UBC Faculty Research and Publications

It's not the money, it's the principle : why user charges for some services and not others? Evans, Robert G.; Barer, Morris Lionel, 1951-; Stoddart, Greg L.; Bhatia, Vandna 1993

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

Item Metadata


52383-Evans_R_et_al_User_charges.pdf [ 1.48MB ]
JSON: 52383-1.0048355.json
JSON-LD: 52383-1.0048355-ld.json
RDF/XML (Pretty): 52383-1.0048355-rdf.xml
RDF/JSON: 52383-1.0048355-rdf.json
Turtle: 52383-1.0048355-turtle.txt
N-Triples: 52383-1.0048355-rdf-ntriples.txt
Original Record: 52383-1.0048355-source.json
Full Text

Full Text

It's Not the Money, It's the Principle:Why User Charges for Some Services and Not Others?Robert G. EvansMorris L. BarerGreg L. StoddartVandna BhatiaHPRU 93:160 DECEMBER,1993.. ,'HEALTH POLICY RESEARCH UNITCENTRE FOR HEALTH SERVICES AND POUCYRESEARCH429·21 94 HEALTH SCIENCES MALLUNIVERSITY OFBRITISH COLUMBIAVANCOUVER, S.C. CANADAV6T1Z3The Centre for Health Services and Policy Research was established by the Board ofGovernors of the University of British Columbia in December 1990. It was officiallyopened in July 1991. The Centre's primary objective is to co-ordinate, facilitate, andundertake multidisciplinary research in the 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 to research, and topromoting wide dissemination and discussion of research results, in these areas. TheCentre aims to contribute to the improvement of population health by being responsive tothe research needs of those responsible for health policy. To this end, it provides aresearch resource for graduate students; develops and facilitates access to health and healthcare databases; sponsors seminars, workshops, conferences and policy consultations; anddistributes Discussion papers, Research Reports and publication reprints resulting from theresearch programs of Centre faculty.The Centre's Health Policy Research Unit Discussion Paper series provides a vehicle forthe circulation of preliminary (pre-publication) work of Centre Faculty and associates. It isintended to promote discussion and to elicit comments and suggestions that might beincorporated within the work prior to publication. While the Centre prints and distributesthese papers for this purpose, the views in the papers are those of the author(s).A complete list of available Health Policy Research Unit Discussion Papers and Reprints,along with an address to which requests for copies should be sent, appears at the back ofeach paper.UBC CENTRE FOR HEALTH SERVICES AND POLICY RESEARCHDISCUSSION PAPER HPRU 93:160It's Not the Money, It's the Principle:Why User Charges for Some Services and Not Others?Robert G. EvansDepartment of Economics, andCentre for Health Services and Policy ResearchUniversIty of British ColumbiaMorris L. BarerCentre for Health Services and Policy Research, andDepartment of Health Care and EpidemiologyUniversity of British ColumbiaGreg L. StoddartCentre for Health Economics and Policy Analysis, andDepartment of Clinical Epidemiology and BiostatisticsMcMaster UniversityVandna BhatiaDurham Region District Health CouncilWhitby, OntarioDecember, 1993This work was funded by the Ontario Premier's Council on Health, Well.Being andSocial Justice. Responsibility for the views expressed herein, and any errors oromissions, rests solely with the authors. R.G. Evans Is supported by a National HealthScientist Award from Health and Welfare Canada, and Is a Fellow of the CanadianInstitute for Advanced Research. G.L. Stoddart and M.L. Barer are a Fellow and anAssociate, respectively, of the Institute. The authors wish to thank the many Individualsboth Inside and outside the health care system who have taken the time to share theirviews on user charges,PrefaceThis is one in a series of articles by the authors about the ongoing debate overuser charges in the Canadian health care system.In this paper we examine why user charges exist for some health care servicesand not for others. The paper analyzes the characteristics of services which appear to,or might, underlie decisious to charge in part or in whole for specific types of services.We propose a framework for evaluating the justification for, and feasibility of, usercharges for particular types of services in particular situatious.Other papers in this series focus on other aspects of the user charge debate,including the popular arguments in support of, the common rationales for, the keysupporters of, and the princlpal gainers and losers from such charges. A briefdescription of each paper follows."The Remarkable Tenacity of User Charges" documents the history of the usercharge debate in Canada. It reviews the participation, positions and rationales ofCanadian interest groups in debates over "patient participation" in health carefinancing."Who Are the Zombie Masters, and What Do They Want?" likens the recurringproposals for user charges to zombies - the so-called 'walking dead' - because althoughthey have been repeatedly rejected by policy-makers and the general public (and thesubstantive claims of their supporters refuted by analyses of the effects of such charges),these proposals refuse to remain buried. This paper examines why that is the case, andwho stands to benefit from the introduction of user charges."User Charges, Snares and Delusions: Another Look at the Literature" reviewsand extends an earlier in-depth analysis of the effects of user charges which three of theauthors published in 1979. The paper assesses whether experience and publishedliterature in the years since then alter any of the (largely negative) conclusions of theearlier study concerning the ability of direct charges to patients to achieve importantpublic policy objectives, including controlling health care costs."Charging Peter to Pay Paul: Accounting for the Financial Effects of UserCharges" outlines a formal and compreheusive analytic framework in which incometransfers - the principal effects of user charges - can be traced between groups in thepopulation (e.g. the healthy and the sick, the rich and the poor), between payers andhealth care providers, and among providers. The framework is used to analyze theincome transfers associated with different types of user charges. ."Why Not User Charges? The Real Issues" examines some of the most frequentlyheard arguments for user charges and looks at what evidence there is for claims andcounter-claims that are often made. Because statements in the "popular" debatesometimes seem inconsistent with each other, or unrelated to or at odds with the facts,2we explore the statements more carefully, asking what they really mean, what valuesthey are based on, and what fundamental issues are at the heart of the user chargecontroversy. This paper in particular is intended for a wide general audience andassumes that most readers will have heard - or perhaps made - the arguments described,but will have little detailed or technical knowledge of the issues involved.In addition, a bibliography entitled "User Charges in Health Care" provides anextensive set of references to articles of relevance to the user charge debate in Canada,drawn from diverse sources including academic research and policy analysis literature,the popular press, government docnments and reports, and the publications and reportsof non-governmental organizations including the professional associations representing avariety of health care providers.User Fees: A Theme with Many VariationsThe expression "user fee", and its various pejorative and euphemistic equivalents~ deterrent charge, or patient participation), are all too often used, by advocates andopponents alike, as if they referred to a well-defined and homogeneous concept or policy.The attachment of a single label tends to create the false and misleading impressionthat there is a corresponding "something" to which it refers. But in fact, there is anextraordinarily wide diversity of charges, policies, and proposals included under thisgeneral term.All "user fees" have the common characteristic that they require the user ofhealth care services to give up some amount of money as a direct consequence of thatuse of services [1]. It is simply erroneous to apply the term to an insurance premium,whether public or private. More generally, it is misleading to apply the term to any taxostensibly to be used for health care funding, whether the tax is labelled a "premium"or a "health care surtax". None of these link an individual's liablIlty to her use ofservices. On the other hand a tax on "health care benefits", administered through theincome tax system, for example, is a user fee for the very simple reason that it doesdirectly link an individual's financial liablIlty to her own use.But apart from this one common feature, user fees are not a single policy, or evena species, but a zoo. Moreover, the diversity continues to grow, as advocates of suchpolicies add proposals to the collection. Few are genuinely new; most are versions orminor variants of old ideas. But the differences do matter, both in principle and inpractice. In principle, the strength of the arguments for and against "user fees" variesconsiderably depending upon who is charged, for what, and how much. And in practice,while no health care system in the developed world, not even that of the Uuited States,relies primarily on user charges for its funding (indeed, except for the Uuited States therest make relatively little use of this method of financing), nevertheless certain types ofcharges appear to be very widely accepted.In Canada, for example, user fees for dentistry and ambulatory prescriptiondrugs, and "claw-backs" in long-term care have never raised much political debate. Thestrong opposition to user fees for hospital and medical care does not extend to thesecategories, which account for most of the roughly one-quarter of total health careexpenditures now coming from private sources. Some analytic work (though not a lot)has been done to try to identify the principles that might underlie this distinction. Theserange from quite obvious in the case of claw-backs, to quite obscure in the case ofprescription drugs. But it is difficult to know how much influence, if any, suchprinciples have had on the evolution of the present pattern.Elsewhere [1] we have identified five dimensions of variation in user fees:how the user pays (the form of the user fee),for what services,how much the user pays,2when the payment is made, andwho collects or keeps the payment.The first three of these can be represented analytically, as components of afunction relating the amount of any user fee to the use of services which generates theliability. The type and amount of services used are represented by the argument(s) ofsuch a function, how the user pays corresponds to the structure of the function itself,and how much the user pays is determined by the values taken by the parameters of thefunction.The last two dimensions could be represented analytically, by dating the variablesand embedding the function determining the fee within an accounting framework likethat developed elsewhere [2]. But this additional complexity would, we believe, add littleto understanding, because in practice there will not be much variation along thesedimensions. User fees can be collected, or at least billed, at the time of receipt ofservices, or can be in some way linked to subsequent tax liabilities [3]. And the ultimatebeneficiaries can be either the providers of services (in the form of higher fees andincomes), or those who would otherwise have carried the burden of payment •. inCanada (and most other developed societies), taxpayers.'The first three dimensions of user fees, how, for what, and how much, can berepresented quite simply (see [2] for more detail). If we designate by q,J the amount q of1 In a system with private insurance, the beneficiarieswould include premium-payers. But very little of health care issupported from true private health insurance [2]. Even in theUnited States, usually identified as the home of privatecoverage, the private system depends on large public subsidies.And the."premium-payers" are predominantly employers, making itvery difficult to identify with precision the gainers fromincreasing the share of health care funded from user fees. wouldit be workers, in the form of higher wages, or shareholders, inlarger dividends, or the purchasers of the firm'S products, aslower prices? Or might it be "future generations," as decreasedcosts for health care benefits reduce pressure on firms' retainedearnings, permitting them to increase investment? The answerdepends on ones' beliefs about structure and behaviour in therelevant markets -- labour, capital, and product -- and thesebeliefs have a large ideological component. Not surprisingly,there