UBC Faculty Research and Publications

Reflections on the revolution in Sweden Evans, Robert G., 1942- Sep 30, 1991

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

Item Metadata


52383-Evans_Robert_Sweden.pdf [ 19.39MB ]
JSON: 52383-1.0048513.json
JSON-LD: 52383-1.0048513-ld.json
RDF/XML (Pretty): 52383-1.0048513-rdf.xml
RDF/JSON: 52383-1.0048513-rdf.json
Turtle: 52383-1.0048513-turtle.txt
N-Triples: 52383-1.0048513-rdf-ntriples.txt
Original Record: 52383-1.0048513-source.json
Full Text

Full Text

REFLECTIONS ON THE REVOLUTIONIN SWEDENRobert G. EvansHPRU 91:90 SEPTEMBER, 1991REFLECTIONS ON THE REVOLUTIONIN SWEDENRobert G. EvansHPRU 91:90 SEPTEMBER, 1991HEALTH POLICY RESEARCH UNITCentre for Health Services and Policy Research429 - 2194 Health Sciences MallUniversity of British ColumbiaVancouver, B.C. CANADAV6T 1Z3The Centre for Health Services and Policy Research was established by theBoard of Governors of the University of British Columbia in December 1990.It was officially opened in July 1991. The Centre's primary objective isto co-ordinate, facilitate, and undertake multidisciplinary research inthe areas of health policy, health services research, population health,and health human resources. It brings together researchers in a varietyof disciplines who are committed to a multidisciplinary approach toresearch, and to promoting wide dissemination and discussion of researchresults, in these areas. The Centre aims to contribute to the improvementof population health by being responsive to the research needs of thoseresponsible for health policy. To this end, it provides a researchresource for graduate students; develops and facilitates access to healthand health care databases; sponsors seminars, workshops, conferences andpolicy consultations; and distributes Discussion Papers, Research Reportsand publication reprints resulting from the research programs of Centrefaculty.The Centre's Health Policy Research Unit Discussion Paper series providesa vehicle for the circulation of preliminary (pre -publication) work ofCentre faculty and associates. It is intended to promote discussion andto elicit comments and suggestions that might be incorporated within thework prior to publication. While the Centre prints and distributes thesepapers for this purpose , the views in the papers are those of theauthor(s).A complete list of available Health Policy Research Unit Discussion Papersand Reprints, along with an address to which requests for copies should besent, appears at the back of each paper.REFLECTIONS ON THE REVOLUTION IN SWEDEN*Robert G. EvansDepartment of Economics, andCentre for Health Services and Policy Research,University of British ColumbiaSeptember, 1991Paper prepared as part of the SNS International Panel Review of theSwedish Health Care System. Stockholm, May-August, 1991*with particular thanks to Goran Arvidsson, Johan Calltorp, Bengt Jonsson,Anders Milton, and my panel colleagues, none of whom can be blamed forremaining errors of fact or interpretation.Author's research supported by a National Health Scientist award fromHealth and Welfare Canada, and by the Canadian Institute for AdvancedResearch.2INTRODUCTION AND DISCLAIMERWhile participating in the SNS review of the Swedish health care system, Ihave also been a member of a Royal Commission which for eighteen monthshas been reviewing the health care system of the Province of BritishColumbia, in Canada. There is a massive contrast between the two groups,in time, staff, recources, and background information brought to the task.And yet, the Commission's resources were also quite inadequate to providea detailed description and analysis of the myriad of issues and problemswhich were presented to it . A useful report, for B.C. as for Sweden, willbe one which identifies certain key issues and fundamental ideas, andmakes suggestions for the broad directions of policy. The specificdetails must inevitably be worked out in actual practice by the personswho know the system, and who have to live and work with the results . Thispaper, and the Commission report , will both be much more general andimpressionistic than a research study for an academic audience. But theymay nevertheless be some help . ("An nescis, mi fili, quantilla prudentiaregitur orbis?" - Count Oxenstierna to his son, 1648)Moreover , most of what the panel have found in Sweden was in truth broughtwith each of us in our own intellectual baggage. Lacking the time andresources really to understand Sweden and the Swedish health care system ,we react to what we are told, or read, on the basis of what we alreadyknow from home. We translate our impressions into familiar contexts so asto grasp and order them. This is inevitable , but also potentiallymisleading for us and for the readers.Of course our understanding of our own systems, when clearly labelled assuch , may be of interest and use . We are on firmer ground in describingwhat our own countries have done, and with what effect, than when we tryto say what Sweden has done or should do. We may also be able to makecertain generalizations which appear to hold across a number of countriesand systems. Such generalizations are already somewhat detached from ourdomestic contexts. But the reader must still approach this work quitecritically . The Swedish interpreter, in the broadest sense , has as largea task as the authors.INTERPRETING FOREIGN EXPERIENCE - TRAPS FOR THE UNWARYFor many years, people from allover the world have come to study theSwedish health care system. Sweden has set a standard and a model thatothers have aspired to, and has been the teacher more than the learner.It is thus understandable that Swedes, particularly those working in thehealth care system, might have come to see themselves as having found theanswers, the "right" way to organize and fund health care. Elsewhere, weknow our systems are at best a more or less adequate set of compromises.Health care organization is an evolving game or contest among interestswhich are partly parallel, and partly in conflict. The broad structure ofa system sets the framework within which this game is played, and theplayers develop and apply their changing strategies.3Each system has particular strengths and weaknesses, and each couldcertainly be improved . When we look at others, we see suggestions as tohow we could do better, and also warnings. Certain issues and problemsare universal, but some systems do work better than others, on quitegeneral criteria. In Canada we derive a certain pride from observing oursuccess relative to the United States , a success judged by ourselves , andalso by many Americans. l But even the U.S. has useful lessons to offer,positive as well as negative.But if Swedes once thought they had all the answers, they no longer thinkso. Now, we are told, there is a crisis of confidence in Sweden whichincludes, but is much broader than the health care system. Some haveswitched from the belief that Sweden has all the answers, to the beliefthat she has none, and that the system requires radical restructuring orcomplete replacement . Some other system, actual or hypothetical, has "theanswers" .We have even been told that the collapse of the centrally plannedeconomies in eastern Europe has had an echo effect in Sweden, and thatsweeping generalizations are being made in some quarters about the"failure" of planning, and the superiority of "market" systems, althoughthe dimensions of superiority are never specific. If this is so, itsuggests a serious and potentially dangerous lack of information aboutforeign health care systems, and a lack of experience in learning fromthem . Indeed this view has disturbing parallels with thinking in theformer Soviet satellites of eastern Europe, where it clearly does arisefrom decades of ignorance about what is really going on elsewhere, andfrom the exchange of one set of ideological blinders for another.THE MARKET FOR IDEAS - AN ORIENTAL BAZAARCaveat emptor applies in the intellectual marketplace as in all others,and international trade provides less consumer protection than do mostdomestic markets. There is a great deal of poor quality merchandise inthe market for ideas about health care systems, purveyed by theideologically committed, the intellectually inadequate, and the down-rightdishonest. Ideas which are not accepted at home, or which have been triedand have failed, are dumped on the world market in the hope of findingless knowledgeable consumers.A principal source of such shoddy goods is the United States. In thefirst place, the very serious and acknowledged flaws of that country'shealth care system make it an astonishingly fertile breeding ground fornew ideas, both good and bad. While the other countries of the GECD havemore or less satisfactory health care systems, albeit with significant1 The comparative literature in North America - academic studies,official reports, and opinion polls, is large and growing rapidly. Arecent survey, which gives many of the references supporting points madein this paper, is Barer, Evans and Hertzman, (1991) .4problems, the U.S. has a uniquely bad combinaton of both extreme inequityand extreme inefficiency which makes others' problems appear trivial bycomparison.Secondly, the sheer size and complexity of the United States makes it verydifficult for outsiders to be aware of what is actually happening. Thusfalse claims of policy success, made from some combination of wishfulthinking, ignorance, and intent to deceive, may have much more influenceoff-shore than they deserve.And finally, while the extraordinarily costly U.S. system places heavyburdens on governments, employers, and consumers, it also creates a verylarge group of income earners with a stake i~ the continuation of theexisting system. Every dollar of expense, whether effective or completelywasted, represents a dollar of someone's income. The most recentestimates place the cost of administrative waste alone in the U.S. systemat between $90 and $120 billion per year, in addition to the cost ofineffective , inappropriate , or overpriced clinical services (~oolhandlerand Himmelstein , 1991). This huge pool of resources is available tofinance the defence of the present s ystem, as the conflict between thosewho pay and those who receive becomes more and more intense.ANALYSIS AND PROPAGANDA - THE DEFENSE OF ECONOMIC INTERESTCanadians are particularly sensitive to this conflict, because in recentyears a number of observers in the U.S. - ranging from individualphysicians and academics to the Congressional Government AccountiungOffice, the American College of Physicians, and influential members of thecorporate elite - have begun to advocate a change to a health care systemmore like that of Canada. The response has been an intense ."disinformation campaign", supported by the American Medical Associationand the Health Insurance Association of America among others, to discreditthe Canadian health care system in particular, but more generally toattack any form of foreign alternative, and to misrepresent the strengthsand weaknesses of the U.S. system. Natural jingoism is powerfullyreinforced by threatened markets and incomes, on a massive scale.This campaign ranges from scurrilous advertising and false or grosslydistorted news stories, to include both pseudo-academic analysis and theselective emphasis and misrepresentation of genuine research findings.$100 billion at risk will buy a good deal of publicity; and academicsenjoy no immunity to temptation. All this misinformation - toxic waste? ­spills over the border through the closely linked news media. Our ownunderstanding of and confidence in our system is thus undermined as aside-effect of the desperate struggle by the beneficiaries of the Americansystem, to preserve their current position.Such language may appear extreme, but so is the phenomenon described . Agigantic industry is fighting a defensive battle for very high stakes,with no holds barred. Anyone attempting to learn from internationalexperience would be well advised to be aware of this massive5dis information effort, and exercise due caution . 2at the San Diego zoo was crushed to death when sheelephants who were having a mild disagreement. Inare now fighting .WHAT SEEMS TO BE THE PROBLEM?Last year , an attendantwas caught between twothe U.S. the elephantsPerhaps the most puzzling feature of our assignment is that we have beeninvited at all. As noted above, the Swedish health care system has longbeen the world magnet for students of health care . The overall indicatorsof health status still show Sweden to be among the healthiest of nationsin the world. It is traditional to apologise for the inadequacy of suchmeasures as life expectancy and infant mortality, but it is salutary tonote that 1n eastern Europe, these have actually deteriorated in recentyears. So they may not be so insensitive after all . On such measures,Sweden continues to improve.The population has access to a modern, well-supported health care system,participating in and contributing to technological advance. Nor did wehear any more than the usual complaints about inadequate provision andregional disparities . Such complaints are universal, and do not seem todepend either on how a system is organized and funded, or on what level offunding is provided. But in terms of equity of access and treatment, aswell as equity of outcomes, the Swedish system seems to be one of the bestin the world . It is not perfect, but looks good relative to the knownalternatives.Health inequalities persist, but are less pronounced than in the U.K.