Physician resource planning: ways and means
Noralou P. Roos, PhD
CMAJ 1997;157:1229-30
[ résumé ]
Dr. Roos is Professor with the Department of Community Health Sciences, University of Manitoba, and Director (on leave) of the Manitoba Centre for Health Policy and Evaluation, Winnipeg, Man.
Reprint requests to: Dr. Noralou P. Roos, Professor, Faculty of Medicine, University of Manitoba, 5101750 Bannatyne Ave., Winnipeg MB R3E 0W3; fax 204 789-3910; nroos@bldghsc.lan1.umanitoba.ca
© Canadian Medical Association (text and abstract/résumé)
See also:
Abstract
There is no simple, widely accepted method for accurately estimating the numbers of physicians needed for a specific population. This creates a stumbling block as medical associations and provincial governments across the country jointly assume responsibility for managing physician resources. In this issue (page 1215) Dr. Noralou P. Roos and colleagues use the example of Manitoba to illustrate the usefulness of a needs-based approach to physician resource planning. In this editorial Roos outlines the key issues that must be addressed in estimating physician requirements and reviews current evidence on the implications of higher or lower levels of physician resources.
Résumé
Il n'y a pas de façon simple et généralement reconnue d'estimer avec précision le nombre de médecins nécessaires pour une population donnée. Cette lacune constitue une pierre d'achoppement au moment où les associations médicales et les gouvernements provinciaux du Canada assument conjointement la responsabilité de gérer les effectifs médicaux. Dans ce numéro (page 1215), le Dr Noralou P. Roos et ses collègues se servent de l'exemple du Manitoba pour illustrer l'utilité d'une stratégie de planification des effectifs médicaux fondée sur les besoins. Dans cet éditorial, le Dr Roos décrit les questions clés auxquelles il faut répondre pour estimer les effectifs médicaux requis et passe en revue les données probantes actuelles sur les répercussions d'effectifs médicaux plus nombreux ou moins nombreux.
The National Ad Hoc Working Group on Physician Resource Planning was formed by representatives of the CMA and of the provincial and territorial medical associations to develop a national framework for physician workforce planning. In its 1995 report, the working group recommended that the provinces modify their approach to determining physician supply requirements.1 The group suggested that standard physicianpatient ratios do not take into account the fact that some services are obtained by patients outside of their home community and urged that needs-based approaches to physician resource planning be examined.
In this issue (page 1215) my colleagues and I present an attempt to estimate the numbers of generalist physicians required in different regions of Manitoba. To identify areas of surplus and deficit, we used the number of visits actually provided by physicians (that is, the actual practice workloads in various regions) and the visits actually received by residents (recognizing the fact that both patients and physicians travel). By focusing on numbers of visits, we include the great majority of services delivered by generalist physicians in both urban and rural settings. After adjustment for population characteristics (i.e., age, sex and socioeconomic indicators), our analysis suggested that there is a relative surplus of physicians in low-need areas of the province and a substantial shortage in high-need northern areas.
The current state of knowledge does not permit an unequivocal formulation of the best method for determining a population's need for physicians. Benchmarking has recently been suggested as a useful approach.2 It involves identifying areas that have relatively low numbers of clinically active physicians without any apparent compromise of patient welfare. After adjustments are made for differences in age and sex across populations, this low deployment level (cited as the number of physicians per 100 000 residents) is proposed as the current best estimate of a reasonable physician workforce. If, in addition, adjustments are made for critical health and socioeconomic differences across populations, benchmarking offers a crude but potentially useful approach to physician resource planning.
