The COMPACCS Study
Questions Left Unanswered

RICHARD A. COOPER

Health Policy Institute, Medical College of Wisconsin, Milwaukee, Wisconsin

	ARTICLE

TOP ARTICLE REFERENCES

The Committee on Manpower for Pulmonary and Critical Care Societies (COMPACCS) recently published the report of a workforce study performed on its behalf by Abt Associates warning of progressive shortages in these specialties over the next 30 years (1). Given earlier predictions that specialists would soon be in short supply (2, 3), there is reason to take heed. However, in my view, the inherent methodological weaknesses of this study prevent it from answering the core questions facing pulmonary and critical care.

The COMPACCS study used the "time and task" approach to workforce analysis originally developed by the Committee on the Costs of Medical Care in the 1920s (4) and adopted fifty years later by the Graduate Medical Education National Advisory Committee (GMENAC) (5). Abt first used it in its 1991 update of the GMENAC report (6), and variations on it were subsequently employed by the Bureau of Health Professions (BHPr) (7) and others (8, 9) and incorporated into the recommendations of the Committee on Graduate Medical Education (COGME) (10). Unfortunately, these various attempts to project the size and composition of the physician workforce largely failed (2, 11). Indeed, COMPACCS characterizes the current disconnect between a need for more specialists and an absence of planning as a "public policy paradox" which it sees as "the legacy of a decade-long debate whose premise was that the US had an oversupply of physicians in general and specialists in particular" (1). What COMPACCS doesn't note is that it was the "time and task" methodology that gave us this legacy.

The basic premise of "time and task" studies is that one can measure all of the tasks performed by physicians and assign times to each. By also estimating the amount of time that physicians work, it should be possible to calculate how many physicians are needed. Future projections require a parallel set of assumptions, like what tasks will be performed, how long will it take to perform them, and how productive will physicians be, specifics that dim across the haze of 30 years.

For the COMPACCS study, the tasks that were analyzed were ICU days, pulmonary inpatient days and pulmonary outpatient visits. Coupled with data and expert opinion about the time required for each and the hours that specialists work, it was simply a matter of algebra. However, it's not so easy. First, the data sources are neither strong enough nor sufficiently concordant to permit this level of quantitative analysis. Second, the practice characteristics of the specialists who were studied are too elusive. For example, intensivists spend less than 40% of their time in ICUs, sometimes working as principal physicians but often as consultants, and they provide only one-third of the ICU care. Indeed, fewer than half of ICUs even have intensivists. Similarly, pulmonologists spend only 60% of their time involved in the defined tasks, and 90% of ambulatory pulmonary care is provided by nonpulmonologists. And then there's the growing presence of hospitalists and of critical care nurse practitioners. So, the specialists who were analyzed relate only partially to the tasks that were quantitated, and the tasks that were quantitated depend only partially on the specialists who were analyzed. Clearly, there is more to this than simple algebra. In fact, it is fair to ask whether it is even possible to construct valid estimates of current need within this methodological framework.

Projections of future demand are even more precarious. For this, the COMPACCS study focused on the disproportionate use of services by the elderly and the greater frequency of ICU use among patients under managed care. By estimating the growth of each, it was reasoned, future demand could be discerned. But why should it be assumed that the elderly will receive more services just because there are more of them if there are so many who now receive little? In fact, the volume of services follows economic patterns, not age patterns. Certainly the elderly require more care, but they receive it not in proportion to how many there are but in proportion to the resources that are available to finance it (12). The relationship to managed care is similar. Why try to make a profit managing few resources? The higher use of ICUs in markets with a lot of managed care is simply a reflection of the fiscal status of those communities. Indeed, our own work has shown strong correlations between the numbers of specialists per capita in metropolitan statistical areas (MSAs) and the per capita incomes of the populations served (13). More prosperous communities demand and receive higher levels of care.

