UBC Faculty Research and Publications

Reflections on the financing of hospital capital : a Canadian perspective Barer, Morris Lionel, 1951-; Evans, Robert G., 1942- Jun 30, 1990

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

Item Metadata


52383-Barer_M_et_al_Reflections_financing.pdf [ 14.05MB ]
JSON: 52383-1.0048510.json
JSON-LD: 52383-1.0048510-ld.json
RDF/XML (Pretty): 52383-1.0048510-rdf.xml
RDF/JSON: 52383-1.0048510-rdf.json
Turtle: 52383-1.0048510-turtle.txt
N-Triples: 52383-1.0048510-rdf-ntriples.txt
Original Record: 52383-1.0048510-source.json
Full Text

Full Text

--- -- -RESEARCH UNITREFLECTIONS ON THE FINANCING OFHOSPITAL CAPITAL:A CANADIAN PERSPECTIVEM.L; BarerR.G. EvansHPRU 90:17D June 1990Discussion Paper SeriesTHE UNIVERSITY OF BRITISH COLUMBIAThis paper was prepared for the Division of Health Policy and Management,School of Public Health, Columbia University, as part of a researchprogram sponsored by the United Hospital Fund. The financial support ofthe United Hospital Fund and the Commonwealth Fund are gratefullyacknowledged . We wish to thank Steve Kenny, Lillian Bayne and Raisa Deberfor providing materials and information, from which we have drawnextensively. Errors, omissions and opinions are, however, ours.I. The Unusual Nature of Health Care CapitalWhat is capital? Capital is commitment. It can take many forms,including bricks and mortar (physical) , trained health care personnel(human), and research and development activities (intangible). Thecommon element is the notion that the using up of resources in the presentwill yield a stream of future benefits from the created capital stock .But all do require that initial "using up" of real resources; hence"commitment". A decision to purchase an item of diagnostic equipment, orto train an additional cardiac perfusionist, is a decision to commitresources 1 to a particular purpose, and therefore to render themunavailable for other uses . This commitment is as important for human asit is for physical capital , although there are differences in the societalcosts of extrication from each type of commitment.The notion of capital as representing a societal commitment, to someactivities rather than to others, is true for any form of capital, in anysector of the economy. Capital investment decisions in the health care"industry" seem however to be relatively inflexible compared with such1 "Resources" is used here in the economist's sense - people's timeand energy , materials and supplies, or the physical capacity of buildingsor equipment. When applied in one context, for one purpose, realresources thereby become used up or otherwise unavailable for applicationin other productive opportunities. The other things that might be donewith the same resources are foregone, and the value of the foregone outputin the most productive alternative use represents the true cost of usingthe resources in the chosen way. Money, while a "resource" to theindividual holding it, is not a resource to the community. While thedistribution of money determines who has command over the real resources ,printing more of it does not add to the pool of such resources.Barer/Evansdecisions elsewhere in the economy .2Within this sector it is difficult,often close to impossible, to reverse decisions or to change course oncecommitted : While other sectors share some of the reasons for this"stickiness", we suggest that none share all.The irreversibility of health care capital has three components .First, health care capital shares the attribute of "rigidity inapplication" with physical capital in all sectors of the economy. By thisis meant that most types of capital (including human) are often designedfor one, or a very few, types of application. A CT scanner does notprovide the means to produce high precision aeronautic parts. Nor can anelectronics warehouse be easily converted to use as an inpatient carefacility. This characteristic is clearly not unique to health care.But a second attribute of health care capital, which is not in factintrinsic to the capital itself but seems, in all western industrializednations, to be a consequence of the institutional environment in which thecapital finds use, is "allergy to dis-use". It has historically been thecase that idling the CT scanner is much more difficult, and done much lessoften, than idling the automobile plant or the pulp mill. Roemer's law isendemic to health care, but much less apparent in other sectors of theeconomy. While this attribute may be viewed as an 'environmental' ratherthan intrinsic characteristic of health care capital, it is no less realand no less important for considerations of health care capital financing .In fact, it is an argument that capital financing decisions in this sectorBarer/Evans3must take into account 'environmental' factors which do not come into playin capital investment decisions elsewhere in the economy.Third, and perhaps most important, new health care non-human capitalbrings with it demands for, or expectations of, new and often quitespecialized human capital . Once the human capital is in place, idlingphysical capital, for example, offers the prospect not only of turning offswitches on machines, but the redeployment, or costly re-tooling, of thecomplementary human resources. Not only is the physical capital theraison d'etre for the human resources, but the reverse also becomes truein practice. And the human capital tends to be the more problematic toidle.This is nowhere more apparent than with the training of physicians inNorth America and elsewhere. Industrialized nations have confrontedphysician surpluses for many years (Viefhues, 1988; Schroeder, 1984;United States, 1980; Lomas, Barer and Stoddart, 1985; Barer, Gafni andLomas , 1989 ; Canada, 1984), without the emergence yet of a comprehensiveand coherent policy to change direction. Not only is it now widelyrecognized that this particular human resource capacity has an importantinfluence on the health care costs of the population it serves, but thiscapacity also creates immense pressure for maintaining what is, for it,complementary physical capital.An additional , important and related consideration that furtherBarer/Evans24distinguishes health care capital is its often low cost relative to theassociated health care operating costs. The cost of training a physician,while high by educational standards, is low relative to the stream ofincome expectations with which it is associated. 2 Similarly, the capitalThis seemingly innocuous statement turns on the distinctionbetween everyday "cash" definitions of cost, and the more sophisticatedconcepts of opportunity cost employed in economic theory. What isreferred to here is the fact that the net present value (NPV) of the(discounted) earnings stream of the average physician - in Canada paidalmost entirely from the public insurance program - is much larger thanthe costs of training that physician, as reflected in the budgets of therelevant university and bospital - again largely paid from public funds.From the taxpayer's perspective (or for that matter the perspective of anAmerican private insurer) each new physician represents an "unfundedliability" substantially larger than the outlays for training.The economic theory of human capital formation, however, remindsus that a major portion of the cost of training any individual is theopportunity cost of that individual's time, the other income/productforegone while the individual is being trained. In a world of perfectinformation and unrestricted, unsubsidized free markets, people willchoose to invest in training up to the point at which the (net presentvalue of the) full cost of training just equals the NPV of the incrementin earnings (or more rigorously the increment in net advantages) resultingfrom training . If, in such a world , training is subsidized but theincrease in income (net advantages) accrues entirely to the individual ,then investment in training will be carried to the point at which the NPVof that increase equals that of the component of training costs borne bythe individual - by hypothesis less than the total, and an "oversupply" oftrained capacity.While "true" on its own terms - i.e . internally consistent - itis not clear that this framework provides us with any insights. In thefirst place it is a theory which describes positions of