is no agreement.If the introduction of user fees leads into more extendedforms of private finance, of course, the beneficiaries willinclude those who will be able to sell increased amounts offinancial and other "overhead" services, both private insuranceand accounting and "management" for providers, thus adding to thetotal cost of health care [2].3some health care service of type j which is used by an individual i, then the user chargewhich individual i has to pay, as a consequence of using that amount q'J' can be referredto as c,J •• the charge C paid by individual i with respect to use of services of type j.The equation:is thus a general form for this relationship. The base (for what) is determined by thetype of service to which j corresponds •• what sorts of services are subject to usercharges? •• while the form (how) and level (how much) are implicit in the generalfunction FO.The simplest form of user fee is a flat charge of $CJ per unit of service of type j,resulting in:Here the choice of C sets the level, and j sets the base. Almost as simple is the flatcoinsurance rate, making the user fee a percentage of total expenditures. If the price orunit cost of services of type j is PJ, then:where c is a parameter between zero and one, and the user fee automatically increasesnot only with the amount of use but also with the price charged by the provider."Deinsurance", or removal of a service from the benefit package, is represented in thisformula by setting c =1.0.The traditional form of user fee imposed by private insurance companies, the"deductible plus coinsurance", requires the user of services to pay their full cost, up tosome fixed amount in a given time period (the "deductible") and then some percentagethereafter. The deductible is typically not defined separately for each type of service,but rather by the total outlays of individual i (or the family associated with him/her).This form of user fee is then represented by:and ~CIJ =c[LiPJ*lJJJ) • D] + D, or=c~(PJ*lJJ.J) + (1 - c)D if LJ(PJ*lJJJ) > D.where LF'J is the total amount of user charges, for all (chargeable) forms of care, paidby person i in the relevant time period. If this total is less than the parameter D set bythe user fee scheme, then the individual simply pays full cost for all such services used.4Once the total exceeds D, only a proportion c of all additional costs must be paid by theuser.Variants on these forms which exempt certain people, such as the aged or thosewith incomes below a certain level, are equivalent to setting Cu to zero for some valuesof i, while letting it be defined by one of the expressions above for the others. H theuser charge scheme places an upper limit L on the amount which an individual mustpay in charges for care received during a given time period, then (assuming theschemeis of the traditional "deductible plus coinsurance" type) we get:CIJ = PJ*lltJ if ~(PJ*lltJ) S D;and kJCIJ =ckiPJ*llt) + (1 • c)Dif kJ(PJ*lltJ) > D but < [L • (1 • c)D)/c;Such a formnlation implicitly assumes that the share of total outlays borne by theuser can be controlled by the reimburser, whether government or private insurer. But ifproviders are free to set their fees at whatever level they wish, then reimbursers mayseek to protect themselves by placing an upper limit on either the per uuit level ofreimbursement, or the total amount reimbursed per time period. The user's liabilitythen becomes open-ended, as the "insurer" in effect protects itself by transferring risk tothe user.Proposals for the integration of user fees into the tax system, or for theintroduction of "tax credits", are minor variants on the above. H the outlays for healthcare services on behalf of each individual are added to that person's taxable income,then the user charge with respect to services of type j will be:where ~ is the marginal tax rate of that individual, lying (like c in the generalcoinsurance case) between zero and one.' It differs from coinsurance only in that theuser fee rate will be higher for higher income individuals. But this by no means impliesthat the introduction of such a user fee would result in higher income persons making agreater contribution to the financing of health care than they do in a fully tax-financedsystem. Quite the contrary. Such a policy will lead to a reduction in the overall rate of2 If an individual's use of care is large enough, it couldpush him/her from one tax bracket to another, and so add one ormore linear terms to the formula, but there would be no change inprinciple.5taxation needed to finance any given amount of health care, and higher income peoplewill gain more by this reduction in tax liability than they will lose through the taxationof health care expenditures (unless they are very high users of care) lZVThe effect of a "tax credit" on the formula depends upon whether the amount ofan individual's credit is linked in any way to her health expenditures. If each taxpayerreceives the same credit of $T, then the actual user charge, the amount of money anindividual must give up as a result of using health care services, is in no way affected bythis Oat "demogrant" (or "taxogrant"). If on the other hand the tax credit is the lesserof $T or actual tax liability resulting from costs incurred on the user's behalf, then itwould be equivalent to a "reverse deductible" of $T.4c, = zero; if LJ(PJ*lLJ) ~ TIt.,and LF'J = t.lLJ(P/lL)]· T; if LiPJ*lL) > TIt.,As an example, consider a taxpayer whose income places her in a 30% marginaltax bracket. If health expenditures on her behalf are added to her taxable income, hermarginal tax rate is the equivalent of a Oat coinsurance rate of 30%. But if she iseligible for a tax credit of up to $300, she will pay nothing with respect to her first$1000 of health expenditures, and 30% ouly of those above $1000. Under the"deductible plus coinsurance" schemes of private insurers, on the other hand, the userpays all of the costs up to the deductible amount, and a percentage thereafter.If her expenses are large enough to push her into a higher tax bracket, however,her coinsurance rate on further outlays will be increased to the higher marginal tax rate.Thus under a tax-linked scheme the share of costs carried by the user increases as herliabilities increase. Risk is transferred back from the reimburser •• in this case thegovernment •. to the nser. The case of an upper limit on liability has already been3 As far back as 1976. the Ontario Economic Councildescribed a number of different ways in which user charges couldbe related to taxable income, and provided examples of theireffects on individuals' liability [3]. This class of schemes wascritiqued in a subsequent OEC publication [4].4. As discussed in more detail elsewhere [2], a "tax credit"for health expenditures which was based on some measure of theexpected health care outlays of each individual or family, takinginto account personal characteristics which are correlated withneed, could have much more complex distributional consequences.But it would also be much more difficult and complex toadminister, if it were possible at all. It would be in effect"reverse underwriting", precisely what private insurers have todo. and that process is very expensive even when most of the highrisk groups in society are excluded from private coverage.6described.The principal point of this algebraic tour is to show that most, if not all,proposals for user fees are in fact minor variants upon one or two basic functionalforms. Apparently "new" schemes can appear very different from the old models inform and impact when presented in language, particularly when presented by advocatesof user fees looking for a way around entrenched opposition. But algebraicrepresentations cut through the verbiage and show that we are actually dealing with"the same old things", or something very like them, with new labels attached.What, then, of our initial claim that there are many different types of user feesobservable in the Canadian and international experience? There are, but the majordimension of variation is in the base to which they are attached. While there arerelatively few choices available as to how to structure the linkage between useof/expenditure on services, and the liability assigned to users, the number of differenttypes of care to which one can attach (and different systems have attached) user charges,is much more diverse. This, in fact, is where the really interesting arguments arise.Rather than attempting to catalogue all types of services to which charges couldbe applied, and to analyze each separately, in what follows we develop a framework oralgorithm against which to appraise present and proposed user charges. We illustratethe use of the framework by applying it to a number of recent user charge initiatives.7An Algorithm for the Classification and Evaluation of User Charge ProposalsAt first sight, there appears to be little if any pattern or logic to the array of usercharges in the Canadian, or any other, health care system. In Canada, the visit to thedoctor is free but the drngs prescribed at that visit may not be .. depending upon ones'age, income, and province of residence. Hospital care is free, but not the ambulancewhich brings one there. Drngs used in hospital are free, but the patient who is stillusing the same drugs after discharge must begin to pay for them. The same is true forbottled oxygen. A tooth extracted by an oral surgeon is a private transaction, unless itis done in hospital, in which case it is free. And so on [5]. It is hard to resist theconclusion that one is observing the more or less random outcome of a number ofcompeting political and administrative pressures.Yet on closer examination it does appear that there is a certain logic to (at leastsome 00 what we observe, and that many » though by no meaus all » of the presentarray of charges can be fitted into a logical framework. That framework categorizesdifferent health care services according to the characteristics which seem, in a numberof cases at least, to serve as a basis for either "free" provision or charges. It givesspecific content to the phrase "medically necessary", and permits us to understand therationale for some of the present .. or proposed .. charges, while identifying others thathave no obvious logic beyond the political and financial pressures of the moment," It is5 There is relatively little overlap between our attempt inwhat follows to identify a rationale for some of the user chargesnow present in Canadian health care (and some possibleextensions), and the "cornmon arguments" discussed in Stoddart etal. [1]. Here we start from the principle that all "medicallynecessary" services should be free, and explore various ways inwhich a service might be "not medical" or "not necessary". Noaccount, therefore, is taken in our algorithm of the patient'sincome, although "those who can afford to pay" would in fact bethe net beneficiaries of a general policy of user charges forhospital and medical care [2].We share some cornmon ground with those who argue for chargesto limit "abuse". But our approach would target charges (andindeed complete deinsurance) to specific and identifiable abuses,rather than basing them on a blanket and unsubstantiated (andimplausible) claim of general abuse.We are not convinced that any manipulation of charges andcoverage is likely to be very effective in controlling overallcosts -- in contrast to a mix of policies to limit budgets, andchange the numbers and mix of, and incentives faced by, those whoactually provide care. But we believe that a serious effort toidentify, and remove from coverage, services which provide eitherno benefit or very little benefit relative to their cost, wouldbe both more effective and more in keeping with wider socialgoals than the rather mindless approach of uniform user fees.laid out as an algorithm In Figure 1.89Rgure 1: A Decision Framework for a UserChargeSTARTDo NotCover(2 )Do NotCover(1)NNIs ItHealthCare?Is ItEffecflve fora well·defined )-__.!:!-__~ClinicalCondition?Not SureYNsomenmesGoToBGoToAFrom C Is ItNecessary ?NDo NotCover(4 )CoverFully(6 )NAre therealternaflve, equall >-__!.l-__~effecflve, ways ofdeliveringIt?Y.. Charge for,or do not cover,more ccsnvoptions .(5 )DiscretlonaryNACanHealth CoreI-------l~ and Non-Health- '>---!.':!.