Go back to SSA PROGRAM

Projected Supply, Demand, and Shortages
of Registered Nurses: 2000-2020 

July 2002
U.S. Department of Health and Human Services
Health Resources and Services Administration 
Bureau of Health Professions
National Center For Health Workforce Analysis

Index
National Supply and Demand Projections

• Chart 1: National Supply and Demand Projections for FTE RNs 2000-2020
• Map 1: States with Shortages of FTE RNs in 2000
• Map 2: States with Projected Shortages of FTE RNs in 2020

Projected Supply of RNs: Driving Forces and Trends

• Chart 2: Total Number of RN Graduates by Degree Program, 1995-2000
• Chart 3: Distribution of RN Graduates by Educational Preparation, 1992-1999
• Chart 4: Age Distribution of RNs, 1980, 2000 and projections for 2020
• Chart 5: New Entrants and Losses from the Licensed Pool of RNs for Selected Periods
• Chart 6: Licensed RNs Not Employed in Nursing, 1988-2000
• Chart 7: Actual Earnings of RNs and Elementary School Teachers and "Real Earning" for RNs, 1983-2000

Projected Demand for RNs: Driving Forces and Trends

• Chart 8: Population Projections 65 to 84 and 85 and Older, 2000-2050
• Chart 9: Per Capita Disposable Income in Constant (1996) Dollars, 1990-1998

Appendix

• Table 1: National Supply and Demand Projections for FTE RNs, 2000-2020 
• Table 2: Supply vs. Demand Projections for FTE RNs by State, 2000
• Table 3: Supply vs. Demand Projections for FTE RNs by State, 2005
• Table 4: Supply vs. Demand Projections for FTE RNs by State, 2010
• Table 5: Supply vs. Demand Projections for FTE RNs by State, 2015
• Table 6: Supply vs. Demand Projections for FTE RNs by State, 2020
• Table 7: Demand for FTE RNs by Employment Setting for Select Years, 2000-2020
• RN Demand Model
• Nursing Supply Model
  

Introduction

The Health Resources and Service Administration, Bureau of Health Professions, National Center for Health Workforce Analysis ("National Center") is the primary Federal agency responsible for providing information and analysis relating to the supply and demand for health professionals. In support of this role, the National Center has recently assumed responsibility from the Bureau's Division of Nursing for conducting the quadrennial National Sample Survey of Registered Nurses and developing supply and demand projections for registered nurses.  Supply and demand projections of registered nurses (RNs) for the period 2000 through 2020 are the subject of this report. Using data on supply trends drawn from the 2000 National Sample Survey of Registered Nurses, these new projections show that a shortage of FTE registered nurses, previously projected to begin around 2007, was already evident in the year 2000. 

This report consists of three sections. The first presents the National Center's projections of RN supply and demand, both national and State-specific, for the years 2000 through 2020. The second identifies and discusses the driving forces and trends underlying the projected supply; the third does the same for the projected demand.  Relevant charts, maps, and tables are presented both in the body of the report and in an Appendix.

National Supply and Demand Projections

In 2000, the National supply of FTE registered nurses was estimated at 1.89 million while the demand was estimated at 2 million, a shortage of 110,000 or 6 percent.  Based on what is known about trends in the supply of RNs and their anticipated demand, the shortage is expected to grow relatively slowly until 2010, by which time it will have reached 12 percent.  At that point demand will begin to exceed supply at an accelerated rate and by 2015 the shortage, a relatively modest 6 percent in the year 2000, will have almost quadrupled to 20 percent.  If not addressed, and if current trends continue, the shortage is projected to grow to 29 percent by 2020.  These numbers are shown in Table 1 of the Appendix.