(Vagero and Lundberg, 1989) and a fortiori than in the U.S. It is notablethat the standardized mortality rate of working-age Swedish males in thelowest social classes is lower than that of British males in the highestclasses, although this reflects much more than the success of the healthcare system.Finally, in broad terms , the system appears to be affordable. During the1970s , health care costs in Sweden rose very rapidly, to the point that asa percentage of national income they took a larger share even than in theUnited States (Schieber and Poullier, 1989). Such a trend cannot besupported indefinitely . As the U.S. experience shows, uncontrolled costescalation results in growing distortions and increasingly severe problemsas payers struggle ever harder to push the burden onto some one else.But this trend did not continue . In the 1980s Sweden controlled, andactually reduced, its share of national income devoted to health care. Itis still expensive, but no more so than Canada, Germany, France or theNetherlands. And Sweden has a relatively elderly population. If thepercentage of national income can be held at its present level, the burden2 An attempt to set part of the record straight is provided in Barer,Welch and Antioch (1991).6is heavy but not unbearable.Sweden appears to have a health care system which is supporting anexceptionally healthy population, with high quality care, at a cost which,while high, is not out of line with what other similar countries arespending. And that cost is at the moment relatively stable, and 'sustainable. All this suggests that there is no real problem; n5vensksjukvard i topp ... n (Landstings-vArlden, 1991).Yet there is. In the first place, if enough people believe there is aproblem, then that is a problem. Furthermore, they may be right . But'wha t is this problem, which refuses to show itself in the aggregate healthdata? Before we can make any useful suggestions for therapy, we need adiagnosis. And for that, we need a medical history.HEALTH CARE COSTS AND THE ECONOMY : THREE PHASES OF EVOLUTIONIn North America, immediately after the Second World War, there was astrong social consensus supporting major expansion of the health caresystem . There were disagreements across the political spectrum as to thebest mechanisms for this expansion, and the relative importance of carefor different segments of the population. Views on these matters differedparticularly sharply across the Canada-U .S . border. But the view thatmore resources should be provided for health care - more people, morebuildings , more , and more sophisticated equipment, and more money - wasvirtually universal.Supported by this consensus, from 1950 to 1970 health care absorbed asteadily increasing share of a national income which was itself growingquite rapidly. The expectation of endless growth - to meet the "needs"which were themselves constantly redefined and expanded - took deep rootin the culture of the health care system. No facility was ever closed,except to be rebuilt, no jobs were ever lost, no budgets reduced - thehorizons were limitless. And this was as it should be, for health waspriceless , and health care was the obvious route to health. The rest ofthe population agreed.By the late 1960s, it was increasingly clear to those responsible foradministering and paying for health care that it was starting to crowd outother social priorities. Public documents and official statements on costcontainment began to appear increasingly frequently in both Canada and theUnited States. Health economics became an identifiable sub-discipline .The difference betweem the two countries was that by 1971 Canada hadcreated a public administrative and payment structure for hospital careand physicians' services which made cost control possible. We had newmechanisms to address new priorities. The United States did not.Starting in the 1970s , Canada began to hold the growth in health carecosts to the same rate, more .or less , as the rise in national income. Inthe United States, health care continued to consume an ever larger share,accompanied by alternative waves of hand-wringing despair, and premature7celebrations of the success of the latest cost-containment fad.During this second phase there was in Canada no explicit consensus tolimit the growth of health care; the idea that "more is better" continuedto dominate public discussion. But there was acceptance of governmentpolicies that held down cost growth, despite predictable grumbling fromproviders. Nor was that grumbling too intense, because the industry wasstill expanding, even if not as fast as before,The third phase began in the early 1980s, when the rate of economic growthfell sharply. At first , health spending was maintained, with the resultthat in a matter of two years its share of national income rose by over apercentage point - the first significant increase in the ratio in tenyears. Much more rigorous control measures were then applied, and for therest of the decade, health care has been held to a constant share of anational income which is itself rising very little.And now, the political pressures and complaints of cut-backs are reallybecoming acute, The thirty year expansion is over . New priorities mustbe funded by moving the funds from somewhere else in the system. Althoughthere are in fact no global cutbacks, and the health care system has morereal resources than at any other time in its history, the politicalpressure for more is more intense than it has ever been . Apparently manypeople in the health care system genuinely believe that it is on the vergeof collapse for lack of resources - because it is not expanding!One has the impression that this three phase history also characterizes anumber of the countries of western Europe. The United States became stuckat the end of phase one, because its payment structure did not permit itto restrain the growth in health care share; with respect to cost controlthey are still in the late 1960s. But in most other countries there hasbeen a period of social consensus for expansion , followed by restraint,followed by much more severe restraint as the overall economic climate hasworsened.In a sense, then, the problems in health care have their source in thegeneral economy . If the growth rates of the post-war decades had beenmaintained, and if health care had simply shared in that growth, would allthe countries of the OECD be struggling with health care reform? It isthe conflict between a low , or no , growth economy, and a health caresystem conditioned to endless expansion, which creates the socialfriction, the political outcry, and the sense that "the system iscollapsing".Our very brief time in Sweden suggests that this has also been the patternhere. Phase one lasted longer than in Canada - up to the late 1970s - andphase two was quite short. Sweden moved almost immediately into phasethree in the early or middle 1980s, and thus may not have had as much timeas others to evolve and become experienced with control mechanisms in aless rigorous environment. In phase three, the political pressure fromthe health care system, and the constant threats of imminent collapse ifmore resources are not available at once, begin to undermine popular8confidence in and support for the existing structures.As that happens. those with greater wealth and better health begin toquestion whether they themselves might not be better off if they could u~esome of their own money to buy superior access to public resources whichthey perceive as increasingly constrainted - i.e. buy their way to thefront of the queue. This could be through either user charges for publicservices, or greater access to privately supplied services subsidized frompublic funds. (The market for truly unsubsidized private care is toosmall to be interesting, in any country; the famous "private" fundingsystem of the U.S. depends on huge subsidies through the tax system.)Some providers will happily participate, if they can directly orindirectly pocket the payments for preferred 'a cce s s . The anxieties of thebetter-off can be used to draw in additional private funds, and keep thehealth care system growing - more services, and higher incomes. And herewe are.NO LIMITS: THE INTERNAL DYNAMIC OF HEALTH CAREBut the deterioration in the external economy is not the whole story, oreven the most important part. It is remarkable that, despite years.decades, of steady expansion, there is no country - not even the UnitedStates - in which the providers of health care have come to feel that theneeds have been met, that enough is enough. Those who pay for the system,taxpayers, premium-payers, employers, respond to the increasing pressureby trying to enforce constraints, with varying success in differentcountries. But the health care system, like a vascularized neoplasm, justwants "MORE"! What has gone wrong with this collection of cells, thisgroup of individuals and organizations, that they cannot seem torecognize, even in principle, the need for limits?The neo-classical economist has , as usual, a simple answer. Demand isinfinite when care is free. And the simple answer is, as usual, wrong.It is, as usual, based on an a priori theory which is psychologicallytrivial, and makes no effort to understand the factors actually affectinghealth care utilization, or even to observe its evolving patterns. 3 Mostobviously , this response ignores the egr'egdous fact that the most rapidescalation of servicing and costs is 'Jccur r i ng in the United States, wherefor most people . care is very defini ceLy not free. but comes withsubstantial out of pocket charges. But there are also important patternswithin health care use to reflec t upon.Oversimplifying a complex reality, these can be summarized as, (1) thenarrowing of the client base, (2) the absence of accountability forresources and outcomes, and (3) the broadening of the definition of3 An attempt is made in Evans (1984) to provide an economic theoreticframework for the analysis of health care which is more consistent bothwith the external world, and in its internal logic.9health. These reinforce each other, and all exacerbate the fundamentalproblem of the erosion of the social consensus and the politicalconstituency, while simultaneously supporting the continuing extension ofthe definition of "need".NARROWING THE CLIENT BASEHealth care use and cost is becoming increasingly concentrated on a smallminority of the population. This is most apparent when one examineschanges over time in age-specific use rates. Per capita use is risingvery rapidly among the very elderly, and much less among the rest of thepopulation. Use of hospitals in particular by the non-elderly is actuallydropping quite fast. We have studied this process in some detail inCanada , and have found similar patterns in U.K. and U.S. data, but theSwedish experience is especially notable.Table 1Health Expenditures per Capita in Swedenby Age Group, 1976 and 1985AgeGroupPopulationDistributionin 1985Expenditures per Capita(in 1985 SEK)1976 1985(1) (2) (3) (lx2) (lx3) (5-4)o -14 18.16% 2691 3049 489 554 6515-44 42.72% 4294 4305 1834 1839 545-64 21.73% 7559 7273 1643 1580 -6365-74 9.96% 13392 15624 1334 1556 22275+ 7.44% 30686 47333 2283 3522 1239Total: 7583 9051 1468Source: H~lsan i Sverige 1987/88 (1988) Table A-3; Gerdtham and Jonsson(1991) Table 4.Swedish health expenditures per capita were 9051 SEK in 1985. But if thepopulation of 1985 had experienced the age-specific use rates of 1975, at1985 prices, average per capita costs would have been only 7583 SEK . Theincrease in per capita use raised costs by 19.36 percent from 1976 to1985, or just under 2 percent per year. But all of this increase isaccounted for by increasing spending for those over 65, and almost all ­nearly 85 percent - by increases for those over 75. For the majority ofthe population, between the ages of 15 and 65, per capita expenditureswere actually falling .10Hospital bed-days for the non-elderly population have fallen 20-25 percentover the same period. But cost per day - after adjusting for inflation ­has doubled for the children. This, and the increase in expenditures forchildren in Table I, may reflect very large increases in costs forneonatal intensive care for low birth weight babies, and for the care ofsubsequent health problems which such "salvaged" children experience overmany years.Table 2Hospital Use and Cost per Capita in Swedenby Age Group, 1976 and 1985, 1985 SEKAge 1976 1985 Change PercentGroup changeo -14Bed-days per capita 0.60 0.47 -0.13 -21.7%Cost/day SEK 1292 2617 1325 102.6%Cost per capita SEK 775 1230 455 58.7%15-44Bed-days per capita 1. 