Challenging Assumptions About the Future Supply and Demand of Physical Therapists in the United States

The American Council of Academic Physical Therapy (ACAPT) is a not-for-profit association that supports a membership including 95% of accredited US programs in achieving excellence in physical therapist education. In support of their mission, ACAPT published an open letter on its website in March 2020 that urged university leaders to consider the changing landscape of physical therapist education before adding new physical therapist professional education programs or increasing class sizes at existing programs.1 On May 3, 2021, ACAPT posted an updated version of this letter2 to reiterate their concern that too much growth in physical therapist graduates could outpace the profession’s growth projections from the Bureau of Labor and Statistics (BLS).3 ACAPT questioned the optimism of the BLS employment projections, which, at the time of the letter, predicted 18% growth for the period 2019 to 2029, with 15,200 new job openings (14,493 full-time equivalents [FTEs]) each year, a much faster growth than that of other occupations.2,3 ACAPT further argued that the actual rise in employment is unlikely to match this projection, based on declining reimbursement for physical therapist services. Instead, ACAPT referenced the American Physical Therapy Association’s (APTA’s) 2020 Physical Therapy Workforce Analysis, which predicted an estimated surplus of 25,000 physical therapists by 2030.4 In our opinion, the APTA Workforce Analysis, upon which ACAPT’s conclusions are based, uses flawed assumptions that result in erroneous conclusions regarding the future supply and demand for physical therapists. Although we fundamentally agree with ACAPT that rising student debt, shortage of qualified faculty, and inadequate number of clinical sites pose a potentially serious threat to physical therapist education, we believe that ACAPT’s caution against the expansion of physical therapist education programs is unwarranted and counterproductive. The purpose of this Point of View is to identify flawed assumptions in the 2020 APTA Workforce Analysis and propose an alternative to ACAPT’s conclusions. Historical visualization of the American Physical Therapy Association’s Workforce Analysis demand projections in 2012, 2015, 2016, 2017, and 2020. Comparison of projected demand for physical therapists between APTA’s 2020 Workforce Analysis and BLS models from 2019 to 2029 (in thousands). BLS demand projections were adjusted to reflect FTEs. The FTE adjustment uses APTA’s last reported assumptions (2017 Workforce Analysis), where 15% of physical therapists were estimated to work part-time, with part-time personnel working 69% of the normal work week. APTA = American Physical Therapy Association; BLS = Bureau of Labor and Statistics; FTE = full-time equivalent. Previous versions of the APTA Workforce Analysis incorporated a shortage at baseline ranging from 19,000 physical therapists in the 2012 model5 to 45,000 in the 2017 model.6 APTA contends that they updated the model in 2020 to align with best practices in workforce planning, which, according to APTA, calls for setting supply equal to demand in the base year (2019) of the model.4 In our opinion, this is a fundamental and fatal flaw. The APTA Workforce Analysis does not provide a reference for this practice. However, APTA’s Physical Therapy Workforce Data webpage7 includes a link to the Health Resources and Services Administration (HRSA) Allied Health Workforce Projections fact sheet8 that presents national-level supply and demand projections for occupational therapists and physical therapists from 2016 through 2030 using HRSA’s Health Workforce Simulation Model.9 For demand modeling, the Health Workforce Simulation Model assumes that demand equals supply in 2016 and applies health care use patterns across future population demographics. To support their assumption to set demand equal to supply at baseline, the HRSA referenced an extensive health workforce planning study10 that reviewed 26 projection models from 18 countries in the Organisation for Economic Co-operation and Development.9 However, rather than supporting this practice, the study argued precisely against it: Ono et al10 criticized workforce planning models that make this convenient and simplistic assumption because the existence of an actual shortage (or surplus) will significantly bias the model throughout the projection period. We believe, therefore, that APTA’s assumption to set supply and demand equal at 212,000 jobs in the base year (2019) of the 2020 APTA Workforce Analysis is not consistent with best practice or reality, resulting in an artificial sharp and sudden decline in expected demand at baseline compared with the previous 2017 report.6 In fact, the assumption essentially resets workforce demand below the original 2012 projection (Fig. 1).5 This revised outlook lies in stark contrast to APTA’s prior 4 reports5,6,11,12 released during the period 2012 to 2017, in which demand projections were more consistent with BLS data. For example, a review3 of the BLS data from 2019 demonstrates a total demand for physical therapists that is closer to 260,000 (246,000 FTEs), which suggests that modeling a baseline shortage closer to 34,000 physical therapists would have been more appropriate (Fig. 