equilibrium in the"long run", and does not address questions of adjustment speed or path.It could therefore only be helpful if one had reason to believe - forreasons quite independent of the theory itself - that actual experiencewere in a more or less close neighbourhood to equilibrium, most of thetime . Given the long time horizons of career decisions in medicine, thelimited information with which they are made, and the high costs of"recontracting", such an assumption is not compelling a priori.Second, the theory describes a relationship which holds for themarginal investor/career chooser; whether or not it has any relevance forthe average depends upon further assumptions about the shapes of utilityBarer/Evans5cost per bed of hospitals is low relative to (even the discounted) per bedoperating costs over the life of the facility. Capital investmentdecisions in health care rarely take account of these downstream costs,perhaps because they are costs "in the future" and they are often (atleast in the U.S. environment) someone else's costs . Yet it is thoseother costs, which become income expectations, that make the initialcapital investments so difficult to reverse.Two points of particular significance seem to emerge from theforegoing. First, trained human resources and scientific advances are asmuch a part of the health care capital stock as are buildings andequipment. While much of the rest of this paper focuses on capital in thetraditional sense of bricks and shiny metal, the complementarity of typesof capital in health care should inform debate over, and policy decisionsand production functions in markets for both training, and the servicesprovided by trained persons.Finally, and perhaps most important, in neither of these marketsdo structures correspond even remotely to those required for, or assumedby , the theory. Entry to training is constrained; there are still manymore applicants than training places in Canadian medical schools. Theprecipitous decline observed in the United States, now down to about 1.7applicants per place, has no Canadian counterpart. This is prima facieevidence that the "marginal" successful entrant regards the net advantagesas grea~er than the cost to herself. Nor do prices in medical servicesmarkets respond to supply and demand in the conventional way. Increasesin supply lead to upward pressure on prices, which in Canada and WesternEurope is more or less constrained by monopoly/monopsony bargainingbetween insurers and physician associations. In the United States, bycontrast, prices/fees just rise. Accordingly, there is no ground forassuming that the resulting pattern of prices, incomes, utilization, andcapacity will bear any relationship at all to that "predicted" by modelsof unconstrained markets with freely adjusting prices, perfectinformation, and rapid adjustment of both output and capacity.Barer/Evans6governing , the commitment of a society's resources to physical health carecapital.Thus, and this second point is particularly important in the healthcare field, decisions about one form of capital constrain or mandatedecisions about others . The decision to equip a coronary artery bypassgraft facility is a decision to train or purchase perfusionists. Thephysical capital cannot be (or rather more accurately rarely is) leftidle, for economic and non-economic reasons. In turn, the pressures forCABG units arise from upstream decisions on the numbers and mix ofphysicians trained. If there were no cardiac surgeons there would be noCABG units. 3This whole feedback loop is triggered by scientific and technicaladvances, a form of accumulation of intellectual capital, which makes newinterven~ions possible. The possiblity compels the fact -- a need formachines, thence people, thence more machines ..... Hence the commonobservation in health care technology, that the dissemination of a new3 This is not meant to imply that there should not be any bypasscapacity. Rather, we are re-emphasizing the notion that the numbers ofsurgical interventions of this type that are performed, and the segmentsof the population on which they get performed, are influenced directly bythe amount of human capital with the enabling skills, independently of thebalance of benefits and costs from the procedure. The fact that one findsmuch more of this procedure in the United States than in Canada, andparticularly among the population over 65 years of age, is notcoincidentally linked to the differences in the two countries in theavailability of the requisite physical and human capital. Yet the juryseems still some way from reporting out on the effectiveness of many ofthe procedures performed on those over 65 (Anderson et al., 1990).Barer/Evans7diagnostic or therapeutic manoeuvre precedes the evaluation of itseffectiveness, and indeed makes objective evaluation extraordinarilydifficult.In sum, the opportunity costs of a capital investment decision arefar more than the immediate costs of that particular decision; theyinclude the subsequent constraints on decisions which will result from theexistence of that capital. The economists tell us that "sunk costs aresunk". The nature of human behaviour, particularly in health care,invalidates this impeccably logical aphorism . Sunk costs may be sunk, butas they sink they create eddies which pull other costs after them!II . Health Care Capital As Public Policy Target. But capital creation has both costs and benefits. It is not byaccident that we see so much human energy invested in the effort. Thecreation of human health care capital - trained personnel - is motivatedby a societal interest in the improvement of the health of populations.Physical health care capital is the productivity enhancer -- it isintended to make more good things possible with any given supply of humancapital. 4 But the market , or what passes for and is presented as a market4 Of course this assumes that the only relevant dimension withinwhich to measure benefits is that of population health status improvement.It is certainly true that certain segments of that population benefit frominvestments in human and non-human health care capital. The owners ofeach type of capital have clear interest in the availability of the other.Different kinds of capital generate quasi-rents , the size of which will beBarer/Evans8in the United States, will not provide anything near an optimal solution.In fact, in the face of the feedback dynamics described above, theoptimization of population health status would be a delightful, andunexpected , coincidence indeed. It is this 'market failure' that makeshealth care capital financing an issue for public policy and scrutiny.The task for public policy is to find a way of ensuring a reasonedbalancing of costs and benefits, in the face of seemingly insurmountableodds -- the research task is daunting, costly and time consuming; thepolitical and vested (human capital) interest pressures are enormous andimmediately present. Unfortunately the very limited research evidence oneffectiveness and efficiency, the sheer mind-numbing, continuouslygrowing, and inter-related numbers of interventions that might benefitfrom evaluative attention, the uneven quality of much of the evidence thatis generated, the difficulty of converting even such 'good' researchevidence as does exist into policy change (e.g. Lomas et al ., 1988, 1989),and the enormous immediacy of the countervailing pressures, ensures thatmuch of this capital creation goes on in innocence of any concreteevidence of its effect on the underlying objective.For example, some conditions clearly benefit from hospital-basedtreatment. CABG surgery does benefit some patients. The hospitalizationof patients diagnosed as suffering from senile dementias is, for some,determined in part by the availability of complementary capital undersomeone else's control.Barer/Evans9clearly preferable to the alternatives. Outcomes in some births areclearly better with casesarian section than without . MRI provides animproved basis for therapeutic decision-making in some cases. And so on.It is not that capital of no value gets created. Rather. the scope ofapplication of capital (that is widely regarded as being of considerablebenefit in some cases, and at least positive (even if small) benefit inmany others). almost universally extends beyond that for which evidence ofits effectiveness (let alone efficiency) has been demonstrated. Thisapplies as much to physician supplyS (Lomas and Stoddart. 