Health Occupations Outlook

Transcription

1 Health Occupations Outlook Prepared for the Louisiana Health Works Commission 2013 Stephen R. Barnes, PhD Director, LSU Division of Economic Development Stephanie Virgets, MA Research Associate, LSU Division of Economic Development

2

3 2013 Health Occupations Outlook Table of Contents Executive Summary... 1 I. Introduction... 2 II. Background... 2 III. Midlevel Providers... 4 A. Physician Assistants... 4 B. Advanced Practice Registered Nurses... 4 IV. Midlevel Providers in the Workforce... 5 A. Primary Care... 5 B. Rapid Growth... 6 V. Using Midlevel Providers for Physician Services... 6 A. Extent of Ability to Provide Physician Services... 6 B. Quality of Care and Satisfaction with Midlevel Provider Care... 8 VI. Reimbursement... 8 A. Medicare... 8 B. Medicaid... 9 C. Other Third-Party Payers... 9 VII. Findings on Savings... 9 VIII. Education IX. Changes Affecting the Future Workforce A. Recommendations from Professional Associations B. Recent Legislation X. Health Works Surveys A. Overview B. Survey Results XI. Employment Forecast XII. Conclusion XIII. Works Cited Appendix A: Professional Survey Instrument Appendix B: Employer Survey Instrument Division of Economic Development i

4 ii LSU E. J. Ourso College of Business Administration

5 2013 Health Occupations Outlook Executive Summary This study combines existing research with original surveys of Louisiana healthcare employers and Louisiana midlevel providers to assess the impact of upcoming demographic and policy changes to the healthcare industry on the healthcare workforce. The current workforce is examined and a forecast developed to show how that workforce with change between now and In particular, this study considers the supply of demand for labor for midlevel providers, which include: physician assistants, nurse practitioners, certified registered nurse anesthetists, clinical nurse specialists, and certified nurse midwives. Key findings from the forecast: The total number of physician assistants will grow from 730 in 2013 to 1,097 in 2022, implying a need to fill 367 new positions over the 9-year period. An additional 153 new PAs will be needed to replace existing workers, bringing total demand over the 9-year forecast horizon to 520. The total number of nurse practitioners will grow from 2,218 in 2013 to 3,363 in 2022, implying a need to fill 1,145 new positions over the 9-year period. An additional 464 new NPs will be needed to replace existing workers, bringing total demand over the 9-year forecast horizon to 1,609. The total number of certified registered nurse anesthetists will grow from 1,171 in 2013 to 1,428 in 2022, implying a need to fill 258 new positions over the 9-year period. An additional 298 new CRNAs will be needed to replace existing workers, bringing total demand over the 9-year forecast horizon to 556. The total number of clinical nurse specialists will grow from 165 in 2013 to 182 in 2022, implying a need to fill 17 new positions over the 9-year period. An additional 80 new CNSs will be needed to replace existing workers, bringing total demand over the 9-year forecast horizon to 97. Finally, the total number of certified nurse midwives will grow from 21 in 2013 to 34 in 2022, implying a need to fill 13 new positions over the 9-year period. An additional 6 new CNMs will be needed to replace existing workers, bringing total demand over the 9-year forecast horizon to 19. Other key findings from surveys: About 30% of healthcare organizations in Louisiana employ midlevel providers Among organizations with midlevel providers, 41% are located in rural areas; 32% of midlevel providers work in rural areas (by comparison 20% of all Americans live in rural areas and only 11% of physicians practice in rural areas) In general, providers work most often at general hospitals (51 percent) and second most frequently in offices of physicians (38 percent) Only 12% of healthcare organizations said they have or will hire new midlevel practitioners due to the Patient Protection and Affordable Care Act 86% of midlevel providers work in primary care (42%) or anesthesia (26%) Most PAs and NPs have prescriptive authority, but about 30% find limitations to their prescriptive authority hinder their practice Over the next few years, the education system is fairly well aligned with demand Division of Economic Development 1

6 Introduction/General Background I. Introduction The healthcare industry is rapidly evolving and expanding in the United States, with growth outpacing many industries including the economy as a whole. Changes in demographics are adding pressure to increase the size of the workforce with the aging of the Baby Boomer generation. These same demographic shifts will lead to an increasing number of retirements among the current health care workforce. On top of that, the recent Patient Protection and Affordable Care Act will contribute additional growth in demand by extending coverage to many Americans who were previously uninsured and drive a disproportionate share of that growth into the primary care setting. These rapid changes combined with a need to have the right mix of providers in the right geographic area to provide adequate care pose unique workforce challenges for the industry, particularly for more highly skilled professions that require multiple years of training. This study examines many of these changes and offers insights from two original surveys to explore the supply and demand for the healthcare labor force in Louisiana over the next 10 years with a special focus on advanced practice registered nurses and physician assistants (i.e. midlevel providers). The first section gives a summary of the challenges facing the healthcare workforce today. The remaining sections of the report focus on one particular group of the healthcare workforce, midlevel providers. Included in the discussion of midlevel providers is background information on each occupation within that group including a review of their role in the workforce and the type of services they provide relative to physicians. The report also offers a review of reimbursement patterns for midlevel providers, evidence on cost savings derived by utilizing midlevel providers, education requirements, and potential policy changes that may influence the use of these providers in the future. New insights on the characteristics of and demand for midlevel providers in Louisiana are discussed based on the results of two original surveys, a survey of advanced practice registered nurses and physician assistants in Louisiana and a survey of healthcare employers around the state. Finally, the report concludes with a forecast of the future demand for midlevel providers in Louisiana. While the forecasts for the healthcare industry imply a steadily increasing healthcare workforce, midlevel providers are expected to grow at an even faster rate than most health related occupations. A variety of conditions are exacerbating the need to expand the capacity of the physician workforce faster than the physician workforce is set to expand. More and more, health care organizations are looking at midlevel providers to satisfy this demand either by substituting their labor to a greater degree for physician tasks or integrating them more heavily into physicianled care teams. As demonstrated in this report, these midlevel provider occupations are poised for extremely rapid growth over the next decade. II. Background Several factors make the healthcare industry unique. In terms of workforce, the rapid growth of this sector of the economy Figure 1: National Health Care Expenditures, $6,000 Total Expenditures ($Billions) $5,000 $4,000 $3,000 $2,000 $1,000 $ Historical Projected Source: Centers for Medicare and Medicaid Services, LSU E. J. Ourso College of Business Administration