---+iL -l Core beseparated?Do Not CoverNon-Health-CoreComponent( 1)Do Not CoverSUbject toFutureEvaluatlonBAre we'Not sure'or IsIt Ettectlve'Sometlmes'?Not sureSome rnesCQnWeTellFor Whom ?PosslblyCover,SUbJect toFutureEvaluatlonDo NotCover( 3 )Not EttectlveEttectlveGotoC11a) Is It Really Health Care?This is the first question that should be asked. Services provided by a recognizedhealth care practitioner, or in a health care institution, are not ipso facto health careservices in the sense of services intended to improve the health of the recipient. Anobvious example is "preferred accommodation" -- private or semi-private rooms -- inhospitals. Like a television set, these are "extra" services which have always carried anextra charge to the patient, and no one seems ever to have questioned the principle. Itis understood that if the responsible physician determines that the patient's conditionmakes such accommodation "medically necessary" <!:.& isolation for certain infectiousillnesses), then it will not be charged for.On the same principle several provincial governments have been moving to"deinsure" certain physicians' services ., the performance of examinations to attesthealth or sickness, for example [6,7]. Such services are not intended to improve health,but to provide administrative information. An insurance company or a prospectiveemployer may want information on an individual's health, before issuing a policy oroffering a job, or an employer may want assurance that an individual is ill before payingfor work loss due to illness.It is understandable that the parties to such a transaction should prefer that thephysician be paid by Medicare. But it is not obvious that the service is, in fact,"medical", let alone "medically necessary", even though the examination requires theapplication of medical skills and knowledge. The examination or documentation ismotivated not by medical need of the patient but by requirements external to the healthcare system imposed upon one of the transactors,"6 The principle in this case is clear, but the practice isnot. Strictly speaking the government of Ontario (or of otherprovinces) cannot "deinsure" or delist insurance examinations andsickness notes -- because they have never been insured benefits.But physicians providing these services have billed OHIP for"office visits" or "partial exams" because they "have not wantedto present a bill to their patients, and have tried to recoup atleast the cost of the office visit through the OHIP plan" [8].Until recently, OHIP has chosen (or has had) to turn a blind eyeto such practices.Indeed it is difficult to see how the Plan could avoidpaying for these services without the cooperation and support ofphysicians. Thus the "new" policy consists of an agreementbetween the OMA and the Ministry of Health to try to enforcerules that have always been in place. But even now one mightexpect that some such billings will continue as physicianschoose, for whatever reason, to protect their patients fromdirect costs.Nor is the principle itself as clear as might appear - ­suppose the practitioner, in conducting an examination, discovers12Ultrasound pictures of the fetus in utero provide yet another example. Thediagnostic value of many such investigations is controversial; we will return to this pointbelow. But the entrepreneur who quite explicitly offers to provide "baby's firstpictures" as a memento for the family, without any pretence of providing diagnosticinformation, is offering a service which while medical in form, is not medical inmotivation, intent or outcome, and has no claim to public reimbursement. (Nor, webelieve, has anyone suggested that it should have.)The distinction becomes less clear in the case of procedures such as electivecosmetic surgery. Again, Medicare has always excluded many such services in principle,although the practice has been less clear. If people have to pay to have tattoos applied,presumably it is reasonable for them to pay to have them removed. On the other hand,"elective" cosmetic surgery for a burn victim falls well within almost anyone's defiuitionof a health care service.Somewhere in between, is the person who is simply dissatisfied with nature. Ifmost of the members of a community believe that facelifts or "tummy tucks" do notaddress a health problem, or fail to recognize micromastia or microphallus as diseases,then presumably the associated services should not be covered by public reimbursementprograms, even though they require the services of a surgeon rather than a"cosmetician"•Our purpose is not to generate a comprehensive list, but merely to give examplesof "health care" services which have in the past or might in the future carry a 100%"user fee", in the sense of being outside the public reimbursement plans, because theyare not in fact health care services. In some cases the distinction is clear •• preferredaccommodation or insurance examinations. In others •• cosmetic surgery •• there is anunavoidable element of judgement or need for application of community standards.'and treats an illness? A service which was intended only togenerate administrative information, might nevertheless representmedically necessary care.'. It does not seem to us that the individual can set thestandard -- one person may suffer acute distress from some aspectof body shape that most others regard as well within the boundsof normality. But an individual might equally well be verydistressed by the condition of his car, or clothing -- this doesnot provide an argument for public subsidy. So if the rest of uswould be quite satisfied, or at least not overly distressed, ifwe shared your appearance, why should we have to pay for you tohave it "remedied"? This too, however, is a political judgement,on which we as analysts deserve one vote each.But might the distress itself be a form of "illness"? Theboundaries between "normal" and "abnormal" psychological statesare very unclear, and one might very well find someone willing to13But there are also forms of service which combine health and non-health care,and which in practice appear to be much more important, quantitatively, than theabove. ("Deinsurance" of minor surgical procedures may generate much public debateover principles, but they are a trivial part of health care activity or costs.) Mostobvionsly, the institutional care of the elderly combines "room and board" and generalmaintenance care along with nursing services. Again, the provincial funding systemshave always recognized the mixture of the two by levying a "user charge" set to recoupthe basic level of public pension received by residents, less a "comfort allowance" [9].But this payment does not generally cover the full costs of such institutional care.If it were possible to isolate with some certainty the costs of nursing and other"health care" services received by persons in long-term care, without placing an undueload on the process of accounting and cost allocation, it might well be argued thatresidents should be charged an amount equal to the costs of the "non-health care"component. At present the health care funding system may be bearing some of thecosts of services which are not, in fact, health care.Such services may go beyond "room and board" to include physical assistanceand support for the frail elderly, and custodial care for those with psycho-geriatricproblems. They may be very necessary, and their provision may be considered in wholeor in part a social responsibility, even though they are not health care services.Accordingly, explicit subsidies would presumably have to be available to pay for suchcare for those who cannot afford it. But for those who do have the resources, it is againnot clear why the public at large should subsidize the non-medical living costs of peoplewho can afford to pay for themselves.Again, our point is not to recommend such a policy -- one can also envision somepotentially significant problems in practice." But the rationale for the per diem chargeslabel surgery to correct an unsatisfactory body shape as"medically necessary" on psychological grounds. Yet this is allQll sequitur. If a person were distressed to the point ofillness by his unsatisfactory possessions, Medicare would not,~, buy him a new car. The distressed individual has apsychological, not a surgical, problem (and it will probably betransferred, not relieved, by surgery) .8 First, such charges generate an obvious incentive forthe institutionalized elderly or their representatives to resisttransfer from "free" acute care to more appropriate long termcare. This is of course an old problem; possible remediesinclude the assessment of long-stay patients in acute care andtheir reclassification as de facto long stay for purposes of bothhospital reimbursement and patient charges. Secondly,charges for long term care which are related to a resident's ownresources will tend to -- and are intended to -- draw down those14we now have in long term care, as well as for proposed extensions, is based on thedistinction between what is, and what is not, "health care"," If that distinction can bemade more explicit, and if the associated cost allocation can be done in a reasonablyreliable manner, then there might be grounds for some extension of charges for the non­health care component.We suspect that the primary effect would be to transfer costs from health tosocial services budgets. There would be some gain for taxpayers, at the expense of (theheirs 00 those people in long term care who have significant incomes or assets. But it isdoubtful (to us) whether there would be any great change in actual patterns of use orcost. Nevertheless, if the collective health care funding system is intended to pay for"medically necessary" care, then presumably such a redistribution of burden would beconsistent with that intent.Moreover it is at least possible that a shift in funding sources, even if ouly fromhospital to social services budgets, might contribute to reversing the "medicalization" oflong term care. A number of observers have pointed out that treating residentialfacilities for the elderly as second-class hospitals, rather than as homes with a carecomponent, leads to dependence as well as extra expense. The B.C, Royal Commissionon Health Care and Costs [101, in choosing its title "Closer to Home", explicitly referredto the need to maintain a more home-like atmosphere, with more personal control andless professional intervention, for people who are residents, not patients, of theinstitution. But we ourselves have no basis for going beyond raising the possibility thatresources. This can foreclose any opportunities for the residentto return to independent living, or threaten the independence ofa spouse.It is not clear how serious such problems may be, inquantitative terms, nor whether effective remedies can bedevised. But they emphasize the potential dangers of simplisticsolutions based on a priori notions, in a complex environment.9. The distinction will not be simple, particularly forthat large proportion of the institutionalized elderly withpsycho-geriatric problems. Is the extra supervision required byan Alzheimer'S patient 'medically necessary", or not? And how!!ill£h care is "necessary", the minimum required to prevent injury,or must there be a "quality of life" component in the "medicallynecessary" package? Again, there is no escaping the making ofjudgements, and the per diems we find in provincial plans may infact be quite pragmatic and reasonable solutions to a verycomplex set of problems.15a change in funding form might help."Where a line can be drawn between "health care", and "non-health care"services, the appropriate policy -- consistent with the fundamental principles ofMedicare .