The projected shortage in 2020 results from a projected 40 percent increase in demand between 2000 and 2020 compared to a projected 6 percent growth in supply.  Demand will grow steadily at a rate of 1.7 percent annually, a relatively modest growth rate when compared to the 2.3 percent annual growth in demand projected by the Department of Labor's Bureau of Labor Statistics.  Factors driving the growth in demand include an 18 percent increase in population, a larger proportion of elderly persons, and medical advances that heighten the need for nurses.  In contrast, the projected growth in supply is expected to reach a peak of only 10 percent by 2011 and then begin to decline as the number of nurses leaving the profession exceeds the number that enter.  (See Chart 1)

Distribution of the RN Shortage:

The national shortage of RNs is not evenly distributed across States.  In 2000, 30 States were estimated to have shortages. (See Map 1)  Because of uncertainties in the estimation process, only States with shortages greater than 3 percent are identified as shortage States.  Affecting the accuracy of State-specific supply and demand projections are situations where:  (a) a State may have many local shortages yet overall have an adequate supply; (b) significant numbers of nurses commute across State boundaries to work; (c) jurisdictions like Washington, D.C. provide service to large populations from outside its boundaries; and (d) demand projection methodology inadequately accounts for the differences in nursing utilization patterns that may exist between States.

Map 1: States with Shortages of FTE Registered Nurses in 2000

By 2020, 44 States and the District of Columbia are projected to have shortages.  (See Map 2)  State-specific projections for the years 2000 through 2020 are found in Tables 2, 3, 4, 5 and 6 of the Appendix.

Map 2: States with Projected Shortages of FTE Registered Nurses in 2020

Projected Supply of RNs-Driving Forces and Trends

An examination of factors affecting the supply of RNs sheds light on what has caused the current shortage. These factors include the declining number of nursing school graduates, the aging of the RN workforce, declines in relative earnings, and the emergence of alternative job opportunities.

Declining numbers of nursing graduates:

Data on the growth in new RNs, as measured by those passing the RN licensing test (NCLEX), show that after growing steadily during the first half of the 1990s the number of new RN graduates fell annually in the last half of the decade, resulting in 26 percent fewer RN graduates in 2000 than in 1995. Declines were seen across all degree programs--diploma, associate degree, and baccalaureate. The decrease in diploma graduates continues a trend driven in the past few decades by the closing of hospital-based diploma programs. In contrast, the declines in associate degree and baccalaureate graduates are a more recent phenomenon, having occurred only since the mid-1990s. (See Chart 2) Further, due to declines in enrollments over the past 5 years, no increase in the number of graduates is expected in the short term. Although the American Association of Colleges of Nursing (AACN) did report a 4 percent increase in baccalaureate enrollment between 2000 and 2001, the relatively longer educational pipeline for baccalaureate students increases the length of time before licensed RNs will emerge.

Associate degree graduates are declining at a somewhat faster rate than baccalaureate graduates, with the net result that baccalaureate graduates now comprise an increasingly greater share of total graduates. (See Chart 3)

This evolving shift from associate degree to baccalaureate-prepared RNs has, as noted earlier, a constraining effect on growth in supply.  Baccalaureate-prepared RNs may need twice as long to complete their education and enter the workforce as those graduating from associate degree programs, thereby increasing the length of time needed for the average RN student to enter the workforce, thus creating a temporary hiatus in the growth of supply

Aging of the RN Workforce:

The average age for RNs has climbed steadily in recent years resulting in a greater proportion of nurses in the older age brackets who are approaching retirement age.  Three factors contribute to this aging of the RN workforce:  (1) the decline in number of nursing school graduates, (2) the higher average age of recent graduating classes, and (3) the aging of the existing pool of licensed nurses.  Graduates of associate degree programs, the largest source of new RNs, are on average 33 years old when they graduate, considerably older than in 1980 when the average age of a new associate degree graduate was 28.  The result has been a significant decline in the proportion of RNs under the age of 30.  Between 1980 and 2000, that proportion declined from 25 percent to 9 percent. (See Chart 4) This slowing of new, young entrants coupled with an accelerating retirement rate for older RNs will produce a national supply of nurses that in 2020 will not only be older but no larger than the supply projected for 2005.  The number of new licenses in nursing is projected to be 17 percent lower in 2020 than in 2002, while the loss from the RN license pool due to death and retirement is projected to be 128 percent higher.

