washington state’s pharmacist workforce through 2020...
TRANSCRIPT
February 2004
Washington State’s Pharmacist
Workforce through 2020:
Influential Factors and
Available Data
Working Paper #90
by
Davis G. Patterson, MA
Susan M. Skillman, MS
L. Gary Hart, PhD
This research was funded by Grant #1 U79 HP 00007-01 from the HRSA Bureau of Health Professions, NationalCenter for Health Workforce Analysis. The authors thank Alice Porter for her superb editing and Seth Blades forvaluable research assistance.
UNIVERSITY OF WASHINGTON • SCHOOL OF MEDICINE • DEPARTMENT OF FAMILY MEDICINE
2
ABOUT THEWORKFORCE CENTER
The WWAMI Center for Health Workforce Studies atthe University of Washington Department of FamilyMedicine is one of six regional centers funded by theNational Center for Health Workforce Analysis(NCHWA) of the federal Bureau of Health Professions(BHPr), Health Resources and Services Administration(HRSA). Major goals are to conduct high-qualityhealth workforce research in collaboration with theBHPr and state agencies in Washington, Wyoming,Alaska, Montana, and Idaho (WWAMI); to providemethodological expertise to local, state, regional, andnational policy makers; to build an accessibleknowledge base on workforce methodology, issues,and findings; and to provide wide dissemination ofproject results in easily understood and practical formto facilitate appropriate state and federal workforcepolicies.
The Center brings together researchers from medicine,nursing, dentistry, public health, the allied healthprofessions, pharmacy, and social work to performapplied research on the distribution, supply, andrequirements of health care providers, with emphasison state workforce issues in underserved rural andurban areas of the WWAMI region. Workforce issuesrelated to provider and patient diversity, providerclinical care and competence, and the cost andeffectiveness of practice in the rapidly changingmanaged care environment are emphasized.
ABOUT THE AUTHORS
DAVIS G. PATTERSON, MA, is a Graduate Research Assistant in the WWAMI Center for Health WorkforceStudies, Department of Family Medicine, University of Washington School of Medicine.
SUSAN M. SKILLMAN, MS, is the Deputy Director of the WWAMI Center for Health Workforce Studies,Department of Family Medicine, University of Washington School of Medicine.
L. GARY HART, PhD, is Director of the WWAMI Center for Health Workforce Studies and Professor in theDepartment of Family Medicine, University of Washington School of Medicine.
The University of Washington Rural Health andHealth Workforce Research Center Working PaperSeries is a means of distributing prepublication articlesand other working papers to colleagues in the field.Your comments on these papers are welcome andshould be addressed directly to the authors. Questionsabout the WWAMI Center for Health WorkforceStudies should be addressed to:
L. Gary Hart, PhD, Director and Principal InvestigatorSusan Skillman, MS, Deputy DirectorRoger Rosenblatt, MD, MPH, Co-InvestigatorLaura-Mae Baldwin, MD, MPH, Co-InvestigatorDenise Lishner, MSW, Center Research CoordinatorEric Larson, PhD, Senior ResearcherHeather Deacon, Program CoordinatorMartha Reeves, working paper layout and productionUniversity of WashingtonDepartment of Family MedicineBox 354982Seattle, WA 98195-4982Phone: (206) 685-6679Fax: (206) 616-4768E-mail: [email protected] Site: http://www.fammed.washington.edu/CHWS/
The WWAMI Center for Health Workforce Studies issupported by the Bureau of Health Professions’National Center for Health Workforce Information andAnalysis. Grant No. 1U76-MB-10006-03; $250,000;100%.
3
Washington State’s PharmacistWorkforce through 2020:
Influential Factors and Available DataDAVIS G. PATTERSON, MASUSAN M. SKILLMAN, MS
L. GARY HART, PhD
EXECUTIVE SUMMARY ANDQUESTIONS FOR REVIEWThis report describes the efforts of the University ofWashington Center for Health Workforce Studies toidentify trends in Washington’s pharmacist workforce.We based our analysis state licensing data, hospitalstaffing data, educational completions data, andpopulation census data. From these sources, wedeveloped models to project supply and demand forpharmacists through the end of the next decade.
In common with other states, Washington isexperiencing a shortage of pharmacists. Our modelssuggest that the current shortage of pharmacists islikely to continue until at least 2008, at which time theprojections diverge. One model projects a growingsurplus; the other projects a worsening shortage. Themodel projections, however, should be considered inlight of the limitations both in reliable data and readilyavailable supporting literature. We offer theseprojections for discussion and critique as anopportunity to explore possibilities for improving datasources and our understanding of this issue.
IMPORTANT FACTORS AFFECTINGTHE PHARMACY WORKFORCESeveral factors may affect the future supply anddemand of pharmacy services, including the following:
• Both pharmacy school applicants and graduatesdeclined in the late 1990s, but increased statefunding has recently expanded enrollment capacity.
• Improvements in information technology andautomation of prescription fulfillment will increasethe productivity of dispensing pharmacists,effectively increasing supply.
• The recent transition to the required Doctor ofPharmacy degree requires more existing pharmaciststo educate new ones, reducing the supply ofpharmacists providing patient services. The longertraining time may also reduce supply by makingentry into the profession more difficult.
• A chronic shortage is likely to increase jobdissatisfaction and exits from the profession asworking conditions become more stressful and lessflexible.
• Women represent an increasing proportion of thepharmacist workforce. Women’s professional livestend to be shorter than men’s, reducing the averagesupply of services provided per pharmacist.
• An aging workforce, combined with an agingpopulation needing more services, could create ashortage of pharmacists.
• Pharmacist roles are expanding from dispensing ofprescriptions to increasing involvement in activitiesto improve patient care, increasing total pharmacistdemand.