7 2013 Health Occupations Outlook is a defining feature. Healthcare expenditures have grown exponentially since the 1960s. Spending on health care has risen from under $75 billion in 1970 to $2.6 trillion in Healthcare spending has experienced much faster growth than the U.S. population. As a consequence, healthcare represents an increasing portion of the U.S. spending, with $1 out of every $20 spent on healthcare in 1970 and $1 of every $6 in 2010 (Centers for Medicare and Medicaid Services, 2012). Growth in healthcare in Louisiana has followed a similar trend to the rest of the nation. Figure 1 shows the rapid growth of U.S. expenditures on healthcare over the last few decades as well as the projected growth to There are several reasons growth in healthcare expenditures is outstripping other areas of the economy. Demand for healthcare services has increased partially due to an increase in real demand for healthcare. As household income increases, a larger percentage of expenditure goes toward healthcare services such as more frequent visits with providers and more elective procedures. Also, health care services have become higher quality due to recent technological advances like MRI machines, laser surgery, robotic surgery, organ transplants, and new treatments for various diseases. The introduction of new highly valued services will increase demand for care and individuals are generally willing to pay more for higher quality services. Finally, health care costs have gone up due to the increased prevalence of third-party payers (insurance) and the substantial increases in demand caused by the long run expansion of governmentfunded programs like Medicaid and Medicare. On top of this growth are two foreseeable increases in the demand for healthcare services. The first is an expected increase in the aged population, which is driven by ageing of the baby boomer generation and increasing life expectancy. The baby boomer generation is the group of people born during the uptick in the birthrate after World War II between the years 1946 and 1964 (at the time of this publication, aged 48 to 67). This demographic bulge means that the United States is seeing a larger aged population now than in the past. The percent of the population 65 years or older is expected to reach 16.1 percent by 2020 and 20 percent by 2040 ( Projected future growth, 2010). The age group of 65 and older tend to have higher medical costs than younger age groups. While this age group made up around 13 percent of the US population in 2002, they consumed 36 percent of total US personal health care expenses, and had an average health care expense over three times that of people aged (Stanton, 2006). One reason for this is that the elderly are more vulnerable and most affected by chronic conditions, which are more difficult and costly to treat than conditions more common in younger age groups (National Academy on an Aging Society, 1999). Meanwhile, the physicians and nurses in the baby boomer generation are retiring, exacerbating the anticipated shortage of doctors, which will be discussed in the following paragraphs. The second foreseeable increase in demand for healthcare comes from the increased coverage from the Patient Protection and Affordable Care Act (ACA). The ACA is expected to extend health insurance coverage dramatically. Some important features of the ACA that will increase demand for healthcare services include the individual mandate, which requires all U.S. citizens and legal residents to purchase health insurance; the requirement for firms with 50 or more employees to offer health insurance coverage for their employees; the creation of state-based standardized health care plans through health insurance exchanges; the expansion of Medicaid to allow people with incomes up to 133 percent of the federal poverty line eligible for Medicaid; and the prohibition for health insurance companies denying coverage due to preexisting conditions. Although Louisiana has no plans to expand Medicaid eligibility in the state, the other provisions of the ACA will increase health insurance coverage in the state. This increased coverage will lead to increased demand for healthcare services, especially primary care services. A recent study by Hofer, Abraham, and Mosovice found that expanded coverage from the ACA will increase the number of annual primary care visits in the United States between and million. If physician productivity remains as it is, between 4,307 and 6,940 new primary care physicians would be needed to accommodate the increase (Hofer, Abraham, and Mosovice, 2011). The ACA recognized the workforce pressure that will be created by this increased demand for primary care, and included some incentives to help alleviate that pressure including higher Medicare fees for primary care physicians in the years 2011 through Most of these changes are expected to take effect in The healthcare industry is also unique due to its specialized occupations. In general, physicians require the most specialized training, which creates a high barrier to entry. Many members of the health care community warn about a coming shortage of physicians. A report by the Robert Wood Johnson Foundation states a shortage of more than 90,000 physicians are predicted to occur within the next 10 years (Rasouli, 2012). The part of the workforce that is older or female, more likely to work fewer hours than younger and male counterparts, is growing. Residency slots are not keeping pace with the increases in medical school enrollment. Prospective physicians drop out due to strict admissions requirements for medical school, high cost of medical school, length of medical school and training requirements like a residency, and the limited number of positions for residencies available to them (U.S. Department of Health and Human Services, 2008). This expected shortage will also impact Louisiana. The 2012 Health Occupations Outlook found that Louisiana has an average annual demand for 362 new physicians in primary care practice areas, which include Family/General Practice, Internal Medicine, Obstetrics and Gynocology, Pediatrics, and General Surgery. That number is greater than the total number of physicians across all Division of Economic Development 3

8 General Background/Midlevel Providers specialties graduating from Louisiana s medical schools on an annual basis (LSU Division of Economic Development, 2012). Also, there are 268 primary care residencies, of which 251 were filled (National Residency Matching Program, 2013). The problem is especially acute for primary care physicians, physicians in rural areas, and physicians that provide care for people covered by Medicare and Medicaid. The difference in median annual income for specialists as compared to primary care is about $135,000 (Rasouli, 2012). Because medical schools leave new doctors with high amounts of debt, higher paying specialties are especially attractive over primary care. For the same reason, physicians are drawn to higher paying jobs and practices in urban settings. For instance, although almost 20 percent of Americans live in rural areas, only about 11 percent of the nation s physicians work there (Ewing & Hinkley, 2013). This lack of primary care physicians is often measured by Health Professional Shortage Areas (HPSAs), regions that have more than 3500 people per primary care physician or more than 3000 people per primary care physician but unusually high needs for primary care. As discussed in the 2012 Health Occupations Outlook, Louisiana has a very high concentration of HPSAs. One growing field in the healthcare arena is that of midlevel providers: advanced practice registered nurses and physician assistants. These providers help bridge the gap between nurses and physicians. They can perform many of the same tasks as physicians with less expense and training. Whether working as part of a physician-led team or as independent practitioners, these midlevel providers can make the provision of health services more efficient and increase the capacity of the health care system. The purpose of this report is to take an in-depth look at these midlevel providers. A more comprehensive review of the state of the healthcare industry including a detailed look at primary care physicians can be found in the 2012 Health Occupations Outlook report prepared for the Louisiana Health Works Commission by the LSU Division of Economic Development. III. Midlevel Providers This study looks at five different midlevel providers in Louisiana to see the occupational outlook of midlevel practitioners and physicians in the state and how these healthcare providers increase access to care. These five occupations are Physician Assistants (PA) and four Advanced Practice Registered Nurses (APRN): Nurse Practitioners (NP), Certified Registered Nurse Anesthetists (CRNA), Clinical Nurse Specialists (CNS), and Certified Nurse Midwives (CNM). A. Physician Assistants A Physician Assistant performs medical services under the supervision of a physician or group of physicians. These supervising physicians define the scope of a PA s practice, which may include assisting in surgery, conducting physical exams, ordering diagnostic and other medical services, injecting local anesthetic, and other medical services within the PA s scope of education, training and experience. The PAs can then use autonomous decision making within the bounds of these delegated practices. Supervision does not mean that a supervising physician must always be present with the PA, only that he or she must be available by phone or telecommunications in the event he or she is needed for consultation in a medical emergency ( Physician assistants: State, 2011). Since May 2004, PAs in Louisiana can be granted prescription authority if they have practiced clinically for at least one year. After this law was changed in Louisiana, Ohio and Indiana were the only states who had not yet granted prescriptive authority to Physician Assistants ( Louisiana PAs earn, 2004). Though they now can be granted some prescriptive authority, PAs in Louisiana lack full prescriptive authority. PAs in Louisiana may only prescribe Schedule III-V controlled medications. The lower the schedule number, the more dangerous and addictive the drug. Drugs classified as Schedule I are considered extremely addictive and determined to have no approved medical use in the United States. Schedule II drugs may only be prescribed by a physician in Louisiana (with a few exceptions in regard to ADHD medication), whereas in some states, PAs can prescribe Schedule II V drugs. Physician Assistants are licensed annually by the Louisiana State Board of Medical Examiners (LSBME). To be licensed, a PA must graduate from an accredited PA program or pass the NCCPA certifying examination (LA. REV. STAT. ANN ). A PA program generally takes 2 2 ½ years to complete and requires at least some healthcare experience before admission. PAs frequently have several years of work experience in a variety of healthcare occupations before completing their education. B. Advanced Practice Registered Nurses Advanced Practice Registered Nurses (APRNs) include the following four specific occupations of interest in this study: nurse practitioners, certified registered nurse anesthetists, certified nurse specialists, and certified nurse midwives. All APRNs are registered nurses that return to school for additional specialized training in order to provide additional services beyond those traditionally done by a registered nurse. APRNs practice autonomously, but must sign a Collaborative Practice Agreement (CPA), a written statement with a licensed physician or physician group, to practice in Louisiana. The CPA outlines the responsibilities of both parties and defines the parameters of the APRN s practice. In addition, it requires that the collaborating physician be available by telephone or direct telecommunications for consultation assistance with medical emergencies, or patient referral. (LA. ADMIN. CODE 46 XLVII-4513). After one year of clinical specialty, an APRN may apply for 4 LSU E. J. Ourso College of Business Administration

