is the priceof affordability (Tuohy, 1986; Evans, Lomas et al., 1989; Evans, 1990a,b).As the American example shows, the price must be paid even for unsuccessfulefforts at control.But it would be quite wrong to conclude, as the American media tend todo, that every funding system has problems and therefore all are in the sameboat. While all struggle with the same problems, some struggle much moresuccessfully than others. Moreover the costs of the struggle are borne verydifferently. In Canada, providers and payers fight, patients are in theaudience. In the United States, the patient (or the employer) is down inthe ring struggling with providers, and it is a much less equal contest.It does not follow, of course, that the Canadian system is ideal andthat Americans should immediately try to import it. Each country has todevelop a system of health care funding and delivery consistent with its ownculture and history, and our histories and cultures are different. But ifAmericans really want to achieve. operating results similar to Canada / e ,controlling overall costs and covering the whole population, then they willhave to, in their own way, develop mechanisms for imposing the kinds oflimitations that exist in Canada. The institutional features may bedifferent, but they will have to do the same tasks.13Everybody Is Doing It - Canada Is Just NearerThis generalization is supported by the Western European experience.As noted above, the maj ori ty of developed countries have succeeded instabilizing their health care costs as a share of national income. Theyhave done so in very different funding systems but all provide more or lessuniversal coverage, either through a single payer, or through a number ofpayers which are then co-ordinated by legislation and regulation. Thecoordinated payment system is then the mechanism through which various formsof controls are applied. The United States is now the outlier, the onecountry that has not succeeded in achieving stability. And the outstandingdifference is that the United States is the one country that has not goneto some form of universal coverage (Abel-Smith, 1985).The critical linkage seems to be between universal coverage, and 501e­source, single-payer funding. As noted, this may be achieved either by asingle payer in fact, as in Canada or Sweden, or by multiple but legallycoordinated payers, as in Germany, or by a handful of payers each withexclusive jurisdiction. One could certainly imagine a system of largenumbers of uncoordinated payers which was extended (at least briefly) toprovide universal coverage; this appears to be the solution advocated by theAmerican Medical Association (AMA) "Universal access, not universalinsurance" - to deal with the large numbers of uninsured (Todd, 1989).Such a system would generate even more rapid escalation of costs - i.e.provider incomes - than the present American system, while preserving boththe financial and the clinical autonomy of providers and the impotence ofpayers. It would add more money to an already over- inflated system, andmore bureaucratic overheads to run yet another program or programs. Butonce it has been decided that everyone is to be covered, the whole apparatusof private insurance (designed in a private marketplace to determine whomto cover, at what price, and whom to exclude) becomes complete waste motion.The higher cost and dynamic instability of such an approach makes clear why,in practice, universal coverage is always associated with sole-source14funding, de jure or de facto.While universality of coverage and sole-source funding are, as far aswe know now, preconditions for cost control, it also appears that costcontrol reinforces universality. The absence of control, in the Americanenvironment, creates strong incentives for those who bear the ever­increasing costs to try to pass them on to others. Governments andemployers are thus tempted, if not forced, to increase the premiums chargedto those covered, while cutting back on the scope of coverage by imposinglarger co-payments on users of care, and/or pushing people off their rolls.Contrary to the naive predictions of market economists, this has not beeneffective in mitigating the escalation of costs, but it does addsignificantly to the human cost of ill-health by combining financial insultwith health injury. When the lifeboat is leaking, one reaction is to throwpeople overboard rather than to try to plug the leak. The universalityadvocated by Todd (1989) would, if ever achieved, begin to crumble in thisway almost immediately.What Seems to Be the Problem, Sam?We began the discussion of health care systems by defining a "good"system as one making a positive contribution to the health of the populationit served, and popular with that population. We then promptly retreated tothe intermediate and somewhat more measureable criteria of affordabiliityand accessibility. What has demonstrably been achieved in Canada andWestern Europe, however, is cost control - at least relative to the UnitedStates - and the almost universal removal of financial barriers to healthcare utilization. As emphasized above, these are not necessarily equivalentto affordability and accessibility. The latter labels imply certainevaluative judgements on the desirability of the outcomes achieved,judgements which do not follow automatically.There is no magic level of expenditure beyond which health care becomesuunaffordable ll • Americans can obviously "afford" to spend over 11 percent15of their national income, unambiguously demonstrated by the fact that theyare spending it now. A number of other countries ~ Canada, France, WestGermany, the Netherlands spend between 8.0 and 9.0 percent, and Sweden hasmoved down to this range since 1980. Britain, Australia, Denmark, andJapan, by contrast, spend much less on health care - between 6.0 and 7.0percent of their national income - and they too worry about affordability.Countries do not spend what they do as a result of some explicitdecision that that level is "right", although Denmark and Sweden seem tohave made fairly broad-based collective decisions to bring their spendingdown, in relative terms, in the 1980s. But for other countries, and Canadain particular, the current spending share is simply the share which ourpreviously escalating costs had reached when we managed to develop botheffective instruments of control and the political will to use them - to putthe lid on. After that, holding the lid on at any level requires constantpolitical struggle with providers who are convinced that, whatever the levelof spending, more would always be better.One cannot necessarily assume that the level of spending is whollyarbitrary; different societies may have different spending propensities, andperhaps Canadians or Germans would not tolerate the health care system thatthey could buy for 6.0 percent of their aggregate incomes. But healthspending in Canada, from 1971 to 1981, remained quite close to the 7.5percent which it had reached when the universal public insurance system wascompleted. It moved up sharply to the 8.5 - 9.0 percent range in 1982, llQlbecause payers or the rest of the community had accepted providers'arguments for more, but simply because in the recession of that year,national income fell sharply. But the increase of one entire percentagepoint of national income in the early 1980s has made no difference whateverto the terms or the tone of the financing debate.The United States is of course in the special situation of having bothby far the world's highest costs, and as yet no effective instruments ofcontrol. But as a matter of arithmetic, normal rates of economic growth16would permit the United States to increase its share of income spent onhealth care for many more years (albeit slowly), and still have growingresources available for other things I consumption or 'investment. So whyshould that country be particularly concerned over the "affordability" ofhealth care?A commonly expressed concern is that the cost of health care borne byAmerican business is both heavy and growing rapidly, making Americanproducts too expensive to compete in international markets - or indeed athome. This is the point mentioned above, in reference to the Canada-UnitedStates free trade agreement, that our less expensive health care systemgives Canadians an lI unf a i r " advantage, and that Americans might argue thatCanadians should be forced to labour under the same handicaps that they haveimposed on themselves.On examination, however, this argument seems too simple. In the firstplace, a general cost disadvantage suffered by American firms can becompensated for through exchange rate adjustment. A decline in the valueof the American dollar can offset a rise in health care premiums - if thatis the source of competitive disadvantage.But secondly, employer-paid health care premiums are part of theoverall compensation package of labour, and it is that package, not anysingle component of it, which represents the cost of labour to the employer.If health care premiums are rising, why can that not be balanced by a lessrapid rise, or indeed a fall, in money wages? After all, surely workerswould realize that their total compensation is rising? If they prefer totake that increase primarily in the form of increasingly costly healthbenefits, why should that raise the firm's overall costs?Yet each of these responses is itself as naive as it is obvious Isuggesting that "affordability" runs somewhat deeper than a simple problemwith labour costs. The weakness of the "currency devaluation" response(apart from the impact of devaluation on relative asset holdings), is that17the growing burden of health care costs is very unevenly distributed amongemployers. It bears most heavily on long-established industries with maturework-forceshighest.older and retired workers whose heal th expenditures areNewly established firms, in new or old industries, have asignificant advantage. Thus a foreign producer of automobiles, for example,which sets up a plant in the United States can hire a younger work force,and will have no obligations to retirees. It will therefore have a built ­in cost advantage which no currency adjustment can touch.The root of the problem is the employer-based financing system.Employers with older work-forces and binding commitments to retirees musteither accept a permanent cost disadvantage, or try to push down the moneywages of their workers as their health care costs increase. This in turnmight be through lowering wages at all ages - resulting in their becomingless competitive in the market for younger workers -·or through reversingthe usual seniority system by paying workers less as they grow older andgenerate higher (expected) health care costs. None of these options is veryattractive. 