• New and increasingly complex drug therapiesincrease the demand for pharmacist services indispensing, patient education, and monitoring.
• Insurance coverage of prescription drugs isexpanding, leading to increased demand.
RESEARCH APPROACH ANDLIMITATIONSTo model pharmacist supply and demand inWashington State, we used four principal data sources:(1) 1998-1999 state licensing data and asupplementary licensing survey from the WashingtonState Department of Health Office of HealthProfessions Quality Assurance, (2) a 2002 study ofstaffing in nonfederal acute care hospitals by theUniversity of Washington Center for Health WorkforceStudies and the Washington State HospitalAssociation, (3) educational completions data forpharmacy programs in the state from 1996 to 2003,and (4) U.S. Census Bureau state population data.
These are the best data available for Washington, butthey are missing critical information needed formaking accurate workforce projections. For example,
4
data are not available on job turnover, providermigration in or out of the state, and exits from theprofession. We had to model total supply and demandfor all sectors by extrapolating from hospital databecause nonhospital data on pharmacists were notavailable. Our assumptions about changes inWashington’s total population and educationalcapacity are probably oversimplified. We are not ableto predict or quantify future changes in the state’shealth services delivery system and health policy. Ouranalysis of aggregate state supply and demand couldmask critical shortages in specific regions andcommunities of Washington State. In addition,projections of the relatively small pharmacistworkforce are more volatile than are projections forlarger workforces (pharmacists number in the 4,000s,compared with nursing, for example, in the 50,000s).
RESULTSThis report shows one method of projectingpharmacist workforce demand and two alternativemethods of projecting supply. The same demandmodel is compared with each supply model to generatetwo scenarios assessing the balance between supplyand demand. All values reported represent persons, notpositions or FTEs. Following are descriptions of thesemodels and their projections:
Demand Model: We used state total populationprojections from the U.S. Census Bureau and hospital-sector vacancies to model demand. We extrapolatedfrom hospital employment and vacancies to estimatetotal state employment and vacancies (both hospitaland nonhospital sectors). There were 514 vacancies in2003, a 10.7 percent shortfall of pharmacists in thestate. The model projects an annual increase indemand, based on population growth, ranging from 59to 63 providers through 2020.
Supply Model I: This model estimates future supplyas a function of recent trends in state licensing ofpharmacists, supplemented by data from two surveys.Supply Model I projects increases of 164 employedpharmacists per year. Supply increases relative todemand, with equilibrium around 2008, and eventuallyoutstrips demand by 20.8 percent in 2020.
Supply Model II: This model estimates future supplyas a function of educational completions and providerretirements, also supplemented by data from twosurveys. Model II assumes that recent expansions ofeducational capacity will be sustained, yielding 154 to155 new graduates annually beginning in 2007. Butprojected retirements and increased demand caused bypopulation growth more than offset this expansion.Unlike Supply Model I, Supply Model II projectsdecreases in supply relative to demand. The shortageof pharmacists increases to 30.2 percent in 2020.
QUESTIONS FOR REVIEWOur models project two divergent trends in the supplyof Washington State’s pharmacists—a future surplus ora shortage. These models were developed with verylimited data. Before such projections can be used toinform policy, they must be reviewed by stakeholdersfamiliar with the environment in which this workforceoperates. These stakeholders can provide subjectiveassessments of how the profession is likely to changewhere quantifiable data do not exist currently, and theycan generate estimates about how these changes mayaffect workforce supply and demand. Below are somequestions for which we seek stakeholder input. Thislist is not exhaustive, and we welcome additionalinsights regarding influential factors and useful trenddata.
(1) Is Washington heading for a shortage or a surplusof pharmacists after 2008? How realistic are thefuture demand estimates in this report, which arebased solely on state population growth?
(2) What are the pressures facing the educationalpipeline to pharmacy in Washington State?
(3) How many of Washington’s pharmacists weretrained out of state? How many of those trainedin Washington stay here to work? What is the netimpact on supply?
(4) This report makes no distinctions betweendifferent types of pharmacy services. How do theprospects for different areas of pharmacy practice(e.g., order fulfillment, primary care, secondaryand tertiary services) differ?
(5) How do the hospital and nonhospital sectorscompare? When only hospital vacancy rates areavailable and these are used to estimatenonhospital vacancies, what kind of error (if any)is introduced?
(6) How equitably are pharmacists distributedthroughout the state? Are there area shortages orsurpluses? What are differences by sector orfacility type?
(7) How will new pharmacy technology and newdrug therapies affect supply and demand?
(8) How can we obtain more recent and accurate datato assess the current pharmacist workforce? Whatare practical long-term strategies for creating thedata needed to monitor pharmacist supply anddemand on an ongoing basis?
(9) What new state and federal policies may changepharmacist supply and demand?
(10) Will economic changes (e.g., recession) causepopulation demand for care to increase ordecrease substantially during the next decade?
5
INTRODUCTIONThe Bureau of Health Professions forecasts a 14.5percent increase in the number of pharmacistsnationally from 2000 to 2010 (Bureau of HealthProfessions, 2000). Applications and enrollments inpharmacy schools declined in the late 1990s, while thedemand for pharmacists continues to rise with anaging population, new drug therapies, and changingpharmacist roles (Bureau of Health Professions, 2000).The supply of pharmacists varies widely from state tostate, but there is evidence for a shortage currentlynationally and locally (Bureau of Health Professions,2000; Knapp, 2002; Skillman et al., 2003).