9 2013 Health Occupations Outlook prescriptive authority. Prescription practices must be specified on the CPA. APRNs are legally authorized to hold hospital privileges in Louisiana. APRNs frequently work as an RN for some time before deciding to further their career with an APRN degree, which entails two or more years to get a master s or doctoral degree in nursing that includes both didactic and clinical components (LA. REV. STAT ). APRNs must renew their license every two years by the Louisiana State Board of Nursing (LSBN). 1. Nurse Practitioners A Nurse practitioner is an APRN educated in a specified area of care to take a larger role in directing patient care than a traditional RN. Among other various services, NPs can obtain medical histories and conduct physical examinations, as well as diagnose, treat, and monitor various diseases, and order diagnostic tests and physical therapy. With a collaborative practice agreement, NPs can have prescriptive authority. They are the most common type of midlevel provider in this state NPs account for about 50 percent of APRN graduates from accredited programs in Louisiana ( LSBN Annual report, 2011). To become an NP, a practitioner must become nationally certified and licensed via an examination or endorsement from another state by the LSBN (LA. REV. STAT ). The typical NP education program takes 18 months to 2 years to complete, and generally requires that the student have a BSN and, frequently, at least a year of experience in nursing before admission. 2. Certified Registered Nurse Anesthetists Certified Registered Nurse Anesthetists are the second largest group of APRNs in Louisiana,. A CRNA is an APRN educated in the field of nurse anesthesia and is authorized to select and administer anesthetics or ancillary services to patients under their care. CRNAs, physicians, and dentists are the only medical practitioners allowed to administer anesthesia to a patient in Louisiana. One recent change in the CRNA field was the announcement by the American Association of Nurse Anesthetists that is requiring that all CRNA programs begin to transition their curriculum from a Master of Science in Nurse Anesthesia degree to a Doctor of Nurse Anesthesia Practice in Programs need to complete this change by 2025, but all nurse anesthetists who obtain the MSNA before 2025 will be grandfathered in and not required to return to school ( AANA announces support, 2013). To practice in Louisiana, CRNA s must be certified by a national body (LA. REV. STAT ). 3. Clinical Nurse Specialists Clinical Nurse Specialists plan, guide and direct care given by other nursing personnel. CNSs are certified by a national body to provide direct nursing care to a select population in a recognized nursing specialty area (LA. REV. STAT ). CNSs have much more specialized education and training than NPs. While NPs take more general courses in their Master s programs that will allow them to treat a wide variety of patients, CNS specialize in one area like pain management or geriatric health. In practice, they combine administrative and education roles with patient care - educating RNs and studying ways to increase quality and cost-effectiveness of care as well as diagnosing and managing care. A CNS can also have prescriptive authority depending on the collaborative practice agreement. There are several education programs for a CNS in Louisiana, but CNS programs enroll much fewer students than NP programs. 4. Certified Nurse Midwife The fourth APRN type in Louisiana is the Certified Nurse Midwife, educated in nursing and midwifery and authorized to manage the nurse midwifery care of newborns and women in the antepartum, intrapartum, postpartum, and/or gynecological periods. This means that for low-risk patients, CNMs may provide counseling and preparation for childbirth, supervision and care during labor and delivery, and care of the mother and newborn in immediate postpartum if the labor meets criteria to be essentially normal (LA. REV. STAT ). The collaborative practice agreement of a midwife will typically authorize the CNM to care for patients with essentially normal criteria, but transfer care to a physician in high-risk cases. There are no midwife programs in Louisiana, but the CNM is still required to have a master s degree in nursing and midwifery as well as be certified by a national body. IV. Midlevel Providers in the Workforce The physician assistant and nurse practitioner professions and training programs were first established in the 1960s, when the United States had a shortage of primary care physicians. Federal support and funding for PAs and NPs was established in the late 1970s, first in rural areas and HPSAs, and then slowly extended to all primary care. By 1997, all restrictions on geographic areas and settings in which PA and NP services could be paid by Medicare had been removed and PAs and NPs could receive an 85 percent reimbursement when billing under their own billing number (Paradise, Dark, and Bitler, 2011). Since then, the midlevel provider workforce has grown and can be seen in many specialties of healthcare, and in many states and different types of facilities. However, a majority of midlevel providers still serve in primary care. A. Primary Care Together, PAs and NPs make up a significant portion of the primary care workforce. Three studies estimate that together they represent percent of the primary care workforce (Bodenheimer and Pham, 2010; Primary Care Professionals, 2008; Larson et al., 2003), while a study by Stange and Sampson (2010) claims they could be as much as 42 percent of the primary care workforce by some measures. NPs are more likely than physicians to serve in HPSAs and to care for large numbers of Medicaid Division of Economic Development 5

10 Midlevel Providers/Using Midlevel Providers for Physician Services beneficiaries and uninsured patients (Paradise et al., 2011). A 2005 survey of Medicare beneficiaries found that patients who reported NPs (as opposed to physicians) as their primary care provider were significantly more likely to be Medicaid recipients. Patients who reported a physician as their primary care provider were more likely to have supplemental insurance (Hooker et al., 2005). Because these providers offer an alternative source of primary care, these providers are especially important to people who do not have regular access to a primary care physician. A study of healthcare recipients in Washington State found that PAs and NPs make up 23.4 percent of the generalist provider population and provide 21.0 percent of generalist outpatient visits. The proportion of PA or NP generalist providers was higher in rural areas representing 24.7 percent of providers in rural areas compared to 20.1 percent of providers in urban areas. This difference in the prevalence of PA and NP providers is especially marked among women providers with PAs and NPs comprising 50.3 percent visits by female generalists in rural areas, but only 36.5 percent of visits in urban areas (Larson et al., 2003). Another study by Ewing and Hinkley (2013) found that as many as 41 percent of rural Medicare beneficiaries saw a NP or PA for all or some of their primary care. These studies provide clear evidence that PAs and NPs are particularly important in providing primary care to traditionally underserved populations like those in rural areas. The importance of midlevel providers in primary care is growing because midlevel providers are the fastest growing segment of the primary care professional workforce, with nurse practitioners in particular seeing the highest growth rate. This is particularly important because the primary care sector is expected to exhibit rapid growth due to changes from the Patient Protection and Affordable Care Act. PAs have seen an average of 4 percent annual per capita growth over the period between the mid 1990s and mid 2000s. Over that same time, per capita growth of NPs was about 9 percent annually, and per capita growth of primary care physicians was only 1 percent ( Primary Care Professionals, 2008). However, demand for those midlevel provider services also faces competition in more specialized areas of medicine, which could limit the ability of these providers to fill shortages in primary care. PAs in particular have seen even more rapid growth outside of the primary care setting with the percent of PAs working in primary care dropping from 50 percent in 1997 to just 31 percent in Just as with doctors, the primary care specialty is becoming less popular for physician assistants (Coplan, Cawley and Stoehr, 2012). PAs in primary care are more likely to be female, Hispanic, or older than those PAs in other specialties. These same demographic characteristics are similar to those of medical students who choose primary care. While the percentage of PAs in primary care is decreasing, the overall growth of the profession has caused the absolute number of PAs in primary care to increase (Cawley, 2012). B. Rapid Growth Midlevel providers have seen rapid growth over past two decades. According to the Bureau of Labor Statistics, the number of physician assistants has also grown rapidly over the last decade. The number of PAs grew from an estimated 56,200 nationally in 2001 to 83,540, a 49 percent increase in just ten years. In Louisiana, the number of PAs grew a bit more rapidly with growth of 59 percent over the same period. Figure 2 shows the number of PAs per 10,000 people in Louisiana and the United States. The data show a significantly lower level of utilization of PAs in Louisiana compared to the rest of the nation. PAs also show faster growth in Louisiana than in the nation, perhaps due in part to the historically low utilization of these professionals in the state that has been gradually coming more in line with national trends aside from the most recent year. According to license data from the National Council of State Boards of Nursing, the number of APRNs in the United States has grown from an estimated 143,904 in 2001 to 228,530 in 2011, nearly a 60 percent increase in one decade. In Louisiana, growth of APRN licenses was a bit slower, with the estimated number of professionals growing by 46 percent. Figure 3 displays the estimated number of APRN per 10,000 people in Louisiana and in the United States. There were an estimated 3,929 APRN in Louisiana in 2011, of which 1,922 had prescriptive authority ( LSBN Annual report, 2011). Figure 4 shows the number of primary care physicians per 10,000 people in Louisiana and in the United States. The change in the number of primary care physicians per person is actually negative over this period, which helps explain why the occupations of midlevel providers are growing rapidly and becoming a larger percentage of primary care providers in the United States. V. Using Midlevel Providers for Physician Services One factor driving expectations that midlevel providers will be used more widely in future years is the potential cost savings because they are willing and able to provide many of the same services as physicians at a lesser cost. Individually, many of these providers are drawn to the profession because the training and education program is significantly shorter and less costly than that of a physician. Because these professionals are able to complete their education at a lower cost, many are willing to provide their services at a lower cost than physicians. A. Extent of Ability to Provide Physician Services It is difficult to determine exactly what portion of medical services midlevel providers currently provide because most insurance systems (including Medicare and Medicaid) allow or 6 LSU E. J. Ourso College of Business Administration