- would seem to be to cover the former and exclude the latter rather thanimposing part charges for either. But there may be services in which the contribution tohealth and that to general well-being are inextricably mixed; if so there may not be anyfirm principle on which to base a decision about nser charges," We suspect, however,that such situations will be relatively unusual, and that more commonly advocates ofuser charges, providers in particular, can draw such distinctions -- how else can they soconfidently refer to patient "abuse" of health care? -- but for various reasons do notwish to do so. We will return to this point below.10 On the other hand, it may simply serve to 'de-integrate'services that are presently integrated. The compulsion tocompartmentalize things into their 'logical' places may, in fact,have quite deleterious effects because of the inability of largesocial agencies to coordinate their efforts.". We should note here, however, that it would be quitemisleading to identify "non-health care" services with servicesthat improve "quality of life". Many, perhaps most, health careservices contribute to quality of life rather than to lifeexpectancy. Relief of pain and suffering, restoration orpreservation of functional capacity, relief of anxiety, are allobviously part of the role of health care. Our rough test wouldbe, if you were authoritatively assured that a particular service(or product) would ~ contribute in any way to your health,would you still want it? If "Yes", then the service must, foryou, have a non-health care component. But for what health careservices would this be true? Contrary to some of the publicrhetoric, anyone who has undergone any significant health careinterventions can say very easily that being healthy, andforegoing the services, would be better -- much better. In theeconomist's jargon, independently of its health effects, mosthealth care lowers utility rather than raising it.16b) Does It Work?Specialists in the evaluation of health care commonly point out that a significantproportion of the care that is provided in any modern health care system is not in facteffective, either at all or in the circumstances in which it is provided, and much more issimply of unknown or incompletely evaluated effect [11,12]Services which are effective in remedying a well-defined clinical condition clearlyshould be reimbursed, and those which are not, should not be. Thus a service may be,in form and intent, health care; but if on the basis of current knowledge it cannot beexpected to do more good than harm (relative either to no treatment at all, or somewell-defined alternative) for those to whom it is offered, then it should not bereimbursed. A good deal of "alternative medicine" -- chelation therapy for heartdisease, for example -- may be excluded (rightly or wrongly) on this princlple,"More commonly, however, the answer to "Does it work?" is either "Sometimes,and for some people" or "Well, possibly, but we're not sure." These two answers havedifferent implications. If the form of care does appear to benefit (more often than not)some people in some circumstances, then the obvious response is not to impose a partcharge equal to some estimate of the proportion of recipients who do not benefit, but tosee if it is possible to identify the persons and circumstances in which the service is(likely to be) effective.This is what protocols are all about. One reimburses the service if it is providedin accordance with an approved protocol -- presumably based on the best availableevidence of effectiveness -- and not otherwise. Thus one might reimbursemammography screening for non-symptomatic women over fifty years of age, forexample, but not for those between forty and fifty. Or one might, and provincial plansdo, place limits on the frequency of reimbursement for periodic health examinations _.again on the basis of judgements about probable benefit. This is obviously a large areafor future research, and any such decisions must be made conditional upon the currentstate of knowledge and subject to revision. But the general principle is clear, and seemsto us hard to argue with. Do not reimburse any service for which there is no reason tobelieve that it will do the patient to whom it is offered more good than harm.In Canada, we have barely begun to address the possibilities in this area,12 Note that ~ are not taking a position on whetherparticular "alternative" therapies are or are not effective,either at all, or as commonly offered. We are only suggestingthat their exclusion from the public reimbursement plans may reston a judgement (By whom? On the basis of what, if any, evidence?)about effectiveness.17possibilities for improving the quality of medical practice as well as limiting costs. Theunderlying logic is, however, quite different from that of proposals for crude, across-the­board partial user fees. Rather, one uses a highly selective, information-based approachto determining what services should and should not be provided, and then uses thecoverage decision -- 100% "user fees" for ineffective services -- to steer patterns of care.If the policy is effective, very few people (if any) should actually be paying a fee, becauseidentifiably ineffective services will no longer be provided. 13Protocols may be developed for particular procedures or services; it is moredifficult to imagine them applying to whole fields of practice. These become problematicwhere the borderline between "alternative" and "maiustream" medicine is contested, asevidenced by the ambiguous treatment of chiropractic or naturopathy. Here coverage isat provincial option; these such services are not included under the Canada Health Actprinciple of "comprehensiveness". Mostprovinces do provide some degree of coverage,but impose limits on the number of reimbursable visits per year, and requirecopayments. Why?There are several possible auswers within our framework. Traditionally, manymedical practitioners have argued that these services "do not work" -- that theyrepresent ineffective and possibly harmfulinterventious based on fallacious, pseudo­scientific "theories". If so the ground of exclusion is obvious -- services which cannot beexpected to contribute to health are not health care services. Partial coverage is simplya concession to political pressure, without any basis in broader principle:'This "absolutist" position, however, appears increasingly difficult to sustain asevidence accumulates suggesting that some of the interventions offered by chiropractors13 The policy may not be effective, however, if thedecision not to provide coverage simply permits private providersto offer the service on any terms they choose. Opportunistic"marketing" by those who profit from providing a service to anill-informed and sometimes desperate clientele could easily swampany "steering" effect. If a health care service is excluded frompublic coverage, it may still require a good deal of otherregUlation; see below.14. An explanation which side-steps our framework entirelyis that provinces impose restrictions and user fees simplybecause they ~ do so, without penalty under the Canada Health~' If not restrained by federal standards, they wouldintroduce user charges for medical and hospital services as well.But this alternative "unprincipled" interpretation fails toaccount for the contentiousness of user fees for hospital andmedical services, and the diversity of provincial approachesprior to the CHA, compared with the comparatively tranquilresponses to such fees for "alternative" services.18in particular do improve the health of some of their recipients. In some cases theyachieve better results than their competitor medical practitioners. Yet the conclusionimmediately drawn by chiropractors, that their services should henceforth bereimbursed by the public plans on the same open-ended basis as those of medicalpractitioners (and if there are user charges, they should apply equally to both), appearsno closer to acceptance either by those who pay for health care, or by the general public.The difficulty seems to be in drawing inferences about entire fields of practice,and their practitioners, on the basis of evidence about particular procedures inparticular settings. The views expressed publically by individual chiropractors, as to thepatterns of services which they regard as appropriate for healthy, non-symptomatic"patients" as well as to the effectiveness of spinal manipulation therapy for e.g.infectious diseases, or coronary disease, leave room for a great deal of concern. Theissue is not so much the efficacy of specific interventions under controlled conditions, butwhat a group of people with diverse therapeutic theories would do in an open-endedenvironment. There are at least grounds to fear that the result might be a substantialincrease in servicing, at public expense, with little or no identifiable health benefit.'![Again, our point is not to pass judgement, ourselves, on the effectiveness orotherwise of the services of such practitioners, but only to observe that there aresignificant restrictions on their reimbursement, and to consider what the basis for thoserestrictions might be.]The question "Who benefits?" can be extended to cover another area in whichuser charges, and proposals for their extension, are alive and well. As noted aboveprovincial ambnlance services typically require users to pay part of the costs," And15 There is, of course, an inconsistency here, in that itis widely believed among the health services research communitythat a significant proportion of the services of medicalpractitioners are also without benefit. There is increasingsupport for, and activity in, the evaluation of services, withthe expectation that (eventually) services which cannot be shownto be effective should not be provided, or at least notreimbursed by public programs. From this perspective theappropriate policy towards chiropractic, and by extension towards"alternative" therapies in general, would be to explore with somecare claims for the effectiveness of specific interventions inwell-defined circumstances. One should reimburse those (and onlythose) which meet a rigorous effectiveness test (subject to thequalification of "necessity" -- see below), rather than openingup yet another field of unevaluated and uncontrollable activity.16. The charge in Ontario is currently $45 for ground or airservices, unless the patient qualifies for some form of socialassistance by reason of age, poverty, or disability, or is being19prior to the Canada Health Act several jurisdictions levied user charges in hospitalemergency wards. The jnstifications offered are that people "abuse" such services,calling ambulances when they could as easily take a taxi, or misusing the facilities of anemergency ward with non-emergent needs (if needs at all) for primary care.The alternative is to draw explicit distinctions on the basis of need, at the time ofservice or afterwards, and in fact emergency ward personnel do this all the time. Theytriage the in-coming patients and care immediately for those in greatest need; the otherswait. One could take this a step further. If the "abusers" are so apparent -- and someof them may be -- the solution is not ouly to assign them a lower priority for care, but tonotify them that they can equally well be cared for in an alternative facility -- aphysician's office or clinic, and perhaps at another time -- but that if they choose to stayat the emergency ward, they will be charged a fee.Similarly the person brought in by ambulance could be charged -- full cost -- ifthose providing the care judge that his condition was not such as to justify the use of anambulance. And indeed that is exactly what happens at present (or is supposed to) inOntario at least. If the attending physician in emergency determines that an ambulancewas unnecessary, the patient is to be charged the full cost (deemed to be $220 for landambulances, and as much as $5000-$6000 for air). Thus the partial user charge appliesonly to those who, in the judgement of a physician, did require the services of anambulance. [The rules do not speak to the question of whether an average individualcould reasonably have been expected to know his own needs -- presumably the physicianis expected to take this into account.]Even if one took the view that some "abuse" continues, perhaps becausephysicians are reluctant to make a determination of unnecessary use, the effect of apolicy of partial user charges would be (is) to continue to subsidize to some degree the"abusers", while in effect taxing the comatose accident or heart attack victim whose useis entirely appropriate, and unavoidable. It is not obvious what social purpose is servedby such a pattern of charges. They will, of course, raise some revenue, but from aparticularly vulnerable group of people."As in the case of care protocols, the principle is that if the user of care could notbe expected to benefit from it, then it should not be covered. Emergency care bytransferred between institutions (or simply refuses to pay) .17 If one finds an accident victim lying on the sidewalk,and removes his wallet, the explanation that one needs the moneyis not usually considered sufficient. It would seem to followthat a hospital, or the Ministry of Health, should also offersome justification beyond their (admitted) need to raise revenue.20ambulance or in hospital benefits the health of some people -- it is often life-saving -- butnot of those who use it unnecessarily."Do not pay for what does not "work", in the sense of contributing to health.Such an approach seems transparently obvious, and indeed at least oneemergency room physician has called for, not a "user charge" but an "abuser charge"[13]. The reluctance to take this step may be rooted in professional discomfort withmaking such explicit judgements of appropriateness in front of the patient, and ofpersonally assigning financial liability.'9 But it may also reflect the fact that muchadvocacy of user fees is not really about "abuse" at all [14,15] and that problems withthe emergency services, real or otherwise, are simply a convenient propaganda device, acover to deflect attention away from the pursuit of other objectives.But supposing that one cannot tell which people, under what