Perspectives of Physicians and Nurse Practitioners on Primary Care Practice

Transcription

1 T h e n e w e ngl a nd j o u r na l o f m e dic i n e special article Perspectives of Physicians and Nurse Practitioners on Primary Care Practice Karen Donelan, Sc.D., Catherine M. DesRoches, Dr.P.H., Robert S. Dittus, M.D., M.P.H., and Peter Buerhaus, R.N., Ph.D. A bs tr ac t From Mongan Institute for Health Policy, Massachusetts General Hospital, and Harvard Medical School both in Boston (K.D.); Mathematica Policy Research, Cambridge, MA (C.M.D.); and the Department of Medicine, Institute for Medicine and Public Health, Vanderbilt University, and VA Tennessee Valley Geriatric Research, Education, and Clinical Center (R.S.D.), and the Center for Interdisciplinary Health Workforce Studies, Vanderbilt University Medical Center (P.B.) all in Nashville. Address reprint requests to Dr. Donelan at Mongan Institute for Health Policy, Massachusetts General Hospital, 50 Staniford St., 9th Fl., Boston, MA 02114, or at kdonelan@partners.org. N Engl J Med 2013;368: DOI: /NEJMsa Copyright 2013 Massachusetts Medical Society. Background The U.S. health care system is at a critical juncture in health care workforce planning. The nation has a shortage of primary care physicians. Policy analysts have proposed expanding the supply and scope of practice of nurse practitioners to address increased demand for primary care providers. These proposals are controversial. Methods From November 23, 2011, to April 9, 2012, we conducted a national postal-mail survey of 972 clinicians (505 physicians and 467 nurse practitioners) in primary care practice. Questionnaire domains included scope of work, practice characteristics, and attitudes about the effect of expanding the role of nurse practitioners in primary care. The response rate was 61.2%. Results Physicians reported working longer hours, seeing more patients, and earning higher incomes than did nurse practitioners. A total of 80.9% of nurse practitioners reported working in a practice with a physician, as compared with 41.4% of physicians who reported working with a nurse practitioner. Nurse practitioners were more likely than physicians to believe that they should lead medical homes, be allowed hospital admitting privileges, and be paid equally for the same clinical services. When asked whether they agreed with the statement that physicians provide a higher-quality examination and consultation than do nurse practitioners during the same type of primary care visit, 66.1% of physicians agreed and 75.3% of nurse practitioners disagreed. Conclusions Current policy recommendations that are aimed at expanding the supply and scope of practice of primary care nurse practitioners are controversial. Physicians and nurse practitioners do not agree about their respective roles in the delivery of primary care. (Funded by the Gordon and Betty Moore Foundation and others.) 1898

2 The U.S. health care system is at a critical juncture in health care workforce planning. The nation has an acknowledged shortage of primary care physicians at a time when the population is aging and the incidence and prevalence of chronic illnesses are increasing. The implementation of the Affordable Care Act will provide millions of previously uninsured Americans with the means to purchase health insurance and access health care services. 1-3 The combination of increased demand and provider shortages has led policymakers to consider increasing the supply of nurse practitioners and broadening their roles in the provision of primary care. 4,5 Nurse practitioners emerged in the 1960s, another period of projected physician shortages. Medicare and Medicaid programs were expanding access to health services, and clinicians were needed to meet the increased demand, especially in underserved areas. Nurse practitioners have since been progressively adopted into the health care workforce and reportedly numbered 180,233 in 2011; of these nurse practitioners, an estimated 30 to 35% worked in primary care. 6 Proposals that focus on the potential for nurse practitioners to help meet current and expected future gaps in the supply of primary care providers have met with wide interest and considerable controversy. 6-8 At the core of the controversy is whether nurse practitioners have the education and experience to provide highquality services and lead clinical practices without supervision by a physician. In 2010, the Institute of Medicine published The Future of Nursing: Leading Change, Advancing Health, which offered an intensive examination of what the nursing profession is now and should become. 4 The authors, a consensus committee, stated that advanced practice registered nurses should be able to practice to the full extent of their education and training. The report proposed that nurse practitioners should be able to admit patients to hospitals or hospices, lead medical teams and medical homes, and be reimbursed at the same rate as physicians for providing the same services. In a response, the Council of Medical Specialty Societies, representing 34 physician organizations, and the American Academy of Family Physicians strongly opposed broadening the scope of practice of nurse practitioners. 