How will the supply and demand of pharmacistschange in the state of Washington, and what factorswill affect the pharmacist workforce? We reviewedliterature on the pharmacist workforce and analyzedexisting data in an attempt to answer these questions.We were able to identify numerous trends in pharmacypractice and health care, but because of seriouslimitations in the availability of data for WashingtonState, we are able to offer only rudimentary andtentative answers to these questions. The overridingmessage of this exercise is that we need much moredata just to understand the current state of affairs, andprojections of future supply and demand should beviewed as exploratory rather than predictive, subject toa number of influential trends that we have few or nodata to quantify.
FACTORS AFFECTINGSUPPLY AND DEMANDOF PHARMACISTSSupply refers to the number of pharmacy services thatcan be provided. Supply is affected over time either bychanges in the number of providers or changes in theconditions of service provision. For example, an agingworkforce decreases the supply of providers—andtherefore the supply of services—through deaths andretirements. Increases in productivity—the unit ofoutput per unit of input—increase the supply. Anincrease in the supply of services does not necessarily
mean an increase in the number of persons providingthose services. For example, new technology thatallows more prescriptions to be dispensed per full-timeequivalent provider, or FTE, causes an increase in thetotal supply of services.
Demand refers to the actual number of pharmacyservices that the population is willing and able to payfor, regardless of financing or whether services arenecessary. Population growth and population aging, allother things being equal, lead to a higher total burdenof disease and thus a higher demand for health careservices.
The pharmacy profession is undergoing rapid changesthat make predicting the future difficult even if datawere available for a perfect reading of the currentsituation. A recent state study showed that pharmacistsare one of the most difficult kinds of personnel forhospitals to recruit (Skillman et al., 2003), echoingincreasing vacancy rates nationally (Bureau of HealthProfessions, 2000).
Table 1 shows the most likely factors that will affectthe future supply and demand of pharmacists.
It is evident from this table that there arecountervailing forces acting on supply levels. At thesame time, all signs point to increasing demand forpharmacist services. A brief explanation of theseforces follows:
Increases in Educational Capacity: Both pharmacyschool applicants and graduates declined nationally inthe late 1990s (Bureau of Health Professions, 2000),but increased state funding has recently expandedenrollment capacity (Health Care Personnel ShortageTask Force, 2004).
Increased Productivity: Improvements in informationtechnology and automation of prescription fulfillmentwill increase the productivity of dispensing pharma-cists, effectively increasing supply (Knapp, 2002).
Increased Length of Training: The baccalaureatedegree in pharmacy has been phased out nationally,and all new pharmacists are now required to obtain theDoctor of Pharmacy degree. This change requiresmore existing pharmacists to educate new ones,
Washington State’s Pharmacist
Workforce through 2020:Influential Factors and Available Data
6
reducing the supply ofpharmacists providingpatient services. The longertraining time may alsoreduce supply by makingentry into the professionmore difficult (Bureau ofHealth Professions, 2000).
Declining Job
Satisfaction: A chronicshortage of pharmacists islikely to decrease jobsatisfaction as workingconditions become morestressful and less flexible.These changes could inturn lead some to abandonthe profession (Bureau ofHealth Professions, 2000).
Increasing Proportion of Women Pharmacists:
Women represent an increasing proportion of thepharmacist workforce (Bureau of Health Professions,2000). Overall, women’s professional lives inpharmacy tend to be shorter than men’s (Bureau ofHealth Professions, 2000). This shift in the gendercomposition of the workforce reduces the total percapita level of services provided when compared withthe historically male-dominated workforce.
Aging Workforce and Population: An agingworkforce, combined with an aging populationneeding more services, could create a shortage ofpersonnel or exacerbate current shortages (Levenson,2002). In Washington, from 1995 to 2025, theproportion of the population age 65 and over willnearly double from 11.6 percent to 20.2 percent (U.S.Census Bureau).
Expanded Professional Roles: Pharmacist roles areexpanding from dispensing of prescriptions toincreasing involvement in activities to improve patientcare, increasing total demand for pharmacists (Bureauof Health Professions, 2000; Knapp, 2002).
New Drug Therapies: New drugs are appearing at afaster rate, and the complexity of drug therapies isincreasing. These trends increase the demand forpharmacist services in dispensing, patient education,and monitoring (Bureau of Health Professions, 2000;Knapp, 2002).
Expanded Insurance Coverage: Insurance coverageof prescription drugs is expanding, leading toincreased demand (Bureau of Health Professions,2000).
DATA ANDMETHODOLOGICALLIMITATIONSThe data and methods used in this analysis suffer fromseveral drawbacks:
Scarcity of Data: Few data relate to the state’spharmacist workforce. The only trend data that existprovide gross numbers of licenses and educationalcompletions. We were able to extrapolate estimates ofa few limited aspects of supply and demand using fourunrelated sources: a survey of licensees that accom-panied the 1998-99 professional licensing and licenserenewal process, a 2002 survey of hospital adminis-trators, educational program completions data from1996-2003, and U.S. Census state population data.1
We used state licensing data from the WashingtonState Department of Health Office of HealthProfessions Quality Assurance. The Department ofHealth also conducted a supplementary survey duringthe 1998-99 licensing and renewal process. Thisprovides the most recent survey data available onWashington’s credentialed health care professionals.(Note: we are currently in the process of analyzing a2003 survey of retail pharmacies.)
Another key source of data for this report is a 2002study of staffing in nonfederal acute care hospitalsconducted by the Washington State HospitalAssociation and the University of Washington Centerfor Health Workforce Studies (Skillman et al., 2003).
Educational completions data come from the NationalCenter for Education Statistics IntegratedPostsecondary Education Data System (NCES IPEDS)and directly from educational program directors in thestate from 1996 through 2003.