33 1.01 -0.32 -24.1%Cost/day SEK 1247 2183 936 75.1%Cost per capita SEK 1659 2205 546 32.9%45-64Bed-days per capita 3.25 2 .53 -0.72 -22.2%Cost/day SEK 1202 1671 469 39.0%Cost per capita SEK 3907 4228 321 8.2%65-74Bed-days per capita 8.20 7.69 -0.51 - 6.2%Cost/day SEK 1050 1414 364 34.7%Cost per capita SEK 8610 10874 2264 26.3%75+Bed-days per capita 27.80 34.27 6.47 23.3%Cost/day SEK 850 1146 296 34.8%Cost per capita SEK 23630 39273 15643 66.2%Source: Gerdtham and Jonsson (1991) Tables 5 and 6.The low, and less rapidly rising, hospital costs per day for the veryelderly, combined with their very high and rapidly rising use rates,suggest a hospital system which is increasingly being used for long-terminstitutional care. As we have found in Canada, it is becoming twosystems in one (Evans et al., 1989) . The "mainstream" population (non­elderly and independent elderly) use fewer and fewer bed-days, but forthem the intensity of servicing and cost per day are increasing rapidly.On the other hand a growing population of dependent elderly is using manymore hospital days, but few services per day, and is de facto permanentlyinstitutionalized within the hospital system.11It is also important to emphasize what these tables do not show, sincethere has been so much erroneous rhetoric, particularly in North America,about the burden of an aging population. They do not show a problem ofincreasing use and costs resulting from increased numbers of elderly.Rather, the increases result from changing patterns of care for elderlypeople - Table 1 uses the 1985 population to calculate per capita costs inboth 1976 and 1985. This is precisely what we have found in Canada.Changes in the age structure of the population have had quite smalleffects on use, which are overwhelmed by the impact of changes in age­specific, per capita rates. The effects of projected future changes onaggregate costs are also quite modest.For the basically healthy, who suffer an injury or develop an illness, thepossibilities of care are limited. But the advance of technologycontinually increases the range of interventions that one can bring tobear on the elderly, the chronically ill, those with serious andultimately intractable problems, whether or not they are near to death.This means that, over time , their care is coming to dominate the healthcare system. As Table 1 shows, those over 75 accounted for nearly 40percent of all health care costs in 1985, compared with 30 percent if theuse rates of 1976 had persisted. Those over 65 took up 56 percent; theywould have received only 47 percent ten years before. If one couldfurther isolate from the costs of the non-elderly, those expenditures forlow birth weight infants and for other persons with chronic disablingconditions, the extent and increase in concentration would be evengreater.The people who not only pay for the health care system (whatever thesocial arrangements for payment), but also support it in some broadersocial and political sense, are receiving a steadily falling share of itsattention , even as they are asked ever more urgently to contribute more toits financing . Since the system as a whole is now growing much moreslowly, this mainstream population are being crowded out or at leastexperiencing increased problems of access. 4From a professional perspective, this may seem inevitable and evenappropriate. When resources are scarce, should not providers respondfirst to those in most need, as defined by professionals? If the generalpopulation feel inconvenienced, or worse, by waiting times, queues, by asystem dominated by professional priorities rather than responsive toconsumers, well, they just have to realize that there are people muchworse off than they. After all, the system does not have enough resourcesto take care of everybody, when and where they might like . "We told youwe were underfunded. If you don't like it, send more money!"4 It may be argued that, viewed over the life cycle, these are thesame people. Most of us will eventually be old. What appears as aninterpersonal transfer , is in the longer view an intertemporal one . Butthe general population are still being crowded out by this major shift inthe emphasis of the health care system , even if in some sense by"themselves" at older ages.12The shift in professional priorities would not be a problem if the generalpopulation believed that its own needs for care were static or declining.But improvements in technology and changes in the definition of "need"have created certain very prominent and specific new areas of concern.These new services range from the generally effective and helpful - kneeand hip replacements - through the sometimes helpful but expensi~e andcontentious - various forms of cardiac surgery - to the clinically verydubious and quite possibly harmful - mass cholesterol screening and drugtreatment.Such activities in fact take up a relatively small part of the overall.he a l t h care system, and of its costs. But to meet these new prioritiesrequires either more resources, or a transfer from other activities andpriorities. And in every country, it seems, the health care system isstrongly entrenched to fight off attempts to reduce capacity in any area.Each new technology or need becomes recruited in support of the strugglefor growth - they are "fiscal can-openers" to open the treasury, and breakthe political will to control costs.These new procedures lend themselves to publicity campaigns which createunease among the general population - "the care will not be there when Ineed it, and I may die!" And there is genuine hardship as well - thefrightened person on the waiting list for cardiac surgery, fearingimminent death and unable to find out when he will be "saved", or thearthritic cripple, immobilized and in pain, waiting to walk again. Theresult is increasing political pressure in controlled systems like that ofCanada, and explosive cost growth in the uncontrolled U.S. The pressureis both to relax constraints on public funds, and to expand the role ofprivate payment and delivery. "For a few dollars more", a parallelprivate system will ensure that the consequences of shortages are borne bysomeone else .ACCOUNTABILITY FOR RESOURCES AND OUTCOMESYet , as our Commission discovered, these highly publicized pressure pointsare very clearly not the result of an overall shortage of resources . Theyare traceable to inadequate management of the resources available. Onecardiac surgeon's patients must wait for months or years , another has fewor no waiting - but no one on the first list is told that the second isavailable! An orthopedic surgeon is unable to do more hip replacements athis hospital because the budget for prostheses is exhausted; he is offeredfacilities at a hospital some miles away, but does not want to travel.Patients are told only that "the system is underfunded", and some of thembecome interested in private care. This pattern, of problems arising frominadequate management and professional discretion being presented to thepublic as inadequate funding, is probably universal.The underlying problem is that in no modern health care system haveproviders been held accountable for either the effectiveness or theefficiency of what they do - either the health improvements achieved, orthe resources used in the process. Public or private funding, public or13private delivery - the same issues and observations emerge. How does oneensure that providers both do the right things, and do things right?The most widely observed consequences of this lack of accountability arethe very large differences from one provider, city, region, or country toanother in the rates of performance of particular surgical procedures, inuse of hospital beds, and in other aspects of medical practice (Andersenand Mooney, 1990). Efforts to relate these differences to measures of thepopulation's needs have not been successful. These large variations inpractice patterns suggest that a good deal of the care that is presentlybeing provided may be unnecessary or inappropriate - providers are notdoing the right things (or are doing the wrong things).And there is powerful supporting evidence for this concern, from theanalysis of particular procedures by clinical epidemiologists.Investigators have been pointing out, for a number of years, that theeffectiveness of most of the medical care being provided in any modernsystem has bever been scientifically demonstrated, and that a significantproportion has been shown to be ineffective or even harmful. Yet suchactivities continue, in the absence of any systematic processes toidentify and discourage inappropriate care.But it is also important to do things right. This includes traditionalconcerns for the "quality of care", but it also extends to the technicalefficiency with which care is provided. Here the relevant evidence is ofsignificant variations in productivity and unit costs from onepractitioner, clinic, or hospital to another. And again, if there are noeconomic or administrative incentives to encourage improved productivity,wasteful and inefficient practices are unlikely to be identified ormodified.The evidence that there are significant potential improvements to be madein both effectiveness and efficiency of care applies both to the care ofthe majority or "mainstream" population, and to that of the small minorityof very elderly or very ilIon whom care is increasingly beingconcentrated. The appropriateness of aggressive surgical and otherinterventions on very elderly patients is as questionable as the massscreening and treatment of the general population for elevatedcholesterol.BROADENING THE DEFINITION OF HEALTHOne might have thought that broadening the definition of health, andplacing more emphasis on preventive services, both of which receive muchrhetorical support internationally, would widen the client base of thehealth care system. Indeed advocates of preventive care, and moregenerally of healthier behaviour and physical and social environments,allege that these will both improve health and lower the use and costs ofcare. Clearly this is not happening.14Drawing on the experience of our Commission, one has the impression thatinstead a variety of physical, emotional and social disabilities anddysfunctions are being absorbed into the definition of "illness", at leastfor the purpose of including their proposed remedies as "health care", andthereby expanding the "needs" which the health care system must meet.There are a number of people in any society whose circumstances areunsatisfactory, and some are appalling. More services, of a variety ofsorts, could make them better off, happier, if not necessarily healthierin a narrow sense of the word. Indeed, probably anyone of us could bemade better off by some services. But we cannot and should not include·a l l of these under the umbrella of health care, with all the regulatoryand financing implications that follow.But many providers seem to try to respond to the absolute circumstances ofthe client, serving the most disadvantaged first and most , with lessconcern for whether the services offered can legitimately be consideredmedical, or even whether they are likely to do much good. Thus thebroadened definition of "needs" results in still more services for peoplewho are indeed in distressing circumstances - the very elderly , thedisabled, the chemically dependent, the mentally ill or retarded, theterminally ill - for whom very little improvement in health narrowlydefined may be possible.The statutory basis for the public financing of health care in Canadastates that all medically necessary care shall be free . It is silent asto who shall decide, on what criteria, what is medical , and what isnecessary . Some that is medical, is not necessary, and some that isnecessary, is not medical . How to decide? The broader ones' definitionof health, the more difficult it is to find a basis for exclusion, and toconstrain the growth of health care.Moreover, as the definition of health is broadened, providers and clientsboth escape from the rigorous discipline of the clinical epidemiologist.Outcomes defined in terms of mortality, morbidity, and functional capacitycan in principle be unambiguously measured . But if through a process of"outcome drift", programs are evaluated simply on whether they makeclients - or someone else - feel happier, then one of the sharpest toolsfor identifying inappropriate care has been blunted. Everything ismedical, and everything is necessary.The rhetoric which begins as health promotion, often ends as needsinflation . This is both ironic and sad. It reflects the twocontradictory avenues of analysis and policy opened by the WHO, inrecognizing that health is much more than simply "the absence of diseaseor infirmity".A broader understanding of the determinants of health leads to healthpolicies which lie outside the sickness care system . This includes, forexample, the studies in Sweden of the