9,10 Such debates are not new and have been played out in many states. Yet, few national data are available on the roles that nurse practitioners play in independent or collaborative primary care practices and how they complement or differ with the practice of primary care physicians. We sought to inform the debate about the roles of nurse practitioners and physicians in private practice by conducting a national survey. Our focus was the role of nurse practitioners in primary care and the likely effects on the health care system of expanding the supply of nurse practitioners and the scope of their practice. Me thods Study Design From November 23, 2011, to April 9, 2012, we conducted the National Survey of Primary Care Nurse Practitioners and Physicians, a postal-mail survey involving 972 clinicians (467 nurse practitioners and 505 physicians). Harris Interactive managed the data collection on our behalf. (The survey instruments are provided in the Supplementary Appendix, available with the full text of this article at NEJM.org.) We defined as eligible for the survey clinicians who were licensed nurse practitioners or physicians, had been trained in a primary care specialty, were actively working in primary care practice, and were providing direct patient care. Samples We obtained samples of nurse practitioners and physicians from the Nurse Practitioner Masterfile and the American Medical Association (AMA) Masterfile, through the Medical Marketing Service. (Sample disposition and response-rate calculations are shown in Table S2 in the Supplementary Appendix.) Physicians were randomly selected from the AMA Masterfile, a comprehensive listing of all licensed physicians in the United States. We selected physicians providing direct patient care in eligible specialties (general practice, family practice, internal medicine, general internal medicine, adolescent medicine, internal medicine pediatrics, pediatrics, and geriatric medicine). Nurse practitioners were randomly selected from the Nurse Practitioner Masterfile, a list that included 165,101 state-licensed nurse practitioners in 2011 (92% of nurse practitioners in the United States that year). 6 We selected nurse prac- 1899

3 T h e n e w e ngl a nd j o u r na l o f m e dic i n e titioners in specialties that were consistent with physician specialties (adolescent medicine, adult medicine, family medicine, general practice, geriatric medicine, internal medicine, pediatrics, and women s health). The nurse-practitioner sample file did not contain a variable for nurse practitioners in a practice providing direct patient care. Our initial sample included 1914 clinicians, 957 each of nurse practitioners and physicians in primary care specialties. Using standards for response-rate calculation and reporting developed by the American Association for Public Opinion Research 11 (Table S2 in the Supplementary Appendix), we determined that our response rate was 61.2%. Instrument Development Our research team developed the survey instrument, drawing on multiple health care workforce surveys published previously by our team and using expert review and pretesting of measures (see the Supplementary Appendix). Domains included scope of work, perceptions of labor supply and nurse-practitioner practice, and personal and clinical-practice characteristics. Data Collection We conducted four waves of mail contact as part of the survey. The first mailing was sent by priority mail and included a cover letter, a questionnaire, a $35 incentive check, and a postage-paid return envelope. Subsequent mailings were sent by first-class mail; the fourth mailing included a $60 prepaid incentive check, since the previous incentive checks had expired. Weighting Among physicians, respondents differed from nonrespondents according to number of years in practice, sex, and region. We created weights to adjust for this differential response, with a weighting range of to Among nurse practitioners, we created weights for sex and region alone, since we did not have a years in practice variable in the nurse-practitioner sample file. Weights for nurse practitioners ranged from to Statistical Analysis We used data from the entire sample of 505 physicians and 467 nurse practitioners for our analyses of all attitudinal measures and personal and clinical-practice characteristics. We used data from the subgroup of 209 physicians and 378 nurse practitioners who reported working in practices with both physicians and nurse practitioners for our analyses of clinical activities and attitudinal measures. The sampling error for the entire sample was ±3.1%. For the sample of clinicians who worked in collaborative practices, the sampling error was ±4.0. The primary focus of our analyses was on the attitudes and experiences of physicians and nurse practitioners in primary care settings. We examined univariate and bivariate relationships, comparing physicians and nurse practitioners with the use of two-sample t-tests for continuous variables and chi-square tests for categorical variables on measures that were posed to the two groups. We also examined differences within each professional group