Table 1. Factors Affecting Supply and Demand of Pharmacists
Factor Effect on Supply Effect on Demand
Increases in educational capacity Increase
Increased productivity Increase
Increased length of training Decrease
Declining job satisfaction Decrease
Increasing proportion of women pharmacists Decrease
Aging workforce and population Decrease Increase
Expanded professional roles Increase
New drug therapies Increase
Expanded insurance coverage Increase
7
We could find no data on job turnover, providermigration in or out of the state, or exits from theprofession; these and other critical individual variablesare not factored into any estimates in this report. Inaddition, we have not incorporated system-levelchanges in health care and economic trends into ouranalysis. In effect, we treated all of these factors asconstants, with no net effects on future supply ordemand. We know that they will change, but availabledata do not allow us to take account of their influencesat the state level. The limited analysis presented hererelies on an extensive set of assumptions that are opento question and revision. For example, our demandmodel extrapolates from hospital sector vacancies toproject demand for the entire state pharmacistworkforce. Are vacancy rates the same in the hospitaland nonhospital sectors? Projections may be highlysensitive to variations in assumptions and factorsexternal to our analysis. In addition, it must be notedthat our demand model is rather simple, based onvacancies. A more sophisticated multivariate economicmodel that simultaneously includes changes in supplyand demand (and accompanying price changes) is evenfarther beyond present data capability.
Exclusion of Geographic Variation: Pharmacists areunlikely to be perfectly distributed according to localpopulation needs. Adequate data do not exist toanalyze regional differences in the pharmacyworkforce over time. An analysis of state supply anddemand in the aggregate showing an apparentequilibrium or surplus of providers can still maskcritical shortages in substate areas.
Size of the Workforce: The pharmacist workforce issmall compared to the largest health occupations in thestate. It is in the 4,000s, as compared, for example, tonursing, which is in the 50,000s. This smaller sizemakes projections more volatile. Small annual changesin educational completions, retirement rates, demandfor services, etc., can cause much larger fluctuationsover time in the balance between demand and supply.
RESULTSOur analysis of available data on pharmacists inWashington yielded the following results:
Demographics: According to Washington Statelicensing data in 1998-99, 79 percent of pharmacists incurrent practice were non-Hispanic white. Bycomparison, 62 percent of 1999-2000 pharmacygraduates were non-Hispanic white, and 22 percentwere Asian or Pacific Islander (Patterson & Skillman,2002). Sixty-four percent of new graduates inWashington were women in 1999-2000 (Patterson &Skillman, 2002).2
Present Shortage of Pharmacists Likely to Continue
Until at Least 2008: We created two projectionscenarios for this report, shown in Figure 1 (seeAppendix for a detailed explanation of methods). Bothscenarios assume that demand for services and rates ofincrease in supply of providers (adjusted forpopulation growth) will continue at current levels.Both scenarios, based on the 2002 hospital vacancyrate of 10.7 percent, suggest that the current statewideshortage is likely to continue until at least 2008. Butthe two scenarios show divergent trends. One scenario,using licensing trends, projects a steady increase insupply, erasing the shortage around 2008 andproducing a 20.8 percent surplus of pharmacists by2020. An alternative scenario projects declining supplyas retirements and increasing demand outstripeducational output, leading to a 30.2 percent shortageby 2020.
8
Figure 1. Pharmacists in Washington State 2003-2020:A Demand Model and Two Alternative Supply Models
Projections include the following state-level data:• Total active professional licenses.• Hospital pharmacist employees and vacancies.• Total general population projections.• Pharmacist program completions.• Retirement projections.
Unavailable data that would improve projections:• Need and distribution of professionals in substate areas.• Trend data on vacancies/turnover.• Practice characteristics (e.g., full- v. part-time, career length, specialty practice).• Job satisfaction and compensation.• Nonhospital employees/vacancies.• Demand differentials by demographic group (burden of disease by age, ethnicity, urban/rural, etc.).• Migration in and out of state.• Regulation and credentialing changes.• Scope of practice changes.• Educational trends (e.g., cost, availability, demand for training).• Technological change (e.g., productivity, new therapies).• Macroeconomic trends affecting health care (e.g., total economic growth, trends in insurance coverage).• Other health care systems/organizational trends.
0
1000
2000
3000
4000
5000
6000
7000
8000
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year
Nu
mb
er o
f p
har
mac
ists
DemandSupply Model ISupply Model II
9
QUESTIONS RAISED BYTHIS REPORTOur models project two divergent trends in the supplyof Washington State’s pharmacists—a future surplus ora shortage. These models were developed with verylimited data. Before such projections can be used toinform policy, they must be reviewed by stakeholdersfamiliar with the environment in which this workforceoperates. These stakeholders can provide subjectiveassessments of how the profession is likely to changewhere quantifiable data do not exist currently, and theycan generate estimates about how these changes mayaffect workforce supply and demand. Below are somequestions for which we seek stakeholder input. Thislist is not exhaustive, and we welcome additionalinsights regarding influential factors and useful trenddata.
(1) Is Washington heading for a shortage or a surplusof pharmacists after 2008? How realistic are thefuture demand estimates in this report, which arebased solely on state population growth?
(2) What are the pressures facing the educationalpipeline to pharmacy in Washington State?
(3) How many of Washington’s pharmacists weretrained out of state? How many of those trainedin Washington stay here to work? What is the netimpact on supply?
(4) This report makes no distinctions betweendifferent types of pharmacy services. How do theprospects for different areas of pharmacy practice(order fulfillment, primary care, secondary andtertiary services) differ?
(5) How do the hospital and nonhospital sectorscompare? When only hospital vacancy rates areavailable and these are used to estimatenonhospital vacancies, what kind of error (if any)is introduced?