relationship between workenvironments and health, looking not just at safety and cleanliness, butat the health effects of job demand and job control, hierarchy and stress ,15and the process of labour market adjustment.But an emphasis only on the broader definition of health, neglecting itsfundamental determinants, can lead to a expansion of the care system toprovide, through a wide variety of therapists, ever more professionalservices which allegedly lead to healthier behaviour and happier lives.This potential for needs inflation is in direct conflict with theimplications of a broader understanding of the determinants of health. Asociety which spends ever more resources on health care will not be ableto afford the improved living and working environments which maycontribute much more to health. Too much health care may be, literally,hazardous to health.A DIAGNOSIS - BUT FOR SWEDEN?The expansion of the system's concept of its own mission, as it focuses onthose in most need - of what? - further detaches it from the mainstreampopulation. Present and would-be providers are trying to meet a wider andwider range of needs, with services which are often of doubtfuleffectiveness, and/or whose claim to the privileged status of "healthcare" is of questionable legitimacy. In an environment of increasinglyconstrained resources, they are concentrating more and more of theirattention on a small minority of the population, with benefits which areoften difficult to determine. And in this environment of intensecompetition for resources, it is suspected that productivity - efficiency- is uneven, possibly falling, and well below what it could be .This is a composite picture, based on facts and selective impressions fromseveral countries, and with a Canadian bias . Does it describe Sweden?Certainly the concentration of care is very pronounced. The Dagmar-50Project (Spri, 1991) has documented variations in practice patterns, andwe understand that physicians in Sweden have resisted managerial intrusioninto professional autonomy and discretion with as much determination asthey have in other countries. We have also been told a number of timesabout low productivity, although this is difficult to prove in the absenceof adequate and consistent measures of output.Moreover , the point has been made very strongly that many users feel thatthe Swedish system is organized by and for providers of care, and providesinsufficient scope for individual patients to choose or influence theirown care. While a certain lack of "user-friendliness" may be a result ofpatients' being assigned to clinics whose budgets do not depend on theiractivity levels, it is likely to be reinforced by the fact that anincreasing proportion of the workload is care of very elderly, very sickpersons who are not in a position to choose very much. As a minority ofthe population use most of the care, and particularly of the care whichproviders themselves think of as most needed, the concerns of the majoritymay be treated as less important.16One has the impression that the Swedish system may be particularlydominated by professionals, rather than managers or politicians, becausehistorically the medical profession and the state have worked together.This pattern long pre-dates the modern "Welfare State", and may have itsroots in an earlier "Warfare State". In other countries of western Europeand especially in North America , physicians have tried to advance theirentrepreneurial interests by controlling the payment process, in directconflict with governments. But in Sweden they have constructed a healthcare system in co-operation with the state.And they have done so sincerely, working to promote (their perception of)the general public interest without the distractions and conflicts ofinterest inherent in fee for service payment. Of course they expect to bewell respected and rewarded, but as highly skilled and conscientiouspublic servants, not as small businessmen. Other health professionalsthen fit themselves smoothly into this same relationship - so long as thesystem continues to expand.But when the money runs out, and the allocative decisions have to be made,a professionally dominated system may be most likely to set prioritieswhich do not match the concerns of the general population. It will thenbe particularly vulnerable to the di~integration of the social consensuswhich supports it . Most of the general public may not be willing to say,at least openly, "Never mind the wretched of the earth, look after me andmine first!" But they may come to feel less committed to a system whichtakes as its primary obligation, caring for someone else. And they maylook for alternatives .Furthermore , a professionally dominated public system will be no moreinterested than a private, entrepreneurial system to inquire too closelyinto the patterns of care being provided by its members. (Interested~system . In both environments there are exceptions, but they face anuphill struggle.) And in the absence of a tradition of conflict, othersare less likely to question the professionals. Perhaps the most startlingcomment we heard came at the Federation of County Councils, when we askedabout concerns over regional variations in procedural utilization rates.The reply was, "Surely that is a matter for the professionals?" Buthealth care is too important to be left to the professionals.LOCAL IMPROVEMENT, BUT GLOBAL CONTROLLike every other country in the developed world, Sweden must find betterways to manage the delivery of health care, to improve not onlyproductivity - service output per unit of resource input - but also theeffectiveness of what is being done. Clinical protocols must bedeveloped, and enforced, to ensure that ineffective care is identified andeliminated. We understand that, despite the comment quoted above, therehas been some work done in Sweden with the development of Medical CarePrograms, but that for various reasons these have not had much impact(Pine et al., 1989). The very recent Dagmar-50 Project may serve as abasis for greater efforts in this area.17But the improvement of productivity and effectiveness - doing the rightthings and doing them right - at the individual practitioner or cliniclevel , will not substitute for overall cost control. The energy andimagination of health care providers can keep the total costs of healthcare growing rapidly despite the best efforts of evaluators and managersto prune out the unproductive and ineffective. Only global mechanisms ofcost control will control global costs; better management will serve toincrease the benefits flowing from the use of those globally constrainedresources.But is global cost control necessary? Yes. In the absence of externalconstraints health care systems will continue to absorb an increasingshare of national income. And it is simply false - quack economics - toclaim that the scale of the health sector does not matter. Such claimsare based on economic theory half-understood or deliberately distorted.·The health care sector is a form of collective consumption which competeswith both consumption and investment in other sectors of the economy, andwhich for structural reasons can have a particular impact on investment.Its expansion, through public or private spending, can thus pose a threatnot only to current living standards but also to improvements inproductivity and economic growth, much like those of an expansion inmilitary spending . .If a society fails to control its health care sector - the U.S. being ourone example at the moment - then the stresses of continuing growth and ofthe ever more frantic responses by payers result in increasingdeterioration of efficiency, effectiveness and equity. But there isreason to believe that a failure to establish limits in a universal publicsystem will eventually undermine the political consensus which supportsit. That may be undesirable in itself, but it can also set up a feed-backloop , a vicious circle, with very serious consequences.As far as we can tell now, the only way to maintain control of health carecosts is through a system with a single payer, as in Canada or the UK, ortightly coordinated ones, who cannot pass costs on to each other, as inGermany. (Almost single . Private insurance and care in the UK andGermany do not undermine the overall control, as long as they operate in aregulatory environment that prevents them from becoming seriouscompetitors to the main system.) Thus a society which abandonsuniversality as "too costly" can expect to see its costs rise even fasteronce it has disarmed itself in the struggle.Decentralized processes of control - market-type mechanisms - have notworked . The endless hymns of praise for user charges which come fromAmerican economists, based on simple-minded theories left over from the1950s, are part of the collection of false statements referred to above asdumped into the international marketplace in ideas. Their currencydepends upon the foreign "buyer" not being sufficiently informed of theAmerican reality . We noted above that the U.S. combines the highest out­of-pocket payments in the DECO with the highest and fastest rising costs.18Moreover the so-called "decisive" RAND Health Insurance Experiment , widelyquoted abroad as showing the power of user charges, excluded persons 62years of age and over, or eligible for Medicare disability benefits. AsTable 1 shows, this group would account for about 60 percent of spendingin Sweden, and all the increase in costs in the last decade. User chargesmay well influence the ambulatory care-seeking behaviour of the ~basicallyhealthy" mainstream population. But this is a shrinking part of thesystem. The high and rapidly growing costs of care are generated afterthe patient has placed himself in the hands of the system, andincreasingly for people whose circumstances are incompatible withrational, informed choice.The RAND results have other serious flaws as a guide to policy. Inparticular the experiment was structured so as to induce no response fromproviders , who would certainly react to a general fall in workloads andincomes by recommending more use. But more generally , proposals forsalvation through private funding are in the main quack remedies, reducingto some combination of increasing total costs while shifting the burden tosomeone else . Those who propose them usually foresee some profit, as withany quack remedy , and indeed the profits are there to be made. But thedamage such ill -thought -out measures can do . economically and ot he rwise ,is one of the strongest arguments in favour of tighter control and moreeffective management. in a global a s well a s a pa r t i a l sense . of thev arious universal ,he a l t h care sys t ems now i n place.That management must start from the general acceptance, by providers aswell as the rest of society, that the total resources available will belimited, for the indefinite future. This in turn means that thedefinition of "needs" is not infinitely expansible, and that some people,in some sad circumstances, will not have everything done for them thatmight improve the quality of their lives. Nor will the rest of us.How does one achieve this acceptance , and maintain this control? Weunderstand that in recent years the federal government in Sweden hasforbidden the county councils from raising their tax rates, in order tolimit the flow of funds into health care . This implies that the politicalbalance at the local level was insufficient to control system growth,which is not surprising. Local county governments have health careproviders as an important part of their electorate and often of theirelected officials , and the provision of health care is their principalfunction. The political constituency for cost control is at the federallevel, and therefore federally-imposed limits will have to be permanent.These limits must ensure an equitable allocation of health resources amongthe counties. An obvious approach would be to follow the U.K. RAWP model,and equalize health spending, on a per capita age-adjusted basis , acrossthe counties, with a rigid cap on this spending . These regional budgetaryallocations, made up of a combination of county taxes and federaltransfers, would pay for the care received by the residents of the region,rather than that provided within the region, so that money would followthe patients. If a patient received services outside the region , thosewould be paid for by a transfer from the budget of the patient's region ,19to that of the provider. sThis principle should , I think, apply to all levels of care, including thetertiary, rather than "top-slicing" funds from the national total tosupport the tertiary care institutions . It is certainly our experience inCanada that, while in theory the tertiary care institutions are expectedto provide care for the whole population, in practice they provide muchmore care for the residents of their home districts. People from out oftown get less. We understand that this is also true in the U.K., despitethe very different funding system, and suspect it