and between the two groups with respect to age, sex, collaborative practice, and region. We hypothesized that physicians and nurse practitioners who worked in collaborative practices would be more similar in their responses than those who did not work in collaborative practices. We tested all outcomes for these relationships and report all those that were significant. A P value of less than 0.05 was considered to indicate statistical significance. R esult s Characteristics of the Respondents In primary care settings, physicians and nurse practitioners differed significantly with respect to several personal and clinical-practice characteristics (Table 1). Nurse practitioners were significantly more likely than physicians to be female; they were also older and had fewer years of work experience, on average, and were also less likely to identify themselves as a member of an underrepresented racial or ethnic minority. On average, nurse practitioners worked fewer hours, saw fewer patients, and earned lower incomes than did physicians. Nurse practitioners were nearly twice as likely to report working in collaborative practice (80.9% of nurse practitioners reported working with a physician vs. 41.4% of physicians who reported working with a nurse practitioner). Scope of Practice for Nurse Practitioners Of the nurse practitioners who were surveyed, 74.9% indicated that they believed they were cur- 1900

4 Table 1. Characteristics of the Respondents and Their Practices.* Characteristic Nurse Practitioners (N = 467) number (percent) Physicians (N = 505) P Value Respondents Female sex no. (%) 432 (92.5) 274 (54.3) <0.001 Race or ethnic group no. (%) Hispanic or Latino 19 (4.1) 41 (8.1) <0.001 White 411 (88.0) 345 (68.3) <0.001 Black 21 (4.5) 24 (4.8) 0.88 Asian 17 (3.6) 90 (17.8) <0.001 Other 18 (3.9) 46 (9.1) Age <45 yr no. (%) 136 (29.1) 165 (32.7) 0.01 Mean no. of years in practice <0.001 Annual income no. (%) $0 $99, (27.2) 28 (5.5) <0.001 $100,000 $149, (40.9) 71 (14.1) 0.05 $150, (29.3) 388 (76.8) <0.001 Missing data 12 (2.6) 18 (3.6) Practices Patient volume Ratio of patient visits to hours worked <0.001 Mean hours worked per week no <0.001 Patient visits in typical week no <0.001 Physicians and nurse practitioners in collaborative practice 378 (80.9) 209 (41.4) <0.001 no. (%) Practice setting no. (%) <0.001 Acute care hospital 63 (13.5) 50 (9.9) Specialty hospital 9 (1.9) 7 (1.4) Ambulatory or outpatient care 251 (53.7) 369 (73.1) Subacute or long-term care 23 (4.9) 12 (2.4) Home or community care 38 (8.1) 23 (4.6) Walk-in or retail-based clinic 18 (3.9) 1 (0.2) School health or student health service in secondary school 21 (4.5) 7 (1.4) or college setting Other 44 (9.4) 36 (7.1) Level of state restrictions on nurse practitioners no. (%) 0.07 Least restrictive 156 (33.4) 150 (29.7) Moderately restrictive 135 (28.9) 181 (35.8) Most restrictive 176 (37.7) 174 (34.5) * All the physicians and nurse practitioners who responded to the survey were licensed clinicians who had been trained in a primary care specialty, were actively working in primary care practice, and were providing direct patient care. Race or ethnic group was self-reported on the survey. Respondents could choose more than one category. Details about the practice setting are provided in item B3 on the questionnaire, available in the Supplementary Appendix. The construction of this variable was based on the rating system used by the annual Pearson Report, which grades states from A to F on measures of autonomy and patient access for nurse practitioners. We grouped respondents into three categories according to the Pearson Report grade: those practicing in the least restrictive states (A or B), those practicing in moderately restrictive states (C or D), and those practicing in the most restrictive states (F). 1901

5 T h e n e w e ngl a nd j o u r na l o f m e dic i n e Agree Neutral Disagree Don t know or blank Not applicable Statement Respondent Nurse practitioners should practice to the full extent of their education and training A primary care practice that is led by a nurse practitioner should be eligible to be certified as a medical home Nurse practitioners should be legally allowed hospital admitting privileges Nurse practitioners should be paid the same as physicians for providing the same services The physicians with whom I work support restrictions on nurse practitioners scope of practice in my state When physicians and nurse practitioners perform the same type of primary care visit, a primary care physician is able to provide a higher-quality examination and consultation Percentage Reporting Agreement or Disagreement with Statement Figure 1. Attitudes about the Scope of Practice of Nurse Practitioners. Survey respondents 505 primary care physicians (s) and 467 primary care nurse practitioners (s) were asked to rate their level of agreement or disagreement with each statement as strongly agree, agree, neutral, disagree, strongly disagree, don t know, or not applicable. All levels of agreement have been combined under Agree, and all levels of disagreement have been combined under Disagree. P<0.001 for all between-group comparisons. rently able to practice to