1985; Denton etal . • 1983) as to MRI . GT scanning, pacemaker insertion (Greenspan et al.,1988), hospital bed capacity (Evans et al .• 1989; Hertzman et al., 1990),and obstetrical ultrasound (Anderson and Allison , 1990). Once theIcapital is created, the feedback loop dynamic is unleashed, and theinevitable result is growing concern over unnecessary utilization, over-capitalization, or uncontrolled costs.One logical response to such pressures is to attempt to imposeexternal limitations on the creation of capital. The forms this takesvary with the nature of the capital . Restrictions on bricks and mortarare virtually universal in North America and Europe -- there are now few5 - We do not intend to 'pick on' physicians. But physician supply isa two-edged sword . Not only does that supply dictate the supply ofcomplementary health human resources (which, because of restrictive rightsto practice. means just about every other category of such personnel ,directly or indirectly), but the over-investment in physicians is also thesingle most important constraint on the development of more efficientsubstitute human capital (Lomas, 1987).Barer/Evans10net additions to per capita hospital capacity, and those that do takeplace are tightly regulated. The diffusion of big ticket diagnostic andtherapeutic equipment is also now relatively tightly controlled everywhereexcept in the United States. 6Controls on human capital (health human resources) are more varied.A number of countries impose the so-called numerus clausus on numbers ofnew medical students (Germany, Netherlands, Canada); others leaveenrolment open (Belgium, Italy, the United States, Israel). Controls onthe training of other complementary health manpoower are even more variedand less rigid. But even countries with a firm numerus clausus findreduction far more difficult than was expansion. Reduction idles medicalschool human and physical capital.Finally, external limitations on the development of scientificcapital seem to be non-existent everywhere. Quite the contrary -- suchadvance is enthusiastically promoted and assisted to some degree invirtually all countries. This leads to the paradoxical situation that theequipment and physical facilities which would house and embody such6 This is an extremely important point. It is not that everyone isstruggling with the same problems, and that no one has developed areasonable solution. Rather, it is that the United States has chosen tobreak all the wind (to use the cycling analogy), is increasingly findingit lonely out in front in the cost race, notices that the pack is makingno effort to 'bring it back', and is now rather more desperately seeking away of taking itself back. The task is made difficult by the fact thatthe bicycle, once created and set in motion, has taken on a momentum andlife of its own, and seems not at all interested in revealing the positionof the brakes to the rider attempting to control it.Barer/Evans11advances are restricted. Given this fundamental contradiction, it is notsurprising that capital financing policy is difficult and contentious.What is missing is some adequate process, formal or informal, forbalancing the fully attributed costs of capital decisions (which arethemselves difficult to quantify), against the often difficu1t-to-identifybenefits.The form or mix of external limitations will influence the mix ofcapital and the manner in which it is deployed. A concrete example isgiven by the contrasting experiences of Canada and the United States inthe utilization of hospital beds. Until the 1980's both countriesdemonstrated the validity of Roemer's Law, that a built bed is a filledbed, and in both countries a steadily increasing number of physicianscreated great countervailing stresses on regulatory policies designed torestrict the supply of beds. The shift to prospective payment in the U.S.led to a rapid fall in the use of hospital beds, a fall which did notoccur in Canada , where Roemer's Law continues to apply. But the pressurefor complementary capital in the U.S. has not abated; it has simply beentransferred from beds and bricks to big ticket equipment. It is bound toemerge somewhere so long as the sources of pressure (medical personnel,techno~~ica1 extension) continue to expand relative to the needs of thepopulation.Thus, there is an inherent dynamic in health care capital formationthat will yield different mixes of resource inputs for addressing theBarer/EvansEven12health care needs of different populations. The mix available to anyparticular population, at any particular juncture, will depend on what'triggering' capital formation decisions have been made along the way.The unregulated, or patchwork-regulated input mix 'solution' will not,except through extraordinary good fortune, bear any resemblance to the mixof resources that would most efficiently address a population's healthcare needs. This is nowhere more vividly illustrated than by the currentsituation in the United States, where virtually everyone now agrees thatthere is excess bricks and mortar, excess big ticket equipment, and wheremost agree that there is excess in the most highly trained human capitalcategories.III. Policy Options for Health Care Capital FinancingBut how does one move toward this efficient mix of resources, andtherefore toward an optimal mix of human and non-human capital?assuming one is able to develop a clear set of population healthobjectives, and an appropriate role for health care in the pursuit ofthose objectives, there are a variety of policy levers that may be used toguide the formation of the health care resources used in that endeavour .Capital creation and financing policy is but one of these. Thearticulation of capital policy with the process of clearly identifyinghealth and health care system objectives has not been markedly successfulanywhere. Hence , one does not find clearly set out objectives for acapital reimbursement policy, beyond some vague notion that it should beBarer/Evans13structured to enhance the efficiency of health care delivery.However one deals with the issue of objectives for a health caresystem, there seem to be three generic approaches to capitalformation/diffusion policy . The first is direct administrative fiatwithin the context of a health care service -- the U.K.'s national system,the Swedish county system, or the U.S.' Veterans' Administration system.The second is arms-length regulation (e.g. American certificate of needlaws). The third is through reimbursement policy (as in Canada).The first approach is irrelevant in the absence of some hierarchicalvariant on a health care service. This seems a remote possibility inCanada, let alone the United States, and thus receives no furtherattention here . Americans have found the second approach to be largelyineffective, at least in the forms and under the conditions attempted.This is not to say that there is something inherently flawed about thisapproach. Either the environment in which it was applied, or theapplication itself, or some combination, led to its general failure.Again, there is a vast literature on this initiative, to which we couldadd nothing helpful for the purposes of the present paper.The Prospective Payment System is an attempt to implement a form ofthe third approach. It has had a major impact on patterns of capitalutilization, if not on total use or aggregate costs . The morecomprehensive form of reimbursement policy applied in Canada appears toBarer/Evans14have had a more substantial impact, not only on the availability and mixof health care physical capital, but on the more broadly defined levels ofactivity and use of complementary capital resources and, ultimately, onaggregate costs.In the remainder of this paper we focus on variants of thereimbursement policy approach to capital financing policy . In particular,we describe the process of capital financing in Canadian, and morespecifically British Columbia, hospitals, offer some thoughts on both theadvantages and pitfalls associated with this form of financing , and closewith some more general reflections on moving toward an 'optimal' capitalfinancing strategy.IV. Financing Hospitals and Hospital Equipment in British ColumbiaMost Canadian hospitals are publicly owned, and those that are notare non-profit (e.g. owned by the Salvation Army). About 95% of theiroperating budgets derive from a single source -- provincial Ministries ofHealth. 