Projected Workforce of Psychiatrists in the United States: A Population Analysis

The projected shortage of physicians in the United States is garnering attention in academic circles, the media, and among health care providers. As of April 2016, the total physician shortfall is projected to be between 61,700 and 94,700 by 2025 (1). Although much attention is focused on primary care and surgical (2,3) specialties, recent data provide evidence for a significant shortage in the specialty of psychiatry. A recent, in-depth analysis of the projected shortage has not been conducted. Previous estimates of the current shortage of psychiatrists vary considerably, ranging from 2,800 (4) to 45,000 (5). To address the shortage in light of the growing U.S. population and the fact that mental illness has become the most costly condition in the United States, at $201 billion annually (6), the estimated shortage must first be quantified. In this review, our objective was to examine the current workforce of psychiatrists, provide an estimate of the projected workforce given the wide range of published estimates, and begin a discussion for addressing the workforce needs of the future.

Background

Increasing Demand for Services as a Source of the Shortage of Psychiatrists

The shortage of psychiatrists has developed for numerous reasons and over decades. First, the United States has a growing population with an already significant prevalence of mental illness, which indicates that the total number of individuals with mental illness is likely to grow. The 2016 U.S. population of approximately 324 million individuals will grow at a rate between .45% and .82% per year (7). Centers for Disease Control and Prevention data for 2004 showed that 25% of U.S. adults reported having had a mental illness in the previous year (8). The lifetime prevalence of mental disorders, as defined by DSM-IV, has been estimated at 46.4% of the population (9). To put this in perspective, the percentage of Americans with diabetes is approximately 9.1% (29 million individuals) (10), and the percentage of adults with hypertension is approximately 29% (70 million individuals) (11). Given the growing population, it can be expected that the total number of individuals requiring mental health treatment will increase. Perhaps more alarming is the fact that in 2015 only 43.1% of adults with a mental illness received mental health services (12). However, lessening of stigma related to mental illness and treatment may contribute to more treatment seeking (13) in the coming years.

Limited Supply of Providers as a Source of the Shortage of Psychiatrists

Despite the growing demand for services, the median number of psychiatrists per 100,000 U.S. population declined 10.2% (14) between 2003 and 2013. The degree of the shortage varies across the country; 96% of counties had an unmet need for prescribers, and 77% of counties had a severe shortage of mental health providers, either prescribers or nonprescribers (15). Exacerbating this decline in the number of practicing psychiatrists is the fact that in 2011, 55% of psychiatrists were age 55 or older, compared with 37.6% of other practicing physicians (16), which leads to the expectation that approximately half of the existing workforce is approaching retirement. The Balanced Budget Act of 1997 has not increased the number of federally funded residency positions, and U.S. psychiatry residency programs are not producing enough psychiatrists to keep up with population growth and the expected rate of retirement. The number of categorical psychiatry residency positions recently returned to the 1990 level of 1,136 in 2012 (1,117 positions) (17). Between 2000 and 2010, the number of U.S. psychiatrists increased less than 6% while the number of U.S. physicians increased nearly 20% (18). In addition, the fact that just 55%−60% of psychiatrists accept insurance (4,14) contributes to maldistribution of services and limits access to mental health care.

Historical Estimates of the Shortage

Efforts have been made to estimate demand for mental health services, with the goal of identifying the number of psychiatrists needed to provide services at a given point in time; workforce projections, however, have not been created. Many have utilized a psychiatrist-to-population ratio to ascertain whether a shortage or surplus was present. A 1980 Department of Health and Human Services (DHHS) report documented that 15.4 psychiatrists were needed for every 100,000 people, inclusive of adults and children (19). This model estimated the required psychiatrist-to-population ratio on the basis of the prevalence of mental illness and the estimated annual provider time required per patient.