the full extent of their education and training. Among those who said they did not have such an opportunity, their comments indicated that state restrictions, hospital regulations, and the type of work setting were the main factors in limiting their scope of practice (comments not shown). Respondents attitudes differed significantly on every measure of policies and practices related to the scope of practice (Fig. 1). A total of 95.6% of nurse practitioners and 76.3% of physicians agreed with the statement that nurse practitioners should be able to practice to the full extent of their education and training. However, physicians were less likely to agree that nurse practitioners should lead medical homes (17.2% of physicians vs. 82.2% of nurse practitioners) or should be paid equally for providing the same services (3.8% of physicians vs. 64.3% of nurse practitioners). We probed to determine whether physicians and nurse practitioners perceived differences in the quality of care provided by physicians. The responses of physicians and nurse practitioners to the statement that physicians provide a higher quality of examination and consultation than nurse practitioners were diametrically opposed: approximately 66.1% of physicians agreed and approximately 75.3% of nurse practitioners disagreed. For all items shown in Figure 1, clinicians did not differ significantly within professional 1902

6 Type of Service Respondent Annual physicals (including screenings and immunizations) Follow-up visits for controlled chronic conditions (e.g., hypertension, CHF, asthma, diabetes) Visits for complex chronic conditions complicated by coexisting conditions or not yet well controlled Visits for acute illnesses not requiring emergency care (e.g., urinary or respiratory infections, otitis media) Patient or family teaching Care coordination at care transitions (e.g., referrals, post-discharge) Follow-up for abnormal screening results Percentage Reporting Provision of Clinical Services by Nurse Practitioners Figure 2. Provision of Various Clinical Services by Nurse Practitioners in Collaborative Practices. Survey respondents 209 primary care physicians (s) and 389 primary care nurse practitioners (s) who worked in collaborative practices were asked, In the practice in which you work, who provides the following services to patients? The response options were provided mostly by physicians, provided mostly by nurse practitioners, provided by both, provided by other specialists or staff not primary care, and not applicable. For analysis of the activity of nurse practitioners in primary care practice, as perceived by s and s, the responses provided mostly by nurse practitioners and provided by both were combined. P<0.001 for all between-group comparisons. CHF denotes congestive heart failure. subgroups according to age, sex, region, or collaborative practice. We asked respondents whether in their work setting they agreed with the statement that nurse practitioners typically defer certain types of patient care services and procedures to the primary care physician. Among respondents in collaborative practice, 88.9% of physicians agreed, as compared with 61.3% of nurse practitioners (P<0.001). Clinicians who agreed with this statement were asked to identify the types of services that were primarily handled by physicians: 43.8% of physicians and 21.1% of nurse practitioners cited care for more complex cases; 11.2% and 15.2%, respectively, cited specific diagnoses or disease groups; and 19.1% and 36.8%, respectively, reported that physicians handled procedures and postoperative care, with the remaining responses accounting for less than 5% of the total. Clinical Services We asked clinicians whether specific clinical services were performed mostly by a physician, mostly by a nurse practitioner, by either a physician or a nurse practitioner, or by someone else (Table S1 in the Supplementary Appendix). Figure 2 shows the responses of physicians and nurse practitioners with respect to the clinical activities that were performed by nurse practitioners. Although physicians and nurse practitioners differed significantly on most items, the majority of the two groups reported that most services were performed by both providers, with the exception that only 28.3% of physicians agreed that nurse practitioners provided services for complex chronic conditions that were complicated by coexisting conditions or were not yet well controlled. No significant differences were observed in bivariate analyses according to age, sex, or region. 1903

