Can the right mix of generalists and specialists produce the needed fix? For good drinks, getting the mix right is surely critical; not so for the physician specialty mix in health policy, because, as Miriam J. Laugesen argues, this is only one factor among many others typically advanced by critics of our health care system. Here I summarize Laugesen's argument, provide some reactions about primary care, and conclude with the usual call for more research to improve policy.
The Argument in Brief
Laugesen delivers an effective critique to those who subscribe to the conventional wisdom that other countries have a better mix of generalists and specialists. Her most important contribution is to refine the policy issue beyond the content of physician specialty training to the broader functions of primary care (Starfield 1994: 1129): “First-contact, continuous, comprehensive, and coordinated care.” Laugesen's cross-national analysis of physician workforce data, allowing for inclusion of specialists who perform some of these functions, suggests that 37 percent of our physicians provide primary care. Thus, once the definition of primary care is adapted to the US context, the grounds for asserting US exceptionalism crumble because, as in other nations, “primary care physicians remain the port of first call more than half the time.”
Add to this conclusion Laugesen's recognition that nurse practitioners, and even some registered nurses, provide primary care, and we have more evidence that the focus on physician specialty mix is a case of what Alfred North Whitehead long ago called the fallacy of misplaced concreteness (on this see also Lawrence D. Brown's contribution to this special issue). As Laugesen suggests, this narrow focus tends to ignore the role of other problems with our health care system: perverse financial incentives, the quest for providing the most technologically intensive services, and prevailing organizational deficiencies leading to lack of coordination across an increasingly specialized system.
I wish that Laugesen had suggested some promising directions for the future of primary care, holding constant the physician specialty mix, for example, the growing use of telemedicine and improvements in integrated medical information systems. I would emphasize three issues that probably affect the delivery of primary care more than physician specialty mix.
First, organizational arrangements are critical in assuring that patients have a “medical home” as the first point of entry. But views differ about the importance of maintaining independent office-based practices, community health centers, or retail medical clinics rather than folding all primary care into integrated health care systems managed by hospitals. No doubt, the relative strengths of these different models depend, in large part, on the quality of information systems and telemedicine, and this reinforces the notion that organization matters.
Second, location is a powerful factor in shaping the delivery of primary care—urban versus rural and, within urban areas, differences among poorer and wealthier neighborhoods. Since the number of primary care doctors is projected to decrease from 2010 to 2025 while the number of nurse practitioners and registered nurses entering the workforce will increase dramatically (Bodenheimer and Bauer 2016), all primary care services, in the future, will necessarily be organized around whatever workforce is available. Different localities, however, will require different strategies that extend far beyond physician specialty mix.
Third, in thinking about the needed fix for primary care, I wonder to what extent discussions about physician specialty mix, and even more broadly, the adequacy of primary care, reflect deeper issues about the inadequacy of social care. Laugesen flags this issue in noting the importance of social care in Germany and the Netherlands. As suggested by the research of Bradley et al. (2011), it may be that greater investments in income maintenance programs and social services contribute far more to effective primary care and population health than all of our spending on health care services.
Although true believers in the virtues of primary care medical homes point to evidence that they can raise service quality, improve health, and contain the growth of health care costs (Jabbarpour et al. 2017), qualifications about differing contexts and variations in findings suggest the need for more research.
Hansen et al. (2015) found that, although primary care is beneficial to people with chronic conditions, there is no evidence that having a well-developed primary care system reduces inequities in health care across groups stratified by years of education. Another approach to the evaluation of primary care access and quality is to assess residence-based hospital admissions for so-called ambulatory-care sensitive conditions. This indicator has often been used to assess primary care in the United States (Chang et al. 2011) but has rarely been used in comparisons of Organisation for Economic Co-operation and Development nations and cities (Gusmano, Rodwin, and Weisz 2006, 2014).
Given the great diversity of models, worldwide and within the United States, that claim to deliver good primary care, it would be useful to know how they perform along a range of important indicators. Instead of hearing slogans about high-performing health systems or case studies on systems that identify themselves in this way, I would like to find evidence about the performance of primary care among Organisation for Economic Co-operation and Development nations, as well as among such health care systems as Intermountain, Geisinger, Mayo, and Kaiser Permanente within the United States.