11 2013 Health Occupations Outlook Figure 2: Physician Assistants per 10,000 People National Louisiana Source: U.S. Census, Bureau of Labor Statistics and author s calculations Figure 3: APRNs per 10,000 People National Louisiana Source: National Council of State Boards of Nursing and author s calculations Figure 4: Primary Care Physicians per 10,000 People National Louisiana Source: U.S. Census Bureau and author s calculations Division of Economic Development 7

12 Using Midlevel Providers for Physician Services/Reimbursement require incident-to billing. Incident-to billing allows midlevel providers to file for reimbursement under their supervising physician s name under certain conditions, so the records may show that a physician performed a service that was actually performed by a midlevel provider. Because of this singularity, it is difficult to tell exactly how much of a physician s work a midlevel provider does or can perform. There have been many studies over the last decade to try and answer this question. A 1992 study by Knickman et al. found that midlevel providers can perform around 80 percent of a physician s work. This percentage is one of the most commonly cited in articles about midlevel providers. A more recent survey by Larson et al. (2011) focused on PAs. It found that PAs performed 61.4 outpatient visits per week compared with 74.2 visits by physicians, for a physician full-time equivalent of Productivity by generalist PAs was even higher than PAs in specialties. In 2012, a study on PAs and NPs set a range to the percentage of services that could be performed by a midlevel provider compared to a physician. It found that hiring a PA is the equivalent of 0.73 to 0.96 full-time family practice physicians. For general internal medicine and geriatrics, a PA can carry out percent of the services done by physicians. NPs can offset the work of percent of a full-time primary care physician. This study also found that PAs and NPs actually saw more preventative care visits than the average physician (Cawley, 2012). B. Quality of Care and Satisfaction with Midlevel Provider Care Research to date on patient health outcomes and satisfaction with care has shown that advanced practice registered nurses and physician assistants generally provide the same quality of care as physicians. Mundinger et al. (2000) looked at health outcomes in cases of 1,316 patients randomly enrolled with either a nurse practitioner or physician. They found no significant difference in patients health status at 6 months after the service was provided between the two provider types, nor was either group of patients found more likely to utilize further health services for the year following the initial treatment. Reviews measuring other clinical outcomes such as mortality, reduction of symptoms, metabolic and pathological parameters, and quality of life show that midlevel providers working as substitutes or supplements for physicians can provide the same quality of care as physician providers (Laurant et al, 2009). Studies of patient satisfaction that compare ratings across practitioner types indicate either no difference or indicate that patients are slightly more satisfied with nurse practitioners than other clinicians. Most studies on this subject compare only nurse practitioners and physicians, such as a Venning, Durie, Roland, Roberts, & Leese (2000) study, which found that nurse practitioner consultations tended to be significantly longer than general practitioner consultations (about 50 percent longer), and patients were more satisfied with nurse practitioners, even after controlling for consultation length. A 2005 survey of Medicare beneficiaries by Hooker, Cipher, and Skscenski, compared patient satisfaction ratings between physician assistant, nurse practitioner, and physician care. They reported PAs and NPs scored within 1 to 2 percent of physicians in patient satisfaction ratings. Employers also seem satisfied with their experiences hiring midlevel providers. One survey of 391 teaching hospitals in the United States found that about one third of all departments that employed physician assistants and nurse practitioners to substitute for or enhance physician work intended to increase the extent of substitution for physician or resident s work. Nearly all others were pleased by the current configuration of work and planned to maintain it (Riportella-Muller et al., 1995). VI. Reimbursement Within the structures of the health care system, reimbursement policies play an important role in determining when and how midlevel providers can provide services traditionally done by physicians. Historically, these providers were reimbursed at a lower rate than physicians, which encourages employers to utilize these services. However, raising those reimbursements to rates closer to physician rates also provides additional incentives for professionals to enter the field. The overall effect of reimbursement on the prevalence of midlevel provider services depends on finding the right balance in efforts to control costs yet offer payments that induce the right level of labor supply among providers. A. Medicare Medicare reimbursement for non-physician providers is a complex structure. Prior to 1990, Medicare reimbursement to non-physician providers was billed incident to a physician s services. These services were billed under the physician s provider number at 100 percent of the physician rate. Incident to billing requires that a supervising physician be in the facility with immediate availability for consultation and that the physician initiate the course of treatment. In 1990, direct APRN reimbursement by Medicare became available in rural areas and skilled nursing facilities. And in 1997, the Primary Care Health Practitioner Incentive Act removed NP billing restrictions on geography and setting, allowing direct Medicare reimbursement to non-physician providers at 85 percent of the physician fee rate ( Reimbursement of advanced, 2012). To file directly for Medicare reimbursement, a midlevel provider must comply with state licensure and scope of practice requirements, provide physician services, and practice in collaboration with a physician (Frakes and Evans, 2006). Recall that in Louisiana, an APRN must have a collaborative practice agreement with a physician in order to practice. In some states, 8 LSU E. J. Ourso College of Business Administration