is true in Sweden aswell. If the funds follow the patients, then (relative) overservicing oflocal residents and under-servicing of outsiders will automatically haveadverse financial consequences for these institutions.Such regional, capitation-based health care budgets are under active-cons i de r a t i on in several Canadian provinces. In American terms, theywould convert counties into geographically-based HMOs. The centralgovernment (provinces in Canada, federal in Sweden) would determineoverall health spending, allocate budgets to counties on the basis oftheir population characteristics, and then use a combination of matchinggrants and limits on tax rates to ensure that counties stayed within thoselimits. Moreover, for obvious reasons, the limits would have to beimposed on private as well as public spending .An outsider can say little about exactly how such long-term, federallyimposed limits might be structured in the Swedish context. But it seemsobvious that either the federal government must take a much larger role inthe process of allocation of funds, or the selection and responsibilitiesof county governments must be changed so as to give them a much greaterstake in cost control. One's impression is that the latter is a moremassive constitutional task, and that the county-level interest in notonly cost control but health care system management can be much enhancedby a more sophisticated structure of federal funding incentives andconstraints.LIVING UNDER RESTRAINTS - EFFICIENTLY AND EFFECTIVELYOne then moves to the even more interesting questions of priority-settingand managerial control at the regional level. Here the problems asidentified to us include uneven productivity, inflexibiity and lack of"user-friendliness". excessive waiting times, questionable utilizationpatterns, and far too much emphasis on specialist and in-hospital care.S This raises some difficult problems of pr~c~ng, since the truecosts of services are usually not known and often not well defined even inprinciple. It will probably be necessary to develop an agreed upon set of"transfer prices" for services across county borders; the federalgovernment may be able to co-ordinate this process . An interregionalsystem of financial transfers may complement, but definitely cannotreplace , a national planning structure .20All of these suggest a system run by and for professionals, supportedthrough a political system which has seen its role as writing chequesrather than asking unpleasant questions.The responses to these problems have several dimensions . Productivity andresponsiveness to patients can, as we have found in Canada, be achieved ina universal, tax-financed system by reimbursing physicians - not hospitals- for their services according to uniform, periodically negotiated feeschedules. Physicians paid by fee for service want patients and want toprovide (billable) services for them. This has created difficult but notinsurmountable problems for overall cost control.But physicians also have a strong incentive to keep visit lengths down,because fees are fixed. The best way of ending a visit is to write aprescr iption or order a diagnos tic tt~st, and then to recall the patientfor f ur the r exam inat ion . And they bl~come a strong source of pressure foradditional pub l ic or private f unding for equipment and facilities whichwi l l incr ea s e their own productiv i ty at someone else's expense - to servethe patien t be tter , of course . Thus the fee for service system createsincentives for increased testing and drug use, even when the physicianreceives no direct economic benefit, as well as for increased rates ofphysician activity. The visit or procedure rate per physician do notmeasure true productivity if the interventions are unnecessary orineffective. But at least the patient has been seen , and may feel betterfor that.These problems arise in every fee for service system which does not permitthe physician to set his own fee, i.e. every system in which the cost ofphysicians' services is not completely open-ended. The financialadvantages of "revolving door medicine" are, however, much reduced in asystem of global and binding expenditure restraints. One may see evermore rapid increases in servicing rates, as practitioners compete forlarger shares of a fixed total, or one may see collective arrangementsamong physicians to co-ordinate their practice patterns. The former maybe the response in Germany ; there is some very preliminary evidenceconsistent with the latter in parts of Canada. Local circumstances andcultures are probably critical.The hospital-based specialties, though themselves paid fee for service,rely on access to globally-budgeted institutions, so here there have beenfrom time to time some serious problems of access and waiting times.There are, moreover, some reported problems of patient inconvenience andprovider arrogance in specialty care, perhaps because specialists areusually chosen by the referring general practitioner, not the patient.But problems of access have arisen only for those procedures whose userates were rising very rapidly - heart surgery, orthopedics, lens implants- and outrunning the hospital resources allocated for their support. Forthe diagnostic specialists who rarely see patients, however, and whoseclients are actually other physicians, the present fee for service systemseems quite unsatisfactory. Some form of salary or sessional payment isbeing encouraged .21For Sweden, some form of volume-based budgeting at the regional levelmight be helpful, combined with a good deal more information aboutregional differences, shared among county councils and among health careproviders. If large differences in productivity per clinic or per surgeonare really a major problem, and if these differences are well documented,then surely it makes sense to link part of the reimbursement of theorganization or the individual to volume of through-put. The problemswith fee for service in North America and in some of the European systemsdo not suggest a wholesale move in that direction. But some volume-basedcomponent of reimbursement might be helpful.The payment per unit of service need not, and perhaps should not, be paidto individuals. Fee for service, with only physicians eligible to be paidfees, creates serious distortions of practice patterns and powerrelationships. Everyone else becomes economically dependent upon thephysician, reinforcing the professional hierarchy, and the recipient offees then sets the terms on which others are allowed to care for his orher patients. But some very large and successful U.S. group practices,the Mayo Clinic in Minnesota, for example, collect fees for the servicesof the clinic, and then pay salaries and other forms of compensation tothe people that work in them. Everyone knows that all incomes areultimately related to the productivity of the clinic, and the internaldivision may also reflect differences in individual productivity. Butthis internal division is up to the group to decide, and individualscannot pursue their own economic interests - or profesional predilections- independently.Could not a County Council develop contracts to make the budgets ofclinics or hospitals dependent in some part on the volumes of specifiedservices provided - in effect develop a fee schedule - andgive them some flexibility in working out internal payment arrangements?Such a system should define the unit of work by what was done, not by whodid it . An initial office visit for an ambulatory problem - about onequarter of all physician costs, in or out of hospital, in our system - canin principle be provided by a general practitioner, as in Canada, aprimary care specialist (internist, paediatrician, ob/gyn .) as in theU.S., or a nurse practitioner, as in some of the U.S. prepaid grouppractices. Outcomes are about the same, costs are not.We have been told that one of the problems with the Swedish system is thatpatients have an excessive faith in and demand for specialists. So doAmericans, in a totally different system. Canadians and British do not.Is this an ethnic or cultural difference, or is it a form of learnedpreference derived from the information generated by the system itself?My bet is the latter.In Canada, a patient who "self-refers" to a specialist is told to go andsee a GP first - because the specialist will get a bigger fee for aconsultation than for an office visit. But the patient does not knowthat. She is told that the GP is competent to decide if the consult isnecessary. In the U.S ., however, the specialist competes by emphasizing22the superiority of his own credentials. So, in Canada GPs make up halfthe physician workforce, and 40 percent of billings. In the U.S., theyhave almost disappeared. Different economic incentives resulting indifferent messages to patients, not "culture", is the key.What messages, verbal and non-verbal, does the patient receive in aSwedish clinic? One might guess that they have to do with the superiorityof specialist medicine. The generalist just cannot keep up witheverything, you know. A good thing you are in my hands - in case yourproblem is serious ...But if the clinic budget is based, not on the numbers of specialistspresent, but on the number of visits, then "crowding out" the generalisthas no economic advantages. A clinic with many generalists, and evennurse practitioners, can get the same budget, if it provides as manyservices . One might even, as in Canada, set the fees to the clinic sothat the specialist consultation carried no fee above an ordinary visitrate, unless there was a prior GP visit and referral.The point is that, if you want people to shift away from what is perceivedas "exce s s i ve " demand for specialists, then arrange the payment ofproviders so as to encourage them to build popular confidence in thegeneralist, not to undermine it. That's all. A former British ministerof health is quoted as saying that if you want to send a message todoctors, write it on a cheque . To this I would add, if you want to send amessage to patients, have their doctors tell them. Now put the two ideastogether, and you will change practice patterns.Volume-based payment does not require either fees charged to patients , oropen-ended bUdgeting. Both Germany and Canada combine fee for service,"free" care at point of service, and global cost control . Germanyprorates payments so that total outlays are determinate, but the share ofthose payments going to different providers is dependent upon theirworkloads. The Canadian provinces use more complex and somewhat lesseffective controls, but some are adopting features of the German approach.Modifications in payment processes cannot be separated from information ,analysis , and regulation. Is comparative productivity data assembled andmade' widely available? Does it get in the newspapers? If not, why not?Are the measures of productivity themselves adequate? Are patientsatisfaction data collected, with particular attention to those who areunhappy? Is it anyone's job to look, to set standards, and to ensurecompliance with them?What about waiting times? Do County Councils collect information onwaiting times for appointments, or for particular procedures, and relatethem to productivity levels? Are these data routinely made available, inan accessible form, to the news media? There might be some powerfuleffects from wide publication of the fact that the staff at clinic X areseeing only half as many patients as those in clinic Y, yet getting paidjust as much. This would be even more interesting if it should turn outthat waiting times for certain procedures are much longer at clinic X.23As noted above, however, activity in health care is not its ownjustification. High productivity and satisfied patients are quitecompatible with high levels of inappropriate and even harmful care. Thiscan only be corrected by rigorous evaluation of current patterns of care,against standards derived fro~ the scientific literature. Departures fromprotocol must be identified and either justified or remedied. And theremedy will require a judicious combination of advice and education,publicity, economic incentives, and regulatory control.Volume-based funding can also be used to address the issue ofappropriateness of care , of which geographic variations in proceduralrates are one indicator . As protocols are developed for particularproblems, covering for example expected length of hospital stay, orpatterns of diagnostic and therapeutic interventions, or drug prescribing,these protocols can be integrated into the volume-bassed payment system ..Those components of volume of activity which do not correspond to theprotocol - excessive lengths of stay, for example - are not included inthe volume base for payment.One does not have to leap immediately (or ever) to a full ProspectivePayment System on American lines, the famous or infamous DRGs . In fact wein Canada believe there are a number of very good reasons not to do so.DRG-based Prospective Payment looks most attractive in a system which isalready