7 T h e n e w e ngl a nd j o u r na l o f m e dic i n e Table 2. Respondents Views on the Effect of an Increased Supply of Nurse Practitioners on the Quality of Primary Care.* Variable Nurse Practitioners (N = 467) Physicians (N = 505) P Value Make Better Make Worse No Effect Don t Know No Response Make Better Make Worse No Effect Don t Know No Response percent of respondents Safety <0.001 Timeliness <0.001 Effectiveness <0.001 Efficiency and cost-effectiveness <0.001 Equity <0.001 Patient-centeredness <0.001 Access to health care for patients without insurance <0.001 Health care costs <0.001 * Respondents were asked, Do you think that increasing the supply of primary care nurse practitioners in the United States will make better, make worse, or have no effect on the following? P values are for the comparisons between physicians and nurse practitioners for all responses. Supply of Nurse Practitioners We asked respondents to consider whether increasing the supply of nurse practitioners in the United States would improve, make worse, or have no effect on multiple aspects of quality of care (Table 2). A majority of respondents said that having more nurse practitioners would result in improved timeliness of care (72.5% of physicians vs. 90.5% of nurse practitioners) and improved access to health services (52.2% of physicians vs. 80.7% of nurse practitioners). However, fewer than one in three physicians said that an increased supply of nurse practitioners would have a positive effect on safety, effectiveness, or equity of care, and about one in three reported that such an increased supply might have a negative effect on safety and effectiveness. We also analyzed these data after adjustment for age, sex, region, and the presence or absence of collaborative practice. (Detailed data for collaborative practices are shown in Table S1 in the Supplementary Appendix.) Discussion Our findings suggest that a substantial number of primary care physicians are unlikely to embrace policy recommendations aimed at further expansion of the roles and supply of nurse practitioners. In particular, physicians concerns about the likely effect of an expanded workforce of nurse practitioners on several aspects of health care quality need to be addressed in discussions of strategy for the development of the U.S. health care workforce. Among respondents to our survey, more than 70% of physicians and 90% of nurse practitioners agreed that nurse practitioners should practice to the full extent of their education and training, and a clear majority of nurse practitioners (75%) reported that they are in fact doing so. However, majority support of this broad principle broke down in a more detailed analysis of the perceptions and experiences of the two groups. Notably, physicians and nurse practitioners disagreed about whether nurse practitioners should lead medical homes or receive equal pay for providing the same services that physicians provide. Although physicians overwhelmingly rejected the statement that nurse practitioners provide the same quality of care that physicians provide, nurse practitioners clearly supported the statement. As we consider these polarized views, it is important to acknowledge that nurse practitioners and physicians come from very different cultures of professional education, are guided by different theoretical perspectives, and often develop their clinical skills in different practice environments. 1904

8 The training of the two groups varies in scope and duration, along with the respective processes of licensure and credentialing. It is not surprising (and indeed may even be expected) that physicians and nurse practitioners emerge without a common vision of their roles in the provision of primary care. Our survey results support calls for increased innovation in interprofessional education of the primary care workforce, encompassing curriculum content, training, and demonstration of competence. 12 Respondents in the two groups were far apart in their views on equal pay for providing the same services. Physicians opposition to equal pay is consistent with their perception, expressed in these data, that for any given service, they provide a higher quality of care than do nurse practitioners. Nurse practitioners support for equal pay is consistent with their majority view that physicians do not provide a higher quality of care for any given service. These survey data cannot provide evidence of the relative value of the training and expertise of these professionals. Nevertheless, the data suggest that physicians do not think that increasing the supply of nurse practitioners would have a positive effect on either the cost or the effectiveness of care, whereas more than 80% of nurse practitioners believe that increasing their numbers would improve the cost savings and quality of health care. From a societal perspective, we might consider whether expanding the supply of nurse practitioners and paying them equally for the same services that physicians provide would negate current savings from the disproportionately lower payments nurse practitioners now receive. More information is needed on the economic implications of the division of work between physicians and nurse practitioners before policymakers can definitively answer the question of whether employing a greater number of nurse practitioners and expanding their role would result in overall cost savings. Our study has several limitations. First, we did not measure experience with, or attitudes about, physician assistants or other health care professionals who provide primary care. Second, our sample source for nurse practitioners, as compared with that for physicians, had a higher rate of inaccurate contact information and did not contain data on activities associated with direct patient care. However, our results are consistent with the findings of the 2009 National Ambulatory Medical Care Survey, which documented the proportion of office-based physicians who employed nurse practitioners. 