PA and MD comparison of hospitalized patients in the Netherlands
Medical unit care has been increasingly reallocated from physicians to physician assistants. Insight into these PAs' roles and tasks is limited. This cross-sectional descriptive study of 34 hospital units provides insight into different organizational models of medical unit care, focusing on the position, tasks, and responsibilities of the involved PAs and MDs. Characteristics of the organizational models were collected from the heads of departments. MDs and PAs in charge for medical unit care (n=179) were asked to complete a questionnaire to measure workload, supervision, and tasks performed. From the data, four organizational models for unit care were identified: medical specialists in charge of admitted patients (100% specialist), medical residents in charge (100% residents), PAs in charge (100% PA), and both residents and PAs in charge (mixed PA/resident). The units with PAs had the highest provider continuity. PAs spent relatively more time on direct patient care; MDs spent relatively more time on indirect patient care. PAs spent more hours on quality projects (P=0); MDs spent more time on scientific research (P=0.03). Across different organizational models for medical unit care, we found variations in time per task, time per bed, and provider continuity.1
Commentary by Roderick S. Hooker: At the heart of labor economics is human capital and the role of individuals in the employment market. One area of growing interest in PA deployment is hospital-based care. In North America, hospitalists are increasing and taking over the role of outpatient-based physicians who traditionally followed their patients into the hospital. At first this was for general hospital care, but now the role is more specialized, including hospital-based obstetricians, pediatricians, and neurologists. The authors have organized a way of examining how well PAs substitute for traditional physician labor on 34 hospital units in the Netherlands. Variables of interest were types of providers, comorbidities of patients, workweek, and patient characteristics. PAs appear to spend more time on direct patient care than the other three models of unit care. Whether this had any effect on outcomes remains to be seen. Further research should focus on the effect of these differences on outcomes and efficiency of medical unit care.
Timmermans MJ, van Vught AJ, Van den Berg M, et al. Physician assistants in medical ward care: a descriptive study of the situation in the Netherlands. J Eval Clin Pract. [e-pub Dec. 23, 2015]
PAs and NPs as procedure service providers
Nonvascular invasive radiology procedures commonly performed by PAs and NPs at two large hospitals were used to identify procedure groups for national trends. Between 1994 and 2012, national Medicare claims by PAs and NPs increased dramatically for nonvascular invasive radiology procedures, including paracentesis from 0 to 17,967; thoracentesis from 119 to 4,141; fine needle aspiration from 0 to 3,921; superficial lymph node biopsy from 0 to 251; abdominal biopsy from 1 to 1,819; thoracic biopsy from 0 to 552; and abdominal drainage from 37 to 410. Although PAs and NPs perform a relatively small portion of commonly performed invasive radiology procedures nationally, paid Medicare claims for those services have increased dramatically over 2 decades, and at a faster pace than that for all providers as a whole. Given the multiple hurdles involved in obtaining Medicare reimbursement, such growth indicates increasing acceptance of PAs and NPs as procedure service providers at the institutional credentialing, state licensure, and payer policy levels.1
Commentary by Joanne Spetz: The fact that PAs and NPs are performing more invasive radiology procedures (as measured by Medicare billing data) is yet another piece of evidence that task transfer is occurring from physicians to PAs and NPs as a means to increase efficiency in healthcare delivery.2 In fact, this paper almost certainly understates the extent of this trend. Services performed by PAs and NPs can be billed to Medicare under their own provider identifiers or under a physician's identifier as “incident-to” services. Perceived financial incentives or legacy practices lead some practices to bill as incident-to, thus undercounting PA and NP services. Even ignoring this likely undercount, a radical power shift clearly is underway in the American healthcare system, from physician-dominated care to care delivered in teams.
1. Duszak R Jr, Walls DG, Wang JM, et al. Expanding roles of nurse practitioners and physician assistants as providers of nonvascular invasive radiology procedures. J Am Coll Radiol. 2015;12(3):284–289.
2. Sibbald B, Shen J, McBride A. Changing the skill-mix of the health care workforce. J Health Serv Res Policy. 2004;9(suppl 1):28–38.
NPs managing older patients with diabetes
The authors compared processes and cost of care of older adults with diabetes by NPs with physicians from a national sample of Medicare beneficiaries (n=64,354) using propensity score matching. Both had similar rates of performing low-density lipoprotein cholesterol (LDL-C) testing and nephropathy monitoring, but NPs had lower rates of performing eye examinations and A1C testing. NPs were more likely to have consulted cardiologists, endocrinologists, and nephrologists. Both NPs and physicians prescribed potentially inappropriate medications and medications with potential interactions although the rate was slightly higher among NPs. No statistically significant difference in adjusted Medicare spending was found between the two groups. In the aggregate NPs were similar to primary care physicians (PCPs) or slightly lower in their rates of diabetes guideline-concordant care. NPs used specialist consultations more often but had similar overall costs of care as PCPs. Studies are needed to evaluate the cost-benefit of resource use in diabetes management by the two types of providers including specialist consultations.1
Commentary by Mark E. Archambault: Studies that NPs are able to provide similar quality of diabetes care at similar costs when compared with PCPs are not new.2,3 However, this study is robust in its methodology in that it uses propensity score matching as a means to mitigate the inherent bias of a nonrandomized, observational study. Typical experimental design seeks to identify, measure, and statistically adjust for variances in confounding variables between the experimental and control groups. Propensity score matching breaks from this methodology by sampling from a large population of potential controls and subsequently selects a control group that is similarly matched in terms of distribution of confounding variables.4 That is, statistical methods are used to determine the proper control group rather than adjust for variances between groups. This methodology improves the quality of research and examines NP care of patients with diabetes through a more accurate lens. The findings may also clarify NP professional intentions: it appears NP propensity is for autonomy in their interdependent relationships with physicians and not necessarily in just seeking a practice stance independent of physicians.
1. Kuo YF, Goodwin JS, Chen NW, et al. Diabetes mellitus care provided by nurse practitioners vs primary care physicians. J Am Geriatr Soc. 2015;63(10):1980–1988.
2. Mundinger MO, Kane RL, Lenz ER, et al. Primary care outcomes in patients treated by nurse practitioners or physicians: a randomized trial. JAMA. 2000;283(1):59–68.
3. Jackson GL, Lee SY, Edelman D, et al. Employment of mid-level providers in primary care and control of diabetes. Prim Care Diabetes. 2011;5(1):25–31.
4. Rosenbaum P, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70:41–55.