a mess; ours is not. But one could phase in specific adjustmentsto the payment base for those activities for which protocols based in theevaluation literature and clinical experience can be developed - Caesariansection and maternal discharge, for example , or acute myocardial infarct.The underlying idea is simple . People and institutions should be paid fordoing things , not for not doing them. But the difficult and criticalcaveat is that they should be paid for doing things which are known orbelieved to do some good, not for those which do not. Often the best careis thoughtful inactivity.There should also be much more public information about both protocols,and departures from them. It is hardly surprising that the people whowork in the health care system do not want to discuss the details of whatthey are doing, except in terms which show dramatic "medical miracles" ordesperate needs for funds - preferably both. But the people who pay themwith public funds, the counties, do have a responsibility to publicize notonly differences in productivity, but also differences in degree ofconformity to protocols and standards. If they cannot, perhaps it shouldbecome a federal responsibility.At present, most of the "information" available on health care isgenerated by the people in the system itself, for their own purposes. Itis retailed to the public by the news media , who themselves ask few hardquestions . All problems are described as the result of underfunding, andpublicized as such . If a medical misjudgement is made, for example, withfatal consequences , it is much less unpleasant for all concerned if it canbe blamed on underfunding and inadequate facilities. When this"explanation" is given to the press, public confidence in the system and24political support for the difficult policies of cost control are bothundermined. This radical imbalance of information must be redressed.There may be several reasons why the imbalance persists, in Sweden aselsewhere. To some extent the County Councils may have been captured bythe health care system, both in that they accept the professionais'definition of their own mission, and in that the political risks from adirect challenge outweigh the benefits. It is, or at least has been,safer just to raise more money. Now that the federal government hasimposed restraints on taxes, it is easier to blame the central governmentfor all problems, rather than to try to manage them locally. And in theend, such open and public displays of professional short-comings areimpolite. Private exhortation is much to be preferred.And it is. Academic economists in particular, isolated in the shelteredworkshop of the university, and rarely if ever called to account for theirown short-comings, vastly underestimate if they notice at all theimportance of style and tone in human relationships. In a well-orderedsociety", more £.Sll be accomplished by a quiet word than by open warfare.Or at least so we Canadians believe, as we look south.But the quiet words are not working, and the conflict between professionalaspirations and economic capacities is escalating. Professionals areincreasingly appealing directly to the public, to undermine thecontainment of public systems or to encourage the deve10pemnt of privatealternatives. The goose that once laid golden eggs is now seen to belaying silver, or even copper, and there is increasing support forpolicies which may be fatal to the goose. Hence the need for more directmeasures - economic incentives, public information, and regulatorycontrols, to protect the goose and to reconcile all parties to a world inwhich silver is the best we can hope for - at least unless and untilgeneral rates of growth of economic productivity turn up again.An increasingly intense struggle over shares, through public or privatemechanisms, in a low-growth environment, can easily dissipate what littlegains are available or result in overall decline. Again the Americanexample is instructive, as an increasing share of its ever escalatingcosts goes to administration - much of which is conflict management ­rather than to the actual provision of care.CAPITAL DECISIONS AND CURRENT CONSEQUENCESThe chances of avoiding a mutually destructive conflict over shares ,whether played out in the public or in the private funding arena, can beseriously reduced by inappropriate longer run public policies towardscapital formation in the health care sector. And capital, it must beunderstood, includes but is much broader than buildings and equipment. Itincludes the numbers , types , and extent of training of health personnel ­human capital. And it also includes "disembodied" capital - technologyand know-how - which like any other form of capital requires effort andresources to create, and then adds to productive capacity.25All capital investments are undertaken in contemplation of profit - futurebenefits - whether measured in money or in some other terms. Sometimesthese expectations are not met; the investment is unproductive. Apeculiarity of capital in the health care sector, however, is that itseems to be able to determine -i t s own use and profitabilily to a muchgreater extent than is possible in other industries .The traditional statement for physical capital is "Roemer's Law" - a builtbed is a filled bed. A number of empirical studies in differentjurisdictions have confirmed that the use of hospitals is determinedprimarily by the availability of beds, independent of any measure ofpopulation needs. But it is equally true of human capital - rapidincreases in the supply of physicians leads not to falling fees andincomes, or unemployed physicians, but to increased volumes of medicalservices, and costs. Having been trained, at considerable expense andsacrifice, physicians expect to make a living. And they have a variety ofways of ensuring that this happens.Technological advance, too, plays a steering role in service delivery.While in principle it is self-evident that a technology does not dictatethe terms of its own use, but only adds new options , yet the commonexperience in health care is that a developed technology tends to be usedin advance of its evaluation. Those who urge careful evaluation beforedeployment, and a weighing of costs and priorities, are met with pseudo­ethical arguments and exaggerated claims of effectiveness - how can we notdo everything possible to save this life? The assessors of technology aregathering strength, in Sweden as elsewhere (Calltorp and Smedby, 1989) ,but there is as yet little evidence that they have had an impact onoverall patterns of care.Any society trying to contain the rate of escalation of its health carecosts , regardless of the specifics of its health care delivery andfinancing system, will have great and increasing difficulty unless it co­ordinates its capital programs with its operating budget . This means notusing hospital construction as a means of regional development in ruralareas, and then wondering why hospital use and costs are so high. But italso means not training an ever larger proportion of the workforce asphysicians and other health care workers, and constantly upgrading thetraining of those workers, and then wondering why the salary and feeclaims keep rising.Several countries, Canada included, have for many years supportedpersonnel policies which are radically inconsistent with cost control. Ifthe physician to population ratio keeps rising, in a no growthenvironment, then to contain the costs of physicians' services one mustreduce average physician incomes at the same rate . Any such strategy willlead not only to increasing political conflict, but also to more and morepressure by physicians to get access to private funds, to convert policiesof cost control into policies of cost shifting, and continuing escalation,through private payment and private insurance. What else can they do?The same elementary proposition holds for all other classes of personnel .26As they increase in numbers, training, and skills, they becomeincreasingly aggressive in seeking outlets for their services - andpayment. These cannot be reconciled with cost containment.Finally, the powerful encouragement given through grants, patents andsubsidies to technological development in health care has effects similarto the training of more personnel and the building of more facilities.People who develop new drugs, equipment, and procedures, often at greattrouble and expense, want to use or sell them, and to be paid. If ones'industrial and technology policy is designed to encourage suchdevelopments , one is going to have great difficulty subsequently refusing"t o finance their use. In principle, of course, such new developmentscould lead to cost reductions - this is common with technological advancein "normal" industries. But that requires that other people lose theirlivelihoods - that is where the cost savings come from - and this does nothappen in health care.Most countries have for some time understood the connection betweenphysical capital and health care costs, and have tried with varyingsuccess to limit hospital construction and regulate the acquisition of newequipment. They are increasingly coming to understand the linkage betweencosts and personnel policies - or the lack of them - but are finding itmuch more difficult to reduce the scale of training programs and denywidespread ambitions for professional upgrading and career advancement ,ambitions which are expressed not just by individuals but by entireprofessional groups. Issues of labour relations, and of gender equity,become entangled with "simpler" questions of health care needs.The third linkage, between costs and the encouragement of technology, hasnot I believe been addressed explicitly anywhere, although countries whichare "home" to multi-national drug companies find that their efforts tocontrol drug costs must be balanced against their desire to improve theirtrade balances by escalating drug costs in other countries. This may bethe toughest of all, as efforts at cost containment can be portrayed asbeing "against science", and harmful to the progress of human knowledge.Moreover the policy issues, and their costs and benefits, can beinternational. For example, some years ago the American-based drugcompanies wanted Canadian patent laws changed to be more favourable tothem, at the expense of Canadian drug buyers . They lobbied the Americangovernment, not the Canadian, which in due course forced the change, andCanadian drug costs are now rising.The general proposition which emerges from these observations is that thebroad range of policies affecting health care capital acquisition must bemade consistent with a stable system, not an ever-expanding one.Otherwise the political strains will become more intense, and eventuallybecome unbearable. Each form of capital, whether physical, human, ortechnological, corresponds to a stream of income expectations, which areby definition also expenditure expectations . If public, or for thatmatter private, policy keeps expanding those expectations, thepossibilities for long run stability are dim .27COMPETITION IN HEALTH CARE: GREAT ENTHUSIASM, WHAT RESULTS?The suggestions made above place various pressures for improvement on theproviders of care. They do not go as far as the "competitive" proposalscurrently receiving a great deal of attention in Europe, with roots in thework of Professor Enthoven in the United States . Those proposals areintended to give individual patients greater influence over the behaviourof providers, directly or indirectly, through their choices of care.Enthoven's essential ideas are much more sophisticated than the crude andcounter-factual market models of the neo-classical economists. Herecognizes the severe limitations on the ability of the individual to"choose" particular services in this environment, which are implicit inthe regulatory structure of every developed society. But he suggests thatindividuals might be able to choose among providers or insurers, from arange of alternatives reviewed and regulated by the state, and perhaps byemployers as well.This choice need not involve different levels of contribution by theindividual . In the U.S., those choosing to enrol in a managed care systemwith restrictions on choice of provider might pay lower premiums thanthose wanting open-ended coverage of costs from any source. But in asystem of universal coverage, the Enthoven approach would attach a certainquantity of public reimbursement to each individual , and then permit herto assign that payment to the provider or insurer of her choice. Theassignee would undertake to provide or pay for all necessary care for thatperson .The level of reimbursement, or capitation rate, associated with eachindividual would be related to her risk status - or else no provider orinsurer would accept high risk persons . But the individual's contributionto the system , in taxes or premiums, would bear no relationship to hercapitation rate.Population-based regional budgets, centrally determined, are similar inthat the budget for