circumstances, willbenefit from a service? This brings us back to the more general question of what to dowhen we are not sure whether a service "works" or not. As indicated in Figure 1,18. As always, an element of judgement must enter here.The individual faced with an unfamiliar health crisis maymisjudge its severity and quite properly play it safe by callingan ambulance when a qualified person would know that it was notnecessary. But those who complain of "abuse" are not referringto such good faith mistakes. Rather they refer to individualswho are persistent abusers, calling on emergency services incircumstances where "anyone" would know that they wereunnecessary. If so, then such persons can be identified andcharged for their unnecessary use.Again the issue arises as to whose standards ofunderstanding should apply. It may be difficult or impossible toestablish whether a particular individual "knew" that anambulance call was unnecessary. But if one instead takes a moregeneral, albeit rough, community standard, that the ordinaryindividual would know better, a "commonsense" standard, then itshould be possible to identify, and bill for, "abuse". Thegeneral principle is similar to that in Note 7 above, concerningwhat is and is not "health care"; the criteria must be those ofthe community, not those of each individual, if the distinctionis to have any meaning.19 It is interesting that while Muran's [13] suggested"abuser charges" are quite consistent with the emphasis on"medical necessity" which underlies our algorithm, he seemsunaware that the rules on ambulance use already permit theemergency room physician to require patients who makeinappropriate use of this service to pay the full cost. This maybe indicative of the difficulties in disseminating sufficientinformation (or incentives) so as to make constructive use oftargeted charges.21branch B, one could reasonably respond to this situation in either of two ways. Onecould reimburse the service, while carrying out further evaluations, or one could refuseto reimburse it, while carrying out further evaluations. Unfortunately both of these"logical" approaches turn out to have significant problems in practice.Reimbursing services on the basis of incomplete or uncertain evidence of benefithas been the past pattern in Canada and indeed in most other jurisdictions •• "when indoubt, cover in order to avoid problems later". But once a funding precedent has beenestablished, it will influence servicing patterns or habits, and those whose incomesdepend upon the provision of the service have a strong and concentrated interest inensuring that it continues to be funded. Those who pay for it have a more diffuse andgeneral interest in controlling overall costs, but it has been hard to focus this interest onany specific services. In a rapidly changing world it is always possible to create somedegree of doubt about the results of any evaluation, and the procedural rules seem to bethat in practice established (or indeed proposed new) interventions get the benefit of anydoubt.This would seem to suggest that services, particularly procedures, with uncertainbenefits, not be reimbursed until conclusively proven to "work". But this may simplyleave the service to the private sector. In the absence of tight restraints on marketingbehaviour, a service which is not demonstrably harmful may be quite rapidly expandedas a private service on the basis of misleading representations of probable benefit. Thegeneral problem of "opportunistic" behaviour in a mixed payment environment is dealtwith in more detail below.In general it appears that the problem of incompletely evaluated services, whilevery real, is not well dealt with by simply manipulating coverage. It may be that whatis needed is some form of intermediate, "conditional" coverage, with the terms of theconditionality spelled out and enforced. But whatever the approach, it seems clear thatuser charges to patients have no particular value in dealing with the technical questionof determining whether or not a service is effective. If the clinicians and the scientistscannot tell, how is the patient supposed to know?22c) Is It Necessarv?One might argue that if a service is clearly medical, and is clearly effective (insome cases at least) in dealing with a well-defined clinical condition, the test of "medicalnecessity" has been passed. But we introduce a further criterion, which seems both tobe. logically required, and to capture an aspect of actual practice, by drawing adistinction between "medical effectiveness" and "medical necessity".The logical requirement arises from the existence of a large class of what PeteWelch has labelled "epsilon" services •. those which have a very small, but nonethelesspositive, health benefit." How small must the benefit be, relative to Its cost, before wedecides that the service is simply "not worth it", "unnecessary"? The answer one givesmay vary depending upon whether one expects to receive the benefit or bear the cost.But if the supply of such services is effectively infinite, as over time it may be, given thepace of technical change in health care, then It follows that some cut-off has to beapplied."The aspect of observed practice which suggests a "necessity" distinction is thatthere is In fact debate over the reimbursement of certain health care services. In vitrofertilization, for example, is clearly a health care service " it is provided by a healthcare practitioner, and is a response to a functional problem. Yet its reimbursementstatus varies across jurisdictions, and there is room for debate about how or whether itshould be covered in a public program. Reversal of sterilization falls Into the samecategory. The conceptual borderline can become thin here, between describing a serviceas non-medical, or as medical but non-necessary, although there are clear examples in20 One must be careful, however, not to allow theeconomist's axiom that "more is better" to slip in the back doorafter it has been booted out the front. A priori arguments forhealth benefit are no more reliable for small than for largegains. In particular, and perhaps unexpectedly, it is not eventrue that "more diagnostic 'information' is better"; for exampleKatz et al. [16] have recently shown that more frequent orderingof mammograms is associated with a significant delay in thediagnosis of breast cancer, apparently because of " ... thepotential disinformation of a nonsuspicious mammogram." (p. 267).21. No society can provide an infinite range of services.Thus attempts to characterize one or other system of health carefinancing as leading to "rationing", when all must do so, ismisleading nonsense.23each case which are inside the borderlme."Much of the public debate seems to have focused on the coverage of particularservices •• whether or not the public should pay for "X". But the "epsilon" issues arelikely to be a good deal more important quantitatively, and like the question of "Does ItWork?", these are patient and context specific. A particular pattern of diagnostictesting, for example, may yield much larger benefits for people in a high risk group,than for the general population. The use of non-ionic contrast media in diagnosticradiology, for example, yields substantial benefits, as well as substantial costs, for anidentifiable group of high-risk individuals. For the rest of ns, the benefits appear to bein the epsilon category and may not be worth paying for.Should one then allow people to pay extra, at their own option, for this "higher"standard of care? (The same issue has arisen with respect to special, higher cost light.weight cast material for broken bones.) The easy answer would appear to be "Yes", butin fact the problem is a bit more complicated, as we shall see below in the discnssion ofopportunistic behaviour.22 The practice, in some provinces, of levying user feesfor (out of hospital) physiotherapy may be based on somejudgement of "necessity". No one seems to have challenged thestatus of physiotherapy as a health service, nor suggested thatits practitioners' interventions were ineffective. On the otherhand, the charges may simply reflect the fact that this serviceis not included in the Canada Health Act principle of"comprehensive", so provinces £§J1 impose charges, withoutpenalty.24d) Is There a Better Way of Doing It?Finally, a service may be health care, necessary, and effective, but may beavailable from different providers or in different settings. If the alternative forms ofprovision are equally effective, then one might charge people who, having access to theless costly form, nevertheless choose to use the more costly. Thus people who self-referto specialists are, in some provinces at least, required to pay the difference between thegeneral practitioner and the specialist fees; the practitioner is only reimbursed by thepublic plan at the generalist rate. Similarly, if an effective and accessible system ofimmuuizations is available through the public health system, one might not wish toreimburse such services when provided by physicians.There is a thin borderline here as well, between the "alternative sources" and the"ineffective care" arguments.The person who self-refers to a specialist with a problem that is welJ within thecapability of a general practitioner, is similar to the person who goes to the emergencyward with a real but non-emergent problem. The extra capabilities of the specialist, orthe emergency ward, are "ineffective" in the sense that they provide no additional healthbenefit. In each case a user charge may serve a steering function, assuming that thealternative, equally effective and less costly services are also equally accessible.But the "alternative source" ground for (highly selective) user charges could beconsiderably expanded in the future, if a wider range of service providers could beintroduced. And here the encouragement of use of general practitioners in Canadaprovides a leading and very successful case. In systems as divergent as theUuited Statesand Sweden, there has been the same move toward an overwhelming predominance ofspecialists in medical care. Canada, like the U.K., has preserved a more equal ratio ofgeneralists to specialists. In both the U.S. and Sweden, it is now generally recognizedthat the over-emphasis on specialization has led both to higher costs and to less effectivecare, but the mix of practitioners is not easy to change,"The Canadian success has been the result of deliberate policy decisions, by bothphysician orgauizations and public regulatory and payment bodies. Left to themselves,23 Indeed, a central plank of the Clinton reform proposalsin the U.S. is an emphasis on primary and preventive care. A keyquestion will be whether the mix of care-givers in the field canbe altered in time to be able to meet the challenge of deliveringon the Clinton blueprint. There is general agreement that thecurrent complement could not do the job, yet one more reason whyhealth care reform in the U.S. will be a slow and painfulprocess.25health care systems seem to drift naturally toward over-spectallzatlon." Thediscouragement of self-referral to specialists has been only a part of the policy"package" to maintain the role of the generalist; educational policies have probablybeen even more important. But the general point is that "steering" user charges mayplaya helpful supportive role in managing the provider or institutional "mix".One might wish to take this idea farther, for example by revisiting the nursepractitioner concept which has been dormant for twenty years and providing more ofprimary care in well-baby or geriatric clinics with a higher ratio of auxiliary staff.Differential charges could form part of such a strategy. The B.C. Royal Commission[10] addressed this same point with its recommendations for an increased nse of perdiem charges in acute care hospitals for people who are receiving de facto long termcare,"24 There are many reasons for this, not the least of whichare the tendency for provincial fee schedules to favourprocedural specialties and sub-specialties, and the failure ofacademic medical centres to promote the prestige and importanceof primary care [17].25 Such a strategy, however, must recognize explicitlythat the employment of less costly, less "human capitalintensive" inputs -- nurse practitioners instead of physicians,for example -- does not necessarily result in more efficientcare. Duplication of functions, or engagement in "feel good"activities which make both staff and patients happy but have nodetectable impact on health, can easily lead to significantlyhigher costs of care. Alternative providers or institutions mustbe managed and budgeted at least as tightly as the presentsystem, if they are not to become just another source of system"add-ons". Again, the