7 These funds, in turn, are generated through a variety ofsources and types of taxation, most notably provincial income taxes andtransfers of federal tax revenue. In British Columbia, as in the rest ofCanada, there are no user charges on hospital services. Residents also7 The remainder comes from a variety of other sources, includingWorkers' Compensation Boards, and private insurance companies (for e.g.private room charges) (Pink et al., 1989).Barer/Evans15are not required to pay hospital insurance premiums. 8The specifics of hospital capital financing vary across provinces(Deber, Thompson and Leatt , 1988; Bayne and Walker, 1989). But thespecifics tend to be less important than the general model, of the majorplanning and funding roles played by provincial Ministries of Health.Furthermore, it is universally true that provincial Ministries alwaysprovide considerably less than hospitals' collective self-perceived'needs', and that the funding arrangement does not necessarily orautomatically ensure that a piece of equipment, once funded in aparticular institution , will be replaced when obsolete. But we arerunning ahead of ourselves.Hospital operatin~ budgets are set prospectively, customarily on thebasis of previous years' budget experience, modified for any new servicesor programs. British Columbia has begun recently to move away from thisreliance on historical precedent as the criterion for intra-provincialallocation of beds and operating budgets, and toward a notion of regionalallocation on the basis of population size , structure and special needs .The operating grants from the B.C. Ministry of Health typicallyamount to 85% 90% of anticipated operating costs. Any given hospital'sgrant will be a function of two key considerations. The first is the8 Only Alberta requires the payment of hospital premiums.Barer/Evans16total funding made available annually to the Hospital Programs Divisionwithin the Ministry of Health. This level of funding is only arrived atafter negotiations between Ministries all vying for larger pieces of thepublic general revenues, and then within the Ministry of Health, whereHospital Programs must compete with the Medical Services Plan, CommunityCare Services , Preventive Services and the other Ministry branches anddivisions.These negotiations, particularly between the Ministry and othercompeting Ministries (e.g. Tourism, Education), are critical components ofthe overall control and management of health care in Canadian provinces.As the Ministry with the largest budget (by some distance), Health comesunder particular scrutiny in negotiations among Ministers, and theMinister of Health must be able to convince his colleagues that givingthem less (him more) makes their constituents better off!The allocation of grants to individual hospitals involves dividing upthe resulting pool of funds from Hospital Programs, partly on the basis ofthe previous year's distribution, but with some discretionary fundsallocated to take account of significant shifts in population, referralpatterns and new programs . The rest of the hospital's operating budgetcomes from revenue from other sources, but again it is important to notethat the major 'other source' is simply a different branch of the Ministryof Health. Hospitals can charge the province's Medical Plan for servicesprovided to outpatients, and this source of revenue commonly covers muchBarer/Evans17of the remaining operating requirement. While such revenues are generallytaken into account in the setting of the global prospective budgets,obvious incentives remain. In fact, hospitals that manage to under-run onthe operating side are, under certain circumstances, able to apply savingsto non-operating cost items such as capital acquisition. 9 As hospitalsare not permitted to run deficits, nor to raise private capital foroperating purposes, they must live within their budgets, appeal to theMinistry to cover over-runs, or absorb over-runs in the operating costs ofthe subsequent year. 10 Furthermore, capital depreciation is not'reimbursed' through the operating side of hospital accounting. Thisoperating cost financing arrangement is critical to the Canadian hospitalcost experience.A key characteristic of the Canadian financing model is that thissame funding agency, the Ministry of Health in each province, is also thesource of the major share of funds for capital construction and equipment.The details of capital cost-sharing (between Ministry, hospital, and othersources) vary across provinces. The fact of majority responsibility9 If a hospital's surplus exceeds 2 - 4% (depending on the size ofthe hospital), it can make application to the Ministry to apply some ofthe surplus to specific capital purchases. However, the hospital must beable to~make the financial case for the purchase. That is, the Ministryrequires that the requested capital purchase must generate futureefficiencies elsewhere sufficient to offset the initial capitalexpenditure.10 As Deber et al. (1988) note, it is often exceedingly difficult forprovincial Ministries not to 'pick up the tab' on deficits run up byhospitals. As often as not (probably more often), the Canadian provinces'policies of not picking up deficits have not been enforced.Barer/Evans18resting with the Ministry does not . In what follows we describe first thearrangements governing the construction, renovation, or expansion ofhospital 'bricks and mortar', then discuss planning and financing for 'bigticket' as well as other capital equipment.Hospital construction and renovation in British Columbia is guided bya five year rolling capital plan, approved by provincial Cabinet.Proposals from hospital boards are vetted by Hospital Programs. But it isimportant to note that regional hospital districts 11 have a role to playas well (albeit not as important a role as they would like) . In theoryat least the Ministry would not approve for funding a request that was notalso approved by the relevant region. 12 The degree to which differentregions' input is taken into account, and at what stages , appears not onlyto vary but to be in the midst of a process of evolution presently.Relevant considerations in the initial Ministry screen are projectedregional needs for beds and other hospital-based services, size andefficiency of existing hospitals in the region, and competing provincial11 The province is divided into twenty-nine official regionalhospital districts (RHD) , plus one area containing one small hospitalwhich receive~ all its capital funding from the Ministry of Health. TheRHD's are geographic designations used for a variety of provincialplanning purposes. They may include as many as twenty hospitals (in thelargest urban area). They receive their operating funds through localproperty assessments.12 Approval at the level of the host regional hospital district isparticularly important in districts such as the Greater Vancouver RegionalHospital District, where a number of institutions may simultaneouslyexpress intent to develop major capital projects . Details of the approvalprocess may be found in published reports of the regional hospitaldistricts (e.g. Greater Vancouver Regional Hospital District , 1990).Barer/Evans19priorities for health care facility construction . How these criteria arearticulated and applied in a consistent manner has not always been clear,and there is considerable feeling at the regional hospital district level,that the process is more one of ad hoc decision-making than theapplication of"criteria consistent with provincial health and health caresystem goals and objectives. On the other hand , there is emergingevidence of 'Mi n i s t r y movement toward a more rational allocation tailoredto regional needs. As we note below, however, there are a number ofdaunting impediments to planning improvements of this nature.A successful proposal emerges from this process with "approval inprinciple", and those submitting the proposal are then expected to developa more detailed 'functional program' and a number of 'physical design'proposals. But once a proposal receives this preliminary approval , fundsare allocated within the capital plan for the planning