In addition to the slower rate at which RNs were added to the workforce in the last half of the 1990s, they now appear to be leaving the RN license pool, through death or retirement, at a faster rate than ever.  Over the four-year periods between the 1988 and 1992 surveys and the 1992 and 1996 surveys, the number of RNs leaving the license pool actually declined, from roughly 30,000 in the first four-year period to 23,000 in the second.   Balancing these losses with the number of new graduates in each of these periods raised the RN license pool in those periods by 10 percent and 14 percent respectively.  However, between the 1996 and 2000 surveys, the loss of RNs from the license pool increased six- to seven-fold, to nearly 175,000. (See Chart 5) Balancing this most recent set of losses with new graduates resulted in an increase in the RN license pool of only 5 percent, one-third to one-half the increase seen in the earlier surveys.  If current projections hold, the situation will worsen as the number of losses approaches and then exceeds the number of new entrants.   After balancing projected losses against projected new entrants, the RN supply is projected to grow 1.3 percent between 2008 and 2012, and by the end of the projection period, to decline by 1.9 percent between 2016 and 2020.

In addition to the number of RNs who give up their license, there are currently almost half-a-million licensed nurses not employed in nursing.  Between the 1996 and 2000 surveys, the number of licensed RNs not employed in nursing grew by 52,000 to over 490,000 (See Chart 6). Unfortunately, little is known about this population.   However, what is known is that 69 percent, or 338,000, of the 490,000 licensed RNs not employed in nursing in 2000 were 50 years or older.   Further, analysis of data from the 2000 RN Sample Survey shows that only 7 percent of the licensed RNs not employed in nursing were actively seeking employment in nursing.

Declines in Relative Earnings:

Salaries are likely playing a role in the declining supply of RNs.  While actual earnings for RNs increased steadily from 1983 through 2000, "real" earnings -- the amount available after adjusting for inflation -- have been relatively flat since 1991.  Thus, on average, RNs have seen no increase in purchasing power over the last 9 years. (See Chart 7) In contrast, the average salary for elementary school teachers has always been greater than that for RNs and is growing at a faster pace. In 1983, the average elementary school teacher earned about $4,400 more than the average RN; by 2000 this had grown to the point where elementary school teachers earned about $13,600 more.

Furthermore, a good portion of the wage growth for these nurses appears to occur early in their careers, then taper off with time.  In 2000, staff RNs employed full-time in nursing, who graduated 5 years earlier, typically earned wages 15 to17 percent higher than those newly entering the field, depending on basic nursing preparation, but only 1 to 3 percent less than nurses who graduated 15 to 20 years earlier.  As their potential for increased earnings diminishes over time, staff nurses may be motivated to leave patient care for additional education and/or other careers in nursing or outside the profession.

Projected Demand for Registered Nurses: Driving Forces and Trends

The major factors and trends behind the growth in RN demand include: population growth, aging of the population, increased per capita demand for health care, and trends in health care financing.

Population Growth and Aging:

The changing demographic nature of the population is a critical factor affecting demand for RNs.  Recent projections show the Nation's population will grow 18 percent between 2000 and 2020, resulting in an additional 50 million people who will require health care.  Much of this population growth can be attributed to advances in science and medicine that have increased life expectancy and resulted in a higher proportion of the population being over the age of 65, a significant source of demand for RNs.

In contrast to the 18 percent growth in overall population, the subgroup 65 years old and older is projected to grow 54 percent between 2000 and 2020, adding 19 million people to the 65-and-over age group. (See Chart 8)  While this amount is less than the 31 million added to those under 65, individuals in the upper age brackets contribute disproportionately to health care spending, spending over 3 times as much on average as those under 65.  Individuals 65 and over have a high incidence of chronic conditions such as: arthritis (50 percent), hypertension (36 percent), and heart disease (32 percent).  Many also have multiple conditions requiring more regular care.  The result is a population that currently has twice as many contacts with a physician as those under 65, accounts for 13 percent of the population but 38 percent of hospital discharges, and has annual per capita health care expenditures of $5,400 compared to $1,500 for those under 65. The greatest per capita demand for health care, and thus the services of RNs, will quite naturally come from the very old, those 85 and over.  This is the fastest growing segment of the population and a major user of long-term care facilities, home health care, and other employers of RNs.