the region is the sum of the capitation amounts foreach of the residents of the region . But the individual and her budgetaryallocation are assigned to one regional agency - in Sweden presumably theCounty Council. This budget-holding agency receives the pre-determinedamount of reimbursement, and determines how to provide or pay for whatevercare is required.But the next step involves establishing several budget-holders, andpermitting people to choose among them . A person might choose aparticular clinic within her region, for reasons of convenience,friendliness, short waiting times, or whatever matters to her. Thatclinic would become the budget-holder for that person, and pay for all hercare - wherever received.Variants of this approach have been very popular with economists inparticular for at least twenty-five years. They seem to embody thepowerful incentives of the market for both efficiency and consumer28responsiveness, while avoiding the obvious problems of extremeinformational asymmetry which make it impossible to maintain any genuinemarket for specific medical services. Their popularity in Europe can betraced to a combination of the increased political friction in the currentpayment systems, the right-ward shift in the Zeitgeist, and theexpositional skills and adaptability to local circumstances of ProfessorEnthoven. (There are several interesting models of care with"competitive" features being tested by some of the Swedish CountyCouncils, but I am not competent to comment on them.)The U.S. experience, however, should moderate our enthusiasm. Consumerchoice or managed care health plans are now very widespread in the U.S .,covering the majority of the working population. They have shown threemajor and related weaknesses: .a. They have had to be selective in their choice of enrollees, a problemvariously described as "cream-skimming" or "favourable riskselection";b . They have had a significant escalation in their administrative costs,both direct and associated; andc. They have not shown any impact at all on overall, system-wide costs.a. Management effort is scarce and costly. Competitive organizationsmust decide whether to put most of their effort into managing the healthcare system itself, or into selecting a healthy population of enrolleesand limiting their own liability for the care this population eventuallyneed. The sad reality in the United States is that it has been moreprofitable to compete by controlling the plans' exposure to risk ­selecting healthy enrollees - than by the very difficult and sometimesdangerous business of trying to manage medical care. European advocatesof competitive systems believe that they can be structured and regulatedto prevent this "cream-skimming". But when the potential profits arehigh , very clever and highly motivated people will be looking for waysaround regulatory restraints. Market games are played by different rules.b . Moreover, management itself is costly for both the payer and theprovider. The notorious situation faced by so many American physicians,of having to argue with a nurse or clerk in Kansas or Minnesota beforegetting authorization to provide some form of care - and perhaps notgetting it - is intensely irritating for physicians and expensive for bothparties. The most rapidly increasing component of health care costs inthe US is not clinical care, but administrative overhead, and this appearsdirectly linked to efforts to promote - and resist - "managed care".c . Finally, it has been known for decades that the capitation-based,prepaid group practices - HIP in New York, Kaiser Health Plans, and thelike - provide effective care at lower cost than the rest of the US healthcare system. All efforts to explain away their advantage as being due tolower quality or healthier patients have failed. But their rates ofescalation of costs are the same as the rest of the system; they do not29show the capping of overall cost growth that one finds in the publicuniversal systems of other countries.Nor have attempts to broaden their experience had any effects on overallhealth costs. Costs are not lower in those parts of the country wheresuch systems are most widespread. Efforts to extend their patterns ofcare to a wider population through Independent Practice Alliances (IPAs)or through commercially sponsored HMOs have largely failed, or mostcharitably have yet to show success. After twenty-five years, the recordis still one of much promise, little performance.In defence advocates claim, with some justice, that they have not beengiven a fair trial. The full theoretical model has never been tested.But, after at least two decades of strong political support for the idea,at least in principle, that claim in itself is a serious criticism. Onecould probably argue , with some justice, that true communism has notreally been tried either. Or Christianity.To date the record on "competitive" forms of organization, relying onchoices by consumers or their employers or other sponsors to improvehealth care delivery, is negative. They have been tried in many differentforms, in many places, and for a long time. They have had no impact onthe overall problems of the U.S . health care system . Perhaps they will inthe future , and perhaps they would in some other system environment .Perhaps.This is not an argument against experimenting with competitive mechanisms.The U.S. experience has very clearly identified some real successes inimproving the quality and lowering the cost of care for particularproblems, people , and places. But it is difficult, on the basis of theU.S . record, to assume that "competitive" mechanisms will make a majorcontribution to resolving the problems of any European health care system.Test out the alternatives by all means, let the rest of us know how youget on, and good luck! But do not bet all your chips.THE CARE OF THE AGED, AND THE LIMITS TO SICKNESS CAREMost of the suggestions and comments above treat the health care systemand the population it serves as aggregates, homogeneous entities. Yet wealso emphasized that the aged, and increasingly the very aged, account fora large and growing share of use and costs, and that the considerableresearch on this topic shows conclusively that this trend owes little tochanges in the age structure of the population.As the Swedish data show clearly, the really powerful effects come fromincreases in per capita use rates by the aged. And this raises importantquestions about the appropriateness of the increasing volume of servicesbeing provided. The magnitude and rate of growth of use by and costs forthe elderly imply that the management of their care is the central problemin health care delivery and finance.30What we are observing is an increasing "medicalization" of the problems ofold age. The increased use by the elderly is not explained by greateraverage age or deteriorating health. Even for the non-institutionalizedelderly who report themselves as healthy, we find in Canada largeincreases in service use, particularly those services accessible only onreferral by a physician - diagnostic procedures, consultations,prescription drugs.Some increases are justified, the procedures made possible by advances inmedical technology and practice which can restore mobility and sight, forexample. But there are certainly grounds for concern that aging itself-h as become a justification for increasingly intensive investigation,regardless of any prospects for improvement. And aggressive medicalinterventions can result in very elderly people who are "repaired" butpermanently institutionalized, having lost the capacity for independentliving . As long as they remain in a hospital they continue to be treatedas patients, not residents, with all the diagnostic and therapeuticinterventions which this implies.The central objective must be to get elderly people £y! of hospitals andback into their homes , and to limit their dependence on medicalinterventions in that setting. For some, this can be done throughexpanded home care. For others, some form of institutional residence isnecessary; but the guiding principle should be that this too is a horne,not a hospital or auxiliary care institution . Aggressive medicalinterventions on a continuing or standing order basis are no moreappropriate there than in a private residence .It is unlikely that we in North America have very many lessons to teach inthis area. Our impression in Canada is that the Nordic countries are wellahead of us in providing more appropriate forms of care. It isunfortunate that in our very short visit we had no chance to examine andconsider this boundary, between sickness care and support of the frail ordemented elderly in institutions or private homes. Certainly this is oneof the major sources of cost pressure in our system, and we understandthat we must get elderly people out of hospitals.Yet the data in Tables 1 and 2 show that high and r~s~ng long termhospitalization of the elderly is a major problem in Sweden as well. Arough calculation suggests that, after adjusting for the different agestructure , hospital use is in fact substantially higher in Sweden than inCanada, and these tables indicate that its share of total health costs isincreasing. We have been told that responsibility for the payment ofhospital costs for elderly people is being shifted from the CountyCouncils to municipal governments, which provide a wider range of housingand social services . This seems like a move in the right direction.But when and how does a person become the responsibility of themunicipality, and what implications does this have for hospital budgets?Elderly people use a significant amount of truly acute hospital care, someappropriate and some not. One cannot transfer all hospital costs.Moreover, the incentives should be symmetrical. Municipalities may have31an incentive, under the new system, to move patients, but will thehospitals co-operate when their budgets are threatened (as they must be)?In the province of British Columbia the financial responsibility forelderly patients in acute care' beds who are assessed as more properlyplaced in long term care - at home or in an institution - is transferredfrom the division of the provincial ministry which funds hospitals, to thedivision administering long term care. This division reimburses thehospital at a much lower rate per bed-day, even though the patientcontinues to occupy an acute care bed.But suppose that all patients who stay beyond some period - say thirtydays - were deemed to be transferred to long term care, unless there werea specific medical assessment indicating that further acute care isnecessary . The onus would be reversed, placing substantial pressure onhospitals to try to move long stay patients into more appropriatefacilities , and to have their medical staff consider carefully theappropriateness of the length of stay .How could one create similar pressures in Sweden? The question must beleft to the people who know that system, because the details of thefunding process are critical in determining what incentives will becreated by any modification in payment. But the general principle is tocreate incentives for the hospital to try to move long-stay patients out,just as the transfer of funding responsibility creates an incentive forthe municipalities to take them.Ultimately the medicalization of aging, and attempts to limit or roll backthe process, blend into the issue of terminal care , and dying withdignity. Our Commission heard a great deal about the need to moderate theintensity of care for the dying, to limit interventions and to increasethe autonomy of the patient, for humanitarian rather than economicreasons. But the patient is often in no position to exercise autonomy atthe time, and there are real problems of interpretation as well asenforcement of any prior instructions.Clear and legally-binding proxy legislation may go part way towardimproving this situation . If a designated person has all the legal rightsof the patient, she can direct or refuse treatment on the patient's behalfas if, legally, she~ the patient. A care-giver who disregards such arefusal would be guilty of a criminal offence.We do not know the present situation in Sweden, in law or actual practice,and we heard nothing about this issue during our visit. But it is anextremely important area, ethically and economically - again the data inTables I and 2 are very suggestive . If increasing quantities of resourcesare devoted to extending lives which hve become a burden to those livingthem, or have ceased to have any meaning at all , then the health caresystem and its patients have a very serious problem indeed, but it is notunderfunding! (And to suggest that it can be remedied by user charges,borders on the insane.)32The appropriate responses are legislative and procedural, not economic.Popular opinion, in North America at least, appears to be in advance ofboth the law and the health care system. In this context less is more,and ordinary people are demanding less, not more.SOCIAL INSURANCE: A SOURCE