13 2013 Health Occupations Outlook this is not required by law, so the APRN can practice without a collaborative agreement but would not receive federal Medicare reimbursement. Physician Assistants and Clinical Nurse Specialists are also paid at 85 percent of the physician rate under Medicare. The Patient Protection and Affordable Care Act of 2010 amended the reimbursement rate only for CNMs to 100 percent of the physician rate, because it was determined that at least 90 percent of visits to CNMs are for primary and preventative care while PA services are more often in specialized care settings. Prior to 2010, CNMs were paid 65 percent of the physician rate ( Medicare information for, 2011). CRNA services are reimbursed through different methods than other midlevel providers. The two most common ways are non-medically directed CRNA services, and medically directed CRNA services. Non-medically directed CRNA services are those performed without an anesthesiologist, and are reimbursed at 100 percent of the Medicare fee through the anesthesia fee schedule. Medically directed CRNA services are reimbursed at 100 percent of the fee, where the anesthesiologist and CRNA are each paid 50 percent of the fee. An anesthesiologist can provide medical direction for up to 4 cases concurrently. This structure offers strong incentives to employ CRNAs. A third way to bill Medicare is through medically supervised CRNA services, where the CRNA is reimbursed 50 percent of the Medicare fee, and the supervising anesthesiologist is reimbursed at a rate less than 50 percent based on the specific service. The supervising anesthesiologist may claim reimbursement for medical supervision in five or more concurrent CRNA cases, but most practices prefer to submit claims as non-medically directed CRNA services to receive the 100 percent reimbursement ( Reimbursement of CRNA, 2010). B. Medicaid The Medicaid program is administered by individual states, so the policies for reimbursement of APRN vary widely. All state Medicaid programs cover medical services by NPs, CNMs, or PAs at either the same rate or a lower rate than that paid to physicians. Even within states, the different provider types may receive different levels of compensation. Common differences include reimbursing CRNAs at a higher rate than other provider types, and giving non-primary care specialties like a psychiatric CNS a different rate of pay (Phillips, 2012). At least 20 states reimburse APRNs at 100 percent of the physician rate and at least another 4 at 100 percent in a large number of cases. Louisiana APRN services are generally reimbursed at 80 percent of the physician fees (Phillips, 2012). They must be billed under the APRN s license, essentially eliminating the incident to billing. The Louisiana Medicaid Professional Services Fee Schedule 1 published July 1, 2012 by the Department of Health and Hospitals, lists the reimbursement costs for different medical services and states Nurse 1 Practitioners, Clinical Nurse Specialists, Certified Nurse Midwives, and Physician Assistants are paid at 80 percent of this fee, except KIDMED medical, vision and hearing screens are reimbursed at 100 percent. C. Other Third-Party Payers States vary in their statutory provisions for third-party payer reimbursement to midlevel providers. Most states mandate reimbursement to midlevel providers; others provide reimbursement in practice even when not required. Some states list midlevel providers as primary care providers; some do not. Many private insurance plans base their fee schedule on Medicare policies, but of course this is not always the case. VII. Findings on Savings Spending in the health industry has greatly increased over the last few decades. One place savings can come from is from labor cost. Physician Assistants and Nurse Practitioners both have average earnings just over $90,000 a year, while the average annual salary for Family and General Practice Physicians is $168,330. Similarly, CRNAs average $154,390 while Anesthesiologists average $232,830 and CNMs average $91,000 while OB/GYNs average $210,000 (Bureau of Labor Statistics, 2013). In hospitals and other medical facilities with a large staff of physicians, the organization can in many cases bill incident to a physician s services, receiving the full 100 percent reimbursement while saving on labor costs by involving more midlevel providers in the provision of care. Medicaid, Medicare, and private insurance plans modeled after the reimbursement structure of the larger public programs reduce total costs when midlevel providers provide the full service and do not bill incident to physicians. Of course, labor cost is not the only component of practice expenses. Another place that the use of midlevel providers can create cost-savings is through lower liability insurance. In most specialties, midlevel providers have much lower premiums than physicians. PAs for example see premiums of about percent of physician premiums (Medicare Payment Advisory Commission, 2002). The Medicare Payment Advisory Commission s paper explains, The lower liability insurance rates for NPs and PAs may partially reflect a lower level of risk for the services they provide. Patients with high-risk factors are given to highly trained physicians, or patients can be transferred to the care of a physician when complications arise. In the short run, at least, this offers a way to provide low risk care from a midlevel provider not paying the same malpractice premium as a physician who sees a mix of higher risk patients. Of course, an actuarially fair malpractice insurance system would eventually react to a change in patient mix with even higher premiums for physicians whose patient mix will become more complex and Division of Economic Development 9

14 Findngs on Savings/Education risky with the increased use of midlevel providers offsetting any system wide cost savings from low insurance rates for midlevel providers. While the payment and insurance structures suggest that midlevel providers can lead to cost savings, it is possible that changes elsewhere in the system offset these savings, including the potential for increased profits or cost shifting. Empirical studies to date have found that midlevel providers cost less than physicians. A study by Mehrotra et al. (2009) found that retail clinics provide less costly treatment than physician offices, urgent care centers, and emergency departments. Retail clinics are usually housed in drug stores and big box stores with pharmacies, and are reputed to provide fast, affordable treatment for routine medical conditions and preventative care. Most are staffed by nurse practitioners. The average cost of care per visit for the same ratio of three common illnesses at the 4 types of facilities were $110 at a retail clinic, $166 at a physician s office, $156 at an urgent care center, and $570 at the emergency department. The study found quality of care between the first three to be similar no adverse effect on quality for patients visiting retail clinics but emergency department quality scores were significantly lower than the other settings. A study by Glance (1999) found that for anesthesia services, which represent 3-5 percent of total healthcare expenditures, the most important cost factor is the anesthesia staff. The use of CRNAs are one area practices can save on anesthesia costs. An analysis by Lock and Rosinia (2009) compares the cost and productivity of physicians and CRNAs. Using a national compensation survey and information on working hour trends, Lock and Rosinia calculated the costs of different anesthesia teams. Physicians cost more per hour; however, there must be a physician on a team to supervise the administration of anesthesia. Lock and Rosinia found that having one CRNA directed by one physician is actually more expensive than a physician alone since they have twice the labor force but can only provide about the same amount of services. Likewise, a ratio of 1 physician to 2 CRNA sees little savings, but a ratio of 1:3 or 1:4 begins to create significant savings. As noted previously, Medicaid reimbursement do not allow physicians to medically direct more than 4 concurrent CRNAs, which is the billing option that allows the physician and CRNA to collect the full reimbursement fee for the anesthesia service. In some areas of the country, especially in the Western United States, physician compensation is lower and the difference between the cost of a physician and a CRNA is small enough that the impact of CRNAs on profitability is not as significant. Lock and Rosinia suggest this is why there is a greater incidence of physician-only anesthesia practices in the West. Another type of midlevel provider that sees significant savings is the CNM. CNM patients see significant savings compared to those using an OB\GYN. This is due in part to cheaper labor costs, but also to other cost-saving measures more common in a midwifery practice than a physician one. CNM patients have an increased use of a birthing room as opposed to the hospital delivery room, less use of anesthesia, and shorter mean length of stay. Physicians tend to use more electronic fetal monitoring and more units of anesthesia during delivery (Knedle-Murray, Oakley, Wheeler and Petersen, 1993). VIII. Education Education programs for midlevel providers typically require less time and expense than those for physicians, which has been an important factor in increasing their numbers historically. Those differences as well as changes to those requirements also play an important role in assessing what will be the future growth in the midlevel provider workforce. Training programs for physician assistants and advanced practice registered nurses generally award at least a Master s degree and take about two years to complete. In , the average in-state tuition for a public university s PA program was $31,210; the average tuition at a private institution was more than double that at $65,573 (Liang 2010). A 2011 study by the Kentucky Coalition of Nurse Practitioners and Nurse Midwives found that costs for APRN programs were similar to those cited for PA programs in the Liang study. The degree can be completed much sooner than a student can finish medical school, and the midlevel provider can begin earning their salary much sooner than a physician. There are currently three physician assistant programs in the state: Louisiana State University Health Sciences Center Shreveport, Our Lady of the Lake College in Baton Rouge, and LSU Health Sciences Center New Orleans. The New Orleans physician assistant program is the newest, having just been granted status in 2012 with the inaugural class accepting 30 students in the spring semester of 2013 (CityBusiness, 2012). All three of these programs require a bachelor s degree and at least 80 contact hours (or 6 months for OLOL) of healthcare experience before admission into the program. The programs take months and awards a Master s degree upon completion. However, it is not necessary to have a Master s degree to be licensed as a PA in Louisiana. There are seven Louisiana schools with graduate programs in nursing: Grambling State University, Intercollegiate Consortium for a Master of Science in Nursing (at McNeese State University, Southeastern Louisiana University, and University of Louisiana at Lafayette), LSU Health Science Center, Loyola University, Northwestern State University, Our Lady of the Lake College, and Southern University. Six of the seven programs offer training for nurse practitioners; four offer training for certified nurse specialists; and two programs offer training for certified registered nurse anesthetists. There are no nurse midwife programs in Louisiana. According to the Louisiana State Board of Nursing annual report, very few nurses enroll in CNS programs 10 LSU E. J. Ourso College of Business Administration