(6) How equitably are pharmacists distributedthroughout the state? Are there area shortages orsurpluses? What are differences by sector orfacility type?
(7) How will new pharmacy technology and newdrug therapies affect supply and demand?
(8) How can we obtain more recent and accurate datato assess the current pharmacist workforce? Whatare practical long-term strategies for creating thedata needed to monitor pharmacist supply anddemand?
(9) What new state and federal policies may changepharmacist supply and demand?
(10) Will economic changes (e.g., recession) causepopulation demand for care to increase ordecrease substantially during the next decade?
10
APPENDIX:A DEMAND MODEL ANDTWO ALTERNATIVE SUPPLYMODELSThis report shows one method of projectingpharmacist workforce demand and two alternativemethods of projecting pharmacist supply. Thesemodels were developed using the best data availablefor Washington. The same demand model is comparedwith each supply model to generate two scenariosassessing the balance between supply and demand. Allvalues reported represent persons, not positions orFTEs. The shaded rows in the accompanying tables arethe raw numbers representing the principalcomponents of provider supply and demand that addup to each year’s projected total surplus or shortage(covered under the Results section of each analysis).
DEMAND MODELThis model uses state population projections and dataon hospital pharmacist employees and vacancies. Weextrapolated from hospital employment and vacanciesto estimate total state employment and vacancies(hospital and nonhospital sectors). This current totaldemand value was then adjusted to take account ofincreasing demand resulting from population growthin each subsequent year.
The following detailed explanations refer to theDemand Model in Tables A1 and A2 where rows arenumbered D1-D3:
(D1) We obtained state population projections for2000, 2005, 2015, and 2025 from the U.S.Census Bureau. We assumed that populationwould grow at a constant rate in each of theyears between these estimates.
(D2) We calculated the total demand in 2003 as thesum of currently practicing (S3, explainedbelow), 4,300, and vacancies (results row 1,explained below), 514. This yields a demand of79 providers per 100,000 population. Weassumed this rate of demand through 2020.Thus demand grows in constant proportion topopulation growth.
(D3) The net annual increase in demand due topopulation growth, maintaining a ratio of 79providers per 100,000, ranges from 59 to 63providers per year through 2020.
SUPPLY MODEL I:LICENSING TRENDSThis model uses recent trends in state licensing ofpharmacists to project future supply. We did not haveinformation about the specific components that led toyearly changes in the number of licenses. Therefore,we assumed (recognizing this is likely anoversimplification) that whatever combination offorces driving these increases historically wouldcontinue at about the same rate.
The following detailed explanations refer to SupplyModel I in Table A1, rows S1-S5:
(S1) 1996-2001 figures are derived from theWashington State Department of Health’sbiennial reports summarizing total activelicenses as of July 30 in odd years. Summarydata were available from 1993 through 2001,inclusive. We estimated even years as themidpoints between numbers of licenses in oddyears. We derived figures for 2002 and 2003 byadding the mean yearly increase for this five-year period of available data. Based on ouranalysis of 1999 licensing data, we know thattotal active licenses overestimate supplybecause these numbers include licensees not inpractice and some duplicate records.
(S2) Yearly net increases in active licensees for 1996through 2000 inclusive are based on actuallicensing data as reported in (S1). We used themean yearly increase for this five-year period,276, as the estimate for increases from 2001 to2020.
(S3) Data come from two sources, for two yearsonly: a survey of licensees that accompanied the1999 professional licensing and license renewalprocess, and a 2002 survey of hospitals inWashington State (Skillman et al., 2003).
The 1999 value was derived from the licensingdata as follows:
— We based all estimates on only activelicensees working or living in Washington whowere currently practicing, up to age 65,inclusive. All others were excluded from ouranalysis.
— 3,787 licensees of 4,090 responding to thesurvey indicated that they were currentlyengaged in nonvolunteer practice (92.9%).
— 326 active licensees (fitting all other criteria)did not respond to the survey. We assumed thatthey were in current practice at the same rate asrespondents, 92.9 percent, yielding anadditional 303 providers.
11
— Currently practicing survey respondents(3,787) and imputation for missing responses(303) total 4,090. Since nonpracticing licenseesmay have been more likely not to respond to thesurvey, this total is likely to be an overestimate.
The 2003 value was derived from hospitalsurvey data as follows:
— The proportion of 1999 licensees indicatingemployment in hospital inpatient or emergencydepartments (excluding outpatient care) was21.0 percent. We assumed that the sameproportion of pharmacists in 2003 wereemployed in hospitals to derive the size of thetotal pharmacist workforce as follows:
— There were an estimated 931 pharmacistpositions in nonfederal acute care hospitals in2003. To adjust for possible overcounting ofpersons occupying multiple positions indifferent locations (or more than one type ofpharmacist position at the same location), weadjusted this value downward based on the factthat 3.1 percent of 1999 licensees giving a worklocation indicated two different hospital sites.Performing this adjustment yields 903 hospitalproviders. Assuming they constitute 21.0percent of the total state workforce, as hospitalproviders did in 1999, the total number ofproviders is 4,300.
We estimated that the number of licensees inthe intervening years (2000-2002) increased at aconstant rate based on the 1999 to 2003 averageyearly change. Using this method, note that thenumber of providers per 100,000 appears tohave remained almost constant from 70.8 in1999 to about 70.5 in 2003.
(S4) We estimated the proportion of active licensees(S1) who are currently practicing (S3) from1999 through 2003 by dividing (S1) by (S3).This proportion declines from 64.8 percent to55.6 percent, averaging 59.4 percent during theperiod. This change over time may be the resultof differences in the data sources used toestimate currently practicing providers for 1999and 2003 or some other source of error.Alternatively, these numbers may reflect a realdecline.