2). This shortage also aligns with the sentiment expressed among large employers that hire substantial numbers of physical therapists.13 Had APTA continued its prior methodology in the 2020 model, the projections would have resulted in a significantly higher demand of 255,000 physical therapists by 2025, more consistent not only with BLS but also with prior year models. Impact of attrition rate on net supply of physical therapists (in thousands). A key driver of any occupational supply model is the attrition rate, which, in this case, is a combination of the number of physical therapists exiting the workforce (ie, retirement) and those transferring to another occupation annually.14 As illustrated in Figure 3, a minor +1% change in the attrition rate assumption results in a disproportionately large 7% decrease in the supply; therefore, the model’s accuracy is highly dependent on attrition. The APTA Workforce Analysis in 2015 stated that there were no known attrition rates available for the physical therapist profession; therefore, 1.5%, 2.5%, and 3.5% attrition estimates were selected,11 similar to those reported in a study by Landry et al15 on workforce projections for the period 2010 to 2020 across the United States. Attrition rate was not reported in the current workforce analysis. However, other studies on nursing16 and physician assistants17 have used 4.2% and 5% attrition estimates, respectively. Therefore, if the actual attrition rate is higher than anticipated in the current APTA model, this would result in an overestimation of future supply (Fig. 3). The APTA Workforce Analysis seriously underestimates future demand, forecasting only 15,000 new physical therapist jobs through 2029.4 In contrast, 2 other reputable sources that produce macroeconomic data estimate closer to 50,000 new jobs (47,675 FTEs), more than 3 times the job growth in 2029 than predicted in the APTA model.3,18 This difference is largely attributable to the flawed assumption to set supply and demand equal at baseline. By using the current ratio of FTE physical therapists to the US population with health insurance to estimate demand, the APTA model also assumes that physical therapist employment will grow linearly relative to population growth. However, we believe that this assumption does not account for changes to patient demographics (eg, aging population), disparities between supply and demand at localized geographies, an evolving payer market, or changes in health care delivery patterns and practice settings, all of which could contribute further to the underestimation of future demand. For example, physical therapist employment continues to shift away from the traditional hospital-based setting to lower-cost settings such as the outpatient setting.19 Further expansion of the physical therapist’s scope of practice (eg, through direct access) is likely, in our opinion, to significantly bolster demand in these settings.20–22 Over 127 million Americans (approximately 50% of adults) experience a musculoskeletal condition that requires medical attention, resulting in $213 billion of direct medical expenditures, lost wages, and decreased productivity annually.23 Accessing physical therapists early in a musculoskeletal episode of care is associated with major improvements in outcomes and decreased costs24–27; therefore, continued efforts to increase the value of health care could result in more patients seeking care from a physical therapist during an episode of care. Alternatively, the BLS model does account for structural changes in the future demand for the occupations it studies, thereby accounting for expected changes to an occupation’s share of industry employment.3 Projected Annual Graduates From Physical Therapist Education Programs Accredited by the Commission on Accreditation in Physical Therapy Education, Relative to Employment Openings by 2023a aFTE = full-time equivalent. bAverage annual job openings/total enrollment. Projected Annual Graduates From Physical Therapist Education Programs Accredited by the Commission on Accreditation in Physical Therapy Education, Relative to Employment Openings by 2023a aFTE = full-time equivalent. bAverage annual job openings/total enrollment. One of ACAPT’s central arguments is that future physical therapist employment is unlikely to achieve the 18% growth projections in the BLS data due to health care cost containment efforts by the federal government and third-party payers. For example, ACAPT cites the recent 3% cut in Medicare reimbursement for physical therapist services and the implication that this results in fewer reimbursable visits per episode and reduced net revenue per visit.2 However, the most important reimbursement change at the Centers for Medicare & Medicaid Services (CMS) significantly cuts payment of services rendered by physical therapist assistants.28 Major reimbursement changes by the CMS drive practice patterns and, therefore, in our opinion, this will increase demand for physical therapists as employers shift staffing from physical therapist assistants to physical therapists. If other payers adjust their payment policies to align with the CMS, as is often the case, an even more dramatic need for physical therapists might occur. Collectively, these arguments suggest that the BLS estimates may be conservative and, therefore, an even greater