phase, and fundswill be tentatively earmarked for future years within the five year plan ,for subsequent design and construction phase costs.The important considerations within the detail, are that the Divisionwithin the Ministry of Health with responsibility for negotiating andapproving operating budgets (Hospital Programs), is also a major player inthe capital construction approval process, and that funds committed toapproval, design and construction must corne from a pool of funds approvedas a five year plan at the most senior political level in the province.In short , the same internal allocative pressures that limit theBarer/Evans20availability of hospital operating funds , also bear on the allocation offunding for capital construction .But the story does not end there, because Ministry approval onlyprovides 60 per cent of the costs. The hospital must raise the other 40per cent, either from its regional hospital district, or through fundraising by its own (hospital) foundation. The exceptions are provincialtertiary care services (e .g. tertiary care components of Children'sHospital, the B.C. Cancer Agency), for which construction costs would befully funded by the Ministry of Health. To this point in time, virtuallyall the provincial tertiary care capital has been placed in one district.Because the capital funding requests to regional districts (forphysical facilities and equipment) generally f~r exceed their resourcecapabilities, and because the degree of capital cost-sharing by theprovince has varied with the type of capital, there has historically beena tendency for the districts to encourage hospitals to apply for the high­Ministry-contribution categories of capital (in particular new physicalfacilities, and diagnostic equipment; see below). This should change,however, with the new policy (effective June 1 , 1990) setting the Ministrycontribution level at 60% regardless of category of capital.At a very general level, the process for major capital equipmentfinancing is quite similar. That is, approval processes are followedwithin the Ministry of Health and within the regional hospital district,and much of the funding comes from the Ministry of Health. An annualBarer/Evans21allotment, presently just under $40 million (Canadian , for a provincialpopulation of just under 3 million) is provided within the Ministry'sbudget, to an Equipment Grants Secretariat, which then has theresponsibility of allocating the funds.Within this pool of funds , about 20% has historically been used foritems of value up to $10,000 but over $500. About 60% of the pool hasbeen available for items costing $10,000 - $100,000, with the r emainder(about 20 per cent) available for the purchase of items costing in excessof $100,000. Allocation of funds for small (under $10,000) items hastaken the form of secondary equipment allowances (SEA's) made toindividual hospitals. The allocation of these has been roughly alignedwith relative hospital operating budgets (Bayne and Walker, 1989) . Thesefunds have been intended to cover roughly one-third of the cost of suchcapital equipment, and could be used at the hospital's discretion foritems of this type . However, the remaining two-thirds has corne from thehospital's regional hospital district , the same source that must providesignificant shares of the cost of more major equipment, or from thehospital's own funding sources. Regional hospital districts haveapparently often "tended to discourage application of the 2/3 secondaryequipment cost contribution" (Bayne and Walker, 1989, 9).Until recently hospitals were required to make application to theMinistry for items valued at over $10,000. The process was similar tothat for capital construction, in that the hospital had to make a caseBarer/Evans22based on its functional role and the necessity of the equipment to thatrole. As of June 1, 1990, this process is now required only for items inexcess of $100 ,000 . The middle band funds will henceforth be consolidatedwith the small equipment allowances , and allocated to hospitals on thebasis of detailed allocation formulae comprised of a fixed amountdependent on role, plus an amount per bed.As noted above, different types of equipment have, until presently,received different levels of provincial financial support (Bayne andWalker , 1989). Three-quarters of the cost of major diagnostic equipment,for example, has been provided by the Ministry of Health. In contrast,the Ministry has funded only one-third of so-called "moveable equipment"(beds, chairs, desks , laundry machines, other non-diagnostic equipment) .The remainder has come from the regional districts or from hospitals' ownprivate foundations (which receive charitable donations and can undertakecommunity fund-raising for specific items or purposes). With the changein categories on June 1 came a change in the cost-sharing arrangement.Henceforth, the Ministry of Health will contribute 60% of cost,irrespective of type of capital.But in general no funding is made available by the Ministry toindividual hospitals unless they have filed a current five year capitalplan with the Ministry. Each hospital must submit an inventory ofequipment purchased at periodic intervals throughout the year. Thisinventory is checked against the five year plan and items approved by theBarer/Evans23Ministry.There is, of course, one other source of funding for capitalequipment -- the capital construction process . The equipment needsassociated with new physical space are approved in the same way asdescribed above, depending on the type of equipment. The cost-sharingarrangements are also the same, but this provides another avenue for theinjection of new equipment into a regional hospital district, and accessesa 'public account' separate from (although in some broader sense competingwith) the equipment fund pools for existing facilities.If new equipment is associated with, for example, a new diagnosticfacility (e.g. MRI service), the hospital must also submit a request foradjustment to its base operating budget to take account of the expandedfunctional role. This is an important consideration, as we see below, interms of the separation of operating and capital accounts. A hospitalcannot expect to receive operating funding for an equipment-based serviceif it has not been approved by the Ministry on the capital side .Conversely, a hospital cannot expect to receive funding from the Ministryfor the equipment unless it can establi sh a case for having the equipmentwithin ~ts regional hospital district.In sum , the funding for almost all forms of hospital capital incomes from capital accounts in the Ministry of Health, the regionalhospital districts , and hospital foundations. The Ministry provides 10 0%Barer/Evans24of capital funding for provincial referral services or programs (e .g. theprovincial cancer agency or tertiary care services within Children'sHospital), and, as of June 1, 1990, 60% for all other categories ofhospital capital (property, construction and 'fixed equipment' (i.e.equipment that cannot generally be moved from one site to another) ,diagnostic equipment, and 'moveable equipment'). It is the single largestsource of capital financing. All capital items in excess of $100,000require Ministry of Health approval, and expenditure on all but thesmallest items under this cut-off is audited against hospitals' five -yearcapital plans. Historically, in fact, items valued at between $10 ,000 and$100,000 have also required piecemeal approval, a labor-intensive andtime-consuming process.There are , then , a number of points of intersection between theoperating and capital sides of hospital financing in British Columbia.They compete for limited institutional funding , Hospital Programs isinvolved in both sides of the negotiation and approval processes, andapproval for adjustments to operating budgets is contingent on priorapproval of capital expenditures .V. Problems and ProspectsBut the planning and allocation processes are far from perfect . Thedecisions are subject to a number of sources of 'end-run' . Furthermore ,they are not well-informed by appropriate population-based data andBarer/Evans25evidence on effectiveness and efficiency. And finally, they are oftenmade in innocence of broader health and health care system implications.First , British Columbia is no more immune to the end-run