Konrad and colleagues (20) based their estimate of the 25.9 psychiatrists required per 100,000 adults on 2001 National Comorbidity Survey Replication (NCS-R) data. Their model estimated demand by combining required provider time per patient with modeled national and county-level mental health prevalence estimates. An estimate of the number of full-time-equivalent (FTE) psychiatrists was then calculated from the results, indicating that 56,462 FTE prescribers were required in 2009 for the 218 million U.S. adults in order to maintain services at the level provided in 2001, the year of the data. On the basis of the NCS-R target of 25.9 psychiatrists per 100,000 adults, Carlat (5) estimated the shortage of psychiatrists in 2010 to be 45,000.

Finally, 2016 research by the Health Resources and Services Administration (HRSA) indicated that the United States lacked just 2,800 FTE psychiatrists (4). It is important to note that this estimate was based on the number of psychiatrists that would be needed to remove an area from the list of those designated as health care professional shortage areas for mental health, a much different approach than that taken in the DHHS report and by Konrad and colleagues. Given that the results driven by DHHS and NCS-R data were based on the prevalence of illness and actual required provider time, demand likely lies somewhere between these ratios.

Using census, physician supply, and board certification data, we sought to clarify the extent of the projected shortage through 2050 and reconcile this with other published estimates.

Methods

Data Sources

Association of American Medical Colleges (AAMC) 2000–2015 data were accessed to obtain the number of psychiatrists in the workforce. American Board of Psychiatry and Neurology (ABPN) 2000–2015 data provided the number of new board certifications granted. U.S. Census Bureau data provided historical and projected 2000–2050 population data.

Data Analysis

The psychiatrist workforce through 2050 was projected by beginning with AAMC data that quantified the number of psychiatrists through 2015 and accounting for additions to and subtractions from the workforce by year. Projected additions to the workforce of psychiatrists were based on a five-year (2011–2015) average of ABPN board certification data, equating to 1,798 certifications annually. Data modeling assumed that psychiatrists entered the workforce at age 30 after obtaining ABPN board certification and after completing a four-year residency program (ages 26–30) and a four-year medical school (ages 22–26). From 2015 AAMC data, the estimated percentage of total psychiatrists over age 55 was calculated to be 56.98% for 2015, equal to 26,479 total psychiatrists over age 55 and 19,993 under age 55. Projected subtractions from the psychiatrist workforce were based on a median age at retirement of 65 (21). Separately, U.S. Census Bureau population data were combined with the established psychiatrist-to-population ratios described above to determine estimates of the required number of psychiatrists. The projected workforce and required workforce were then compared to evaluate the shortage by year.

Results

Initial Period of Workforce Contraction

The psychiatrist workforce will contract through 2024, to a low of 38,821 total psychiatrists. The inflection point will occur as a direct result of the retirement in 2024 of the last of the 26,479 psychiatrists who are currently over age 55. This represents the final year in which the number of retirees is expected to surpass the number of newly board-certified psychiatrists entering the workforce (Table 1). In 2024, there will be 11.3 psychiatrists per 100,000 population and 14.4 psychiatrists per 100,000 adults, which equates to an estimated shortage of between 14,280 and 31,091 psychiatrists, on the basis of the established ratios determined by the DHHS and NCS-R data noted above.

Workforce Expansion

Beginning in 2025, the psychiatrist workforce will begin a gradual expansion, because the expected annual number of granted ABPN board certifications in psychiatry (N=1,798) will supersede the expected number of annual retirees (N=800) (Table 1). Table 2 shows a projected growth by nearly 10,000 psychiatrists per decade during this period of expansion. Projections indicate that by 2050, there will be 16.3 psychiatrists per 100,000 population and 20.3 psychiatrists per 100,000 adults. This equates to a range between a surplus of 3,428 and a shortage of 17,705 psychiatrists, depending on the psychiatrist-to-population ratio used (Table 2 and Figure 1).