7In a Canadian-style system, by contrast, the increasing health costs ofolder workers, like those of all other older individuals, are spread overthe community as a whole through the general tax system. The province ofQuebec also raises part of its revenue from payroll taxes, and Ontario hasannounced its intention to follow suit, but the tax rates are invariantacross employers. They do not impose a differential burden on particularfirms or industries. Thus the Canadian advantage from a lower cost systemoverall is accentuated in industries with mature work-forces.7 As Reinhardt (1989) points out, there is another option - writingdown the shareholders' equity to reflect the capitalized value of theprevious commitments. This is also unpopular, and in any case is only aone-time response. Even if there were no commitments to retired workers,firms with older workforces would still be faced with a choice betweenhigher costs, lower money wages, or lower benefit coverage relative to theircompetitors. The "perfectly competitive" marketplace would, one way oranother, impose lower take-home wages on older workers.18But what, apart from long-established industrial relations tradition(and the consequent probability of severe industrial unrest and associatedcosts), is wrong with reversing the seniority profile and paying olderworkers lower money wages as their health care insurance costs rise? Thisleads into the second point above I the "over a I L compensation package"argument. Older workers would not really be earning less, only taking theirearnings in a different form.Indeed this argument is more general. From the "total compensation"perspective Americans collectively are not worse off as their health caredividend of economic growthcosts escalate. They are simply- in thetaking their increasedform of health benefitsincome therather than asother types of consumption. Some analysts - economists mostly - have goneso far as to suggest that an empirical correlation between per capitanational income and share of income spent on health care indicates that,contrary to the traditional interpretation, health care is a luxury good onwhich wealthier nations Hchoose ll to spend relatively more.Far from being a problem, increased health spending is on this view thenatural consequence of growing wealth. As a subwtext, other countries withlower spending levels are then not ahead of the United States in being moresuccessful at control, but behind in that, when they are as wealthy, theywill spend as much. 8 Furthermore, this interpretation also implies that thewidespread American concern over the affordability of health costs isunjustified, and presumably that all those who share it are simplymisinformed or confused. Rather than wringing their hands, Americans shouldhappily open their wallets and celebrate the increased well-being whichhealth spending brings.8 However gratifying to American national pride, this interpretationhas had considerable difficulty with the international spread of successfulcost control in the 1980s, especially the pronounced fall in the share ofspending to national income in Sweden and Denmark, and its stability inJapan. The argument never did look very strong in Canada, where except forthe 1982 recession the health care share has been more or less stable since1971.19Can't Pay? Won't Pay? Don't Want to Pay! (And Can't Stop)Providers of care are in the main in enthusiastic agreement with thisline of argument, but few other Americans seem impressed. Just as workersstrongly resist accepting lower wages as their health premiums rise - hencethe competitive disadvantage of their employers ~ so Americans in generalseem by their behaviour to have rejected the idea that their increasedhealth spending is adequate compensation to induce them to give up otherconsumption. This could reflect a belief that additional health spendingis not in fact yielding "value for money", but is being dissipated in higherprovider incomes, overhead costs, and ineffective interventions.Alternatively, it may be that even "effective lr care, which results in someform of health benefit, is no longer considered worth the price. This isin fact a perfectly reasonable position, for low enough benefits and highenough prices - but very few are willing openly to admit it.Either way, the real source of distress is not that Americans cannotafford their health care, but that they do not want to. The socialpriorities of the United States, and the private priorities of individualAmericans, are in conflict with the amounts that are spent on health care.But the American institutional framework does not permit the balancing ofhealth care against those other priorities, or generate effective pressuresto promote "value for money". Instead it encourages or forces the expansionof health care, which is not valued as much as the other opportunities whichare foregone in consequence. The absence of any mechanisms for thecontainment of overall costs, or for the more equitable distribution ofthose costs over the whole community, means that Americans remain unhappywith the overall result. If that is not what the concern for affordabilitymeans, it is hard to think of any other logical content which it might begiven.At the level of the individual firm, the result is that profitabilityand competitive advantage corne to depend not just on the value of theproduct or the skill and effort of management and labour, but on the20historical accident of the age and health status of the work-force, and therelative conservatism or extravagance of the local health care providers.At the national level, the unwillingness of Ameripans collectively toforego other consumption as their health care costs rise may be part of theexplanation for the particularly anaemic American savings rate, relative notonly to that of Japan and other Pacific Rim countries, but even to that ofCanada. A difference of several percentage points of national income spenton health care - 3.0 percent more than most other industrialized countriesand 5.0 percent more than Japan· leaves room for a great deal of differencein savings. Business spending on health benefits has risen from 14.4percent of after-tax profits in 1965 to 94.2 percent in 1987 (Levit et al.,1989) .In summary, the American health care financing system seems mostresponsive to the priorities of providers of care, for whom ever~growingexpenditures represent ever-growing incomes. The users of and payers forthis care do not seem to value it as much; this is expressed both in theirwidespread complaints about "unaff or dabd Ldtiy " and in their resistance toreducing other forms of consumption to pay for this supposed benefit. Theirresistance, in turn, may be part of the explanation although thetentativeness of this part of the argument must be emphasized . for thedecline in American savings rates} such that investment levels can only bemaintained with increasing foreign borrowing. In this way the long-rungrowth of the American economy is mortgaged, in part to pay for theexpansion of health care. Such a scenario can certainly be described as"unaffo r dab i Li t.y" .On the other hand, the shrill cries of "unaffordabili ty" which ariseregularly from governments and other payers for health care in all the othercountries of the developed world are, ironically, part of the process ofcontrol. Since cost control is always and everywhere achieved in the teethof the providers of care, who are constantly struggling for quiteunderstandable reasons for expansion, it is necessary to mobilize a21political constituency for control. This is done, not by arguing indefiance of providers and usually patients as well - that more spending oncare would not be a good idea, but only that the cost pressures are sosevere that it is for the moment lI unaf f or dabl e " ,It is "unaffordable" in the United Kingdom, where 6.1 percent ofnational income is spent (Schieber and Poullier, 1989), or in' Canada , where8.6 percent is spent, or in the United States, where 11.2 percent Thedifference is that in most countries other than the United States there areinstitutional mechanisms capable of imposing control. The principal problemis the maintenance of political will, which in turn ultimately depends uponpopular support or at least acquiescence."You Don't Want Your Baby to Die, Do You?" .,. Doctor Knows BestAcross the political and rhetorical trenches from the advocates ofaffordability are the defenders of "accessibility" - again a confrontationobserved everywhere in the developed world. These are the beneficiaries ­providers and to some extent patients - of increased expenditure. Outsidethe United States, universal financing systems have largely removed theissue of individual ability to pay for care; the arguments over access nowturn on the adequacy of the total resources mobilized through the healthcare system, its capacity and level of output. All health care systemsoutside the United States are "underfunded" according to the officialspokesmen of those who work in them; this includes in particular theCanadian system which, according to the DECD statisticians, is the secondmost expensive in the world. None, it is claimed, have sufficient resourcesto meet the needs of those for whom they are supposed to care.The structure of the argument has become familiar, during 1989, to anyAmerican interested in health care. Defenders of the status quo in Americanmedicine have responded to the increased interest in universal publicfunding, and particularly the Canadian example, by charging that theCanadian system fails grievously in meeting the needs of the Canadian22population. or does so under conditions which would be unacceptable to mostAmericans. The process of cost control is alleged to result in long waitinglists and queues for care, unavailability of the most modern technology,depreciation of the physical plant, and a general deterioration of standardsin a stagnant, bureaucratic, "public utility l1 style of medical care.The apparent universal accessibility of health care in Canada is thusportrayed as a hollow boast; care may be II free 1\ at t.he point of service, butthe services are not really there when needed. Affordability has beengained, but at the cost of genuine accessibility; in this context financialaccessibility is simply a sleight-of-hand. An air of artisticverisimilitude is then added by selected anecdotes of particular Canadianpatients suffering, and perhaps even dying, as a result of care delayed ordenied, or fleeing to the United States for the services their own countrycannot or will not provide.This argument draws on two powerful rhetorical traditions. First,italleges implicitly that everyone is out of step but Uncle Sam. All thecountries of western Europe also have public or quasi-public fundingsystems, covering all or almost all of their populations. And all have nowsucceeded in limiting the growth of costs to a proportion of their nationalincome equal to or less than that in Canada. It follows that they must be"underfunding" their systems, and subjecting their populations to inadequatecare, to an even greater degree than is Canada. This sort of argument, thatAmerican differentness implies American superiority, has always been popularin the United States. A foreigner such as myself can only ask, if Americansreally are convinced that they have the world's finest health care system,or even an adequate one, why are so many of them so unhappy with the result?Secondly, the "underfunding"/"urunet needs II argument follows verysmoothly from an ancient medical tradition which can be expressedalternatively as "Your money or your life". Only professionals are capableof determining how much and what kinds of care are needed by a population ­and in the sub-text, only professionals should decide how much they23themselves are entitled to be paid in the process. s The third party payer,public or private, has no right to interfere in this process, its onlylegitimate function is to pay the bills. To the extent that it fails to doso, the patient should be required to make up the difference, but theoverall size of the bill is a matter for professional judgement alone.That professional judgement is, by definition, exercised only andwholly on the patient's behalf. It follows that any attempt to limit theflow of resources into health care must lead to harm to patients - needlesssuffering and perhaps even death. After all, if the care were not needed,professionals would not be recommending and providing it. And the pricethey demand for their services will be both fair, and necessary, tocompensate for their effort, responsibility, and training. Again, one canrely on professional responsibility for that.The argument is quite circular, and is intended to be so. Since Canadais containing costs. relative to what American and many Canadianproviders would demand, then a priori Canadian patients must be sufferingas a result. Interestingly this circular argument has an exact parallelwhen naive neo~classical economic analysis is applied to the question, aparallel in which "the market" plays the central role which professionalsassign to professional responsibility.The Economic Variant: Doctor Pangloss Goes To