Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2007 August 4; 335(7613): 218–219.
PMCID: PMC1939779

Expanding the medical workforce

David C Goodman, professor of pediatrics and of community and family medicine

Is unlikely to be cost effective or produce better outcomes for patients

In a policy marked more by controversy than consensus, medical schools in the United States have joined the United Kingdom in embarking upon the largest expansion in the training of doctors in more than a generation. In this week's BMJ, Weiner outlines the rationale for this initiative while questioning the Association of American Medical Colleges' evidence to support such an expansion.1 Given the differences between the US and other countries within the Organisation for Economic Cooperation and Development, can the US healthcare system provide any generalisable policy lessons?

The question of whether there is or will be a surplus or shortage of doctors is usually answered by projecting today's healthcare system into the future. Using this method the answer is simple—because of population growth and ageing we need more doctors. However, when we consider how to improve the health and wellbeing of growing populations with limited national funds, a more pertinent question is whether expanding the medical workforce is an evidence based investment.

The most striking feature of the medical workforce is the large difference in per capita supply across countries.1 Equally important is variation within countries such as the US, where the workforce varies two to threefold across different hospital referral regions.2 These large regional differences can help us investigate whether increasing the number of doctors improves health care. Recent studies on this subject had two main findings.

Firstly, analysis of healthcare markets in the US has shown that doctors do not generally settle where healthcare needs are greater—as measured by patient indicators.2 One example is the lack of association between the regional supply of US neonatologists and measures of perinatal risk (such as birth weight and mother's education), despite the availability of detailed perinatal statistics.3 Such misdistribution of doctors also occurs in countries with more centralised workforce planning, including the UK.4

Secondly, epidemiological studies in the US found no association between the number of doctors per person and health outcomes, quality, measures of access, or patient satisfaction.5 6 7 The exceptions to this finding are that a very low supply of doctors can be harmful, and higher rates of primary care provision are generally associated with higher quality and more efficient healthcare organisations.5 8 9 10 Having more medical specialists per person resulted in higher volumes of medical care and greater costs, with little difference in patient outcomes.6 7

How can a greater supply of doctors have so little affect on the wellbeing of patients? In regions with higher supply, care is disordered—with higher rates of hospital admission, more specialist referrals, and more diagnostic tests and procedures. Where supply is lower, care is more likely to be dominated by tightly integrated doctor-hospital healthcare systems that can prospectively plan staffing and capital investments.

As the total number of doctors grows, their distribution will probably be shaped by the same forces that led to today's pattern of regional variation. From 1979 to 1999, the number of US doctors per person grew by 51%.11 Three out of four new doctors settled in regions that already had a high workforce level. Adding more doctors to systems of care that are already inefficient (higher cost, no better quality) is a poor investment of resources.

If the benefits of enlarging medical schools are unclear, the cost side is robust. Training doctors is a long and expensive process. Weiner suggests that the additional US public sector investment would be $3-5bn (£1.5-2.5bn; €2.2-3.7bn) each year, but this is probably a conservative estimate. Doctors are also the most highly paid members of the healthcare team. In addition to the increase in salaries, the supply of doctors greatly affects regional spending on inpatient care, time in the intensive care unit, diagnostic tests, and procedures.6 7

Most countries assume that increased doctor training rates will improve health care, but evidence indicates that national healthcare funds are better spent on directly improving the delivery of care or promoting evidence based care. The Cochrane Collaboration has identified many examples of effective medical interventions that are incompletely implemented and many instances of ineffective care. Another promising investment is to improve the organisation of care.

Some healthcare systems in the US have already achieved high standards of patient care with a low number of doctors and low costs. The Mayo Clinic in Minnesota is one of many such examples.12 The result of larger doctor training programmes will probably be disappointing. Implementing what we know can help patients may be a more sound and evidence based investment in our national healthcare systems.


Competing interests: None declared.

Provenance and peer review: Commissioned; not peer reviewed.


1. Weiner J. Expanding the US medical workforce: global perspectives and parallels. BMJ 2007. doi: 10.1136/bmj.39246.598345.94
2. Center for the Evaluative Clinical Sciences. Dartmouth Atlas of Health Care. Dartmouth College, 2007.
3. Goodman D, Fisher E, Little G, Stukel T, Chang C. Are neonatal intensive care resources located where need is greatest? Regional variation in neonatologists, beds, and low birth weight newborns. Pediatrics 2001;108:426-31. [PubMed]
4. Gravelle H, Sutton M. Inequality in the geographical distribution of general practitioners in England and Wales 1974-1995. J Health Serv Res Policy 2001;6:6-13. [PubMed]
5. Goodman D, Fisher E, Little G, Stukel T, Chang C, Schoendorf K. The relation between the availability of neonatal intensive care and neonatal mortality. N Engl J Med 2002;346:1538-44. [PubMed]
6. Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending. Part 1: the content, quality, and accessibility of care. Ann Intern Med 2003;138:273-87. [PubMed]
7. Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending. Part 2: health outcomes and satisfaction with care. Ann Intern Med 2003;138:288-98. [PubMed]
8. Krakauer H, Jacoby I, Millman M, Lukomnik J. Physician impact on hospital admission and on mortality rates in the Medicare population. Health Serv Res 1996;31:191-211. [PMC free article] [PubMed]
9. Gulliford MC. Availability of primary care doctors and population health in England: is there an association? J Public Health Med 2002;24:252-4. [PubMed]
10. Macinko J, Starfield B, Shi L. The contribution of primary care systems to health outcomes within Organization for Economic Cooperation and Development (OECD) countries, 1970-1998. Health Serv Res 2003;38:831-65. [PMC free article] [PubMed]
11. Goodman DC. Trends: twenty-year trends in regional variations in the US physician workforce. Health Aff (Millwood) 2004;(suppl web exclusives):VAR90-7.
12. Goodman DC, Stukel TA, Chang CH, Wennberg JE. End-of-life care at academic medical centers: implications for future workforce requirements. Health Aff (Millwood) 2006;25:521-31. [PubMed]

Articles from The BMJ are provided here courtesy of BMJ Group