B.C. Royal Commission emphasized thispoint in its recommendations for expansion of home care services.26e) But If Not, 0 King".For services which make it all the way through our algorithm, however, the casefor user charges appears rather weak. If a service is incontestably medical in intent,and is effective, and is regarded by the community as necessary, and can be provided inno other, less costly way, why would one want to impose a user charge? At that pointone of the standard arguments against user charges, that they tax the sick, seems whollyjustified. Such charges may be highly effective as a revenue raising device, but whywould one regard the experience of lIIness, and the use of effective care, as an indicatorof taxable capacity or ablIlty to pay? No answer ever seems to have been offered.In Ontario, oxygen therapy for out-patients provides an example of just such anJlloglcal pattern of coverage and charges. It is fully covered for seniors, at present,under the Ontario Drug Benefit Plan, as well as for people on social assistance, and (upto six months) for those in palliative care. Those born after July 1, 1963 are coveredunder the Assistive Devices Program, but must payout of pocket 25% (currently about$1100 per year) of the approved rates negotiated between vendors and the Ministry ofHealth. (Those between the ages of 30 and 65 are apparently left out.)Oxygen therapy is clearly a health care service. Both programs reqnire medicalcertification of chronic hypoxia, and no one appears to have raised a question as to theeffectiveness of the therapy, so the question of medical necessity presumably does notarise. So what is the principle underlying the part charges in the ADP, and theexclusion of some age groups entirely? Is there some evidence that people are morecareless with their use of oxygen when it is free? Or is there some collective judgementthat people with hypoxia should make a larger contribution to the financing of healthcare?On a much larger scale, the basis in principle for the survival of user charges inpharmacy seems to be equally obscure. The political, legislative, and administrativehistory is clear enough, but what possible ground is there in logic for charging for drngswhich are available only on a physician's order? There may be widespread misuse ofprescription drugs; certainly there is a great deal of expert opinion, and not a littleevidence, to that effect. But the whole point of a prescription regime is to take theutilization decision out of the patient's hands, and place it with a competent expert.As Hurley and Johnson [18] point out, user charges are more likely to reduce theoverall use of prescription drugs than of hospital or medical services, because thoseproviders whose incomes would be reduced as a result cannot as directly influence theutilization decision. And indeed the evidence does indicate that user charges reduce use,but they do not selectively reduce inappropriate use [18,19]. Students of the use andabuse of prescription drugs have consistently concluded that inappropriate prescribingbehaviour is the source of both excessive use and the resulting cost escalation, and thatthe roots of this problem lie far deeper than can be reached with user charges directed27at patients."Though co-payments are a tempting quick fix, they must be viewed as the band­aid solution they are. They can help slow the bleeding of funds from provincialtreasuries, but at the price of diverting attention from the cause of the wound."[18, p, 487]It is hard to resist the conclusion that user fees are a way whereby payers canshift the escalating costs of drugs onto patients while avoiding the politically difficultactions •• addressing prescriber behaviour and the marketing practices of drugcompanies •• that would be necessary to limit their inappropriate use. The federalgovernment has taken this process of evasion to a higher level still, by passing legislationdeliberately designed to raise the prices of prescription drugs, and then reducing itsfinancial contributions to the provinces and pointing with concern to the high costs of"social programs".User charges in pharmacy thus provide a good example of the way in which suchcharges can lead to increased costs of health care, by shifting the financial pressureaway from those in a position to take effective action, and onto those who cannot.Effective action is politically difficult, as it necessarily threatens both incomes andprofessional prerogatives. Cost shifting, particularly once the user pay principle hasbeen accepted, as in pharmacy, is much safer.A partial exception is provided by a user charge policy introduced some years agoin B.C. Seniors, whose prescriptions were previously fully reimbursed, are now requiredto pay a proportion of the dispensing charge, up to an annual ceiling. The intent was toreduce the nnmber and amount of dispensing fees, by encouraging patients to demandlarger quantities of drugs in each prescription and to seek out pharmacies with lowerdispensing charges. This, we believe, falls under the category of a "steering" charge,intended to encourage the choice of an equally effective but less costly style of care,"Policies to encourage the use of lower priced generic drugs have sometimes reliedon user charges for similar "steering" effects. Going beyond efforts to inform26 Pharmacists, whose incomes were the deliberate target ofthis policy, argued that for a variety of reasons the large­prescription "alternative" was not equally effective. We takeno position here on this point; those introducing the policyclearly believed that they were encouraging an equally effective,less costly form of care -- making less use of whatever servicesare provided during the dispensing process. Apparently thepolicy did reduce payments for dispensing, but it is not clearthat total costs fell because more was paid out for ingredients.The user fee itself, however, transferred part of these costsfrom public to private budgets -- taxing the sick again.28prescribers about the relative costs of different versions of a drug, public programs havepermitted pharmacists to dispense generic equivalents to prescribed brand-name drugs,and in some cases limited reimbursement to the price of the lowest-priced equivalent. Ifthe drug actually dispensed costs more, the patient pays the difference.But for such steering charges to guide behaviour, the patient mnst both havealternatives, and know that they exist. In general patients are not informed as to thechoices available. And even if they were, the intent and effect of present federal drugpatent legislation is to restrict the availability of competitive equivalents and enhance themonopoly power of suppliers of patented drugs. Where there is no choice, there is nosteering effect; nser charges can only shift costs, not lower them.Dentistry is also a bit of a puzzle in principle. Do we regard dental care as not ahealth care service? Is it not effective? Are there alternative sources of care? Or dowe as a society simply judge it "not necessary"? Again the realpolitik aspects are clear.Provincial governments did not want, in the early 1970s, to take on an additional largespending commitment, and dentists feared, quite rightly, that a public plan wonld limittheir freedom in pricing and perhaps also in patterns of practice.The existing system preserves professional autonomy and incomes, and limits theovert financial exposure of governments, although private dental insurance, which isquite widespread, is subsidized by the federal government through the income taxsystem. As a result, dentistry in Canada is a good deal more expensive and less efficientthan it could be, but the resulting extra costs are widely dispersed among taxpayers,employees, and purchaser of products, very few of whom know that they are bearingsuch costs. The beneficiaries, on the other hand, are highly concentrated and self-aware,making change unlikely. The outcome reflects interest, if not principle.In terms of our algorithm, virtually all of dentistry falls comfortably within therealm of health services. Questions arise at the next step, as much of what is offered as"preventive" dentistry may be of questionable effectiveness. The "six monthly check­up" apparently never had any evidentiary foundation, and the benefits of topicalfluoride treatment for populations whose drinking water is fluoridated, are alsodoubtful.At the third step, the borderline between "needed" and elective cosmetic care isas always thin. But there seems little doubt that a substantial share of orthodontictreatment is purely cosmetic, and indeed some dental organizations have begun tomarket it on this basis. On the other hand, the provision of comprehensive dentaltreatment for children (which in practice requires a school-based, public program, notsubsidized private care) yields not only immediate payoffs in dental health, butimportant preventive benefits over adult life.Finally, there is extensive evidence, and has been for many years, that there are29much less costly alternative sources of care than the general (or the specialist) dentist,which dentists collectively have used their self-regulatory privileges to suppress.In general, dentistry provides an excellent example of the irrelevance of user fees,their powerlessness to induce structural change and improve efficiency and effectivenessin a sector of health care which remains under the tight regulatory control of theestablished providers. It also illustrates the strength of the phenomenon of "regulatorycapture" - governments which are not themselves at financial risk from providerbehaviour, are responsive to concentrated provider interests, not diverse public ones.Just as in the case of the federal patent drug legislation, providers are able to secure alegislative and regulatory environment in which patients are denied both information,and access to alternatives. The "free market", isn't; patients can only respond to userfees by paying (or doing without) [20,21].There are some remaining anomalies in the relationship between dentistry andgovernment. H dental care is outside the public health care funding system, why is itsubsidized by the federal government through tax expenditures? Dental insurancepremiums paid by an employer are a deductible expense, a labour cost, but (unlikemedical insurance premiums in those provinces which retain them) they are not taxablein the hands of the employee. This form of subsidy is more valuable to those in higherincomes and higher marginal tax brackets; it also provides absolutely no influence forthe donor government over either dental fee levels or practice patterns. It would beinteresting to know how much dental insurance coverage would continue if this subsidywere withdrawn (by making employer-paid premiums a taxable benefit) and what theimpact would be on dentists' fees and incomes.Another form of public subsidy is provided for those dentists able to use freepublic facilities -- hospitals -- in their work. Most dentistry requires a considerableinvestment in equipment and support staff. The dentist who has access to a hospitaloperating room need not bear these costs, yet he can charge the same fees as hiscolleague in external practice, thus pocketing a substantially higher net fee. Moreoverthe oral surgeon, who works at the intersection of medicine and dentistry, can bill thepublic medical insurance program for certain surgical services if they are provided in ahospital setting, as well as having the costs of anaesthesia and other support servicescovered from public budgets. And while the fees from the public plan may benegotiated, the dentist retains the right to bill the patient in any amount he chooses forassociated office services _. in effect open-ended extra-billing.30Patrolling the Borders, and BeyondSuch questions lead ns Into a much wider area, the general problem of"opportunistic behaviour" In an environment of multiple funding sources, public andprivate. What opportunities for profitable manipulation or malfeasance are created byalternative forms of financing? And In a market or quasi-market environment, what arethe possibilities for controlling them?As Wennberg [22] has pointed out, even In an environment hedged around withprofessional norms and restrictions on opportunities for overt profit-seeking behaviourthe fertile imaginations on practitioners and their natural optimism and urge tointervene will generate plausible new manoeuvres (or variants on old ones) faster thanthese can be tested for safety or effectiveness. When these expansionary tendencies arecombined with direct economic incentives, the problems of regulation and controlbecome an order of magnitude more difficult. The current American experience Intrying to combine "market forces" with regulatory protection of the public do notencourage optimism [23,24].Public regulation of health care, whether direct or through delegation toprofessional bodies, has always been justified as providing protection for "vulnerableinterests" .