than is anyother jurisdiction. There are a number of potential sources of pressurewhich can result in decisions being made independent of, and in fact inspite of, the formal bureaucratic Ministry and regional district planningprocesses . We can label these , for expository convenience, political, andlocal/institutional/professional.As an example of the former, an equipment manufacturer may be able toexercise 'moral' suasion with senior Ministry bureaucrats or politicalrepresentatives, for placement of new equipment on a trial, experimental,or research basis (see , e.g. McLintock, 1990). In the extreme case, suchequipment might find its way into a hospital in a 'politically friendly'regional district. But irrespective of eventual site, the effect is muchthe same. A new facility is established without having gone through theinternal planning/review process, funds are committed to it (leaving lessfor other applications in the pipeline), and even if it is introducedostensibly for research purposes, the potential remains for it to becomean add-on to the clinical arsenal over time. What distinguishes this typeof end-run from the others described below is that it ends up having callon the pool of capital funds within the Ministry of Health. Inparticular, if it is new big ticket diagnostic equipment, sixty percent ofthe capital cost would be the Ministry's responsibility, with theBarer/Evans26expectation that the regional district would make the remainder a highpriority!But there are no statutes prohibiting a hospital or a regionalhospital district from raising 100% of the capital requirement for, say, aCT scanner. The Ministry of Health is under no obligation to providecapital or incremental operating funds, but if the facility is able tooperate the equipment from within its approved operating budget, or findssome other source of operating revenue, the Ministry has no means toprevent the operation of the facility.13 Furthermore, there is nothingl~precluding physicians who use the facility from billing the provincialMedical Services Plan for the professional component1~ of (e.g. CTdiagnostic) services provided to outpatients.This is more than simply a theoretical curiosus . In one recentinstance, a hospital in British Columbia raised the funds for a CTscanner, the regional hospital district is providing all operating costs,13 In theory, at least, it could appoint a public trustee to replacethe existing board of trustees, perhaps on the grounds of mismanagement .In practice, however, it seems unlikely that this would occur (it hasnever yet occurred for this reason in this province).In some cases where non-hospital-staff physicians use hospital ­based diagnostic or therapeutic equipment, the hospital may bill theMedical Services Plan, and then pay the physician a 'professional fee',retaining the remainder as the technical, or overhead , component. If apiece of equipment has been purchased without Ministry approval, and ifthere are separate technical and professional components of an item withinthe provincial fee schedule , the physician could bill the Medical ServicesPlan for the professional component.Barer/Evans27and the Medical Services Plan pays professional fees for scans provided tooutpatients! In other situations , influential members of the public havemobilized sufficient community support to raise the necessary capitalfunds for the purchase of the equipment. Generally, hospitals havemanaged to find the operating funds for such facilities from withinexisting allocations, which raises questions about slack within theoperating budgets of institutions that are perpetually complaining aboutunder-funding!More generally , the pressure for this sort of facility or equipmentmay derive from the community, from hospital-based or hospital-usingprofessionals within the community, or from hospitals attempting toattract professionals by providing particular diagnostic or therapeuticcapacity. Direct capital funding may not be made available by '.heMinistry , but over time the operating cost imperatives eventually have away of working themselves into baseline budgets . Furthermore, Ministriesin some provinces have indicated that such initiatives should not bediscouraged because they "reflected and helped to strengthen thehospital's relations and interaction with its cor-s t Ltiuerrt s " (Bayne andWalker, 1989, 10).The source of the pressure is important only in80far as it serves toreinforce the notion that capital begets pressure for more capital. Thepolitical or true community initiative is the exception. It is probablysafe to surmise that 'community' pressure rarely emerges independently ofBarer/Evans28professional or institutional influence. Similarly, a philanthropicsource rarely offers unsolicited support for particular types of capital.The more common route is for the institution and/or key members of theprofessional community to mobilize the community to support the need for aparticular resource , or to develop innovative ways of raising funds fromsources other than the provincial Ministry (e.g. see Pink et al., 1989,for an Ontario example of expansion of physical capital without directMinistry approval or funding).For example , individual eminent clinicians may make independentarrangements with equipment manufacturers, for the testing and evaluationof new technology (i.e. for research purposes). Of course once suchequipment is in place, over time it tends to find its way into clinicalpractice, creating considerable pressure on the Ministry to provideoperating resources . 15The 'end-run' phenomenon is very clearly a hole in the cost controland technology diffusion net, and one that can be extremely tempting bothto the recipient hospital and to fiscally strapped Ministries of Health.For the hospital , the attractive prospect is of bypassing the time-consuming and often unsatisfying process of planning and approval; for the15 Two British Columbia examples from recent years are theintroduction of a lithotripter and an MRI machine. In both cases theyentered the province as research facilit ies, and in both cases one, or agroup of, academic physicians were influential in their placement. Inboth cases the equipment is now used for clinical practice.Barer/Evans29Ministry, the short run path of least resistance is to have someone elsefinance the capital, and to not worry about the potential operating costimplications. As noted earlier , however, the capital costs are almostalways a small share of the (discounted) total operating costs over theuseful life of the capital, and it is much more difficult for the Ministryto avoid eventually picking up those operating costs.'End-runs' are also often striking examples of human capital sourcesof pressure for these sorts of facilities, and of the demands such un-approved facilities place on the publicly funded training of more humancapital. Not only does new equipment create new demands for technicians,but it also may attract more clinicians . 16This leads us into the issue of lack of integration , in the Canadianprovinces, of phys ical and human capital planning decisions . Not only do'end-runs' of all varieties create pressures for the training ofcomplementary human capital, but physical capital decisions are alsolargely divorced from decisions about the mix and size of trainingprograms, and about health care personnel availability more generally.This leads to two different types of situations. One finds both instanceswhere the physical capital creates pressures on human capital (e .g.