supply of physical therapists will be needed to meet future demand. In addition to establishing the future demand for physical therapists, it’s also important to consider ACAPT’s concern regarding the increasing supply of physical therapists. There has been a significant expansion in the number of accredited and developing physical therapist education programs over the last 20 years. The Commission on Accreditation in Physical Therapy Education (CAPTE) reports that there are 261 accredited physical therapist education programs and 10 additional programs in development (Candidacy).29 It’s true that some existing programs are considering or have already increased their class size. However, ACAPT is likely overstating the extent to which this expansion in class size is occurring. Based on the raw data available on the Physical Therapist Centralized Application Service, 44 (18%) of programs reported a growth in class size expectations in 2021, 33 (13%) reported an expected decline, and 171 (69%) anticipated enrollment to remain flat.30 Therefore, although more programs expect enrollment to grow in their upcoming cohort, it is difficult to anticipate the material impact that this incremental level of expansion—which we believe to be modest—will have on the future supply of physical therapists. Perhaps most important, current data support the need for further expansion in the supply of physical therapist graduates to meet current and future employment demand. CAPTE’s most recent aggregate data show that there are approximately 11,331 graduates from accredited physical therapist education programs in the United States in 2020.29 However, the BLS data indicate 16,500 (15,733 FTEs) current employment openings for physical therapists annually.3 Even ignoring any employment growth over the next 10 years, only 67% of the current demand is being satisfied, resulting in the need for approximately 5200 additional graduates annually. If we conservatively assume that 90% of the 43 developing physical therapist education programs will eventually become accredited with an average class size of 45 students,30 these programs will still only yield an additional 1700 graduates over the next 7 to 10 years. A significant gap in the future supply relative to demand remains, even accounting for future growth. Analyzing the current class size data as reported by the vast majority of institutions who use the Physical Therapist Centralized Application Service (and inputting the average class size for those who do not), there will still be only 12,500 graduates annually by 2023.28 The same trend is true regionally, in particular for the Western part of the United States where less than 50% of demand is being satisfied (Table).3,28 A large gap between current demand and the supply of physical therapists exists even if one assumes the BLS data are overly optimistic. The available employment and education data suggest that demand for physical therapist employment in the United States is more likely to continue outpacing the growth in physical therapist education for at least the next 10 years. We believe that the current APTA Workforce Analysis4 is fundamentally flawed and significantly undermines the validity of at least the most recent ACAPT position statement, the conclusions of which are heavily based on the APTA Workforce Analysis.2 In fact, in our view, the appropriate officials should consider retracting or revising both the current APTA Workforce Analysis and the most recent ACAPT position statement that urges academic administrators not to develop new education programs or expand class size in current programs. At best, such a recommendation is an overreach, because no professional association (eg, APTA, ACAPT) or regulatory body (CAPTE) has the authority to limit the expansion of physical therapist education. At worst, this recommendation could harm the future of our profession if enough physical therapists are not available to meet an evolving and growing demand. Although not covered in this Point of View, we share ACAPT’s concerns regarding rising student debt, the shortage of qualified faculty, and the inadequate number of clinical sites in physical therapist education and recommend exploring innovative approaches to address these important challenges. Concept/idea/research design: J.D. Childs, L. Benz, A. Arellano, A. Briggs, M.J. Walker Writing: J.D. Childs, L. Benz, A. Arellano, A. Briggs, M.J. Walker Data collection: A. Arellano, A. Briggs, M.J. Walker Data analysis: J.D. Childs, A. Arellano, A. Briggs, M.J. Walker Project management: J.D. Childs, A. Briggs Consultation (including review of manuscript before submitting): J.D. Childs A. Arellano, A. Briggs, J. Childs, and M.J. Walker are employees of Evidence In Motion (EIM), which works with colleges and universities in the development and operation of hybrid, accelerated graduate health care programs (eg, physical therapy, occupational therapy, physician assistant). L. Benz is employed by Confluent Health, which owns and operates physical therapy and occupational therapy centers under various brand names and owns EIM and Fit For Work. The views, opinions, and content expressed in this Point of View are those of the authors and do not necessarily reflect the official policy or position of EIM or its client universities/colleges and their faculty.