Expanded Advanced Practice Registered Nurse Scope of Practice Laws: Can We Ignore Economic Motives and Effects?
Feb 24, 2012, 1:00 PM, Posted by Mark Pauly
By Mark V. Pauly, PhD, Bendheim Professor of Health Care Management, Wharton School, University of Pennsylvania and co-director, Robert Wood Johnson Foundation Interdisciplinary Nursing Quality Research Initiative (INQRI)
The recent and useful article by Patricia Pittman and Benjamin Williams, “Physician Wages in States with Expanded APRN Scope of Practice,” examined annual earnings of salaried physicians in states with and without expanded Scope of Practice (SOP) laws. It found that there is no smoking gun to support the hypothesis that laws which permit registered nurses to do more result in lower primary care physician salaries. While there may still be reasons why physicians and others may be skeptical of such laws, and whatever suspicions people may harbor, proven financial harm to physician salaries would not appear to be one of them; primary care physician salary levels were not different, either absolutely or relative to surgeon salaries, in states with expansion laws.
In my view, these results definitely help in the debate about expanded SOP rules, potentially moving it to a different (and probably more appropriate) level where it is about quality and access, not about economics. However, the results are, as the authors note, far from definitive as proof that no causal impact exists. In addition, they raise more questions than they answer about how primary care workforce markets actually function. Paradoxically, if there are no financial impacts, there may be no cost savings either. Here I first lay out the questions which these results raise for understanding how markets are working, and then specify some empirical issues that would need to be addressed for more definitive empirical conclusions.
A simple economic model that suggests an impact of SOP laws on physician incomes is one that assumes that primary care physicians and advanced practice registered nurses (APRNs) potentially provide substitute sources of supply for primary care. The simplest story would be a “fixed patient, workforce, and revenue” model. This model assumes that there is a fixed number of primary care services to be rendered in a market, a fixed amount of revenue those services will generate, and a fixed number of primary care physicians and APRNs. Then by definition a larger share of total services supplied by APRNs in expanded SOP markets means fewer services, and a smaller share of the revenue pie, per primary care physician. In this first simple story there is no effect on aggregate spending. Any effects one would care about, such as effects on health or consumer satisfaction, would come from potentially differential quality of—or easier access to—services provided by APNs relative to physicians.
Even this simple model shows one possible alternative interpretation to the “no wage effect” conclusion: if expanded SOP causes some primary care physicians to leave the market, possibly in anticipation of lower earnings, the average income for the remaining physicians could be unaffected. So a full analysis would want to control for any effects of SOP laws on the supply of primary care physicians and, for that matter, on the supply of APRNs.
These considerations also hint at how expanded SOP could be a win-win-tie result. If there was excess demand for primary care services in states with limited SOP laws, expanding SOP might allow the total volume of services and total spending both to increase (contrary to the simple model), potentially making patients better off in terms of access, APRNs better off in terms of income, and primary care physicians to be no worse off. Only insurers lose, but if increased primary care has high enough value, society gains.
As noted in the article, however, some have advocated expanded SOP laws as a way of reducing insured medical spending on primary care, imagining that APRNs will render similar services at lower prices. In this alternative model, even if physicians maintained their prices at the same level, the lower APRN fees might increase their share of the market (perhaps as preferred provider organizations channel patients their way), thus reducing the physician share and their net incomes. Since this did not occur in the data, one must conclude that there is also no evidence here for cost containment effects arising directly from competitive pressures, a result which may be disappointing to some. The data in this study only measure total physician salaries, not unit prices, however, so we cannot really tell what is or is not happening. Given the prevalence of insurance and the desire of APRNs to be paid at the same rate as physicians for the same services (re-enforced by state laws and advisory bodies), the absence of a competitive effect may be no surprise, but more direct evidence on this point is needed.
In terms of the empirical work, there is clearly a potential problem with measuring physician financial interest by the incomes of salaried physicians. It is true that they are a large and increasing share of the total physician workforce. However, the net earnings of self-employed physicians and earnings of partners or owners in group practices with salaried physicians may have been adversely affected, and that does not show up in this measure. Indeed, if salaried physicians are thought to be mobile across states, one would expect their incomes to be equalized regardless of peculiar state laws, so that differences in aggregate physicians income would be concentrated in what happens to the net incomes of self-employed, partner, or owner physicians. The authors have done the best they can with the best data available, but it still leaves some unanswered questions.
The other empirical question is how to disentangle any effect on physician income of variations in SOP rules across states (which one expects to be relatively modest) from other influences. The authors attempt to control for such differences by using salaried surgeon income as a control group, imagining that the effects of other unmeasured potential confounders would be the same on surgeon incomes as on primary care physician incomes. This was reasonable given the data available, although one might worry that things like malpractice premiums, insurance coverage, and consumer income might have different effects across specialties and could be correlated with the dependent variable.
There is a potentially better alternative strategy that might be employed, should the data permit it, to provide unbiased estimates of the causal effects of SOP laws. This is what is called a “difference in difference” approach, and is now very popular as an econometric technique in health economics. The idea would be to identify states that change their SOP laws, see what happens in those states to the dependent variable (average primary care physician wages) before and after the laws, and compare that change to what happened over the same time period in states that did not change their laws (that either stuck with limited SOP laws or had long since changed their laws). This requires time series data, and presumably endogenous state legislatures to generate sufficient natural experiments of this type for us to observe, but some version of it might be possible. (There is an alternative version that would compare changes in primary care incomes in states that changed laws with changes in surgeon incomes in those states.)
There is an additional finding in the paper that the rate of change in salaries in states with expanded SOP laws over the period 1999 to 2009 was not significantly different from that in other states or from surgeon incomes in the same states. The bad luck here is that apparently no states changed their laws during this period. Hence, the failure to find any difference, even accounting for the somewhat contaminated data, is at least consistent with the ‘no smoking gun’ hypothesis only if we think that any response to laws passed in the early 1990s was having a lagged effect.
To sum up: the evidence is circumstantial rather than definitive, but it does not appear that any obvious and serious economic mayhem (for good or ill) has occurred. I am pleased that, for once, we may be able to keep money out of what should be an important debate about productivity, quality, and consumer satisfaction.
Read the article, “Physician Wages in States with Expanded APRN Scope of Practice.”
Read the Institute of Medicine report, “The Future of Nursing: Leading Change, Advancing Health.”
INQRI supports interdisciplinary teams of nurse scholars and scholars from other disciplines to address the gaps in knowledge about the relationship between nursing and health care quality. It is helping to advance the recommendations of the Institute of Medicine’s landmark report, The Future of Nursing: Leading Change, Advancing Health, which include fostering interprofessional collaboration and preparing and enabling nurses to lead change. By requiring research teams to include a nurse scholar and at least one scholar from another health care discipline, INQRI not only fosters interprofessional collaboration, the Initiative also ensures that diverse perspectives are brought to bear in research. Learn more about INQRI.
This commentary originally appeared on the RWJF Human Capital Blog. The views and opinions expressed here are those of the authors.