15 2013 Health Occupations Outlook in 2011 only 4 students were admitted to CNS programs, and in 2012 only 5. The number of students admitted to CRNA programs in Louisiana has decreased by 48 percent between 2011 and 2012 (from 155 students to 81), while the number of students admitted to NP programs stayed pretty steady (420 in 2011 and 412 in 2012). NPs made up 83 percent of the students admitted to Louisiana APRN programs in 2012 (LSBN Annual Report, 2012). Another important consideration in midlevel provider education is the national shortage of nursing faculty. In their survey Enrollment and Graduations in Baccalaureate and Graduate Programs in Nursing, the American Association of Colleges of Nursing (AACN) reports that two thirds of nursing schools experienced a shortage of nursing faculty and were forced to turn away qualified nursing degree applicants. The Special Survey on Vacant Faculty, also by the AACN, found a national nurse faculty vacancy rate of 7.6 percent. Of these vacant positions, 88 percent were positions for which a doctorate was required or preferred. There are multiple reasons for these nurse faculty shortages. One significant contribution is the higher compensation available to advanced practice registered nurses in clinical and private-sector settings. Another factor is the aging population: the average retirement age of a nurse educator is 62.5, while the average retirement age of the nurse educator is 53.5 (AACN, 2012). IX. Changes Affecting the Future Workforce Debate over the role of the physician assistant and advanced practice nurse focuses on two issues: the scope of medical services that can be provided by midlevel practitioners, and the degree of autonomy midlevel providers are allowed. In 2010, after a two year initiative on the future of nursing, the Institute of Medicine (IOM) recommended to Congress and state legislatures that scope-of-practice barriers on Advanced Practice Registered Nurses should be removed so that they can practice to the full extent of their education and training (Institute of Medicine, 2010). However, there is still resistance from many groups who believe that the more highly trained and educated physicians should retain control of patient care. One fast-growing model of care, the patient-centered medical home or PCMH, emphasizes team-based care led by a primary care physician who makes first contact and provides an initial complex diagnosis and some ongoing care. Frequently, this involves occasional checkups by the physician so that he can continue monitoring the condition, and more frequent use of other team members like PAs, APRNs, other nurses, pharmacists, nutritionists, social workers, educators and care coordinators. The PCMH is especially important for patients with chronic conditions, such as diabetes or hypertension. Using midlevel providers in this team-based environment helps lower costs and increase access to care (American Academy of Family Physicians, 2012). The growth of this model of care coincides with the growth of the 65 and older population, the group most vulnerable to and most affected by chronic conditions. At the same time, the expansion of Medicaid and the ACA will cause a large increase in the number of people who seek primary care services. Whether operating in physician-led teams or independent practices, midlevel providers are needed to ensure every person s access to care. A. Recommendations from Professional Associations Understanding how the role of midlevel providers may change in Louisiana requires a sense of what has been done in other states and what specific types of changes have been proposed in Louisiana. The remainder of this section discusses several policies that are being promoted by professional organizations including the American Academy of Physician Assistants and the National Council of the State Boards of Nursing with a focus on where Louisiana falls compared to other states in some key areas of practice. Physician Assistants Each state has the power to regulate PA practice and there are large differences between states in the scope of practice and limitations of the PA. In January 2011, the American Academy of Physician Assistants, the national professional society for PAs, identified six key elements of modern PA practice: 1. Licensure as the Regulatory Term: There are traditionally three terms used to regulate PAs: certification, registration, and licensure. Out of 51 states (including DC), 50 states use the term licensure 2. Full Prescriptive Authority: As discussed previously, full prescriptive authority is the ability to medically diagnose and prescribe Schedule II-V drugs. Alabama and Florida PAs are restricted only to legend drugs, the most common and harmless medications, while the 36 states give full prescriptive authority to PAs and the rest only allow Schedule III-V drugs to be prescribed by PAs. 3. Scope of Practice Determined at Practice Level: Removing scope of practice limitation laws is not equivalent to leaving PAs completely to their own devices or take over all medical activities performed by a physician. This regulation will allow individual practices to give PAs regulations so that they can be allowed to practice in a manner commensurate to his or her education and experience, rather than determining it by a central board. This element is met by 36 states. 4. Adaptable Supervision Requirements: State laws that allow each practice to decide how to implement physician Division of Economic Development 11

16 Changes Affecting the Workforce supervision maximizes team effectiveness. This element is met by 21 states. Figure 5: Number of Key Elements Met by State 5. Chart Co-Signature Requirements Determined at the Practice: Like the scope of practice and adaptable supervision requirements, chart co-signature requirements should be determined at the practice level. This element is met by 22 states. 6. Number of PAs a Physician May Supervise Determined at Practice Level: Practices differ in specialty, setting, and patient population. The number of PAs to supervising physicians should be set at the practice level. This element is met by only 11 states. American Academy of Physician Assistants, 2013 Figure 5 displays the number of elements that have already been adopted by each state. Louisiana meets only two of these modern elements of practice: (1) the almost-universal element of licensure as a regulatory term; and (2) the scope of practice determined at practice level. Figure 5 shows how other states regulate their PAs. Advanced Practice Registered Nurses One of the campaigns for more fully utilizing APRNs in the United States is the Consensus Model for APRN Regulation of the National Council of State Boards of Nursing. This model states seven goals to create uniformity between states in the recognition and practice of Advanced Practice Registered Nurses. The following are the goals of the National Council of State Boards of Nursing: Source: American Academy of Physician Assistants, 2013 Figure 6: APRN Independent Practice and Independent Prescribing Authority 1. Require all APRNs be nationally certified (48 states) 2. Require that APRNs obtain a graduate degree or postgraduate certificate before they can practice in an advanced role (47 states) 3. Recognize all four roles of APRN that is, NP, CRNA, CNS, and CNM (41 states) 4. Use the title advanced practice registered nurse to refer to this group (29 states) 5. Use the term license to authorize APRN practice (26 states) 6. Independent Practice: APRNs can order and interpret diagnostic tests and administer treatments without written agreement with a physician, supervision, or other conditions for practice. (24 states) 7. Independent Prescribing: APRNs can prescribe medications without written agreement with physician. (17 states) National Council of State Boards of Nursing, 2012 Only six states Utah, New Mexico, Montana, North Dakota, Idaho and Vermont have implemented each of the seven goals of the Consensus Model. These last two are the most contentious. Louisiana meets the first five goals, but APRNs in Louisiana must have a Collaborative Practice Agreement with a physician in order to practice or prescribe. Figure 6 depicts in which states NPs can diagnose, treat, and prescribe medication to patients without a written agreement with a physician, like the Louisiana collaborative practice agreement. Other APRNs have similar restrictions but those may differ slightly state to state. B. Recent Legislation Physician Assistants Louisiana has recently seen a multi-year debate on the scope of practice of the PA. In July 2011, the Louisiana State Board of Medical Examiners (LSBME) proposed a rule that would require certain services be performed only if the supervising physician is physically next to the PA and others that require the supervising physician to be in the suite and available to be 12 LSU E. J. Ourso College of Business Administration