(S5) For the years 1999 through 2002, the increasein the number of licensees in current practice isderived from the annual increase in totalemployment estimated in (S3). As explainedabove, the values in (S3) were based on twodifferent sources of data, one yielding anestimate of employment for 1999, the other for2003. The average annual increase in currently
practicing licensees between these two timepoints was 52.
To project future growth in currently practicinglicensees from 2003 through 2020, the modelbegins by looking at the growth trend in totallicensees from 1996 through 2001. During thisperiod, total licensees grew at a mean annualrate of 276 (S2). We used this historical meanannual growth rate in total licenses to estimatethe future annual rate of increase. To obtainestimates of only those pharmacists currentlypracticing, total licensees (which include bothpharmacists in practice and those not in practicewho continue to maintain their licenses) mustbe adjusted downward. We adjusted the annualincrease in licensees of 276 to reflect that onaverage, only 59.4 percent of active licensees(estimated in S4 above) were employed aspharmacists from 1999 through 2003. Thisadjustment yields an annual increase of 164providers from 2003 to 2020.
Results: The following detailed explanations refer tothe Results Section of Table A1, rows 1 and 2:
(1) Hospital administrators surveyed in 2002-2003(Skillman et al., 2003) reported an estimated108 vacancies. We adjusted this value toaccount for possible coverage of more than oneposition by a single provider and then estimatedvacancies in all settings (hospital andnonhospital sectors), as in (S3) above. Theseadjustments yielded 514 total vacancies. Weprojected vacancies in each subsequent year byadding the annual increase in demand (D3) andsubtracting new providers (S5).
(2) Vacancies are expressed as a percentage of totaldemand in each year. A positive numberrepresents a shortfall of providers; a negativenumber represents a surplus.
Summary of Supply Model I: The number ofpharmacist licensees increased during the years forwhich we have data (1995 to 2001) at an average rateof 276 per year. We were able to derive estimates ofthe proportion of active licensees in practice duringeach year from 1999 to 2003. From analysis of statehealth professions licensing data, we know that anaverage of 59.4 percent of active licensees were inpractice during each year of this period. We appliedthis proportion to our estimates of future annualincreases in licenses. Using this method, we projectedincreases in employed pharmacists of 164 per year.The annual increase in demand, based on populationgrowth, ranges from 59 to 63 through 2020. Beginningwith 514 vacancies in 2003, a 10.7 percent shortfall ofpharmacists in the state, Supply Model I showsincreases in supply relative to demand, with
12
equilibrium around 2008. In this model, supplyeventually outstrips demand by 20.8 percent in 2020.Policy interventions would probably ensure that animpending surplus of this magnitude would nevercome to pass.
SUPPLY MODEL II: EDUCATIONALOUTPUT AND RETIREMENTSThis model uses data on educational completions andprovider ages to project future supply. We attempted toestimate net change in supply by taking account ofnewly educated entrants to the profession and exitsdue to retirement.
The following detailed explanations refer to SupplyModel II in Table A2, rows S1-S8:
(S1) Same as Supply Model I, (S1).
(S2) Same as Supply Model I, (S3).
(S3) Same as Supply Model I, (S4).
(S4) Same as Supply Model I, (S5), years 1999through 2002, for comparison purposes only.Supply increases based on actual employmentestimates are similar to what would be expectedbased merely on educational completions net ofretirements, the method we employed in thismodel.
(S5) We obtained completions data from twosources. NCES IPEDS data for Washington’stwo pharmacy programs are publicly availablefor the years 1996-98 and 2000. We obtaineddata for 1999 and 2001-03 directly from theprograms.
A shortage of pharmacists in Washington hasresulted in the appropriation of funds toincrease enrollment capacity in pharmacyprograms (Health Care Personnel Shortage TaskForce, 2004). This increase will result in 16additional graduates annually beginning in 2007(William E. Fassett, Ph.D., Dean, WashingtonState University College of Pharmacy, personalcommunication, February 17, 2004). Weprojected educational completions to continueat the higher rate beginning in 2007 to takeaccount of this capacity increase. The imputedvalue of 138.5 completions per year from 2004through 2006 is the annual mean of thecompletions for the eight prior years ofavailable data (1996-2003). We assumed that allprogram completers sit for and pass thelicensing exam.
(S6) At any given time, some proportion of programcompleters will not be in practice. Ourestimates, based on available licensing andpractice data, suggest that about 59.4 percent ofcurrent license holders are in active practice (as
in Supply Model I, row S4). We adjusted valuesdownward by this proportion to yield activeproviders resulting from yearly programcompletions.
(S7) The 1998-99 state licensing survey askedlicensees their age. We had no data on exitsfrom the profession due to death, outmigration,change in occupation, etc., and thereforeattrition in our model is captured exclusivelythrough aging out providers surveyed in 1998-99 as they reach age 65.3
(S8) The net annual increase in supply is simply thedifference between the gain from completions(S6) less retirements (S7).
Results: The following detailed explanations refer tothe Results Section of Table A1, rows 1 and 2. Thesemethods are the same as those used to derive theresults for Supply Model I:
(1) Hospital administrators surveyed in 2002-03(Skillman et al., 2003) reported an estimated108 vacancies. We adjusted this value toaccount for possible coverage of more than oneposition by a single provider and then estimatedvacancies in all settings (hospital andnonhospital sectors), as in (S3) above. Theseadjustments yielded 514 total vacancies. Weprojected vacancies in each subsequent year byadding the annual increase in demand (D3) andsubtracting new providers (S5).