MarketThe economic argument begins from the accounting identity noted above,that total expenditure on health care necessarily equals total incomesearned from providing health care. If that total is reduced - or its growthrestrained, it follows that either fewer goods and services must beprovided, or lower prices must be paid, on average, for them. But by9 In the United States the professional rhetoric may make a politicallyexpedient reference to the forces of the competitive marketplace at thispoint, but the formal and informal institutional arrangements of the medicalprofession have heretofore been sufficiently powerful to ensure that themarket works weakly, if at all, even when supply is increasing rapidly.24hypothesis, the price of the services is a reflection of their "quality" ­that is ensured by the competitive marketplace. It follows that costcontrol reduces either the quantity or the quality of the care~rovided, orboth - exactly what the AMA would - does - say.Both forms of analysis demonstrate conclusively that accessibility !!ll!tlbe reduced, in either quantity or quality" terms, as a consequence of costcontrol. Moreover they do so on purely a priori grounds, in a totally data­free environment. (Actual information would be an irrelevant distraction.)Nifty, though alas fraudulent. The trick is worked, as in all a prioriarguments, by careful choice of assumptions and definitions. In particular,both medical and economic arguments sidestep completely the question ofoutcomes, or the effects of care on the health of patients.Most people interested in health policy, most patients, mostphysicians, most of us, judge the quality and appropriateness of care by thelikelihood that it will do more good than harm to someone's health. TheCanada Health Act, which lays out the conditions that provincial healthinsurance plans must meet to be eligible for federal financialcontributions, explicitly states that the purpose of the system is tomaintain and improve the health status of the population.Accessibility is then judged in terms of whether people can in fact getthe care that they need, in the sense of care that is likely to improvetheir health. And accessibiliity of higher cost care is only worthwhile ifthe higher cost purchases higher expectation of benefit. How much higher?That is a touchy policy judgement. But most of us can agree that there hadbetter be some extra benefit.The B priori arguments, on the other hand, both professional andeconomic, carefully avoid explicit consideration of this issue. The firstimposes the assumption that whatever is provided must have been needed,otherwise expert and responsible professionals would not have provided it.The second modifies this to the assumption that "consumers" of care (not25patients) will use only those services which they value - and that theirvaluations, not health outcomes, are the legitimate standard against whichto judge accessibility.As an aside, on this argument the American health care system must befaulted for making access to laetrile, or quack remedies generally, moredifficult, because a number of "consumers" obviously want to buy them. Itcannot he faulted, however I for failing to provide care to those "unwilling"to pay for it. That is right and proper, because they obviously do notvalue care sufficiently to justify its cost of provision. That theirunwillingness may be rooted in absence of insurance or personal resources,is irrelevant.This is not merely a debating point. II Consumer II willingness to pay,unadjusted for differential resources or imperfect information, is byassumption the fundamental test of value in the intellectual framework ofmarket economics. It is the foundation stone. on which "are based allnormative statements, all policy recommendations as to what. "should" bedone. The well-organized economy provides whatever people want - if theyhave the resources to pay - and does not provide commodities for which theywill not or cannot pay. The full implications of this assumption are rarelyhighlighted by neo-classical economists. But the advocate of "free market".approaches to health care delivery and finance, who does not simultaneouslyadvoca t e open access to "quack." practitioners and remedies of all kinds, issimply being intellectually inconsistent. In the free market there are noquacks; the concept has no meaning.It is important to be clear about the fundamentally circular nature ofsuch critiques of the Canadian health care system, since otherwise a gooddeal of time and energy can be wasted in discussions which by design gonowhere. But the question of the accessibility of needed care in differentfunding systems remains a very serious one, quite apart from its misuse inpublic relations exercises. Americans know full well that a substantialproportion of their population has access to either sub-standard, or no,26care, as a result of economic barriers. But it is certainly possible,though not self-evident, that the accessibility which the Canadian fundingsystem gives with one hand, by removing financial barriers, it takes awaywith the other, by providing insufficient resources to meet populationneeds. What do the data show?The Price of Paying Less: What Do Canadians Give Up?But what data? It was suggested above that the tests of a health caresystem were its contribution to the health of the population it serves, andits acceptability to that population. Does it work, and do they like it?Both are linked to accessibility. If a system is "unde r funde d" in a realsense, not just in that the people working in it would like higher incomesand more gadgets to play with, then the resulting restriction of accessshould be visible in either or both of adverse health outcomes - mortalityand morbidity - or increased time and trouble for patients in gaining accessto care. Health status and/or public satisfaction should suffer.One could add a third criterion: the degree of equity of access withinthe health care system. Some, myself included, believe that a good healthcare system provides care on the basis of need rather than ability to pay,and treats all members of society equally in this respect. All systems"r ation ll care, in the obvious sense of the elementary economics textbooks.But a system which denies or impedes access for those with greater needsand lesser resources, while responding with alacrity and enthusiasm to thosewith minimal or imaginary needs but ample resources, is on this criterionsignificantly inferior, in terms of accessibility, to one in which allcitizens with equivalent needs are treated (more or less) equally, even ifthe latter does not meet all needs which providers can imagine andcommunicate to their patients. Nor is this inferiority compensated for bya higher level of provision overall; inequitable aCcess is not mitigated byproviding even more services to those who do not need them.27It is not, however, appropriate to insist on this criterion in thepresent discussion. In the first place, it is probably not as widelyshared, particularly in the United States, as the first two. (Although itmay be more widely shared, even there, than is reflected in currentpractice; why else would the existence of so many uninsured and underinsuredbe the occasion for such public hand-wringing, even by those who have nointention of doing anything about the situation? See also Taylor, 1990.)And secondly, such a criterion rigs any comparison with Canada so heavilyagainst the United States as to amount to settling the accessibility issue8 priori, a strategem which was just criticized above.Despite the political controversy which forever surrounds health carefunding in Canada, the popular support for that system remains, as noted atthe outset, overwhelming. Moreover, that support has been demonstrated inthe most unambiguous fashion possible; it is not merely inferred from theconversations of visiting academics with taxi drivers. But that stillleaves open the second question: "Does it work?U ·Or does universal publicfunding with cost containment result in impeded access to needed care, andconsequent adverse outcomes?Ideally, we would wish to be able to measure the patterns of morbidityand mortality in Canada and the United States, and attribute them to thecontributions of the respective health care systems. We would then be ableto determine, for example, whether the fact (if it is a fact) that theUnited States has more CT scanners than 7-Eleven stores, and Canada doesnot, pays off, all else equal, in greater health for Americans. Those whoargue that the Canadian system is "underfunded" are implicitly assertingthat this is the case. Unfortunately, they do not have the evidence tosupport this claim. Nor do I, and nor does anyone else.It is notorious, throughout North America and Western Europe, thatminimal data are available on the health status of populations, let aloneon the relationship between that health status and the provision of healthcare. Mortality data are available but, as everyone knows, there is much28more to health than life alone, and anyway, many other factors affectmortality. The rather idiosyncratic approach which Americans take to guncontrol, for example I clearly has a bearing on their relative mortalitystatistics. Indeed, the country which is currently showing both the bestand the most rapidly improving life expectancy statistics, at all ages, isJapan. Its health care system has recently been described by a respectedexternal observer as "anachronistic" (Iglehart, 1988), and the OECDstatisticians report that Japan spends a bit more than half as much as theUnited States on health care, relative to its total income.For what it is worth, the comparative data available on mortality andmorbidity in North America show Canadians as slightly healthier thanAmericans, but very little different (Battista et al., 1986). There is nonecessary connection with the effectiveness of our respective health caresystems. One can certainly say that there is no indication, at theaggregate level, that the health of Canadians has been affected as a resultof our spending less on health care. Whether Americans are beginning tosee, in their infant mortality and life expectancy trends, the consequencesof unequal access to care is another matter, but fortunately not one whichneed be dealt with here.While it may be impossible to assess directly the relative healthcontributions of entire delivery systems, clinical epidemiologists maketheir livings carrying out such investigations on particular diagnostic andtherapeutic manoeuvres. They very commonly find that such interventions,offered in good faith and carried out competently, turn out to do no good,and sometimes even harm, to some or all of those who receive them.Accordingly, there is no a priori reason to assume that less care, in total,implies less health - Canada may simply provide less ineffective care. Butthen again, maybe not.One can, however, get at the accessibility question indirectly, andproduce a partial answer which goes a good deal of the way, by examiningjust what it is that Canadians spend less on. If it could be shown that the29difference in overall spending were accounted for by items which have nodirect connection with health outcomes, that would support the inferencethat accessibility was not in fact being impaired in' Canada, As it happens,not all but most of the difference is accounted for by such items.The discrepancy between health spending in Canada and that in theUnited States can be measured in several different ways, but the most commonis through comparison of the percentages of national income. -because thisavoids problems of adjustment for both exchange rates and differentialinflation rates. (It also introduces some problems of its own, but forcomparisons between economies so closely interlocked as those of Canada andthe United States, these are minor.) At present, the gap is nearly threepercentage points of GNP, implying that Canada spends about three-quartersas much as the United States. This