In our study, if we had taken physician supply in Manitoba's healthy rural south as a benchmark, at least for Brandon and Winnipeg, we would have identified a substantially greater surplus of generalist physicians in urban areas than we report in our study. Alternatively, the higher levels of physician deployment in Quebec, Nova Scotia, Ontario and British Columbia could have been used as benchmarks for Manitoba; in this case it is likely that most areas of the province would have been shown to have too few physicians. However, if the lower deployment of physicians in Saskatchewan or Alberta had been used as benchmarks, greater surpluses would have been identified. In reality, there is little evidence to suggest that higher benchmarks would be justified. Manitobans enjoy good access to physicians: 84% of the general population and 95% of the elderly population make at least 1 physician contact per year. In fact, it appears that a lower percentage of Manitobans report delays in obtaining health care services than do Canadians generally (4% v. 7%).3
Critics of needs-based approaches to physician resource planning have asked for evidence that increased supply to high-need areas would actually improve the health status of the population. This is a reasonable concern, and one that raises an equally important question: Will reducing physician resources in low-need areas threaten the health of these populations? We have little evidence that a greater-than-adequate supply of physicians results in better health. Winnipeg has the most abundant supply of physicians in the province, and low-income Winnipeg residents have the highest physician contact rate in the province. Nevertheless, residents in low-income areas also have the highest rates of hospital admissions for conditions that physicians have identified as those for which medical intervention is believed to be effective in prevention,4,5 early detection or treatment.6,7
Others have found that access to a greater supply of physicians did not reduce hospital admission rates.8 We found that areas with a greater physician supply did not have lower rates of premature mortality.9 Much of what physicians do cannot be detected by changes in the mortality rate in the communities they serve. However, we have also found that routine physician contact does not appear to lead to improvements in quality of life.10 Thus, elderly Manitobans who had regular physician contact over a 12-year period were no more likely to age successfully -- that is, to reach an advanced age, to continue to function well at home, and to remain mentally alert -- than those without regular physician contact.
Despite the likelihood that, at the margin, adding or subtracting physician resources has little impact on health outcomes within a population, it makes sense in the context of capped funding to shift physician resources away from areas of relative surplus toward areas of deficit. Eyles and colleagues,11 using a different approach but a similar logic, suggest that almost twice as much per capita should be spent on delivering physician services to residents in Sioux Lookout (given their poor health profile) than on typical Ontario residents. Even if high rates of contact with physicians might not lead to "better" health, it seems reasonable to provide people in poorer health with more physician resources than people in better health. Our analyses make it clear that it is both opportune and possible for the medical associations and the provinces to recognize factors that influence a population's need for physicians and to use this information to develop policies aimed at areas in which there is a clear shortage or surplus in the physician supply.
The Manitoba project was undertaken as part of a 5-year contract between the University of Manitoba and Manitoba Health to establish the Manitoba Centre for Health Policy and Evaluation and by a grant from the Medical Research Council of Canada (HEALNet). Dr. Roos is a Career Scientist with the National Health Research and Development Program and an Associate of the Canadian Institute for Advanced Research.
References
- National Ad Hoc Working Group. Report of the National Ad Hoc Working Group on Physician Resource Planning. Ottawa: Canadian Medical Association; 1995.
- Goodman DC, Fisher ES, Bubolz TA, Mohr JE, Posage JF, Wennberg JE. Benchmarking the US physician workforce: an alternative to needs-based or demand-based planning. JAMA 1996;276:1:811-7.
- The 1991 Canadian general social survey. Ottawa: Statistics Canada; 1994. Cat no 11-612E(8).
- Roos NP, Mustard CA. Variation in health and health care use by socioeconomic status in Winnipeg, Canada: Does the system work well? Yes and no. Milbank Q 1997;75(1):89-111.
- Billings J, Zeitel L, Lukomnik J, Carey TS, Blank AE, Newman L, et al. Impact of socio-economic status on hospital use in New York City. Health Aff 1993;12(1):162-73.
- Poikolaine K, Eskola J. The effect of health services on mortality: decline in death rates from amenable and non-amenable causes in Finland 196981. Lancet 1986;1:199-201.
- Desmeules M, Semenciw R. The impact of medical care on mortality in Canada, 19581988. Can J Public Health 1991;82:209-11.
- Krakauer H, Jacoby I, Millman M, Lukomnik JE. Physician impact on hospital admission and on mortality rates in the Medicare population. Health Serv Res 1996;June:191-211.
- Roos NP, Fransoo R, Bogdanovic B, Friesen D, Frohlich N, Carriere KC, et al. Needs-based planning for Manitoba's generalist physicians. Winnipeg: Manitoba Centre for Health Policy and Evaluation; 1996.
- Roos NP, Havens BJ. Predictors of successful aging: a twelve year study of Manitoba elderly. Am J Public Health 1991;81:63-8.
- Eyles J, Birch S, Chambers S. Fair shares for the zone: allocating health-care resources for the native populations of the Sioux Lookout zone, Northern Ontario. Can Geographer 1994;83(2):134-50.
 |
Send a letter to the editor responding to this article
Envoyez une lettre à la rédaction au sujet de cet article |
Editorial