(2) Vacancies are expressed as a percentage of totaldemand in each year. A positive numberrepresents a shortfall of providers; a negativenumber represents a surplus.
Summary of Supply Model II: The number ofeducational completions from 1996 to 2003 averaged138.5 pharmacists per year. Recent expansion ofcapacity will add 16 graduates annually beginning in2007. For purposes of this projection, we assumed thatall completers of pharmacy programs would obtain alicense to practice. From analysis of the state healthprofessions licensing data we know that an estimated59.4 percent of active pharmacist licensees are inpractice at any given time. We adjusted educationaloutput in each year using this percentage, whichresulted in increases to supply of 82 to 92 providersper year. Our projections of retirements, based on agesof licensed providers, result in annual reductions tosupply that grow from 45 in 2003 to 161 in 2020. Ourestimates of the annual increase in demand, based onpopulation growth, range from 59 to 63 through 2020.Beginning with 514 vacancies in 2003, a 10.7 percentshortfall of pharmacists in the state, Supply Model IIshows steady decreases in supply relative to demand.The shortage of pharmacists increases to 30.2 percentin 2020.
13
Tab
le A
1. P
roje
ctio
ns
Usi
ng
Su
pp
ly M
od
el I:
Ph
arm
acis
t L
icen
sin
g T
ren
ds
Su
pp
ly M
od
el I
1996199719981999200020012002200320042005200620072008200920102011201220132014201520162017201820192020
(S1)
Active lic
ensees
5,8
03
5,8
55
6,0
84
6,3
13
6,7
48
7,1
83
7,4
59
7,7
35
(S2)
Net in
cre
ase in
active lic
ensees
52
229
229
435
435
276
276
276
276
276
276
276
276
276
276
276
276
276
276
276
276
276
276
276
276
(S3)
Tota
l curr
ently
pra
cticin
g(p
er
100,0
00)
4,0
90
(70.8
)4,1
42
(70.7
)4,1
95
(70.6
)4,2
47
(70.6
)4,3
00
(70.5
)
(S4)
Pro
port
ion o
factive lic
ensees
curr
ently p
racticin
g64.8
%61.4
%58.4
%56.9
%55.6
%
(S5)
Net in
cre
ase in
curr
ently p
racticin
g52
52
52
52
164
164
164
164
164
164
164
164
164
164
164
164
164
164
164
164
164
164
Dem
an
d M
od
el
1996199719981999200020012002200320042005200620072008200920102011201220132014201520162017201820192020
(D1)
Sta
te p
opula
tion
(in 1
000s)
5,5
16
5,6
02
5,6
87
5,7
73
5,8
58
5,9
38
6,0
18
6,0
98
6,1
78
6,2
58
6,3
38
6,4
18
6,4
98
6,5
78
6,6
58
6,7
38
6,8
18
6,8
98
6,9
78
7,0
58
7,1
33
7,2
08
7,2
83
7,3
58
7,4
33
(D2)
Tota
l dem
and
(79 p
er
100,0
00)
4,8
14
4,8
77
4,9
40
5,0
03
5,0
66
5,1
29
5,1
92
5,2
55
5,3
18
5,3
81
5,4
44
5,5
07
5,5
70
5,6
33
5,6
92
5,7
51
5,8
10
5,8
69
(D3)
Annual in
cre
ase
in d
em
and
63
63
63
63
63
63
63
63
63
63
63
63
63
59
59
59
59
59
Resu
lts
1996199719981999200020012002200320042005200620072008200920102011201220132014201520162017201820192020
(1)
Tota
l vacancie
s514
413
312
211
110
9-9
2-1
93
-294
-395
-496
-597
-698
-799
-904
-1,0
09
-1,1
14
-1,2
19
(2)
Vacancie
s /
tota
l dem
and (
%)
10.7
8.5
6.3
4.2
2.2
.2-1
.8-3
.7-5
.5-7
.3-9
.1-1
0.8
-12.5
-14.2
-15.9
-17.5
-19.2
-20.8
14
Tab
le A
2. P
roje
ctio
ns
Usi
ng
Su
pp
ly M
od
el II
: P
har
mac
ist
Ed
uca
tio
nal
Ou
tpu
t an
d R
etir
emen
ts
Su
pp
ly M
od
el II
1996199719981999200020012002200320042005200620072008200920102011201220132014201520162017201820192020
(S1)
Active lic
ensees
5,8
03
5,8
55
6,0
84
6,3
13
6,7
48
7,1
83
7,4
59
7,7
35
(S2)
Tota
l curr
ently
pra
cticin
g
(per
100,0
00)
4,0
90
(70.8
)4,1
42
(70.7
)4,1
95
(70.6
)4,2
47
(70.6
)4,3
00
(70.5
)
(S3)
Pro
port
ion o
f
active lic
ensees
curr
ently p
racticin
g
64.8
%61.4
%58.4
%56.9
%55.6
%
(S4)
Net in
cre
ase in
curr
ently p
racticin
g in
Model I ro
w (
S5)
52
52
52
52
(S5)
Educational
com
ple
tions
161
164
68
136
150
137
138
154
138
139
138
155
154
155
154
155
154
155
154
155
154
155
154
155
154
(S6)
New
com
ple
ters
curr
ently p
racticin
g81
89
81
82
91
82
82
82
92
91
91
92
91
91
92
91
91
92
91
91
92
91
(S7)
Retire
ments
26
38
36
53
45
45
54
57
93
92
91
87
108
120
105
123
121
127
149
157
156
161
(S8)
Net in
cre
ase in
curr
ently p
racticin
g55
51
45
29
46
37
28
25
-1-1
05
-17
-29
-13
-32
-30
-35
-58
-66
-64
-70
Dem
an
d M
od
el
1996199719981999200020012002200320042005200620072008200920102011201220132014201520162017201820192020
(D1)
Sta
te p
opula
tion
(in 1
000s)
5,5
16
5,6
02
5,6
87
5,7
73
5,8
58
5,9
38
6,0
18
6,0
98
6,1
78
6,2
58
6,3
38
6,4
18
6,4
98
6,5
78
6,6
58
6,7
38
6,8
18
6,8
98
6,9
78
7,0
58
7,1
33
7,2
08
7,2
83
7,3
58
7,4
33
(D2)
Tota
l dem
and
(79 p
er
100,0
00)
4,8