differential is almost entirelyaccounted for by differences in administrative costs, in the rate ofescalation of physicians' fees, and in the intensity of servicing ofpatients in hospitals (Evans, Lomas et al., 1989; Evans, 1986).Pruning Private Bureaucracy: Canada's Teeth-to-Tail Ratio Is HigherOf these, the first category is the most unambiguous with respect toaccessibility. Canadians have, through their health care system, much lessaccess to the services of accountants, administrators, insurance salesmen,specialists in public relations and marketing, and management consultants.The whole panoply of services provided by the private insurance industry,and charged for in the form of the net revenues of health care insurers,costs between five and six times as much in thc United States as in Canada.The reason is simple. When the whole population is cover ed-vf'o r everything,the costs of designing and selling policies, determining eligibility, makingrates, all disappear. Much of the effort of a for-profit insurer must bedevoted to determining who not to cover, and what not to cover, in order tohold down losses (consider AIDS). This is not inherent meanness; insurancecompanies make profits by collecting premiums, not by paying claims. Thecompetitive marketplace forces them to try to increase the former while30minimizing the latter; and they quite understandably devote a good deal ofhigh-priced talent to both. But in a universal system, these functionsvanish.In addition, the costs of providers, both hospitals and otherinstitutions, and professional offices and clinics, are significantlyreduced because the staff required to deal with the payment system areminimal. The large financial apparatus of an American hospital has noCanadian counterpart, and the physician's office staff can be reduced aswell (Himmelstein and Woolhandler, 1986). In the business services sector,the whole field of employee benefits is significantly simplified. TheUnited States maintains a vast private bureaucracy whose function is to pusharound the bits of paper associated with health care (Reinhardt, 1988).Ca~ada does not. The total impact of this bureaucracy on health care costsis difficult to estimate with precision, but the order of magnitude isconservatively about one percentage point of GNP, or about one-third of theCanada-United States differential - over 50 billion dollars.Only part of this total bureaucratic cost, however - perhaps about halfshows up explicitly in the differential costs of the insurance system ­prepayment and administration. The remainder is buried in the budgets ofhospitals, and to a lesser extent of physician practices and clinics, whereit takes the form of costs of administrative and financial servicesnecessary to establish patient eligibility for coverage, submit and justifyclaims, collect bills, and generally meet the demands of the payment system.These costs are recorded as costs of hospital and medical care, althoughthey are really costs of the insurance system.The gain in American health status from such activities is, however,easier to estimate - nil. The administrative overhead of the Americansystem contributes nothing at all to health outcomes, and contributesne~atively to patient well-being. The compliance costs, of choosing andmaintaining coverage (or trying to discover it!) and struggling overreimbursement entitlement, and the associated anxieties, are again simply31non-existent in Canada. If there is a question about coverage, orappropriateness, which there rarely is, that is for the provider and thepayer to sort out. The patient is not involved.In military terms I the "teeth-to-tail ration is much higher in theCanadian system. A substantially higher proportion of resources is devotedto providing care and a lower proportion to pushing paper. Nor is theeffectiveness of the system reduced by leaner administration, because mostof the "tail" in the American system does not in fact support the functionsof the II teeth II component I the actual providers. Rather it is involved inan elaborate game of cost redistribution, of determining who will pay.One could certainly imagine an administrative support system which didmake a significant contribution to the effectiveness of care, for exampleby monitoring and evaluating its impact and improving the knowledge basewhich lies behind clinical decisions. There is plenty of room forimprovement in this area, everywhere in the world, and the Canadian recordis not particularly impressive in this respect. But that is not, in fact,how most American administrative resources are now spent. This activity maybe greatly expanded in future (Roper et a1., 1988), and that will be all tothe good, but it will also represent yet a further cost item.A Restaurant Analogy: Who Chose This Place, Anyway?The current American situation can be represented by the well-worneconomists' analogy of a group of people going to a restaurant for lunch,and agreeing to split the .cheque equally. This is meant to represent theincentives in an insurance system ~ public or"private ~ in which care is"free" - what you eat is mostly paid for by others. The usual argument isthat all the diners will eat more than they really want - or would eat ifthey had to pay the full cost - and the bill will be distressingly high asa result. Everyone wil be unhappier than if they had all paid their own32bills .'0But the story is incomplete. The maitre d ' presents the bill, andindeed it is very high. In the American system, the diners immediatelybegin to argue about what their respective shares should be, and to try torecontract out of their prior agreement to pay equal shares. As the disputeintensifies, they each bring in t.he Lr accountants to justify their claimsto a smaller share. Matters escalate, and soon the lawyers begin to arrive.All these back-up experts are paid by the hour. The lunch becomes veryexpensive indeed, though the costs of arguing over the bill leave no onebetter fed, and do significant harm to the digestion.The Canadian approach is different. There too, the bill for lunch israther distressing, including items the diners are not sure were everprovided, others which seem to be overcharged, and still others which werenot very good. But instead of arguing among themselves, to the relief ofthe maitre d' and the profit of their accountants and lawyers, the Canadiansappoint a spokesman (the provincial government) and call in the maitre d'to negotiate the bill.These negotiations may become acrimonious, and the maitre d' frequentlyinsists that it is unprofessional for him to have to justify the bill. Theconsequences will be demoralization in the kitchen, deterioration ofstandards, and conceivably (though not yet) food poisoning. Sometimes hewishes he worked in the American restaurant next door, though he reallywould not want to put up with the shouting, pushing, and crowding, and heknows that staff over there occasionally get hurt in the melee.The net result, however, is that the restaurant bill is lower than nextdoor, and the total cost of lunch is much lower. But do the Canadians get10 Even on its own terms the analogy is inconsistent. If the dinersare completely selfish, as the example assumes, and take no account of theimpact of their behaviour on others, then why did they agree to go to therestaurant together in the first place, and split the cheque?33less food, or lower quality? Well, what else gets cut out of the bill? Asnoted above, the second and third components of the Canada-United Statescost differential are physicians' fees and hospital servicing intensity.Controlling Physicians' Fees: Processes and ConsequencesIn Canada, as noted above, physicians in each province are paidaccording to a uniform fee schedule, negotiated at periodic intervalsbetween their provincial medical association and the provincial government.They cannot extra-bill the patient, and the schedule includes bothprocedural definitions and values, and a set of rules of payment whichdefine the circumstances under which particular fees are payable. These feeschedules have risen, averaged over time and across the country, more orless in line with overall inflation rates, Divergences in both directionsare observed from time to time, depending upon the relative skill andbargaining power of the negotiators, and particularly upon their success (orlack of it) in forecasting general inflation rates. Over time, however,these divergences tend to average out (Barer et al., 1988).This is in sharp contrast to American experience where, with theexception of the early 1970s, physicians' fees have consistently outruninflation. Thus a significant part of the difference in health care costson the two sides of the border arises because, when physicians have tonegotiate their fees with a single payer, those fees rise less rapidly. The"private market", at least in its present American form, supports a steadyescalation in fees in real terms adjusted for inflation. Bilateralnegotiations do not.Under these circumstances . "price controls II in an industry of self­employed practitioners - economic theory predicts unambiguously that thequantity of services offered by providers will go up, or down, or (less34likely) remain the same." Each of these possibilities has been forecastby participants in the American debate on physician fee schedules.Cross-border comparisons suggest that in fact the increase in services,or at least billings, per practitioner has been slightly more rapid inCanada over the past two decades (Barer et a1., 1988). The difference hasnot been large, however, and has not offset the difference in £ee trends,so that overall costs have gone up more slowly in Canada. This is in partdue to the rules for payment associated with the schedules, which havelimited the opportunities for providers to expand their billings throughprocedural multiplication. In addition, some provincial governments havein recent years negotiated fee schedules in which fee increases are phasedin over time, and may be reduced if utilization rates rise too fast (Lomaseta1., 1989). In a more open-ended system of payment such as the UnitedStates, one might well find that attempts to limit fees were met by off-setting increases in servicingto limit their growth.unless corresponding measures were takenIn any case, the Canadian and the American evidence suggests thatcontrols on fees will tend to increase, not lower, the volume of servicesoffered by practitioners. To the extent that utilization of services is aproxy for access, fee controls at least do not impede access, and are muchmore likely to enhance it. The real problem is not impeded access, but"hyper acces s " - overuse. There are however two possible qualifications,only one of which can confidently be dismissed.Most obviously, if fees were set in a hypothetical competitive market,11 One cannot base predictions on a positively sloped supply curve,because the opportunity cost of the professional's own labour is thepredominant component of "fLrm'' costs I and this cost is positivelycorrelated, through income levels, with output prices. Furthermore, a largeshare of the return to labour is in fact a quasi-rent to the human capitalembodied in the professional; entry to the field is not free; and inputmixes are constrained by regulation. Under these constraints, a backward­bending supply curve both of own- time and of total output is not onlypossible, but quite likely.35with free entry and fully informed participants, the long run effect of feecontrols would be to discourage people from taking up medical careers.Supply would eventually dry up, in the way that rent controls are allegedto reduce the supply of rental housing. But of course medicine does noteven remotely approximate the conditions in such a market. And what we findin reality is that in Canada medical school places are over-subscribed toan even greater degree than in the United States.There are as many physicians per capita in Canada as in the UnitedStates, and