That leaves us in a bit of a quandary. Students, educators and payers all need some guidance. I believe that a more open and objective way to proceed is through an examination of trends (11, 13). The major ones are (1) population growth (since the demand for physicians is sensitive to the size of the population); (2) training and attrition rates (since the former are falling while the latter are rising); (3) physician work-effort (which also is declining as more women enter the workforce and as many physicians choose to practice fewer hours); (4) substitution among disciplines (since few specialists are unique providers of their specialty); and (5) the economy (since it is the principal driver of health care spending). In general, as the economy grows, the utilization of advanced medical specialists grows. Over the past 40 years, the percentage increase in specialist utilization has been approximately equal to the growth of GDP, which, on an inflation-adjusted per capita basis, is expected to increase about 2.0% per year (13). A good rule of thumb is to start with this and to adjust it upward if work effort falls and downward if specialists spend more time in their own specialties or if other providers assume more of their responsibilities.

How does the COMPACCS study mesh with these trends? First, the demand that it projects over the next 10 years is actually less than the growth of the United States population, leading one to question whether the current supply may be too great. Only in the out years past 2020 does demand reflect the needs of an expanding economy. But who will meet those needs? The COMPACCS study omits the consideration of others who might, an issue that is particularly germane because pulmonologists and intensivists provide the minority of services that fall within their arenas of special interest. And, finally, how will the economy influence the quantity and organization of care? Thus, while it is clear that specialists as a group will be in short supply over the coming years (3, 13), the magnitude of this shortfall for the particular specialties of pulmonary and critical care is a question that, in my view, COMPACCS has left largely unanswered.

	Footnotes

Correspondence and requests for reprints should be addressed to Richard A. Cooper, M.D., 8701 Watertown Plank Road, Milwaukee, WI 53226. E-mail: rcooper{at}mcs.edu

	References

TOP ARTICLE REFERENCES

1. Angus DC, Kelley MA, Schmitz RJ, White A, Popovich A Jr.. Current and projected workforce requirements for care of the critically ill and patients with pulmonary disease: can we meet the requirements of an aging population? JAMA 2000; 284: 2762-2770 .

2. Cooper RA. Seeking a balanced physician workforce for the 21st century. JAMA 1994; 272: 680-687 .

3. Cooper RA. Perspectives on the physician workforce to the year 2020.  JAMA 1995; 274: 1534-1543 .

4. Lee RI, Jones LW. The fundamentals of good medical care. Chicago, IL: University of Chicago Press; 1933.

5. Graduate Medical Education National Advisory Committee. Report of the Graduate Medical Education National Advisory Committee to the Secretary, Department of Health and Human Services. Washington DC: US Department of Health and Human Services; 1981. DHHS Publication No. (HRA) 81-652.

6. Abt Associates Inc. Reexamination of the adequacy of physician supply made in 1980 by the Graduate Medical Education National Advisory Committee (GMENAC) for selected specialties. Final report. Washington DC: US Department of Health and Human Services; 1991. DHHS Publication No. (HSRA) 240-89-0041.

7. Politzer RM, Gamliel SR, Cultice JM, Bazell CM, Rivo ML, Mullan F. Matching physician supply and requirements: testing policy recommendations. Inquiry 1996; 33: 1810194 .

8. Lee PP, Jackson CA, Relles DA. Demand-based assessment of workforce requirements for orthopedic services. J Bone Joint Surg 1998;80-A:313-326.

9. Weiner JP. Forecasting the effects of health reform on US physician workforce requirement: evidence from HMO staffing patterns. JAMA 1994; 272: 222-230 .

10. Council on Graduate Medical Education. Recommendations to improve access to health care through physician workforce reform. Washington, DC: US Department of Health and Human Services; 1994.

11. Cooper RA, Goodman DG, Menken M, Salsberg ES, Whitcomb ME. Evaluation of specialty workforce methodologies: Council on Graduate Medical Education. Washington, DC: Health Resources and Services Administration; 2000.

12. Getzen TE. Population aging and the growth of health expenditures. J Gerontology: Social Sciences 1992;47:S98-S104.

13. Cooper RA. Forecasting the physician workforce: Proceedings of the 11th Federal Forecasters Conference 2000. Washington, DC: US Department of Education. (In press)