- ~situations such as cardiac transplant queues exacerbated by insufficient16 In fact one of the arguments one often hears, after capital is inplace , is that in fact one requires a clinical critical mass to make itspresence worthwhile!Barer/Evans30numbers of cardiac perfusionists), and situations where human capitalcreates pressures on physical capital (e.g. the physician communityarguing that the hospital sector is underfunded).Finally , Canada certainly does not lead, and perhaps lags otherjurisdictions, in the extent to which evidence on technologicaleffectiveness and efficiency are generated and, more importantly, used toguide decisions about temporal and spatial diffusion. While its sizeperhaps precludes it taking a lead in some of the more expensiveevaluations, the lack (until very receantly) of a national, or a series ofprovincial , 'clearing houses' for the considerable information that isgenerated outside the country, and the lack of a structure for ensuringthat such information informs the policy, planning and review processes,are serious problems with the Canadian approach to capital financing(Deber, Thompson and Leatt , 1988; Feeny, Guyatt and Tugwell, 1986) .VI . Some Final ThoughtsOn the other hand, it would be naive to think that in the foreseeablefuture , in Canada or anywhere else, evaluation ~ill always precedeclinical application. First, evaluation is often itself relativelycostly, and it seems unlikely that sufficient resources would ever be madeavailable to provide evaluative results for all possible newinterventions. Nor would such resource allocation necessarily beBarer/Evans31desirable. But, second, the nature of technological innovation, and thenature of evaluative research, militate against the emergence of researchresults in advance of clinical diffusion. Evaluations are not onlycostly, if done properly they can be quite time-consuming, particularly ininstances where the effects (or lack thereof) of particular newinterventions may not become evident for long periods of time.Technological capital does not stand still while evaluations proceed.There are countless examples ' of technology becoming obsolete long beforethe evaluation has concluded (Deber et a1., 1988, describe precisely sucha situation in the early days of CT scanners in Ontario).We do not mean to imply that evaluation of the effectiveness, andparticularly of the efficiency, of new capital is unnecessary. Rather, wewould suggest, first, that such evaluations should not always be assumedto be worthwhile (the costs of the evaluation may outweigh the fu11y­allocated costs of the capital) so that the evaluation of everything newshould not become a social objectve; and, second, that appropriate rolesfor evaluative research in the process of capital diffusion need to becarefully developed.While the development of such roles is beyond the scope of thispaper, we offer some very tentative thoughts here. First, it seems worthnoting the obvious, that technology assessment will take place even in theabsence of any effort to see diffusion decisions informed by efficiencyevidence. The process in Canada whereby provinces make decisions toBarer/Evans32finance some capital acquisitions but not others , represents a form ofimplicit technology assessment (Evans, 1982). This process can be more,or less, informed.Second, the promotion of an information diffusion process,articulated with the capital planning and funding processes, that wouldinform those latter processes at a variety of stages, would seem to be ofparamount importance. Where results become available before capitalequipment, facilities, or the like, become integrated into clinicalsettings, those results should playa role in decision-making , not onlyregarding whether they will be available, but in what numbers/amounts andto serve what populations. But, and this would seem critical, even ininstances where diffusion precedes evidence, mechanisms need to bedeveloped for 'turning off' the use of facilities or capacity if, as abalance of good evidence emerges, there is reason to question theefficiency of such capital. This may take the form, for example, ofeliminating both capital and operating funding for procedures shown to domore harm than good, and of providing financial incentives for replacingless efficient capital with more . Along the same lines, if capitaldiffusion has been restricted on the basis of research evidence, and thebalance of that evidence changes over time, the means for ensuring thatsuch information becomes available to those making decisions regardingcapital acquisition must be developed.Thus the process of planning, approving and funding must beBarer/Evans33sufficiently flexible to adapt to new information. Unless items orprocesses are shown unequivocally to do more harm than good, the healthcare systems on both sides of the border have demonstrated clearly that,if left to their own devices, efficient capital substitutions will not beadopted.Presently both Canada and the United States are faced with situationswhere existing information systems and research efforts are rarely able tocallan fundamental cost and outcome information as would be provided byanswers to such basic "What if" questions as "What if British Columbiadoubled its number of MRI machines?", or "What if Washington State halvedits available coronary artery bypass graft capacity?". In the absence ofsuch information, each country has chosen its own more rough-edgedapproach to rationalizing the availability of hospital capital.But whatever the processes by which capital is funded, the certaintyis that some or all of clinicians, patients and institutions will believethat not enough is available, irrespective of how much is in factavailable. The financial incentives driving the development of newtechnological and scientific capital are not only exceedingly difficult toalter, but such alteration would be undesirable. One does not want tothrow out the baby with the bathwater; one might, however, like to see alot less bathwater . Unfortunately part of the development incentive isthe ease of the sale . The more easily new capital is 'sold' to an eagerhealth care system, components of which stand to benefit from itsBarer/Evans34deployment, the more resources will be devoted to the generation of suchcapital. The more of such capital generated, the more pressures we willfind for the complementary human capital . Health care has traditionallynot been an industry in which innovation is driven by efficiencyconsiderations. Nor can one rely on the recipient professionals orinstitutions to provide arms-length estimates of capital requirements.Add-ons far outnumber the situations of pure substitution. Aging healthcare capital never dies ; it just fades away, very slowly.While Canada and the United States share an interest in seeingcapital diffusion guided by population needs and evaluative research onthe relative efficiency of alternative ways of meeting those needs,realistically we are some way from having timely research results for morethan a small subset of the possible evaluative targets (pharmaceuticals,diagnostic, and other therapeutic technologies, hospital vs. other typesof facilities or care, etc .). Furthermore, it would take more than thepresence of the necessary research resources to ensure that the clinicalapplication of new capital was always informed by research evidence.Nevertheless , the cost implications of having unevaluated capitalequipment applied in clinical practice are enormous.This suggests a need for some combination of centralized fiscalcontrol and needs-based planning to rationalize the temporal diffusion ofcapital, a population-based information system (including information ongeographic care-seeking patterns, particularly for tertiary care) to guideBarer/Evans35its spatial diffus'ion, and a framework for bringing such research evidenceas becomes available, to bear on both. All of which is a rather long­winded way of suggesting that leaving the control of capital diffusioneither to the 'market' or to evaluative/appropriateness research, willimprove neither the cost experience nor necessarily the appropriateness ofcapital diffusion.The situation in the United States at present, at least to observersfrom beyond its borders, appears to be one of a reluctant relinquishing of'market' solutions (which virtually everyone agrees have been a dismalfailure), in favour of more intrusive interventions . The development ,promotion and sale of regulatory initiatives such as protocol analyses,utilization management and quality assessment (e.g. Roper et al., 1988;Field and Gray, 1989; Brook and Kosecoff, 1988) are examples of newhea1th-care-re1ated growth industries. The only thing one can say aboutthese sorts of initiatives with any sort of certainty, is that they willincrease (direct and hidden) health care administrative costs in theUnited States, by creating a new set of incomes and income expectations inand around that country's health care industry.L~s certain by far is their effect either on the overall levels ofcare (and costs), or on the mix of appropriate and inappropriate care.One may be able to identify