17 2013 Health Occupations Outlook at the PA s side within five minutes. Currently, physician assistants are delegated medical services by their supervising physicians, each to the extent of the education and experience (Louisiana Hospital Association, 2011). The Louisiana Academy of Physician Assistants, the Louisiana State Medical Society, the Louisiana Hospital Association, and other groups opposed the new law. The proposed rule was dropped, and on May 17, 2012, the final version of the rules were published requiring that physicians provide a level of supervision appropriate to the specific service, allowing the scope of practice to be determined at the practice level (American Academy of Physician Assistants, 2012). Advanced Practice Registered Nurses In May 2012, a bill was proposed to exempt nurse practitioners with certain qualifications from the CPA requirement if practicing in medically underserved areas. The bill was backed by the Louisiana Association of Nurse Practitioners ( Nurse practitioner bill, 2012). However, the bill ultimately failed and no changes were made in this most recent legislative session. X. Health Works Surveys Overview To more fully assess the current midlevel provider workforce and identify changes that will impact the future workforce, the 2013 Health Occupations Outlook included two original surveys. A survey of all Louisiana-licensed physician assistants and advanced practice registered nurses included questions about the supply of midlevel providers, their education and experience, and solicited information about their future labor force plans. A survey of healthcare employers in Louisiana captured the current and future demand for midlevel providers and some key characteristics of their workplaces. Table 1: Distribution of Louisiana Midlevel Provider Licenses License Type Louisiana Licenses Percent of Total Sample Physician Assistant % Nurse Practitioner 2,409 50% Certified Registered Nurse Anesthetist Clinical Nurse Specialist Certified Nurse Midwife 1,343 28% 177 4% 30 1% Multiple Licenses 51 1% Source: Louisiana State Board of Medical Examiners, Louisiana State Board of Nursing Professional Survey The professional survey went to all physician assistants and advanced practice registered nurses that were licensed by the Louisiana State Board of Medical Examiners or the Louisiana State Board of Nursing. LSBME and LSBN provided names and addresses of these providers, which provided a basis for the survey. The survey attempted to reach these professionals in three different ways: First, a copy of the survey was sent to each midlevel provider at a work address (for PAs) or home address (for APRNs). Second, APRN s were available for most licensees so an message with a link to the survey was sent to those who did not reply to the mail survey. Finally, the LSU Public Policy Research Lab called those providers that had not yet responded to mail or attempts to encourage their participation and complete surveys over the phone or offer additional follow-up via fax or in order to maximize survey responses. At the time the survey was conducted, the population included 768 licensed PAs and 4,010 licensed APRNs. Many of these APRN have multiple licenses making it difficult to determine how many APRN professionals are practicing with a primary focus in each APRN area. However, the breakdown is as follows: Nine percent of those surveyed have an address from outside of Louisiana, with the highest percentage of out-of-state addresses from CRNAs and CNMs. These midlevel providers were included in the survey sample because midlevel providers who live in other states may still practice in Louisiana, especially if they keep their licenses current. However, because the survey is focused on the Louisiana workforce, a question was included in the survey to determine if they practice in Louisiana. Other questions focused on education and prior experience, practice area or specialty, and number of hours worked. In addition, there were questions about age and future hours used to gauge the future labor supply. The survey also included questions about prescriptive authority, limits to scope of practice, and supervising or collaborative physicians. The final set of questions focused on the provider s employer. The survey includes 2,066 responses from the professionals surveyed. After accounting for bad addresses, the overall response rate was calculated to be 43.9 percent. Table 2 breaks out the response rates by provider type. To ensure the results best Table 2: Provider Response Rates by Type PA NP CRNA CNS CNM 35% 49% 38% 54% 66% Division of Economic Development 13

18 Health Works Surveys Table 3: Health Organizations by Size and Industry Industry Code Industry Code Description Size 1-4 Size 5-49 Size Offices of Physicians (except Mental Health Specialists) Total Establishments 1,742 1, , Family Planning Centers Outpatient Mental Health and Substance Abuse Centers Kidney Dialysis Centers Freestanding Ambulatory Surgical and Emergency Centers All Other Outpatient Care Centers Home Health Care Services All Other Miscellaneous Ambulatory Health Care Services General Medical and Surgical Hospitals Psychiatric and Substance Abuse Hospitals Specialty (except Psychiatric and Substance Abuse) Hospitals Nursing Care Facilities Source: U.S. Census County Business Patterns Residential Mental Retardation Facilities Totals: 2,102 2, ,555 represent Louisiana s midlevel provider workforce, weights were creating using provider type, Louisiana state residence, and the number of years the license was held. These three characteristics were available for the entire sample and help ensure responses are representative of the entire workforce. Employer Survey The employer survey is a critical source of data on workforce needs including future demand for APRNs and PAs working in Louisiana. Every organization was asked for their total number of employees, number of physicians, and total level of midlevel providers, while the ones that indicated they currently employed midlevel providers or expected to do so in the near future were asked additional questions about their practice. The survey was sent to every healthcare employer in Louisiana by mail with options for responding with prepaid mail, fax, or online response options. Organizations that did not respond to the initial mailer were called by the LSU Public Policy Research Lab to complete the survey over the phone or offered additional follow-up via fax or . Table 3 provides counts for the number of health care employers according to the U.S. Census. The survey sample file was compiled from the Louisiana Hospital Association, a DHH list of accredited health care facilities, and mailing information purchased from InfoUSA for health care employers in other industries. Responses were received from 2,285 organizations. The sample was then weighted up to represent the total count of Table 4: Employer Survey Response by Industry Group Industry Group Responses Response Rate Ambulatory Health Care Services 1,796 39% Hospitals 95 39% Nursing and Residential Care Facilities % Table 5: Employer Survey Response by Size Size Responses Response Rate 1-4 Employees % 5-49 Employees 1,373 50% 50 + Employees % 14 LSU E. J. Ourso College of Business Administration

19 2013 Health Occupations Outlook Table 6: Active Midlevel Providers in Louisiana Provider Type Number of Licenses Unique Providers by Primary Type Estimated Active Providers Percent Actively Working Physician Assistant % Nurse Practitioner 2,460 2,449 2, % Certified Registered Nurse Anesthetist 1,347 1,347 1, % Clinical Nurse Specialist % Certified Nurse Midwife % Total 4,830 4,778 4, % organization type and size in Louisiana according to data from the U.S. Census County Business Patterns. The response rate was 41.7 percent, and varied by industry and size of organization, as can be seen in Table 4 and Table 5. Survey Results Current Workforce The starting point for assessing workforce needs is identifying an accurate count of the current workforce, which accounts for unemployed and retired license holders as well as those who maintain their license but who have a primary job function outside of direct patient care. The estimated numbers of active midlevel providers for each type are shown in Table 6. A total of 51 APRNs have multiple license types, including one with three separate licenses. The first column describes the total number of licenses of this type held in Louisiana, while the second column has assigned providers with multiple licenses to one provider type based on their renewal category or the license they listed first. Next, the table estimates the number of every midlevel provider licensed in Louisiana, regardless of home or business address that is actively practicing in Louisiana. This excludes any providers that responded that they were not currently practicing in Louisiana, or that worked 0 hours last year. Responses were weighted up to represent the entire sample. The column Percent of Total Licenses shows the estimated percentage of midlevel providers licensed in Louisiana that are actively practicing in healthcare. Overall, 90 percent of Louisiana midlevel provider license holders are actively practicing in the state. The percent of Certified Nurse Midwife license holders actively working seems particularly small compared to the number of licenses; however, many of these midwives live out of state or otherwise consider themselves not practicing in Louisiana. Five midwives have addresses out of state; an additional three considered themselves not currently practicing because their primary job was in education, and the remainder are not practicing due to retirement or being out of work. These are all reasons that other provider types may elect not currently practicing in Louisiana or otherwise not be counted, but the effect is greatest among CNMs. Data on the number of midlevel providers were also collected in the employer survey. The purpose of this was to identify the relationship between physicians and midlevel providers as well as provide a benchmark for future growth. Including direct employees and contract workers, the estimated number of providers was 5,352, which is higher than the total number of providers due to the fact that contract employees can work for more than one employer. Considering only those directly employed by a health care organization provides a lower estimate of 3,614 midlevel providers, which excludes those self-employed individuals. When including contract employees, health care organizations indicated that the ratio of physicians to midlevel providers was 2.9. Including only direct employees, that ratio is 2.4. The employer survey also identifies how concentrated midlevel provider employment is across different types of organizations. There are an estimated 1,598 healthcare organizations in Louisiana that directly employ at least one midlevel provider in Louisiana, or close to 30 percent of all health care organizations. Table 7 shows the percentage of organizations that employ at least one midlevel provider or expect to within the next few years, broken down by industry group. Hospitals are more likely than employers in other industries to hire midlevel providers, and also more likely than employers in other industries to use them as contract employees rather than direct employees with nearly half of all hospitals using midlevel providers. Table 8 provides a similar summary with results stratified by employer size. Large organizations are also more likely than small organizations to hire midlevel providers. This finding is driven largely by the concentrations of hospitals in the large employer section, however the steadier pattern shows that large organizations in other industries are also more likely to utilize the services of midlevel providers. Characteristics of Workforce Supply The professional survey asked many questions of midlevel providers to understand certain workforce supply characteristics to assess how these factors are likely to affect the forecast for these providers. The survey gathered information about hours worked and retirement to help characterize the current supply Division of Economic Development 15