this ratio is rising at between 1.5 percent and 2 percent peryear (an increase several times the increase in It ne ed lt represented by thechanging population age structure). By the end of 1989 the number of peopleper physician had fallen below 450, and the decline will continue for theforeseeable future.The principal concern of those responsible for manpower policy is whatto do to control the numbers? Some physicians do go to the United States,but not enough to affect the overall stock and, in any case, many come back.Thus fee controls have not impeded access to physicians' services byreducing their numbers any more than by reducing their work incentives.A more subtle effect, however, might be to induce physicianspreferentially to provide more remunerative services, or to adopt a styleof practice - short visits, frequent recalls - with higher payoff per hour.Servicing rates per capita would rise, but if the effectiveness of care werereduced (which it need not be), "access" might be interpreted as reduced.Indeed, access to needed services would be unambiguously reduced ifphysicians' time were all taken up with the increased provision of lessneeded but more remunerative care.This line of argument, which it must be emphasized is pure speculation,takes us back to the fact that the linkage between utilization and outcomeis distinctly shaky, in every health care system. Since we have so littleinformation on the effectiveness of health care services, we would be hard36put to know how to test this possibility. But there is equally no warrantfor assuming that, if Canadian fees increased more rapidly, any resultingchanges in patterns of practice if they occurred - would result inimproved outcomes. The assertion that physicians must be given whateverfees they ask for, or they will react in ways which will harm theirpatients' health, is an interesting commentary on the professional standardsof practitioners, as well as yet another example of a circular argument, ina data-free environment, that any attempt at cost control must lead to harm.But it does provide further support, if such were needed, for gaining morehard information on the connection between servicing patterns and patientoutcomes.So the second thing that Canadian patients give up, in addition to theservices of insurance salesmen, accountants, and management consultants, issome part of the lifestyles of their doctors. Canadian physicians are, liketheir American counterparts, at the top of the occupational income scale,but they do not earn quite as much, absolutely or relatively. The impactof this form of "reduced access" on the health of patients is ratherdifficult to detect, though it goes far to explain the concern of Americanphysicians' organizations to protect their patients (and even those whocannot afford to be) from the disaster of universal public insurance.Hardware and Hard Questions: How Will I Know When I'm Better?The third major area of expenditure differences is in the acute carehospital sector, and that is where the interesting questions of differentialaccess and associated outcomes arise. It is also from here that the talesare carried south of the border, about long waiting lists for electivesurgery, insufficient and out of date equipment, and patients suffering oreven dying for lack of care. Queues for medical services form, not becauseof a shortage of physicians, but because of insufficient provision offacilities, equipment and personnel for physicians to work with - not toofew cardiac surgeons, but too little surgical capacity. The situationsdescribed or alleged are multi-dirnrnensi<;mal and complex, and cannot be37easily assessed with the sort of evidence which we brought to bear on thefirst two sources of cost differences.To begin with, some of the stories are true. Waiting lists forelective surgery do build up at some times in some parts of Canada, and theavailability of advanced diagnostic and therapeutic equipment, on a percapita basis, is less and sometimes substantially less, than in the UnitedStates. There are periodic crises of access, and more often allegations ofcrises, and some people do go to the United States for care. But theexplanation of these observations is much more complex than simply a globalshortage of resources imposed by stingy or impecunious governments I andtheir implications for the health or well-being of patients is by no meansunarnbiguous .Like physicians, hospital beds are in ample supply in Canada, and areheavily used. Canadians use one-third to one-half more patient days percapita in acute care than do Americans 1 about twelve hundred days· perthousand population per year, and occupancy rates average abou~ 85 percentacross the system as a whole. These compare with American averageoccupancies in the 65 percent range. Thus Canadians may appear to have lessaccess to hospitals than Americans do, because with much higher averageoccupancy rates a randomly chosen Canadian hospital is much more likely tobe full on any given day.Yet per capita rates of hospital admission are remarkably similar inthe two countries, just under 150 per thousand population per year,indicating that perceived shortages in Canada reflect higher propensitiesto hospitalize and/or more intensive use of facilities, rather than lowerrates of admission. Canadians do get into hospital, and at about the samerate as Americans. And once admitted Canadians stay longer on average; theyhave substantially greater "access" to days of care.There are several possible explanations for this greater utilizationof patient days. The usual official story is that, because the public38insurance programs were introduced for hospital care in the late 1950s, andonly 10 years later extended to medical care, Canadian physicians andpatients both became used to an institutional style of care which haspersisted to this day. But examination of American payment data shows thatmost hospital expenses there are also covered by some form of insurance ­about 90 percent - while physicians' services are much more commonly paidout of pocket. Yet American patient day utilization rates -are much lower.Another incentive arises from physician fee schedules, which do notcover the technical component of costs for many of the more expensive formsof diagnostic and therapeutic equipment - lichotripters, for example, ordiagnostic imagers such as MRI, PET, or CT scanners. This limitsphysicians' ability to expand their incomes by setting up free-standingfacili ties and self - dealing by referring their patients. The expensiveequipment is provided to the hospitals, where operating costs are fundedthrough the annual global budgets. This both restricts the availability anduse of such equipment, and channels patients through the hospital.But that does not explain the use of inpatient beds, since hospitalscan and do provide a range of ambulatory diagnostic and therapeuticservices. Just because the hospital owns, and is paid for, a particularfacility or piece of equipment, is no reason for physicians who referpatients to that facility to admit them as inpatients first. CertainlyCanadian hospitals do not require this; after all they are not paid fees forservice, and the hospitals with the high-technology equipment are not ingeneral troubled by low occupancy. And in any case, if admission wererequired for access to high- tech equipment, that should be reflec ted inhigher admission rates, not longer lengths of stay.A third argument, frequently heard from physicians, is that acute carebeds in Canada are being "blocked" by de facto long-stay patients, who wouldmore appropriately be cared for in some form of extended care facility.High levels of acute care utilization are alleged to reflect an inadequatesupply of such facilities - again an "underfunding" problem.39Detailed analysis of the trends in hospital utilization, and ofreported reasons for hospitalization, provides some support for thisposition, but not very much. There has been an increase, over the lastdecade, in the numbers of acute care days identified as "patient awaitingplacement", but this appears to be due in part at least to changes in thediagnostic coding systems. The ninth revision to the InternationalClassification of Diseases, which was adopted in Canadian hospitals at theend of the 1970s, introduced this category for the first time, andphysicians have learned over time to use it (Hertzman et a1., 1990).Furthermore, very large increases have taken place in long-term bedcapacity in Canada, without alleviating the alleged pressure. Canada hasa rate of institutional utilization which is among the highest in the world.And finally, even if reported acute care hospital use is reduced byarbitrarily removing all patients with lengths of stay of 60 days or longer,this still leaves per capita use rates well above comparable American rates(Evans, Barer et a1., 1989)."It's Just Our (Clinical) Policy"One comes back to explanations in terms of the more conservativepractice styles of Canadian physicians, and the lesser incentives for themto care for patients out of hospital. These are reinforced by thedifferential incentives bearing on hospital managements; global budge ts areless strained when patients are kept in longer, while item of servicereimbursement rewards high turnover and plenty of servicing.Canadian hospital utilization by acute care patients is in fact movingslowly downwards.squeezes. WhileBut the adjustment is taking place through administrativein the United States the Prospective Payment Systemprovides financial incentives to reduce inpatient use, in Canada provincialgovernments achieve the same result by providing fewer beds than the medicalstaff would like, or encouraging (pressuring) hospital administrations toconvert acute beds to extended care, and to set up alternative ambulatory40facilities. The relentless increase in the supply of physicians, pushingagainst a relatively stable (per capita) bed supply, not only holds upoccupancy rates but generates increasing pressure for individual physiciansto economize on beds. Bed- to-population ratios are high and relativelystable; but bed-to-doctor ratios have been falling steadily for a long time,and this is forcing changes in practice patterns.But the problem is often notconflict, of shortages,a shortage ofand waiting lists forfacilities in absolutesqueezesadministrativeThethe process.claimslikenotPhys ic ians dogenerate politicalcare.terms, but rather a conflict between government policies to encourage moreuse of ambulatory facilities, or simply less bed use, and physicians wishingto keep putting patients in beds because they have always done so. Henceone sees the paradoxical combination of "shortages ll and waiting lists in anenvironment of apparent overutilization of inpatient care at leastrelative to American practice. Over time, however, the necessaryadjustments have been occurring, and inpatient utilization has been driftingdown, though these trends have to some extent been masked by thesimultaneous expansion of extended care wards within acute care hospitals.This latter development makes it difficult to interpret the cross­border comparative data on hospital costs. Hospital expenditure per capita,adjusted for hospital input prices, has been rising substantially faster inthe United States than in Canada, for many years (Barer and Evans, 1986).This is consistent with the argument that, even if there is plenty ofhospital space in Canada, much less in the way of diagnostic and therapeuticservices is provided to hospitalized patients. The real problems are ofaccess not to beds or doctors, but to.up-to-date technical services. Thelimitations on free-standing facilities in Canada point in the samedirection.But hospital accounting systems in Canada do not permit one toidentify, on a system-wide basis I the share of acute care hospitalexpenditures which are going to acute care patients. And we know that the41mix of hospital patients, and especially patient days, has changed towardsa higher