13 The Department of Health and Human Services recently conducted a national survey of nurse practitioners, and the results are expected to expand the national database on primary care and specialist nurse practitioners in the United States. 14 Third, in samples of this size, it is difficult to control for all personal and clinical-practice characteristics that differ between the two professional groups. The lack of diversity with respect to sex and racial or ethnic group in these samples limits our analysis. However, differences between the two groups were highly significant on virtually every outcome, even with samples of this size. As we consider these findings, we cannot help but reflect on the effect of these attitudes and practices on patients and patient care in the U.S. health care system. As changes are proposed in the size and configuration of the health care workforce and in the education and preparation of physicians and nurse practitioners, can the public be assured that the quality of health care will be maintained or improved? Furthermore, from a practical perspective, patients may need objective guidance in selecting health care professionals who are the most appropriate for them. Currently, the confidence that 82% of nurse practitioners have expressed in their ability to practice independently as leaders of patient-centered medical homes is not reflected in the attitudes of the majority of physicians we surveyed. Reasoned discussion about the education and roles of both physicians and nurse practitioners is needed to ensure that patients are receiving appropriate health care services. Our data provide evidence to inform ongoing public debates among physicians and nurse practitioners about their roles, responsibilities, and scope of practice. Both physicians and nurse practitioners will be needed to address the many challenges of developing a workforce that is adequate to meet the need for primary care services. It is our hope that the stark contrasts in attitudes that this survey reveals will not further inflame the rhetoric that has been offered by some leaders of the two professions but rather will contribute to thoughtful solutions for health care workforce planning and policy. 1905

9 Supported by the Gordon and Betty Moore Foundation, the Johnson & Johnson Campaign for Nursing s Future, and the Robert Wood Johnson Foundation. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. We thank Sandra Applebaum of Harris Interactive, Annie Farris of Vanderbilt University, Tessa Kieffer and Swaati Bangalore of Mathematica Policy Research, and Paola Miralles of Mongan Institute for their contributions to the management of the project and to the analysis and reporting of the data. References 1. Ku L, Jones K, Shin P, Bruen B, Hayes K. The states next challenge securing primary care for expanded Medicaid populations. N Engl J Med 2011;364: Primary care professionals: recent supply trends, projections, and valuation of services. Washington, DC: Government Accountability Office, February (Publication no. GAO T.) 3. Cooper RA. New directions for nurse practitioners and physician assistants in the era of physician shortages. Acad Med 2007;82: Institute of Medicine. The future of nursing: leading change, advancing health. Washington, DC: National Academies Press, Medicare Payment Advisory Commission. Report to Congress: reforming the delivery system. Washington, DC: National Academies Press, September American Journal for Nurse Practitioners. The 2011 Pearson report ( Iglehart JK. Medicare, graduate medical education, and new policy directions. N Engl J Med 2008;359: Fairman JA, Rowe JW, Hassmiller S, Shalala DE. Broadening the scope of nursing practice. N Engl J Med 2011;364: CMSS response to the future of nursing report. Chicago: Council of Medical Specialty Societies, Primary care for the 21st century: ensuring a quality, physician-led team for every patient. Leawood, KS: American Academy of Family Physicians, September Standard definitions: final dispositions of case codes and outcome rates for surveys. 7th ed. Deerfield, IL: American Association for Public Opinion Research, The Josiah Macy Jr. Foundation. Grants: interprofessional education and teamwork ( grantees/c/interprofessional-education -and-teamwork). 13. Park M, Cherry D, Decker SL. Nurse practitioners, certified nurse midwives, and physician assistants in physician offices. NCHS Data Brief 2011;69: Performance report for grants and cooperative agreements. Washington, DC: Health Resources and Services Administration Bureau of Health Professions, Copyright 2013 Massachusetts Medical Society. clinical trial registration The Journal requires investigators to register their clinical trials in a public trials registry. The members of the International Committee of Medical Journal Editors (ICMJE) will consider most reports of clinical trials for publication only if the trials have been registered. Current information on requirements and appropriate registries is available at