OF FUNDS, PRIORITIES, OR INCENTIVES?The relationship between health care and social insurance was raisedseveral times. It was suggested that the current provisions for incomesupport in case of illness and injury, particularly short-term, were too.gener ous , and that there could be overall advantages from reducing thesebenefits and transferring funds into the health care system for specificpurposes.For example, a person might need a procedure - e .g. removal of kidneystones or cholecystectomy - requiring a substantial period of hospitalcare and subsequent recovery. Newer, more expensive techniques couldreduce in-hospital time and speed return to work, thus lowering outlaysfor income support. A transfer of funds from social insurance to healthcould lead to lower costs overall, and better outcomes.Or, a person with a disabling condition - arthritis, say - might be on thewaiting list for restorative surgery. More money for health care couldpay for increased surgical capacity, lowered waiting times, earlierrestoration of function, and lowered income support outlays. Total costsgo down, and outcomes are better. Everybody wins.This is a dangerous simplification. Consider another way of looking atthe same facts. The health care system, given a very large quantity ofpublic funds, sets its own priorities for the use of those funds, decidingwho to treat, when and how . Those priorities lead to inappropriate ordelayed treatment of someone who could benefit significantly - they areinconsistent with more general public priorities. It is then necessary toprovide still more money to induce providers to respond to those externalpriorities. (The issue of whether income support benefits are toogenerous is irrelevant. If they are, then they should be reduced, and soshould social insurance contributions. If not, not.)If the external priorities - treat quickly the person who can recoverquickly and get back to work, for example - are compelling, then the realproblem is the priority-setting process within the health care systemitself. If that system can, simply by setting internal priorities whichdiffer from those outside (whether they be those of governments, payers ofincome support claims, or private citizens), extract additional resourcesfrom other pockets to expand its activities, the incentives are obvious."We will do what you want, but only if you give us more money." "Whatabout the money we are already giving you?" "Well, we have alreadycommitted that to other, very important things. There are so manyneeds ... " The fundamental problem is the accountability of the healthcare system itself - management again.•33A more interesting policy, however, would transfer purchasing power to thesocial ~nsurance system within the framework of an overall fixed healthbud&et . The funds flowing to health through the County Councils would bereduced by the same amount as the increase coming from social insurance(How?). The social insurance system would then be in a position toinfluence treatment priorities and patterns directly.Still more interesting are proposals to fund part of physicians' servicesfrom social insurance funds. Purely as a funding system, this makeslittle sense. To a Canadian, most of the European (especially British)arguments for social insurance rather than tax funding of health care areso obviously specious that one suspects they are really support for a moreregressive funding system, weaker control of costs, and increased marketsfor providers of overhead services. But the Swedish argument isdifferent .In essence, the alternative funding source would support an alternativemore efficient process of delivery. The point made to us is that atpresent Swedish clinics operate with a very high ratio of other staff tophysicians. This would be very efficient if these other staff substitutedfor physicians in providing clinical services . There is extensiveevidence that nurse practitioners can provide many of the services ofphysicians at lower cost and to equivalent or better quality standards.(They are rarely permitted to do so.) But instead , we understand thatthese other personnel are additional to physicians, complements ratherthan substitutes. Sweden has a relatively high ratio of physicians percapita.If the non-physician personnel are providing services additional to thoseof physicians, they may be the source of part of the increasing cost andnarrowing client base suggested above as a major economic and politicalproblem. Most people contact the health care system infrequently, withwell-defined problems. They want to see a physician, and do not usuallyneed the services of a diversity of other professionals. The latter willprimarily be engaged in caring for the minority of high-use patients withcomplex, multi-factorial and largely intractable problems , whose needs arepotentially unbounded.If this picture is accurate, then a more efficient approach might be tosupport, with social insurance funds, a leaner form of ambulatory carewith a much lower ratio of other personnel to physicians, more suited tothe needs of the majority of the population. This should be funded withinthe overall limit on total health spending, and its providers might bepaid in part on a volume of service basis, but it definitely should ~involve differential user charges. (A parallel private or semi-privatesystem for the better-off would be a disaster .)This could both improve the efficiency of the public system, and shift itspriorities . Changing the mix of providers might reduce the system'svulnerability to needs inflation and outcome drift, and the resultingcrowding out of the general population. It would also lower the overalllevel of employment in health care, which would be helpful in a strained34labour market. But it must be emphasized that this suggestion is aresponse to a description of Swedish ambulatory care which we are notourselves in a position to confirm.SUMMARYOn the broad aggregates the Swedish health care system looks very good.It provides comprehensive and high-quality care, on an equitable basis, ata cost which, while high, is not out of line with other majorindustrialized countries. And it is supporting a relatively elderly andtherefore relatively expensive population. Moreover Sweden's major healthindicators will stand comparison with anywhere in the world.The world is changing, and every country, Sweden included, is having tolook much harder at the way its health care system is managed. But it isimportant to recognize the very real achievements, and not give upsubstance for shadow. Many so-called "reform" proposals amount to turningback the clock, and adopting ideas and institutions from systems muchinferior to the current Swedish system - more costly, less efficient, lessequitable, and associated~with higher morbidity and mortality .The Swedish problem seems at root similar to that in most other countriesof western Europe, Canada, and the South Pacific. A professionally­dominated system generates both internally and among the publicexpectations of continuous growth, while resisting both constraints onthat growth, and accountability for the effectiveness and efficiency ofwhat is done . Health care is about meeting needs, as defined byprofessionals, and in ways chosen by professionals . And if that leads toan ever larger system, well, what higher priority could there be?This set of attitudes has, in the 1980s, run into the slow-down in generaleconomic growth. The expectations of continuing expansion in health careare now much more intensely in conflict with other social priorities. Thisis a general problem.What may be different in Sweden , is that there appears (to an outsider) tohave been less obvious conflict prior to this between the health caresystem and the rest of society . In other countries, it has been clearsince the beginning of public payment for health care that at least overeconomic issues, political conflict between the providers of health careand the rest of society, usually represented by government, was normal andinevitable. Ways of managing it have been developed. Negotiations overfees for service, in particular, focus this conflict for the generalpublic in an easily understandable manner.But in Sweden, the conflict between the state and the health care systemappears to be more recent, and "to lead to the impression that the systemis unsustainable, is falling apart, and must therefore be transformed orreplaced. Not so - at least not if the political consensus foruniversality can be maintained . If you let the toothpaste out of the tubeby permitting "privatization" and mult iple funding and delivery systems ,35you will have the devil's own job getting it back in again. A smallprivate sector, without public subsidy, and restricted from cream­skimming, may satisfy a small minority which sees private care as anindicator of privileged status - and for whom equality is not a value buta cost - but it will not solve any of the central problems of themainstream system either. A large private system, sustaining itselfthrough public subsidy and shedding the bad risks, is an American-styledisaster.To protect other social priorities, it is essential to place a limit onthe overall cost of health care, and that is a responsibility of centralgovernment. And this must be done in a manner which is both equitable,and creates incentives to improve efficiency and effectiveness of care,rather than simply to blame all problems in the system on "underfunding"resulting from arbitrary limits imposed at the centre.Within those overall , bindin~ constraints, there are a number of measuresthat one might take to improve productivity and effectiveness. We havesuggested volume-based budgeting for delivery organizations to rewardhigher productivity, better evaluation of patterns of care and developmentof clinical protocols linked to funding, modifications to physicianpayment to encourage more reliance on general practitioners as providersand gate-keepers , and more specific financial incentives to both hospitalsand municipalities to get people out of hospitals. Further, counties andthe central government should provide a good deal more information to thepublic about what is going on , and why . If you do not educate yourmasters, they will not support you. The suggestions are none toospecific; few outsiders know enough to be specific. This one does not .The suggestions are embedded in capitation-based regional budgets, withthe funding following the patients, not the providers. But they do notinvolve competition for patients among capitated provider or insurerorganizations. That approach has very great theoretical attractions , butalso serious theoretical concerns. More to the point, it has yet to showsuccess, anywhere, on a systemic basis .There are no magic solutions, but there are ways to improve, and theproblems are primarily political and managerial, not technical or eveneconomic. Proposals based on professional considerations alone are likelyto be quite expensive . Those based on economic theory alone areguaranteed to be wrong.Don't panic.36REFERENCESAndersen, T.F., and G. Mooney, eds . (1990), The Challenges of MedicalPractice Variations, London: MacMillanBarer , M.L., R.G. Evans, and C. Hertzman (1991), "The 20-Year Experiment:Accounting for , Explaining, and Evaluating Health Care Cost Cont a i nment inCanada and the United States", Annual Review of Public Health 12 :481-518Barer, M.L., W.P . Welch and L. Antioch (1991 forthcoming), "Canadian(U.S.Health Care: Reflections on the HIAA's Analysis", Health Affairs FallCalltorp, J., and B. Smedby (1989), "Technology Assessment Activities inSweden" , Internaltional Journal of Technology Assessment in Health 5:263­297Evans, R.G. (1984), Strained Mercy: The Economics of Canadian Health CareToronto: ButterworthsEvans , R.G., M.L. Barer , C. Hertzman, et a1. (1989), "The Long Goodbye:The Great Transformation of the British Columbia Hospital System", HealthServices Research 24(4) :435-459Gerdtham, U.-G ., and B. Jonsson (1991), "Health Care in Sweden, AnInternational Comparison", (unpublished ms.) Centre for Health Economics ,Stockholm School of Economics , Stockholm, 19 JuneWoolhandler, S ., and D.U. Himmelstein (1991) , "The DeterioratingAdministrative Efficiency of the U.S. Health Care System", New EnglandJournal of Medicine 324(18) :1253 -1258Landstings-v.!l.rlden (1991), "Snabbast for.!l.ndring av svensk sjukvard" Nr.12, 29 august:4-7Pine, L., et a1. (1989), "The Swedish Medical Care Programs, an InterimAssessment", Health Policy 10 :155-176Schieber, G.J. and J.-P. Poullier (1989), "Overview of InternationalComparisons of Health Care Expenditures", Health Care Financing Rev iewAnnual Supplement , 1-7Spri (1991), Dagmar-50-, Sprirapports 307-312 , StockholmSveriges officiella statistik (1988) , H.!l.lsan i Sverige , H.!l.lsostatistiskArsbok. 1987/88 , Statistiska centralbyran, StockholmVagero , D., and O. Lundberg (1989) "Health Inequalities in Britain andSweden", The Lancet, July 1


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