proportion of long term care use. Consequently the intensity ofservicing of the truly acute care patients 'may well be going upsubstantially faster than is reflected in the aggregate data.If one were to remove from both countries' data the proportion ofhospital costs accounted for by financial and administrative activities Jwhich is much larger and faster growing in the United States, and then focusonly on acute care patients, it is not clear that there would be asubstantial difference between treatment patterns on the two sides of theborder. The analysis has not been done, but there is some supportive expertopinion from clinicians and administrators with cross-border experience.Moreover, recent c ros s -bc r de r comparative studies of the rates ofperformance of particular surgical procedures have shown that although theUnited States has higher rates for some complex procedures (e.g. coronaryartery bypass surgery) the rates for other complex procedures (e.g. repairand replacement of heart valves and major peripheral vascular procedures)are as high or higher in Canada (Anderson et al., 1989). Procedural studiesalso show that in Canada, as in the United States, there are large andunexplained regional variations in performance of high intensity procedures.Rates of performance of certain specific procedures carotidendarterectomy, pacemaker implantation, caesarian section, and in someregions cardiac bypass grafts - are at levels which justify concern aboutover-servicing and possible harm to patients.That said, however, one still comes back to the fact that on a simplecount of maj or, high- technology equipment, "there is substantially moreavailable in the United States than in Canada. Even though the Canadianfacilities tend to be used more intensively, and partly in consequence havelower unit costs, it seems undeniable that Americans in and out of hospitalsreceive a number of such procedures and services (not all) at a higher ratethan do Canadians.42Cardiac bypass grafts have been a leading example. Although the ratesof increase are similar, many more procedures per capita are done in theUni ted States. For bypasses in particular, complaints of insufficientcapacity, long waiting lists, and patients going to the United States, arecommon in the media.This, finally, is where we come to the hard edge of the accessibilityquestion. It is clear that improved access to the services of insurancesalesmen and management consultants is not the primary objective of a healthcare system. Access to higher physicians' fees and incomes is also oflesser immediate priority, unless one happens to be a physician. But is notaccess to the services of MRI machines and lithotripters, or to cardiacbypass grafts, a more plausible primary objective? Well, in fact no, or atleast not necessarily.The Politics of "Saving Lives", On Camera and OffThe key point to remember is that nobody in his right mind wants healthcare services for their own sake. And the phrase llin his right mind lt isused advisedly, because there is a mental illness, known as Munchausen'ssyndrome, whose victims want health care when they are not sick. The samepoint is made by the wisecrack that anybody who wants health care when heis not sick, is sick. It is access to needed care which is critical, accessto care which is effective, which has a demonstrable (positive) impact onpatient outcomes.But it is well known, and has been extensively demonstrated by studentsof health care utilization, that one cannot infer need from use. One cannotassume that, simply because Canadians use fewer of certain types ofservices, they necessarily suffer from a reduction in access in the senseof access to health outcomes. And that is what we are really interested in,not activity, however technically impressive, for its own sake.43Furthermore. there are adaptation processes in the Canadian fundingsystem. It is by no means as stagnant and as starved for funds as it issometimes portrayed in the American media. There, Canada is frequentlybracketed with the United Kingdom as virtually equivalent "horribleexamples" of "socialized medicine", but the parallel is without merit exceptfor propaganda purposes. As noted above, Canada has socialized insurancesuperimposed on a private delivery system, and spends as much per capita asany other country in the world, outside the United States, on health care.Accordingly, when pressure points develop more resources are available toremedy the situation.The process of resource mobilization is, however, overtly political.The theatre of shortages and unmet needs, what we have called elsewhere"or che s t r a t ed outrage II (Evans, Lomas et a1., 1989) I creates politicalpressures which define social priorities 'and determine where the resourcesare most needed - or where the advocates can mobilize the greatest politicalpressure.Coronary artery bypass grafts are a case in point. Waiting lists andshortages are not the result of a refusal by provincial authorities toprovide facilities, on the contrary, capacity and utilization are expandingrapidly. But cardiac surgeons are bringing people to surgery even morefrequently, and particularly very elderly people. This growth in "demand"by surgeons is outstripping the growth in facilities and utilization.Cardiac surgeons have in effect decided to re-allocate public resources intothis field (and to themselves) through a powerful political campaign,including elements of lTdisinformation".But as in the United States, clinical practice in this area shows widegeographic variations which seem unrelated to patient needs (Anderson andLomas, 1989). There are thus good grounds, reinforced by the equivocal orabsent evidence from clinical epidemiology, for believing that some, perhapsmuch, of the surgery is inappropriate. Knowing this, provincial ministriesof health have deliberately tried to restrain the growth of surgical44capacity. But the political costs are high.No one would pretend that such a process is perfect in its ability tomatch resources to actual needs. Indeed in the case of cardiac surgery thepolitical process is looking quite vulnerable. But overall, this approachdoes not look too bad when one considers the known alternatives.And it would be quite misleading for outsiders to imagine, as manyAmericans do imagine, that the political theatrics indicate a system incollapse or even under markedly more strain than any other in the world.On the contrary, the on-going political controversy is itself a form ofsolution to the inherently very difficult problem of setting socialpriorities with respect to health care, and giving those priorities effect.Unlike the current situation in the United States this solution, imperfectthough it inevitably is, appears both acceptable and stable for the mediumterm at least. The American combination of rising costs and fallingcoverage, by contrast, suggests a system which is not dynamically viable;projection of the current trends indicates steadily increasing conflict andmisery for a growing proportion of the American population.The Missing Links: Utilization, Need, and Health OutcomesIt is in this context that one must consider the issue of relativeaccessibility of particular medical procedures and interventions on the twosides of the border. Substantial differences in utilization, for some atleast, are readily demonstrable, but the significance of these forcomparative health outcomes is unknown. A number of American researchershave concluded that certain procedures are greatly overutilized in theUnited States, far beyond what either scientific evidence or even expertopinion supports as beneficial or appropriate (Brook and Vaiana, 1989).It is quite possible that Canadians are better off with less, instraight-forward health outcome terms. At least one knowledgeable Americanobserver (Enthoven) has conjectured that more Californians die in the course45of unnecessary or inappropriate heart surgery, than Canadians die fromdelays. But it must be admitted that, in North America and everywhere inthe world, we know much less than we should about the positive and negativeconsequences of health interventions.Until we do, it is not possible to say with confidence that no Canadianever suffers as a result of inadequate access to health care - and indeedthe statement is almost certainly not true. Would it be true in any othercountry? What is much more sustainable is the statement that the Canadianhealth care system suffers not from underfunding but from undermanagement(Rachlis and Kushner, 1989), so that the problems of access which do existwill not be remedied simply by throwing in more resources. Again theinternational evidence is supportive; health care systems in all developedcountries display on-going conflicts over costs and access, regardless ofhow much is spent on care. Would those Americans who feel that Canadianssuffer from lack of access to certain services want to claim that their ownmuch higher· level of expenditure has solved, or even significantlymitigated, access problems in their country?For that matter, it is a gross over-simplification to refer to levelsof access "Ln Canada", or "in the United States" t as if the nationalaverages were representative of the entire of two very large and diversecountries. American researchers have clearly documented the wide diversityof patterns of care in the United States, diversities which show up betweenregions or states, and also among very small regions within states, butwhich cannot be shown to bear any relation to patient needs or outcomes.Recent work by Wennberg et e I , (1989) has even shown very largedifferences in average utilization and costs between Medicare populationsin Boston and New Haven, each served by one of the most prestigious healthscience complexes in the world. Mortality patterns are the same in bothareas, but it costs twice as much to die in Boston. Per capita use ratesfor particular procedures are even more variable. Which represents"American medicine at its best ll ? If per capita use patterns and costs in46Boston were somehow brought into line with those in New Haven, a great dealof money would be saved for the American taxpayer. Would this represent thecatastrophe of "rationing ll in Boston, but not in New Haven? Does"rationing" mean nothing more than holding providers accountable for whatthey do, and spend?But let us not pretend that the Canadian approach to funding representsan adequate response to this situation of apparently arbitrary patterns ofuse and cost. Inter- and intra-provincial variations are just as prominentnorth of the border. The most costly province in Canada - Ontario - may bequite similar to many states in the United SCates, while if one comparedpatterns of care and cost in Boston with those in, say, Quebec or BritishColumbia, really spectacular differentials would emerge. Yet each isconsistent with acceptable levels of care for a modern population.Recognition of the extent of regional variation in each countryunderscores, heavily, the essential arbitrariness of patterns of medicalcare. This in turn demonstrates the patent absurdity of the claim -endlessly repeated by provider representatives - that any attempt at controlmust threaten the health of patients. This arbitrariness is the other sideof the coin from the observation that a high proportion of the care actuallyprovided, in any modern health care system, is of unevaluated, or no,beneficial effect in the circumstances in which it is given.The Role of Research: Guide or Alternative to Action?The pervasive lack of knowledge about the effectiveness of health careprovides strong support for a maj or expansion in re.search on thedeterminants of health outcomes. As a sub-species of that, research on thedifferences in patterns of care, and in outcomes, on the two sides of theCanada-United States border might be particularly interesting. Americanresearchers appear to be well out in front of the rest of the world in sucheffectiveness research, though there are also several