Trends in Health Care Financing:

Demand for health care services, and by extension for registered nurses, is also driven by the ability to pay for health care, either with insurance or through out-of-pocket expenditures.  Since 1990, an average of 85 percent of the population has been covered by some form of health insurance, making health care available to the vast majority of the population. At the same time, real per capita disposable income has increased steadily, growing 16 percent between 1990 and 1999, making it easier to pay for non-covered health care with out-of-pocket resources, thereby increasing the demand for such care. (See Chart 9)

Distribution of Demand by Setting:

Between 2000 and 2020 the demand for RNs will continue to grow in all employment settings, but some will grow more rapidly than others, resulting in changes in the distribution of demand by setting.  Hospitals have been and will continue to be the major source of demand for RNs but while the total number of nurses in hospitals will continue to grow, the hospital sector's share of total RN employment will remain stable at about 62 percent.  (See Appendix Table 7) Employment settings closely associated with service to the elderly are projected to increase their share of the total demand for RNs.  For example, the demand for RNs in nursing homes is projected to increase from 8 percent of total demand in 2000 to 10 percent in 2020.  Similarly, growth in the home health care sector will result in an increase in demand for RNs from 6.5 percent to 9 percent of total RN demand.   These increases will naturally be offset by a corresponding decline in the proportion of demand in ambulatory and "other" settings.

APPENDIX

Table 1: National Supply and Demand Projections for FTE Registered Nurses; 2000 through 2020

Limitations of Supply, Demand, and Shortage Projections:

Supply and demand projections at the State level should be carefully assessed.  Estimates of supply and demand by State are to some extent biased by national averages and arbitrary State boundaries.  In particular, demand may be underestimated for States that are rural and mountainous, have higher than average elderly populations, and utilize more RNs than the national average in their health care systems.  Because the model averages the nurse staffing across 50 States, the projected demand increases for those States may be underestimated.  Conversely, for States that utilize lower RN staffing ratios than the national average, demand estimates may be overestimated.

Also, State boundaries, especially for small States may bias the demand and supply estimates.  Hospitals in the District of Columbia (DC), for example, serve Maryland and Virginia.  Thus, estimating demand for RNs, based on current population in DC alone may underestimate the demand for RNs there.  On the supply side, the same bias exists.

Table 2: Supply versus Demand Projections for FTE Registered Nurses by State--2000

* No definitive shortage. Estimate is -3% or above.

Table 3: Supply versus Demand Projections for FTE Registered Nurses by State--2005

* No definitive shortage. Estimate is -3% or above.

Table 4: Supply versus Demand Projections for FTE Registered Nurses by State--2010

* No definitive shortage. Estimate is -3% or above.

Table 5: Supply versus Demand Projections for FTE Registered Nurses by State--2015

* No definitive shortage. Estimate is -3% or above.

Table 6: Supply versus Demand Projections for FTE Registered Nurses by State--2020

* No definitive shortage. Estimate is -3% or above.

Demand for Full-Time Equivalent RNs by Employment Setting for Select Years 2000-2020

Table 7: Employment Distribution by Setting

Percentage Distribution by Setting

Registered Nurse Demand Model (NDM)

The model used to project the demand for registered nurses is the Nursing Demand Model (NDM).  The NDM combines empirical analysis with input from healthcare experts regarding how the health care system operates and the role of nurses in the delivery of care.

The NDM forecasts future demand for healthcare services in six settings:  inpatient, outpatient, and emergency departments in general and short-term hospitals; non-general and long-term hospitals; nursing facilities; and home health.

The workload measures used to model demand for healthcare services are inpatient days (for inpatient settings in short-term hospitals and for long-term hospitals); visits (for short-term hospital outpatient and emergency department settings, and for home health); and nursing facility residents.  Data limitations prevented the modeling of State-level estimates of the demand for ambulatory care in physician offices, occupational settings, at schools, and in public health clinics.

To forecast future healthcare demands of the population and where patients are likely to receive services, the NDM first divides the population into 32 subgroups based on age, sex, and urban or rural location.  For each subgroup, the current national per capita use of health care services was estimated for each of the six delivery settings.  The NDM then applies these national rates to each State's population to obtain initial estimates of demand for healthcare services at the State level.  Then, the NDM multiplies this initial extrapolation of healthcare utilization with a scalar that inflates or deflates the initial extrapolation to incorporate the projected impact on utilization of trends in the healthcare operating environment, economic conditions, and the overall health of the population.