ineffective and inappropriate care, establishstandard practice protocols, and promote the adoption of more appropriateand efficient capital. It is a long, and suspect, leap from there toBarer/Evans36population-wide changes in care pattern appropriateness, and systemefficiency. Not only must the new protocols and guidelines be widelyadopted and applied, but where they imply lower (or no) incomes for somesegments of the human capital -, these segments must be assumed not torespond in income-protecting ways. Furthermore , this result is dependenton the continued uptake of new information. We suggested earlier, and itseems worth re-emphasizing here, that it is unrealistic to assume that allthe necessary evidence can be available when it is needed. The beliefthat we can identify and eliminate inappropriate care, and that this willyield both better health status and lower costs is laudatory, and wrong.It may also be an inefficient way of attempting to simultaneously improvequality and lower costs.This is not to imply that we should not try . Ineffective, andharmful, care are just that. Whenever we can identify and eliminate them,we should, but not at any cost! Something more is needed. Betterscience will not solve the problem of capital financing. In fact, it isneither necessary nor sufficient for an improved capital financingprocess. What seems necessary is a process that will provide a balancingof the costs of different types of errors. There is an unfortunatetendency when dealing with health care capital to apply, or succumb toapplication of, the 'law of asymmetric justification'. Expansion does notrequire justification , contraction does, because (so the argument mustBarer/Evans37go), errors of omission are more costly than errors of commission. 17 Yetas we have attempted to demonstrate in this paper, the errors ofcommission can be very costly indeed, because of the 'multiplier effects'unleashed by the interlocking types of health care capital .The Canadian model of capital financing might be described as"certificate ' of need with financial teeth". But that would be impartingperhaps too much refinement to the actual process. Canada does not yethave the balance right. There is plenty of scope for developingallocation criteria , informed by such research evidence as may beavailable, and applied at the regional level. In the United States, onthe other hand, what appears to be needed is a philosophical separationfrom the notion of quick fixes. The 'market' (a 'quick fix' because noone need do anything other than 'watch') does not provide an acceptablesolution; nor will the generation of ever more evidence on appropriatenessof care be a magic bullet. What is required is a large and bluntinstrument to go with the scalpels. Sometimes one will be right, othertimes one will be wrong. On balance, the cost of errors made by applyingthe stick is likely to pale (both in clinical and economic terms) beside17 This philosophy has pervaded debates over physician supply inNorth America during the past two decades. Medical associations have hadno trouble identifying shortage specialties and geographic areas, whilefinding it virtually impossible to establish criteria for theidentification of surpluses. Yet the implicit criteria underlying theestablishment of shortages have been every bit as arbitrary as thecriteria used by planners to identify surpluses. These asymmetric rulesof evidence playa critical role in the 'stickiness' of health carecapital decisions.Barer/Evans38the cost of errors made by attempting to be too cute .While the provincial processes in Canada are far from perfect, so farCanada still has more 7-Eleven stores than CT scanners, and there seems tobe widespread agreement that American capital surpluses are much more of a(clinical and economic) problem than Canadian 'shortages'.Barer/EvansReferencesAnderson, G.M. and D. Allison (1990), 'An Assessment of the Value ofRoutine Prenatal Ultrasound Screening. A Review prepared for the CanadianTask Force on the Periodic Health Examination" , HPRU Discussion Paper90 :6D, Vancouver: University of British Columbia, Division of HealthServices Research and DevelopmentBarer, M. L., A. Gafni and J. Lomas (1989), "Acconunodating Rapid Growth inPhysician Supply : Lessons from Israel, Warnings for Canada", InternationalJournal of Health Services 19:1, 95-115Bayne, L. and M. Walker (1989), "Capital Equipment Acquisition: ADiscussion Paper", Vancouver: Stevenson Kellogg Ernst and WhinneyBrook, R.H. and J .B. Kosecoff (1988), "Competition and Quality", HealthAffairs 7 :3, 150-61Canada (1984), "Physician Manpower in Canada", A Report of the Federa1­Provincial Advisory Committee on Health Manpower, Ottawa: Health andWelfare CanadaDeber, R.B., G.G .Thompson and P. Leatt (1988), "Technology Acquisition inCanada : Control in a Regulated Market", International Journal ofTechnology Assessment in Health Care 4, 185-206Denton, F .P., A. Gafni, B.C. Spencer, and G.L. Stoddart (1983), "PotentialSavings from the Adoption of Nurse Practitioner Technology in the CanadianHealth Care System", Socio-Economic Planning Sciences 17 , 199-209Evans, R.G. (1982), "The Fiscal Management of Medical Technology : The Caseof Canada", in H.D. Banta, ed., Re s our ces fo r Health. Technolo gyAssessment for Policy Making , 178-99 , N.Y.: PraegerEvans, R.G ., M.L. Barer, C. Hertzman , G.M . Anderson, I.R. Pu1cins and J .Lomas (1989) , "The Long Goodbye: The Great Transformation of the BritishColumbia Hospital System", Health Services Research 24:4, 435-59Feeny , D., G. Guyatt and P. Tugwell, eds. (1986), Health Care Technology:Effectiveness . Efficiency and Public Policy, Montreal: Institute forResearch on Public PolicyField, M.J. and B.H. Gray (1989), "Should We REgulate 'UtilizationManagement?', Health Affairs 8 :4, 103-112Greater Vancouver Regional Hospital District (1990), The Fact Book 1990 ,Burnaby: GVRHDGreenspan, A.M., H.R. Kay, B.C . Berger, R.M. Greenberg, A.J. Greenspan,and M.J. Spuhler Gaughan (1988), "Incidence of Unwarranted Implantation ofPermanent Cardiac Pacemakers in a Large Medical Population", New EnglandJournal of Medicine 318(3): 158-63Hertzman, C., I.R. Pulcins , M.L. Barer, R.G. Evans, G.M. Anderson and J.Lomas (1990), "Flat on your Back or Back to your Flat? Sources ofIncreased Hospital Services Utilization Among the Elderly in BritishColumbia", Social Science and Medicine 30:7, 819-28Lomas, J. (1987), Health Manpower In Ontario: Distribution. Planning andPolicies, prepared for the Ontario Health Review Panel, Hamilton: HealthEconomics and Policy Analysis Group , McMaster UniversityLomas, J., G.M. Anderson, E. Vayda, R. Roberts, and B.MacKinnon (1988),"The role of Evidence in the Consensus Process: Results from a CanadianConsensus Exercise", Journal of the American Medical Association 259,3001-5Lomas, J., G.M. Anderson, K. Domnick-Pierre, E. Vayda, M.W. Enkin and W.J.Hannah (1989), "Do Practice Guidelines Guide Practice? The Effect of aConsensus Statement on the Practice of Physicians", New England Journal ofMedicine 321:19, 1306-11Lomas, J., M.L. Barer, and G.L. Stoddart (1985), Physician ManpowerPlanning: Lessons from the MacDonald Report , Toronto: Ontario EconomicCouncilLomas, J. and G.L. Stoddart (1985), "Estimates of the Potential Impact ofNurse Practitioners on Future Requirements for Physicians in Office-basedGeneral Practice", Canadian Journal of Public Health 76:119-23McLintock, B. (1990), "'Freebies' for hospital board couple", TheVancouver Province, June 6, p . 6Pink, G.H., R.B. Deber , J.N. Lavoie and E. Aserlind (1989), "InnovativeFund Raising: The St. Michael's Hosp ital Health Centre", unpublishedmimeo, Toronto: University of TorontoRoper, W.L., W. Winkenwerder, G.M. Hackbarth and H. Krakauer (1988),"Effectiveness in Health Care: An Initiative to Evaluate and ImproveMedical Practice" , New England Journal of Medicine 319(18): 1197-1202Schroeder , S. (1984), "Western European Responses to PhysicianOversupply" , Journal of the American Medical Association 252, 373-84United States (1980), Report of the Graduate Medical Education NationalAdvisory Committee, vols. 1-7, DHSS Pub. No. (HRA) 81-651-657,Hyattsville , Md. : Health Resources AdministrationViefhues, H., ed. (1988), Medical Manpower in the European Community,N.Y.: Springer-Verlag


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