strong groupselsewhere.47But while a certain humility in the face of the vast unknown is bothseemly and prudent for the scientist and the scholar, it can be remarkablydangerous for those responsible for public policy. The researcher willalways assert that more research is needed; it is sometimes hard todistinguish modesty from marketing. One does not have to go over theCanadian experience with a fine- tooth comb I and turn it inside out, todecide whether it offers, in Enthoven's (1989) compact phrase, "politicallyfeasible incremental changes that have a reasonably good chance ofmaking things better."Those who argue that, until the differences between Canada and theUnited States are mapped and understood in much more detail, no secureconclusions are possible, are both marketing their own services, andproviding a very powerful defence for the status quo. That might be a moreplausible position, if there were fewer problems with the American statusquo.Thus while there is clearly much more which can be learned fromcomparative research on health care patterns between the two countries, itis quite wrong and dangerously misleading to suggest that such detailedresearch is either necessary or sufficient for the design of American policybased on Canadian, or other international, experience. The fundamentalissues in health care policy are political, not technical, and an attemptto portray them as amenable to "scientific" solutions is simply part of thatpolitical process, often with researchers as conscious or unconsciousparticipants.Moreover I even in the technical sphere t "life is the art of drawingsufficient conclusions from insufficient premi.se s ?. Hegel t s comment thatMinerva's owl flies only in the darkening twilight, can be interpreted tomean that by the time the facts are all in and the situation fullyunderstood, the game is long over and players and spectators have all gone"home. ItMore research" can easily be a stratagem for delaying action untilthe window of political opportunity has closed.48I think we now understand how and why the Canadian health care systemworks, after watching it for 20 years, much better than did the people whodesigned and established it. They were by no means totally ignorant; theyhad spent a lot of time drawing inferences from the examination of otherexperience, and thinking pretty hard. But they certainly did not have thequality of data that would be published in the New England Journal ofMedicine. What they did have was quite a lot of courage and the will tobegin.Do Right"and You May Be Right, But Be Prepared for the Long HaulPerhaps even more important, the architects of the Canadian system hada moral vision of what a good health care system, in a decent and humanecommunity, should look like. That moral vision carried them through a greatdeal of technical uncertainty, to the national legislation, unanimouslyadopted, which one of the leading Canadian commentators has called "a leapin the dark". And it has paid off, with a system which is (relatively)affordabIe and accessible, apparently sustainable, and remarkably popular.Universal coverage, in a single-class, single-payer system, with thefinancial burdens spread according to ability to pay, through the taxsystem, rather than according to needs for care, has turned out to be notonly morally but economically sound, even if the latter was not central tothe original intention. Americans may not wish to adopt, may not be ablein their context to adopt, an exactly similar system. But as far as we cantell, any successful funding system will have to have those samecharacteristics.Furthermore, any funding system must have built into it a combinationof adaptive intelligence with a fairly stable framework. There is no once­for-all set of rules which can be established, after which the funding ofhealth care can become an automatic process like eighteenth centuryclockwork. Instead, health funding is an on-going game among parties withinterests which are inevitably opposed but who are committed to the game.49Their strategies will evolve, as they react to each other and as theexternal world changes. But they must also have enough continuity instructure and personnel to learn how to play the game without tipping overthe board. The conflict between payers and providers must be channelled andcontained, managed as constructively as possible. It will never go away.The High Noon scenario in which the bad guys (government bureaucrats? theAMA?) are confronted and blown away is a story for children; the dream ofan obj ective, scientific solution to an inherently political problem isequally mythical.And finally, this funding game is everywhere a collective process,managed through organizations which pool financial and political interests.In Canada, as everywhere else in the developed world, this process ismanaged to a greater or lesser degree by the state. Only in the UnitedStates, for historical and ideological reasons, is there so firm acommitment to finding private structures within which to manage the fundingof health care.One cannot say that this is impossible; there are in the United Statesa number of very good and innovative minds working hard on the problem. Thefertility of their ideas explains why European countries, Canada included,maintain such an interest in organizational developments in the UnitedStates, even though the operating characteristics of the American system aregrossly inferior to their own. But one can, I think, say confidently thatno other country has tried to run its health care funding system through theprivate sector, and that the American record to date is not one of success.If the United States ever pulls the trick off, the international interestin American models will rise several fold. But at the moment the dominantAmerican approach looks like a very long shot indeed, being pursued onlybecause the ideological constraints are so severe that the obvious is notpermissible.Meanwhile, as the pressures build, recall that not too far away, in acountry more similar to the United States than any other in the world, a50pretty decent system is functioning to general satisfaction. If the art ofthe possible should become more attractive than ideological purity ortechnical virtuos i ty, some of our experience may be helpful. For themoment, despite the widely trumpeted inadequacies of the current Americanarrangements, I suspect that not enough (politically relevant) people arereally suffering. But that is probably only a matter of time.REFERENCESAbel-Smith, B. 1985. Who is the odd man out? The experience of westernEurope in containing the costs of health care. The Milbank Quarterly: Healthand Society 63(1):1-17.Anderson, C.M., and J. Lomas. 1989. Regionalization of coronary artery by­pass surgery: effects on access. Medical Care 27(3):288-296.Anderson, C.M., J.P. Newhouse, and L.L. Roos. 1989. Hospital care forelderly patients with diseases of the circulatory system: a comparison ofhospi tal use in the United States and Canada. New England Journal ofMedicine 321(21):1443-1438.Banta, H.D., C. Behney and J.S. Willems. 1981. Toward Rational Technologyin Medicine: Considerations for Health Policy. New York: Springer.Barer, M.L., and R.C. Evans. 1986. Riding north on a south-bound horse?Expenditures, prices, utilization and incomes in the Canadian health caresystem. pp. 53-163 in Medicare at Maturity: Achievements, Lessons &Challenges, R.C. Evans and C.L. Stoddart, eds. Calgary: University ofCalgary Press.Barer, M.L., R.C. Evans, and R.J. Labelle. 1988. Fee controls as costcontrol: tales from the frozen north. The Milbank Quarterly 61(1):1-64.Battista, R.N., R.A. Spasoff, and W.O.patterns of medical practices. pp.Achievements, Lessons and Challenges,Calgary: University of Calgary Press.Spitzer. 1986. Choice of technique:181-213 in Medicare at Maturity:R.C. Evans and C.L. Stoddart, eds.Blendon, R.J. 1989. Three systems: a comparative survey. Health ManagementQuarterly 11(1):2-10.Brook, R.H., and M.E. Vaiana. 1989. Appropriateness of Care: A Chart Book.Washington, D.C.: The Ceorge Washington University National Health PolicyForum.Canada, Health and Welfare Canada. 1987. National Health Expenditures inCanada 1975-1985. Ottawa: Health and Welfare Canada.Enthoven, A. 1989 forthcoming. What can Europeans learn from Americans aboutfinancing and organization of medical care? Health Care Financing Review,Annual Supplement.Evans, R.G. 1986. Finding the levers, finding the courage: lessons from costcontainment in North America. Journal of Health Politics, Policy and Law11(4):585-616.Evans, R.G. 1990a forthcoming. Tension, compression, and shear: directions,stresses, and outcomes of health care cost control. Journal of HealthPolitics, Policy and Law.Evans, R.C. 1990b forthcoming. Life and death, money and power: the politicsof health care finance. In Health Politics and Policy (second edition), T.J.Litman and L.S. Robins, eds. Albany, N.Y.: Delmar.Evans, R.C., M.L. Barer, C. Hertzman, et al. 1989. The long good-bye: thegreat transformation of the British Columbia hospital system. HealthServices Research 24(4):435-459Evans, R. G., J.expenditures: the320(9) :571-577.Lomas, M.L. Barer,Canadian reality.etNewal. 1989. Controlling healthEngland Journal of MedicineFeeny D., C. Cuyatt, and P.Effectiveness, Efficiency andResearch in Public Policy.Tugwell. 1986.Public Policy.Health Care Technology:Montreal: Institute forHertzman, C., I. Pulcins, M.L. Barer, et al. 1990 forthcoming. Flat on yourback or back to your flat? Sources of increased hospital utilization amongthe elderly in British Columbia. Social Science & Medicine.Himmelstein, D.D. and S. Woolhandler. 1986. Cost without benefit:administrative waste in D.S. health care. New England Journal of Medicine314(7):441-445.Iglehart, J.K. 1988. Japan's medical care system - part two. New EnglandJournal of Medicine 319(17):1166-1172.Levit, K.R., M.S. Freeland, and D.R. Waldo. 1989. Health spending andability to pay: business, individuals and government. Health Care FinancingReview 10(3):1-12.Lomas, J., C. Fooks, T. Rice, and R.J. Labelle. 1989. Paying physicians inCanada: minding our Ps and Qs. Health Affairs 8(1):80-102.Rachlis, M., and C. Kushner. 1989. Second Opinion: What's Wrong with theCanadian Health Care System and How to Fix It. Toronto: Collins.Reinhardt, D.E. 1988. On the B-factor in American health care. WashingtonPost. August 9:20.Reinhardt, D.E. 1989. Health care spending and American competitiveness.Health Affairs 8(4):5-21.Roper, W.L.,Effectivenesspractice. NewW. Winkenwerder, C.M. Hackbarth, and H. Krakauer. 1988.in health care: an initiative to evaluate and improve medicalEngland Journal of Medicine 319(18):1197-1202.Schieber, C.J. and J.-P. Poullier. 1989. International health careexpenditure trends: 1987. Health Affairs 8(3):169-177.Short, P.F., A. Monheit, and K. Beauregard. 1988. Uninsured Americans: A1987 Profile. Rockville, Maryland: National Center for Health ServicesResearch and Health Care Technology Assessment.Taylor, H. 1990. Public Opinion, Some International Comparisons and theFuture of the U.S. Health Care System. Address given at the Epcot Center,Orlando, Florida, Feb. 1.Todd, J .S. 1989. It is time for universal access, not universal insurance.New England Journal of Medicine 321(1):46-47.Tuohy, C.J. 1986. Conflict and accommodation in the Canadian health caresystem. pp. 393-434 in Medicare at Maturity: Achievrnents, Lessons andChallenges, R. G. Evans and G. L. Stoddart, eds . Calgary: Universi ty ofCalgary Press.Wennberg, J.E., J.L. Freeman, R.M. Shelton, and T.A. Bubolz. 1989. Hospitaluse and mortality among Medicare beneficiaries in Boston and New Haven. NewEngland Journal of Medicine 321(17):1168-1173.HEALTH POLICY RESEARCH UNIT429 - 2194 Health Sciences MallUniversity of British ColumbiaVancouver, B.C. CANADAV6T 1Z6DISCUSSION PAPER SERIESPhysician Supply: Lessons fromFebruary 1988. (M.L. Barer, A.HPRU 88:1D Accommodating Rapid Growth inIsrael, Warnings for Canada.Gafni, J. Lomas)HPRU 88:1R Barer, M.L., Gafni,.A.and Lomas, J.