To forecast the level of nursing services that patients will require and the mix of nurses to provide those services, the NDM forecasts future nurse staffing intensity.  Staffing intensity is defined in terms of FTE nurses per unit of healthcare services provided for the six delivery settings where future healthcare use is estimated (e.g., FTE RNS per 1,000 inpatient days at short-term hospitals).  Staffing intensity is defined in terms of FTE nurses per population in the delivery settings where future healthcare use is not estimated (e.g., FTE RNs in public health per 10,000 population.)  Forecasts of future nurse staffing intensity are based on extrapolations of current staffing intensity and take into account projected trends in the healthcare operating environment, in acuity, and in economic conditions that affect staffing intensity.

Twenty-four equations in the NDM quantify the relationship between healthcare utilization and its determinants and between nurse staffing intensity and its determinants. Twenty of these equations were estimated using multiple regression analysis with State-level data from 1996 through 2000, although most regression equations were estimated using a subset of these years based on data availability.  Six equations describe the relationship between healthcare utilization and its determinants for the six healthcare settings modeled.  Eighteen equations describe the relationship between nurse requirements and its determinants for each nurse type in each setting modeled.  In four settings, poor data or poor regression results led to the decision to model future requirements using an alternative approach.  RN and LPN requirements in the "all other" category are modeled as a fixed ratio of FTE nurses per population.  RN requirements in the "school health" category are modeled as a fixed ratio of  FTE RNs per population age 5-17.  RN requirements in nurse education are modeled as fixed percentages of total RN requirements in the other 11 settings.

Nursing Supply Model

The model used to project the supply of registered nurses captures the age-specific dynamics of the flow of nurses in and out of licensure and the work force, their progression from one educational level to an-other, and their State-to-State migration. Data are developed for each State on: (1) the population of nurses - all those with licenses to practice on a given date; (2) the supply - all those employed or available for employment (if sufficient positions are not available at the time being considered); and (3) the full-time equivalent supply that expresses employment independently of full- or part-time status.  The United States summary is an aggregation of the State data. 

Projections of first-time licensees from United States schools are derived from projections of graduates.  For these current projections of graduates from each of the three main types of basic nursing programs - diploma, associate degree, and baccalaureate - statistical regression models were used to determine those factors most likely to predict the numbers of graduates.  Historical data on graduations, from the National League for Nursing annual surveys of schools of nursing, provided the trend information on the number of graduates from each type of program.  Independent variables representing the availability and attractiveness of nursing as a career were considered for each of the models.  These included measures of job availability, salaries, and health care expenditures.  Also considered were data on the female population within the age groups found among nursing students and on the proportion of female high school graduates enrolling in higher educational institutions.

In order to capture the changes brought about by registered nurses taking additional education after their initial entry into nursing, statis-tical regression models were developed to project annual graduations from post-RN baccalaureate and master's degree programs.  National League for Nursing annual survey data provided the historical data on graduations used in the models.  Projections of graduations from post- RN baccalaureate programs were based on graduations from basic AD and diploma programs and data from the National Sample Survey of Registered Nurses on the length of time between when nurses gradu-ated from the generic program and the baccalaureate program. Similar variables on prior education graduations and intervening time were used in projecting post-RN master's degree graduates.

New licensees represent additions to the nurse population.  Of critical importance to determining the overall nurse population in any one year is a measure of those who no longer have licenses as registered nurses.  Most of the population consists of those who have entered nursing over a period of years and have continued their licensure.  Losses to the RN population occur through death, failure to renew licenses, or in some instances, disciplinary action that would revoke all licenses.  The model accounts for deaths by using measures derived from the life tables of white females developed by the National Center for Health Statistics.  However, there is no direct measure of nurses who failed to renew their licenses.  Therefore, "net losses" are derived.  They meas-ure the net change in the RN population that is caused by those who drop all licenses minus those RNs who become re-licensed after having dropped all their licenses.

To identify the supply of registered nurses, "activity rates" are devel-oped using data from the National Sample Surveys of Registered Nurses on the proportion of the nurse population within a specific age group that is employed in nursing.

The derivation of the full-time equivalent RN supply was dependent upon data in the March 2000 National Sample Survey of Registered Nurses that identified the nurses working full-time and those working part-time and the number of scheduled hours for each nurse.  A ratio was developed of the number of nurses within each age group cohort who were working on a full-time basis plus the full-time equivalent of those working on a part-time basis to the total number of working nurses.  The full-time equivalent for part-time nurses was determined by the ratio of average scheduled hours of part-timers to average scheduled hours of full-timers.

Go back to SSA PROGRAM