The Eye of the Beholder.

Abstract

The paper by Kuo et al.1 uses Medicare billing records data to examine differences in clinical performance between nurse practitioners (NPs) and primary care physicians (PCPs). Medicare billing and claims data are known to be wide but not deep. The first challenge is determining who provides primary care. The attribution appears to rest on having two or more evaluation and management visits from one type of practitioner and none from the other. Such a definition does a weak job of identifying meaningful primary care, but it segregates the care according to provider type. The study focuses on individuals with a diagnosis of diabetes mellitus who were taking diabetes mellitus medications. Using these criteria, they found that nearly 25% of those in the study were being cared for exclusively by NPs. In a separate article, these authors noted that the percentage of Medicare beneficiaries having an NP as their primary care provider increased from 0.2% in 1998 to 2.9% in 2010.2 The discrepancy between that growth pattern and the sample in this study suggests that NPs are playing a substantial role in chronic disease care. Not surprisingly, the use of NPs is higher in rural areas and states that have supportive legislation. Even after balancing the groups, individuals seeing NPs had fewer comorbidities, but NPs are more likely to provide primary care services to individuals who have social determinants of health that influence processes of care and increase the complexity-of-care decisions.3 Using Medicare claims data provides information on a large group of Medicare recipients with links to International Classification of Diseases, Ninth Revision, diagnostic codes and procedures, but the claims data do not include all NPs who are caring for Medicare beneficiaries. The efforts to distinctly identify “pure” NP and PCP cases inevitably obscure the areas of collaboration, which is an important element of comprehensive care. In many instances, NPs may have provided services, and PCP billing codes were used under “incident to” billing mechanisms. The authors do not identify whether they were able to discriminate “incident to” billing from direct service billing for the PCP codes, thereby not indicating that the PCP billing codes may have been for services that NPs provided. Statistics can sometimes mislead. When examining the results of this study, it is important to keep in mind that, with a sample size of 25,000, many small and clinically unimportant differences can be statistically significant. Moreover, odds ratios can be hard to interpret without examining the actual base rates. So what are the differences? Individuals who saw NPs were more likely to see specialists. There was a small difference in medication adherence rates (means of 74% for NPs and 75% for PCPs). The use of inappropriate medications differed by less than 1 percentage point. NPs did an eye examination 3 percentage points less often than PCPs. Both rates were approximately 60%, which seems quite high for repeat visits in a year. Lipid tests were ordered at about the same rate; again 85% seems high. Glycosylated hemoglobin test rates differed by less than 1 percentage point (~93% for both groups). Nephropathy monitoring tests again differed by less than 1 percentage point (both ~85%). All of these data compare favorably with national performance standards. Adjusted Medicare spending was basically the same. NP charges were lower than for PCPs but were offset by specialist costs. Here again, the differences are modest. The biggest differences are for outpatient facility charges, which reflect the work location of the clinicians. Here is a case of cup perspective. The authors use these findings to suggest that the quality of care that NPs provide is less than that of physicians, but the performance seems similar. Moreover, the analysis does not take into account the effects of other important geriatric factors, such as physical function, frailty, cognitive status, and social determinants of health, all of which can limit the ability to interpret quality of care differences. At the very time when society faces a boom in chronic disease, there is a bust in primary care. NPs represent an important resource to help address this imbalance. These results offer a basis for optimism. It would be good to see more teaming between NPs, PCPs, and other health professionals, but it seems clear that NPs bring something of value to the table. This observation comes as no surprise to geriatricians who have worked successfully with NPs in delivering long-term care and complex primary care services to older adults in various settings.4, 5 These two types of clinicians seem to be working in different settings. There is reason to believe that environment influences clinician performance. A study of nurses long ago showed that environment was a big predictor of performance.6 A more-recent randomized control trial of NPs and physicians had a similar finding.7 Why continue to study differences only between two types of providers? Other factors are salient. Processes and outcomes of care for older adults with diabetes mellitus need to be examined for quality and individualized outcomes in a world of coordinated care, accountable care organizations, transitional care services, and value-based payment, approaching care from the perspective of those who live with diabetes mellitus and their families to elaborate on what is known about diabetes mellitus care—that personal engagement and use of teams including nutritionists leads to better outcomes. The concept of primary care remains elusive, and efforts to use the best-available administrative data to identify processes and outcomes of primary care have been frustrating, particularly so for older adults with complex chronic illnesses. Too often, specialists deliver the bulk of care to many older people and may not be able to address the range of concerns encompassed in the underlying concept of coordinated, continuous, comprehensive primary care. In many instances, NPs work with specialists to fill some of those gaps. At a time when it is a struggle to fill the primary care gap, NPs should be viewed as valued partners. Kuo et al. have embraced the challenge to ascertain answers to these important questions using available, if incomplete data, leaving us to recognize that better data and better measures are needed to understand how to achieve the best outcomes. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Author Contributions: Dr. Robert Kane drafted the work; Eileen M. Sullivan-Marx edited the material and made contributions. Sponsor's Role: No sponsor.