(1989), "AccommodatingRapid Growth .in l'hysician Supply: Lessons from Israel,Warnings .for Canada", .International Journal of Health Services19(1):95-115HPRU 88:20 The Long Goodbye: The Great Transformation of the BritishColumbia Hospital System. March 1988. (R.G. Evans, M.L.Barer, C. Hertzman, G.M. Anderson, I.R. Pulcins, J. Lomas)HPRU 88:2R Evans, R.G. ,·Barer, M.L., Hertzman, C. ,Anderson, G.M.,Pulcins, I.R. and Lomas, J. (1989), ".The Long Goodbye: TheGreat Transformation of the British Columbia Hospital System",Health Services Research 24 (4): 435-459HPRU 88:30 Reading the Menu With Better Glasses: Aging and Health PolicyResearch. April 1988. (R.G. Evans)HPRU 88:3R Evans,R.G. (1989),UReading the Menu With Better Glasses:Aging and Health Policy Research", in S. J. Lewis (ed.), Agingand Health: Linking Research and Public pollcy, .LewisPublishers Inc., Chelsea, 145-167HPRU 88:4R Barer,M.L.(1988), "Regulating Physician Supply: TheEvolution of British Columbia's Bill 41", Journal of HealthPolitics, Policy and Law 13(1):1-25HPRU 88:50 Regionalization of Coronary Artery Bypass Surgery: Effects onAccess. May 1988. (G.M. Anderson, J. Lomas)HPRU 88:5R Anderson, G.M. and Lomas, J. (1989') ,"Regionalization ofCoronary Artery Bypass Surgery: Effects on Access", MedicalCare27(3):288~296HPRU 88:60 Diagnosing Senescence: The Medicalization of B.C.'s Elderly.July 1988. (M.L. Barer, I.R. Pulcins, R.G. Evans, C.Hertzman, J. Lomas, G.M. Anderson)HPRU 88:6R Barer, M.L. ,PulcinS,I.R. ,Evans, R.G., .Hertzman, C., Lomas,J .. and :Anderson,·G.M.(1989), "Trends in Use of MedicalServices by the Elderly in· British Columbia",· Canadian Medical. Association Journal 141 :39-45HPRU 88:70 The Development of Utilization Analysis: How, Why, and WhereIt's Going. August 1988. (G.M. Anderson, J. Lomas)HPRU 88:80 Squaring the Circle: Reconciling Fee-for-Service with GlobalExpenditure Control. September 1988. (R.G. Evans)D = Discussion Paper R =ReprintHEALTH POLICY RESEARCH UNITDISCUSSION PAPER SERIESHPRU 88:9D Practice Patterns of Physicians with Two Year Residency VersusOne Year Internship Training: Do Both Roads Lead to Rome?September 1988. (M.T. Schechter, S.B. Sheps, P. Grantham, N.Finlayson, R. Sizto)HPRU 88:10R Anderson, G.M. and Lomas,J. (1Q88), ".Monitoring the Diffusionof a Technology: Coronary Artery BypassSurgelCY in Ontario",American Journal of Public Health 78(3) :251-254HPRU 88:11R Evans, R.G. (1988), ""We'll Take Fare of it For You": HealthCare in the Canadian Community",Daedalusl17HJ:155:189.HPRU 88:12R llarer,M.L., Evans,R.&.~;;d£.ilieile,R.S.(i988),,,F~~controls as Cost Control: Tales From the Frozen North",· The. Milbank Quarterly 66(1):F64 .HPRU 88:13D Tension, Compression and Shear: Directions, Stresses andOutcomes of Health Care Cost Control. December 1988. (R.G.Evans)HPRU 88:13R Evans, R.G. (1990),"Tension,Compression,and)Shear:Directions, Stresses and Outcomes of Health Care CostControl", Journal of Health Politics, Policy and Law15(1) :101-128HPRU 88:14R Evans, R.G., Robinson,G.C.and Barer,M;L. (1988),'!WhereHave All the Children Gone? Accounting for the PediatricHospital Implosion", inR.S •. Tonkin and·J.R.Wright (eds.),Redesigning Relationships in Child Health Care, B.C.Children's Hospital, 63-76 .Long TermJanuaryMcCashin, N.HPRU 89: 1DHPRU 89:2DPhysician Utilization Before and After Entering aCare Program: An Application of Markov Modelling.1989. (H. Krueger, A.Y. Ellencweig, D. Uyeno, B.Pagliccia)Flat on Your Back or Back to Your Flat? Sources of IncreasedHospital Services Utilization Amo~g the Elderly in BritishColumbia. January 1989. (C. Hertzman, I.R. Pulcins, M.L.Barer, R.G. Evans, G.M. Anderson, J. Lomas)HPRU 89:2R Hertzman, C. ,Pulcins,·I.R., .Barer,M.L. ,Evans,.·R.G.,Anderson, G.M. and 'r.omas , J. ·0.990) ·"Flaton ¥ourBack or 'Backto·yourFlat? Sources .0fIncreased:HospitaLServicesUtilization Among· the Elderly in BritishCol\llllbia", Social.. Science .end Medicine30{7):819~§28·' ..HPRU 89:4DHPRU 89: 3RBuhler,L.. , Glick,N/and Sheps ,S.B.I1988J,i"prenatalC<lre..• A Comparative Evaluation of: Nurse",Midwives·and'Family :.:,: ..". "'.·•.·,.Physicians ..,.Canadian~edicalAssCJc~a~ion;ro!!rnal'139r397.,403Recent Trends in Cesarean Section Rates: Ontario Data 1983 to1987. February 1989. (G.M. Anderson, J. Lomas)nPRii ·89.:.tR •••:.Md~r~bn;:::·G .fol.··~rid'·'i.b~~·~";i5iJ1989·i,j'R~8~;;{i±r~nd~j.:;ii,)·'··, Cesarean SectionRatesiriOntario".,iCanadiani~edical:·Association Journal141:10~9-1053·· . .. . .HPRU 89:5D The Canadian Health Care System: A King's Fund Interrogatory.March 1989. (R.G. Evans)D = Discussion Paper R =ReprintHEALTH POLICY RESEARCH UNITDISCUSSION PAPER SERIESHPRU 89:6D Benefits, Risks and Costs of prescription Drugs in Ontario: AScientific Basis to Evaluate Policy options. April 1989.(W.O. Spitzer, G.M. Anderson, U. Bergman, J.L. Blackburn, E.Wang, M.C. Weinstein)HPRU 89:7D The Dog in the Night Time: Medical Practice Variations andHealth Policy. June 1989. (R.G. Evans)HPRU 89:8D Life and Death, Money and Power: The Politics of Health CareFinance. June 1989. (R.G. Evans)Trends in Australian Medical CareAugust 1989. (M. Barer, M. Nicoll,From Medibank to Medicare:Costs and Use, 1976-1986.M. Diesendorf, R. Harvey)HPRU 89: 9R Barer,M. L., Nicoll,M.j ,Diesendorf ,M. and Harvey,R. ' (1990) r"From Medibank to Medicare: Trends in Australian Medical CareCosts and Use From 1976 to 1986", Community Health StudiesXIV(1J:8-18HPRU 89:9DHPRU 89:10D Cholesterol Screening: Evaluating Alternative Strategies.August 1989. (G. Anderson, S. Brinkworth, T. Ng)HPRU 89:11R Evans, R.G., Lomas, J., Barer, M.L., Labelle, R.J., Fooks, C.,Stoddart, G.L.,Anderson,G.M., Feeny, D., Gafni, A.,Torrance, G.W. and Tholl, W.G. (1989), "Controlling HealthExpenditures - The Canadian Reality", New England Journal ofMedicine 320(9) :571-577HPRU 89: 12D The Effect of Admission to Long Term Care Program onUtilization of Health Services by the Elderly in BritishColumbia. November 1989. (A.Y. Ellencweig, A.J. Stark, N.Pagliccia, B. McCashin, A. Tourigny)HPRU 89:13D Utilization Patterns of Clients Admitted or Assessed but notAdmitted to a Long Term Care Program - Characteristics andDifferences. November 1989. (A.Y. Ellencweig, N. Pagliccia,B. McCashin, A. Tourigny, A.J. Stark)HPRU 89:14D Acute Care Hospital Utilization Under Canadian National HealthInsurance: The British Columbia Experience from 1969 to 1988.December 1989. (G.M. Anderson, I.R. Pulcins, M.L. Barer, R.G.Evans, C. HertzmanJHPRU 90:1R Anderson, G.M., Newhouse, ,J.P. and Roos,L.L. (1989),"Hospital Care for ,Elderly Patients with 'Diseases of theCirculatory System. A Comparison' of ,Hospital Use in theUnited States and Canada'~,New,England Journal of Medicine321:1443~1448 ',HPRU 90:2D Poland: Health and Environment in the Context of SocioeconomicDecline. January 1990. (C. Hertzman)HPRU 90:3D The Appropriate Use of Intrapartum Electronic Fetal Heart RateMonitoring. January 1990. (G.M. Anderson, D.J. Allison)HPRU 90:4D A Comparison of Cost Sharing Versus Free Care in Children:Effects on the Demand for Office-based Medical Care. January1990. (G.M. Anderson)D = Discussion Paper R =ReprintHEALTH POLICY RESEARCH UNITDISCUSSION PAPER SERIESHPRU 90:5D Does Family Practice Certification Affect Practice Style? AnAnalysis of Office-based Care. February 1990., (G.M.Anderson)HPRU 90:6D An Assessment of the Value of Routine Prenatal UltrasoundScreening. February 1990. (G.M. Anderson, D. Allison)HPRU 90:7R N~etz, ·l'.N./Baliard,D.J.,Beard,C.M.;Lucl~:i.~,J.tTangalos, E.G., :Kokmen,E., Weigel,K.M., .Belau,>P.G .., . Bourne,W.M. and Kurland, L.T. (1989) "An Anatomy of the ·Autopsy,Olmsted County ,1935 through 1985", Mayo .C:linicProceedings64: 1055-1064 . . ...HPRU 90:8R. Nem~tz,l'.N;, Beard, C;M. r .. Banard,i~.J,,£~~~i;,§.,··Tangalos,E;G. ,Kokmen,E. ,·weigel, ·K.M•.,·Belau,·· P.G.,'BO\lrne,W. M. and Kurland, .L.T.. ·( 1989) "Resurrecting the Autopsy:Benefits and Recommendations" Mayo Clinic Proceedings64:1065-1076HPRU 90:9D Technology Diffusion:Study of Low and High1990. (A. Kazanjian,The Troll Under the Bridge. ATechnology in British Columbia.K. Friesen)PilotMarchHPRU 90:10R Sapphires in the Mud? The Export Potential of American HealthCare Financing. Enthoven, A.C.(1989),"what Can EuropeansLearn from Americans?'.' r Evans, R. G., Barer ,:M. L. "(1989) rComment. Heal th Care Financing Review, .AnnualSupplement 1989HPRU 90:11D Healthy Community Indicators: The Perils of the Search and thePaucity of the Find. March 1990. (M. Hayes, S. Willms)HPRU 90:12D Use of HMRI Data in Nineteen British Columbia Hospitals andFuture Directions for Case Mix Groups. April 1990. (K.M.Antioch)HPRU 90:13D Producing Health, Consuming Health Care. April 1990. (R.G.Evans, G.L. Stoddart)HPRU 90:14D Automated Blood Sample-Handling in the Clinical Laboratory.June 1990. (W. Godolphin, K. Bodtker, D; Uyeno, L.-O. Goh)HPRU 90:15D Hospital-Based Utilization Management: A Cross Canada Survey.June 1990. (G.M. Anderson, S. Sheps, K. Cardiff)HPRU 90:16D Hospital-Based Utilization Management: A Literature Review.June 1990. (S. Sheps, G.M. Anderson, K. Cardiff)HPRU 90:17D Reflections on the Financing of Hospital Capital: A CanadianPerspective. June 1990. (M.L. Barer, R.G. Evans).HPRU QO:18D Accessible, Acceptable and Affordable: Financing Health Carein Canada. september 1990. (R.G. Evans)D = Discussion Paper R =ReprintHEALTH POLICY RESEARCH UNITDISCUSSION PAPER SERIESHPRU 89:6D Benefits, Risks and Costs of Prescription Drugs in Ontario: AScientific Basis to Evaluate Policy Options. April 1989.(w.O. Spitzer, G.M. Anderson, U. Bergman, J.L. Blackburn, E.Wang, M.C. weinstein)HPRU 89:7D The Dog in the Night Time: Medical Practice Variations andHealth Policy. June 1989. (R.G. Evans)HPRU 89:8D Life and Death, Money and Power: The Politics of Health CareFinance. June 1989. (R.G. Evans)Trends in Australian Medical CareAUgust 1989. (M. Barer, M. Nicoll,From Medibank to Medicare:Costs and Use, 1976-1986.M. Diesendorf, R. Harvey)HPRU 89:9R Barer 'M .L., Nic8n,M. ,>Diesendorf,·M. and Harvey,R. (1990),"From Medibank to Medicare:· Trends in Australian Medical CareCosts and Use From 1976 to 1986", Communi ty Heal th StudiesXIV(1):8-18HPRU 89:9DHPRU 89:10D Cholesterol Screening: Evaluating Alternative Strategies.August 1989. (G. Anderson, S. Brinkworth, T. Ng)HPRU89:11REvans, R.G., Lomas, J.,Barer, M.L., Labelle, R.J., Fooks, C.,Stoddart, G.L. ,Anderson,G.M., l'eeny,·D., Gafni, A.,Torrance,G.W. and Tholl, W.G.(1989),"Controlling HealthExpenditures - The Canadian Reali ty",New England Journal ofMedicine 320(9):571-577HPRU 89: 12D The Effect of Admission to Long Term Care program onUtilization of Health Services by the Elderly in BritishColumbia. November 1989. (A.Y. Ellencweig, A.J. Stark, N.Pagliccia, B. McCashin, A. Tourigny)HPRU 89:13D Utilization Patterns of Clients Admitted or Assessed but notAdmitted to a Long Term Care Program - Characteristics andDifferences. November 1989. (A.Y. Ellencweig, N. Pagliccia,B. McCashin, A. Tourigny, A.J. Stark)HPRU 89:14D Acute Care Hospital Utilization Under Canadian National HealthInsurance: The British Columbia Experience from 1969 to 1988.December 1989. (G.M. Anderson, I.R. Pulcins, M.L. Barer, R.G.Evans, C. Hertzman)HPRU 90:1R Anderson, G.M., Newhouse, ,J.P. and Roos,L.L. (1989),"Hospital Care for .Elderly Patients with Diseases oLtheCirculatory System.· A Comparison of ,Hospital Use in theUnited States and Canada",NewEngland Journal of Medicine321: 1443-1448. HPRU 90:2D Poland: Health and Environment in the Context of SocioeconomicDecline. January 1990. (C. Hertzman)HPRU 90:3D The Appropriate Use of Intrapartum Electronic Fetal Heart RateMonitoring. January 1990. (G.M. Anderson, D.J. Allison)HPRU 90:4D A Comparison of Cost Sharing Versus Free Care in Children:Effects on the Demand for Office-based Medical Care. January1990. (G.M. Anderson)D =Discussion Paper R = Reprint11111111111111111111111111


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items