20 Health Works Surveys Table 7: Prevalence of Midlevel Providers by Industry Percent with Midlevel Providers Industry Group Total Direct Employees Only Ambulatory Health Care Services 32% 26% Hospitals 46% 26% Nursing and Residential Care Facilities 23% 9% Table 8: Prevalence of Midlevel Providers by Size of Organization Percent with Midlevel Providers Size Total Direct Employees Only 1-4 Employees 15% 13% 5-49 Employees 38% 32% 50+ Employees 52% 28% of providers and better anticipate how the supply of labor might change in the coming years. The survey also included questions about the education and experience of each provider to help better characterize the career pathway for these providers. Unless otherwise specified, the following results include only those midlevel providers who are currently practicing in Louisiana. The current characteristics including age, hours, and retirement plans of the current workforce offer a helpful starting point in assessing future workforce needs. The first of these questions is the age of the midlevel providers, which provides a starting point for forecasting the future supply of providers by determining working and retirement patterns. The age distribution of the current workforce is displayed in Figure 7 for PAs and APRNs. The large number of PAs in their early years and, to a lesser extent, the number of APRNs in their late thirties and forties is due in part to the growth of these professions in recent years. There is a much larger percentage of PAs in the early professional years. One reason for this may be that PAs tend to have less experience in the medical field before they become licensed. APRNs must first become an RN before they can become and APRN, and often Figure 7: Midlevel Provider Age Distribution 30% 25% 20% 15% 10% 5% 0% APRN PA 16 LSU E. J. Ourso College of Business Administration

21 2013 Health Occupations Outlook work for many years as a nurse before pursuing further education. Another notable pattern is the increase in APRNs in their late fifties and sixties, which is analogous to the US population, showing the baby boomer bump in population. Differences in ages between the four types of APRNs were not significant enough across the entire distribution to warrant comparisons as in Figure 7, however the average age for professionals with each license type did differ as shown in Table 9. In particular, the average age of the CNS is much older than other groups. One reason for this pattern is that CNSs are the most likely of all provider groups to obtain a doctoral degree as will be discussed below. Another reason for the differences in average age is the timing of graduations from advanced practice nursing programs. According to LSBN, only 1 percent of students graduating from APRN programs in 2012 were in a CNS program. (LSBN Annual Report, 2012) However, CNSs make up 5 percent of the APRN population, so a significant portion of the workforce must have entered the workforce years ago when CNS programs were relatively more prevalent. The next set of questions focused on the number of hours worked by midlevel providers as well as future changes to labor supply. The estimated average of weekly hours was 41.3, which includes regular hours and the portion of on-call time in which providers were actively providing medical services. Differences in hours worked by provider type are shown in Table 10. PAs work significantly more hours than other midlevel providers. One contributing factor for this may be the relatively larger population of males, who are less likely to work part time (the average hours across all midlevel providers for males is 44.3 hours compared to 39.1 hours for females). PA work is also organized differently than APRN; PAs usually work more closely with a physician, and the 2012 Health Occupations Outlook on primary care physicians found that physicians work an average of 50.5 hours each week. Also, although PAs work longer hours across all age groups, younger PAs do work longer hours than older ones, and there are a greater percentage of young PAs than APRNs. While the average hours for all provider types indicate that most providers are already working full time, the survey asked about the ability of the providers to see more patients. Figure 8 shows the percent of respondents who indicated they must reduce patient load, cannot take on any new patients, and could add new patients. In general, most providers indicated they were able to take on new patients, and no provider type deviated significantly from the pattern. Another way of interpreting these results is that about 25 percent of midlevel providers indicated limited additional capacity, and the organizations where they work may be interested in hiring new practitioners to accommodate growth in the demand. Table 9: Average Age by Provider Type Provider Type Average Age PA 37.6 NP 44.6 CRNA 45.9 CNS 55.6 CNM 50.4 Table 10: Average Weekly Hours by Provider Type Provider Type Avg. Weekly Hours PA 44.6 NP 41 CRNA 40.2 CNS 38.1 CNM 38.4 Figure 8: Ability of Midlevel Provider to See New Patients 15% 58% 8% 19% Must Reduce Cannot Add Any Could Add a Few Could Add Many Each type of provider expected to work fewer hours in the future than they do currently, changing from last year s 41.3 hours to only 31.6 hours ten years from now, in This trend is driven by providers that cut back hours as they age and by people who Division of Economic Development 17

22 Health Works Surveys Table 11: Average Weekly Hours in the Future Survey Age Current Hours In One Year In Five Years In Ten Years All Ages plan to retire within the next 10 years (thus changing their hours to 0). Table 11 displays these average hours now and in the future for different age groups to illustrate the effect of aging on hours. Hours are relatively stable for workers in their thirties and forties, but in their fifties and sixties, people begin to answer they intend to reduce hours significantly (including zero) in the next 10 years. Next, the survey asked if the midlevel provider planned to stop practicing, retire, or move out of state in the next 10 years and if so, when. Around 24 percent planned to stop practicing in the next 10 years. The percentage of providers planning to leave the Louisiana workforce is highest among CNSs, at 52 percent. Table 12 breaks down the results to this question by provider type and calculates the number of jobs that will need to be replaced in the next 10 years. The average midlevel provider has worked for his or her primary place of employment for 6.7 years. Recall that the average number of years a midlevel provider has been practicing is 10 years, suggesting a relatively low turnover rate. Table 12: Expected Retirements by 2020 Type Percentage Count PA 20% 148 NP 21% 477 CRNA 27% 310 CNS 52% 84 CNM 31% 7 Figure 9: Louisiana Education and Training To better understand the relationship between the location of education and training and the state where the provider currently practices, the survey asked whether the provider had any PA or APRN degree from Louisiana and, if not, if they had performed their clinical requirements in the state. Only about 64 percent of Louisiana midlevel providers received at least one degree for their license in Louisiana, though many of the remainder did complete their clinical requirements in the state. A total of 85 percent of active Louisiana midlevel providers completed either their degree or their clinical requirements in the state. Figure 9 helps illustrate these relationships. These numbers show that the workforce relies heavily on Louisiana education and training. Unsurprisingly, CNMs had the smallest number to receive education or training in Louisiana (57 percent) because there are currently no CNM schools in Louisiana. NPs had by far the highest percentage (91 percent). The other provider types had between 77 percent and 81 percent that had received their education or training in Louisiana. To inform discussions related to minimum education standards, the survey asked what the midlevel provider s highest level of education was. Just over 82 percent of midlevel providers 18 LSU E. J. Ourso College of Business Administration

23 2013 Health Occupations Outlook Figure 10: Education by Provider Type 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% PA NP CRNA CNS CNM Diploma Bachelor's Master's Doctorate indicated their highest degree was a master s degree. This result fits well against the current education requirements in the state because APRNs are required to have at least a master s degree and the three PA education programs in Louisiana all offer a master s degree. However, it was not always the case that APRNs were required to have a master s degree, and many with a bachelor s degree or diploma were grandfathered in. Furthermore, PAs are still not required to obtain a master s degree and may continue to move into the state after completing a bachelor s degree elsewhere. Figure 10 shows the distribution of degrees for each provider type. On the survey, diploma was an answer that must be volunteered, so the bachelor s degree may be slightly overrepresented and diploma underrepresented in the graph. PAs are the most likely providers to have a bachelor s degree (about 48 percent), while CNSs are the most likely providers to have a doctorate (about 18 percent). As discussed later, the high prevalence of CNSs in education and research helps explain the large incidence of doctorate degrees. Because CRNA schools are transitioning to doctorate programs starting in 2015, the survey asked providers if they would have chosen the career path if required to obtain a doctorate. Fortyeight percent of all midlevel providers, including 38 percent Figure 11: Years Experience Prior to Becoming Midlevel Provider What does this graph mean? contains approximately 90% of data } contains 50% of data Division of Economic Development 19