Discussion

We aimed to quantify and assess the projected workforce of U.S. psychiatrists through 2050. The analysis led to several important findings. First, in the absence of actions to address it, the shortage of psychiatrists will worsen before it improves. With 55% of the total psychiatrist workforce expected to retire within the next ten years, the number of psychiatrists leaving the workforce exceeds the number entering by a multiple of two. The contraction of the workforce will continue through 2024. Second, there will be an inflection point at the end of 2024 wherein this trend reverses, because the large cohort of pending retirees will have left the workforce. At that time, there will be an estimated 38,821 psychiatrists and a shortage of 14,280–31,091 psychiatrists, depending on which established psychiatrist-to-population ratio is used (Figure 1). Third, beginning in 2025, the supply of psychiatrists will begin to expand because the flow of new psychiatrists into the workforce will exceed the number exiting due to retirement. This entails a gradual process, however. For example, a return to the estimated 2017 shortage of psychiatrists (5,529–20,701) is not expected to occur until approximately 2039.

The inflection point marked by the end of 2024 is critical because it delineates two time frames to target in addressing the shortage: the present through 2024 and 2025 onward. Some strategies may be implemented with clear intent to curb the extent of the shortage at the inflection point, whereas others may aim to curtail the shortage over the next 30 years. As an example, a strategy such as increasing the number of psychiatry residency positions would be expected to build the workforce over decades given the time required to attract and train psychiatrists. In contrast, innovative models of care (for example, collaborative care), an increase in prescribers (for example, nonphysician behavioral health care providers), or use of technologies (for example, telepsychiatry and new interventions) may affect workforce needs prior to 2024.

Fourth, given that these projections extend more than 30 years into the future, it is difficult to determine whether the shortage will be completely alleviated by 2050 without any intervention. Projections based on the proposed DHHS ratio indicate a minimal surplus by 2050, whereas Konrad colleagues’ proposed ratio would result in a continued shortage. Fifth, it is important to note that when viewed both in isolation and in the context of previous studies on this topic, these results are consistent with the existence of a pending and significant shortage of psychiatrists in the United States. The precise amount of the shortage, however, varies among studies. Compared with the 2016 U.S. DHHS HRSA study (22), our 2013 data show a similar number of psychiatrists (45,580 and 45,533, respectively) and project a similar number of psychiatrists entering the workforce by 2025 (20,470 and 22,947, respectively). However, our data project fewer psychiatrists retiring through 2025, which accounts for the differing projections of supply of psychiatrists (45,210 in the DHHS HRSA study and 39,819 by our estimate) as well as for the overlapping but different projections of the shortage in 2025 (6,080–15,400 and 13,671–30,712, respectively) (22).

As with any set of projections, this analysis was limited by the assumptions built into the model. We assumed that the number of new entrants into the workforce each year was equivalent to the number of newly granted ABPN psychiatric board certifications. We acknowledge that physicians who are not board certified may operate as psychiatrists and thus increase the effective workforce. However, 2010 data showed that 75% of physicians with an active license were in fact certified by one or more American Board of Medical Specialties boards and that this percentage was “dramatically” higher (to 86%) among physicians age 30 and older as they enter the workforce (23). The projected number of annual ABPN psychiatric board certifications was also held constant in our model. On the basis of the literature, it was also assumed that psychiatrists would retire at age 65; however, some psychiatrists instead may elect to reduce their commitment to practice over time. U.S. Census Bureau population projections were also assumed to be accurate. Finally, we assumed that the psychiatrist-to-population ratio will remain constant. However, we acknowledge that changes in models of care, demand for new services, new technologies, treatment innovations, and payment and policy reforms could alter the psychiatrist-to-population ratios used in this model.

Conclusions

Because of steady population growth and the expected retirement of more than half the current psychiatrist workforce, the workforce of U.S. psychiatrists will continue to contract through 2024—to a projected shortage of 14,280–31,091 psychiatrists—if no interventions are implemented. Beginning in 2025, the workforce will begin a gradual process of expansion, but it is unclear whether the shortage will have completely resolved by 2050. Future research should focus on developing strategies to address this quantified shortage, in an effort to curb the worsening shortage through 2024 and to address workforce needs in the coming decades.