- the interests of those who, for a variety of reasons, would be unable toprotect themselves in arm's-length transactions, or who are simply not represented In atransaction which nevertheless may have significant effects upon them. Indeed, theprotection of vulnerable interests underlies all professional regulation,"Regulation of health care services has tended to emphasize the protection ofpeople whose interests are vulnerable because their capacity is impaired, either bytemporary or permanent physical or mental disability, or more commonly because theydo not have and cannot reasonably obtain the Information necessary to make decisionson their own behalf [26]. Despite the rhetoric of consmnerlsm, most health care isprovided to people who are in fact patients, heavily reliant on the expertise and goodfaith of physicians or other professionals to guide their decisions in their own bestinterests.The emphasis on good faith is critical; the mere possession of expertise Is noguarantee that it will be used In someone else's interest. In an arm's-length transactionbetween parties each pursuing their own interests, knowledge Is power •• or wealth -­and the party who Is Ill-mformed may expect to come out on the short end. Professionalethics backed up by professional regulation is intended to discourage (though in an .imperfect world It cannot wholly prevent) opportunistic behaviour by professionals using27 One of the best expositions of the justifications forand forms of professional regulation is provided by Trebilcock ~al. [25].31their superior information to exploit their patients (or clients, or consumers).In this process there is always a tension between professionalism and economicincentives. There is money to be made (or other benefits to be gained) by exploitingvulnerable interests, otherwise there would be no problem. In general, and most of thetime, professionals do seem to behave ethically, that is, they forego opportunities forpersonal gain which would involve exploiting the informational or other vulnerabilitiesof their patients. But all systems of payment, of whatever form, embody patterns ofeconomic incentives; and professional behaviour is the resultant of a shifting balancestruck between these incentives and prevailing ethical norms. (The balance shiftsbecanse, even if the payment system itself is static, the changing technology of healthcare is constantly changing the incentives embodied in that system.)Changes in the payment structure, in particular changes in the mix of publicreimbursement and user payment, may also shift that balance. Thus the importance ofconsidering the possibilities for opportunistic behaviour implicit in such changes. Ingeneral, the interests which may be vulnerable fall into two categories: (i) patientsthemselves, who may be at greater risk of financial exploitation in a user charge system,and (ii) other citizens, who may find themselves carrying extra costs as a result of the"private" transactions of patients and providers.32Privatize the Profits, Socialize the CostsThe latter form of opportunism, threatening the interests of other citizens, isperhaps the easier to recognize and to deal with. We mentioned above the example ofthe private dentist who gains access to public hospital facilities, in which he performsprocedures for which he bills the patient (or the patient's insurer) privately, and forsome services may in addition bill the public medical insurance plan. The overheadcosts are paid by the taxpayer; the revenues accrue to the dentist.28A similar "cost spill-out" has been identified in the case of in vitro fertilization.There has been some contention as to whether such services are "necessary" in theterms of our algorithm above, or whether they should be left to the private sector andprivate payment. The recent Royal Commission of New Reproductive Technologies [27]has recommended that coverage for IVF be included within the Canadian healthinsurance programs, but only for demonstrably effective services.But could IVF services be treated as a purely private transaction, even if onewanted to? IVF patients have an increased risk of multiple births, and of low birthweight babies. These can be exceptionally expensive to care for; at present such costswould be covered by the public reimbursement plans. The fees for the service arecollected privately, but (some of) the cost is borne by the public. Could one impose atax on the private clinics, or require them to post a bond or purchase private insurance,to cover any downstream health care costs arising out of the "private" medicaltransaction of IVF? H so, those costs would have to be passed on to patients, and onecan confidently predict protests not only by private providers of such services, but alsoby their patients. But if IVF is to be treated as a private transaction, with private, for­profit provision, why should the rest of ns cover part of the costs?28 Physicians also use hospital facilities to care fortheir patients; several specialties could hardly practice at allwithout such free "overhead". In (Canadian) medicine, the publicprovision of facilities for private practice is the norm. Thedifference, however, is that physicians' fees are negotiatedwith, and paid by, the same public agency -- typically theMinistry of Health of the province -- which funds the hospitalservices. That agency is in a position to take into account, inits negotiating position, the fact that it is through anotherbudget providing these overhead services for free. Privatedentists, on the other hand, set their own fees unilaterally, andthose fees include a substantial overhead allowance. A dentistwho enjoys free access to public facilities thus collects alarger net fee. (As noted above, fees for oral surgery servicesprovided in hospital are negotiated and paid by the provincialinsurance plans, but the fees for any associated services not soreimbursed are completely under the practitioner's control.)33Yet another example: it can be (and has been) argued, that seniors with their ownmeans should be allowed, or even required, to purchase long-term care privately. If awell-off elderly person chooses to pay for particularly comfortable and expensivesurroundings, that is surely no one else's concern. But the facts of biology are thatelderly people, like everyone else, become progressively older, and their capacitiesprogressively fail. And incontinent or deranged patients are not welcome in comfortableand expensive facilities, either by the operators (they are expensive to look after) or bythe other residents (they are unpleasant to have around). Thus the "private" contractbetween facility and resident may well specify that residents whose physical capacitiesdeteriorate past a certain point must leave -- the facility is not equlpped to handle"medical" problems.This makes perfectly good commercial sense, and is probably quite acceptable toresidents as well. But it opens the very real possibility of a form of private "spend­down", in which the aging person is welcome in private facilities until her health andher money run out, after which she is "dumped" into public facilities, at publicexpense."Private contracts which have annuity elements, specifying an initial large capitalpayment in return for care until either death or deterioration, would be particularlyopen to this form of opportunism. Like the private insurer, the private facility operatoris very interested in "customers" with ample resources of health and money, but mustshun those without. The financial transition may be speeded up by offering expensiveservices early on in the institutional stage of life.29 The term "spend-down" is drawn from the AmericanMedicaid program, where patients must become impoverished toqualify for public coverage of long-term care. The Americanexperience illustrates how easy it is to go down this road, andhow difficult to get back, precisely because all the privateincentives push in this direction.34Are Markets "Free", When Services Are Not?Perhaps the most difficult problems of protecting vulnerable interests, however,arise from the opportunism which threatens patients themselves. There is a deepinconsistency between long-established practices and habits of thought in the health caresector, and those in the private marketplace. In the marketplace, the fundamentalprotections for "consumer" interests are information and choice, Knowing your ownneeds, knowing what products are available at what prices, and being able to chooseamong alternative and competitive suppliers or products, permits you to identify andavoid opportunistic behaviour •• over-charging, faulty products, false advertising •• bysuppliers. In addition, of course, there are various regulatory restraints and legalremedies to discourage such activities.But in the health care field, both the regulatory framework and the behaviour ofproviders has always worked deliberately to block the operation of market forces, on thegrounds that the patient's interests were better protected by a combination ofprofessional ethics and public regulation. Thls argument should not be dismissed lightly.There is neither experience nor a priori grounds to support a casual assumption that"the market" •• whatever that means •• would necessarily lead to a better functioninghealth care system [26].But if, through "de-Insurance" or user charges, we move services outside thepresent framework, how do we mobilize the alternative forms of protection for consumerinterests through the market? H we do not, then users are open to exploitation,financial or otherwise, by providers. The problem is two-fold. Do consumers havechoices available to them, and do they have enough information to make the right (forthem) choices?As an example, consider a situation in whlch opportunism has not been aproblem, although it could be. "Preferred accommodation" in hospitals, that is, privateaccommodation for whlch there is no medical indication, carries an extra charge whlchis retained by the hospital. Suppose a hospital administration adopted a policy oflimiting access to regular ward beds, such that patients unwilling or unable to pay extrafor "preferred accommodation" had to wait longer for admission. H the hospital couldalso set its own preferred accommodation charges, this could be quite a profitablestrategy.As far as we know, this has not happened in Canada. The limitations on suchbehaviour include physicians, who would object to limitations on their patients' access tobeds, hospital trustees, representing the community interest as well as the hospital, andpatients, who would be able to create a political response. Most hospital managers inCanada would also regard such opportunistic behaviour as improper, even if profitable.35But "the market" does not impose any restraint. Patients go to the hospital atwhich their physician has privileges; in many communities there is only one hospital,and in any case hospital managements have many avenues for collusion, which theyregard, quite properly, as co-operation. When Canadian hospitals are presented withthe opportuuity,~ a foreign patient with insurance -- or even without -- they do pricequite "aggressively", and the patient has little or no recourse.Fear of hospital and medical opportunism also underlies resistance to proposalsto permit Canadian hospitals to expand their resources by offering services to U.S.patients, on a fee for service basis [28]. If this could be done without in any wayreducing access by Canadian patients, it is hard to see a basis for objection. But it isequally hard to see how hospitals and physicians, presented with the opportunity toincrease their incomes by caring for Americans, would not begin to favour the moreprofitable clientele. Assurances to the contrary are worth very little without an effectivemonitoring and enforcement system; such does not now exist and it is not clear how onemight be designed.Coming back to the "easy" cases -- lIIness certificates, or tattoo removal,forexample, if they are "not health care" for payment purposes, are they health care forlicensure purposes? Must the "customer" receive the care only from a specificallylicensed person, or could others compete for the market? Could licensed practitionersnot only set their own fees, but advertise them, and would collusion over fees orrestrictions on advertising be in violation of federal competition legislation?The core of the problem is, how does one encourage, and indeed enforce,competitive behaviour by professional providers in a small set of "markets" whilesimultaneously protecting the pattern of ethical, regulated, and necessarily highlycooperative behaviour in the rest of their work? But if providers of services for self­paying "consumers" are not under the discipline of a competitive market, how can oneensure that fair prices are charged, and an appropriate range of services is madeavailable?The problem may not be insoluble, but if a larger range of services is to be left to"the market", it must be understood that free markets do not come into existence orpersist automatically. They are a human creation, an institution, not a law of nature,and there is always a strong incentive for some of the participants in a market, usuallythe sellers, to seek out ways of manipulating the market so as to limit consumer choiceand raise prices (as Adam Smith pointed out long ago, though many of his more recentadmirers have conveniently forgotten).In both dentistry and pharmacy we find strong and generally quite successfulefforts by professional organizations to limit consumers' access to alternative providers,to restrict the availability of comparative information on service prices, and toencourage price collusion. In both of these fields, public authority has been co-opted by36professionals to serve their private interests by protecting their prices and their markets,at considerable cost to the consumer. Concentrated professional interests have carriedmuch more weight, politically, than the diffuse interests of consumers, becausegovernments themselves are not paying the bills.One possibility for overcoming the natural professional reticence about overtmarket competition, is to encourage entry by new forms of provider organization [21].Thns one might permit for-profit corporations to offer for a fee those services which areexcluded from the public reimbursement plans because they are judged "not healthcare", or "not necessary", or simply not yet proven effective. And indeed suchorganizations are now attempting to move into the "market" for IVF, as well as offeringsome forms of prostate and eye surgery whose effectiveness is as yet open to question.There are also moves afoot to offer for-profit diagnostic services over and above thosepublicly available in the hospital system.Such developments, however, can also represent serious threats to the vulnerable.interest of the uninformed patient -- threats of several different types [21]. The firmstand taken by the Report of the recent Royal Commission on New ReproductiveTechnology [27] against for-profit enterprise in this field, and in support of nationallicensing and stringent regulation of providers of fertility services, is based on an explicitrecognition of these threats and a judgement as to their serionsness. But the sameconcerns underly the long-standing and widespread unease about, if not out-righthostility to, the operation of "normal" commercial enterprise in the health care field.The traditional concern is for "quality of care" in the narrow sense of "doingthings right". It is feared that a concern for profits will lead to cutting corners, relaxingstandards of procedure, in a way that endangers patients." Such behaviour is notunheard of in not-for-profit organizations; hence the need for external regulation andaccountability. But profit-making is believed to add another powerful motivationcompetitive with quality maintenance.Professional organizations have over the years developed a number of institutionsand procedures intended to maintain "qnality" in this narrow sense. More recently,however, students of health care have begun to draw attention to the equal importanceof "doing the right thing" -- of providing services which are effective for the purposesclaimed, and which match the needs of the particular patient. Again, the RoyalCommission [27] identified and raised concerns about the widespread provision, in theprivate sector, of fertility services which are ineffective or frankly experimental, topatients whose likelihood of benefit is minimal at best. They concluded that the onlyfeasible remedy was to eliminate for-profit provision of fertility services, while providing30privatewithoutThe Royal Commission, for example, found cases ofclinics providing artificial insemination servicestesting the donor semen for HIV infection.37coverage under the public insurance plans for those known to meet some minimalstandard of effectiveness.The same "marketing" problem arises with respect to the suggestion that privateproviders be permitted to offer services parallel to the public system. The patient willbe told she "needs" a particular procedure that is either not available through thepublic system, or available only after an unacceptable wait. The provider who has anequity interest in a for-profit facility, has a powerful economic incentive to oversell thevalue of the service, and to minimize its risks." This is not to say that there are noethical restraints remaining to restrict such behaviour, but only that the balance ofinterests is shifted. The greater the economic interest, the greater the marketing effort.This is not a hypothetical possibility. The special problems with newreproductive technologies are new, but the extraordinary marketing efforts of the for­profit drug industry are notorious, and have been for decades. The medical literaturealso contains clearly documented examples of patients undergoing higher, sometimesvery much higher, rates of diagnostic investigation when the practitioner has an equityinterest in the facility [29]. And the public advertising for certain new for-profitsurgical facilities in the Toronto area, contains no warnings of the professionaluncertainties associated with the procedures they offer. Leading figures in the medicalprofession, world-wide, have sounded very direct warnings about the threat to bothphysicians and patients from increased reliance on economic incentives to motivateprofessional decisions [30].Yet "marketing" in the encounter between provider and patient is virtuallyimpossible to control from outside. The provider may be offering aservice performed in a "high-quality" manner, to a patient who is unaware of theevidence (or lack of it) of effectiveness and emotionally vulnerable. Hence the heavyreliance placed by professional organizations on inculcating and maintaining professionalethics, backed up but not substituted for by regulations and external oversight. Thus ifservices are simply excluded from public coverage and left to the private marketplace,the problem of acquisition of valid information on which to base decisions is likely tobecome even more acute than it is in a professionalized system.This may not be a conclusive argument against withdrawing coverage -- imperfectinformation is a problem for consumers of other products as well. But it emphasizes yet31. If in addition that provider has access to, and someinfluence over, the corresponding public facilities, thepossibility arises that s/he may deliberately manipulatepatients' access to those facilities, so as to encourage the useof the private service. This has been, prior to the recentreforms, a long-standing complaint about the behaviour of privateconsultants in the British National Health Service.38again that if one expects any forms of deinsurance or user fees to change consumerbehaviour, to encourage them to make informed and rational choices, then thetransactions which they participate in have to be structured such that they can get validand relevant information, and have the opportunities to use it. Otherwise chargingpatients !§..merely a "tax on the sick", with or without opportunities for providers tocircumvent public controls on prices or marketing behaviour.39REFERENCES[1] Stoddart, G.L., M.L. Barer, R.G. Evans and V. Bhatia (1993), "Why Not UserCharges? The Real Issues" Toronto: The Premier's Council on Health, Well­being and Social Justice Discussion Paper.[2] Evans, R.G., M.L. Barer and G.L. Stoddart (1993), "Charging Peter to Pay Paul:Accounting for the Financial Effects of User Charges" Toronto: The Premier'sCouncil on Health, Well-being and Social Justice Discussion Paper (forthcoming).[3] Ontario Economic Council (1976) Health: Issues and Alternatives Toronto:Ontario Economic Council.[4] Barer, M.L., R.G. Evans, and G.L. Stoddart (1979), Controlling Health CareCosts by Direct Charges to Patients: Snare or Delusion? Toronto: OntarioEconomic Council.[5] Curry, Adams and Associates (1992) Public and Private Health Care Financing:Literature Review and Description (October 1) Ottawa: Curry Adams.[6] Papp, L. (1992). New deal with MDs costly for workers, labor says. The TorontoStar. October 8:AI8.[7] Bochove, D. (1992). OMA ratifies removal of OHIP coverage of some services.The Globe and Mall. November 9:A8.[8] MacDonald, J. (1992). Deinsured services [Letter to Editor]. The Globe and Mall.December 7:AI8.[9] Barer, M.L., C. Hertzman, R. Miller and M.V. Pascali (1992), "On Being Old andSick: The Burden of Health Care for the Elderly in Canada and the UnitedStates", Journal of Health Politics. Policy and Law 17(4):763-782.[to] British Columbia, The Royal Commission on Health Care and Costs (SeatonCommission) (1991) Closer to Home [Report, Three Volumes] Victoria, B.C.: TheProvince of British Columbia.[11] Brook, R.H. and M.E. Vaiana (1989) Appropriateness of Care: A Chart BookWashington D.C.: National Health Policy Forum.[12] Lavis, J.N., and G.M. Anderson "Hospitals in Canada: The Appropriateness ofSetting and Services Provided" Work in Progress for Submission to the Queen's­University of Ottawa Economic Projects.40[13] Muran, K. "The Case for 'Abuser' Fees in the Medical System" Toronto StarJuly 20, 1993 p, A17.[14] Evans, R.G., M.L. Barer, G.L. Stoddart and V. Bhatia (1993) "Who Are theZombie Masters, and What Do They Want?" Toronto: The Premier's Council onHealth, Well·being and Social Justice (forthcoming).[15] Barer, M.L., V. Bhatia, G.L. Stoddart and R.G. Evaus (1993) "The RemarkableTenacity of User Charges", Toronto: The Premier's Council on Health, Well·being and Social Justice Discussion Paper (forthcoming).[l6] Katz, S.J., T.G. Hislop, D.B.Thomas, and E.B. Larson (1993) "Delay fromSymptom to Diagnosis and Treatment of Breast Cancer in Washington State andBritish Columbia" Medical Care 31(3) pp. 264·268.[l7] Barer, M.L. and G.L. Stoddart (1991), Toward Integrated Medical ResourcePolicies for Canada: Background Document, Discussion Paper #91:6D (June),Centre for Health Services and Policy Research, University of B.C.[18] Hurley, J., and N.A. Johnson (1991) "The Effects of Co-payments within DrugReimbursement Programs" Canadian Public Policy Vol. XVII(4) pp. 473489.[19] Foxman, B., R.B. Valdez, K.N. Lohr, et aI. (1987) "The Effect of Cost Sharing onthe Use of Antibiotics in Ambulatory Care: Results from a Population-BasedRandomized Trial" Journal of Chronic Diseases 40(5) pp. 429·437.[20] Evans, R.G., and M.F. Williamson (1978) Extending Canadian National HealthInsurance: Options for Pharmacare and Denticare Toronto: University ofToronto Press for the Ontario Economic Council, 1978, 235 pp.[21] Evans, R.G. (1980) "Professionals and the Production Function: Can CompetitionPolicy Improve Efficiency in the Licensed Professions?" in S. Rottenberg, ed.,Occupational Licensure and Regulation, Washington, D.C., American EnterpriseInstitute, pp. 225·264.[22] Wennberg, J.E. (1990) "Outcomes Research, Cost Containment, and the Fear ofHealth Care Rationing" New England Journal of Medicine 323:1202-1204.[23] Rodwin, M.A., (1993) Medicine. Money & Morals: Physicians' Conflicts ofInterest New York: Oxford University Press.[24] Lindorff, D. (1992) Marketplace Medicine: The Rise of the For.Profit HospitalChains New York: Bantam.41[25] Trebilcock, M.J., CJ. Tuohy and A.D. Wolfson (1979) Professional Regulation: AStaff Study of Accountancy, Architecture, Engineering and Law in OntarioPrepared for the Professional Organizations Committee, Toronto: Ministry of theAttorney-General of Ontario, January.[26] Evans, R.G. (1984), Strained Mercy: The Economics of Canadian Health CareToronto: Butterworths, 1984.[27] Canada, Royal Commission on New Reproductive Technologies (1993) ProceedWith Care (Final Report) (Patricia Baird, Chairman) Ottawa: GovernmentServices Canada.[28] Priest, L. (1993), "Hospitals' plan to seek U.S. patients worries minister", TorontoStar April 15, p. 10.[29] Hillman, BJ., C.A. Joseph, M.R. Mabry et aJ. (1990) "Frequency and Costs ofDiagnostic Imaging in Office Practice -- A Comparison of Self-referring andRadiologist-Referring Physicians", New England .Journal of Medicine 323(23) pp.1604-8.[30] ReIman, A.S. (1992)"What Market Values Are Doing to Medicine" The Atlantic269(3) (March) pp. 98-106.


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