14
4,8
77
4,9
40
5,0
03
5,0
66
5,1
29
5,1
92
5,2
55
5,3
18
5,3
81
5,4
44
5,5
07
5,5
70
5,6
33
5,6
92
5,7
51
5,8
10
5,8
69
(D3)
Annual in
cre
ase
in d
em
and
63
63
63
63
63
63
63
63
63
63
63
63
63
59
59
59
59
59
Resu
lts
1996199719981999200020012002200320042005200620072008200920102011201220132014201520162017201820192020
(1)
Tota
l vacancie
s514
531
557
592
630
694
758
821
879
959
1,0
51
1,1
27
1,2
22
1,3
15
1,4
09
1,5
26
1,6
51
1,7
74
(2)
Vacancie
s /
tota
l dem
and (
%)
10.7
10.9
11.3
11.8
12.4
13.5
14.6
15.6
16.5
17.8
19.3
20.5
21.9
23.3
24.8
26.5
28.4
30.2
15
NOTES1 The Labor Market and Economic Analysis Branch ofthe Washington State Employment SecurityDepartment has produced projections and job vacancyestimates, but because their estimates of employmentare significantly higher than those suggested by anyother data source, we did not incorporate its estimatesinto this analysis.
2 Sex is missing for 72 percent of pharmacists inWashington State 1999 licensing data, and notreported here.
3 Substituting a 2.6 percent retirement rate every yearfrom 2003 through 2020, suggested by William E.Fassett, Ph.D., Dean, Washington State UniversityCollege of Pharmacy (personal communication,March 19, 2004), changes the supply projection verylittle. This rate of retirement yields a total of 1,873vacancies by 2020, a 31.9 percent shortage, comparedto this model’s estimated vacancy rate of 20.2 percent.
REFERENCESBureau of Health Professions (2000). The pharmacist
workforce: a study of the supply and demand for
pharmacists. Rockville, MD: U.S. Department ofHealth and Human Services, Health Resources andServices Administration, Bureau of HealthProfessions.
Health Care Personnel Shortage Task Force (2004).Progress 2003: a report of the Health Care Personnel
Shortage Task Force. Olympia, WA: WorkforceTraining and Education Coordinating Board.
Knapp, D. A. (2002). Professionally determined need
for pharmacy services in 2020. Retrieved February 17,2004, from http://aacp.org/site/pdf.asp?TP=/Docs/MainNavigation/Resources/4634_needsconferencefinalreport.pdf.
Levenson, D. (2002). Hospitals struggling to fill
pharmacy, radiological technology staff positions.Retrieved September 18, 2003, from http://www.hospitalconnect.com:80/jsp/article.jsp?dcrpath=AHA/NewsStory_Article/data/AHANEWS2B241&domain=AHANEWS.
Patterson, D. G., Skillman, S. M. (2002). Health
professions education in Washington State: 1996-2000
program completions statistics. Working Paper #73.Seattle, WA: Center for Health Workforce Studies,University of Washington.
Skillman, S. M., Hutson, T., Andrilla, C. H. A. (2003).Washington State hospitals: results of 2003 staffing
survey. Working Paper #79. Seattle, WA: WWAMICenter for Health Workforce Studies, University ofWashington.
16
RELATED RESOURCESFROM THE WWAMICENTER FOR HEALTHWORKFORCE STUDIES ANDTHE RURAL HEALTHRESEARCH CENTER
PUBLISHED ARTICLESBenedetti, T. J., Baldwin, L. M., Andrilla, C. H., Hart,L. G. (2004). The productivity of Washington State’sobstetrician-gynecologist workforce: does gendermake a difference? Obstetrics and Gynecology, 103(3),499-505.
Larson, E. H., Hart, L. G., Ballweg, R. (2001).National estimates of physician assistant productivity.Journal of Allied Health, 30(3), 146-152.
Larson, E. H., Palazzo, L., Berkowitz, B., Pirani, M.J., Hart, L. G. (2003). The contribution of nursepractitioners and physician assistants to generalist carein Washington State. Health Services Research, 38(4),1033-1050.
WORKING PAPERSBaldwin, L.-M., Fay, M. M., Larson, E. H., Lishner, D.M., Mauksch, L. B., Katon, W. J., Walker, E., Hart, L.G. (2003). Modeling the mental health workforce in
Washington State: using state licensing data to
examine provider supply in rural and urban areas.Working Paper #80. Seattle, WA: WWAMI Center forHealth Workforce Studies, University of Washington.
Benedetti, T. J., Baldwin, L.-M., Andrilla, C. H. A.,Hart, L. G. (2003). The productivity of Washington
State’s obstetrician-gynecologist workforce: does
gender make a difference? Working Paper #82. Seattle,WA: WWAMI Center for Health Workforce Studies,University of Washington.
Larson, E. H., Palazzo, L., Berkowitz, B., Pirani, M.J., Hart, L. G. (2001). The contribution of nurse
practitioners and physician assistants to generalist
care in underserved areas of Washington State.Working Paper #64. Seattle, WA: WWAMI RuralHealth Research Center, University of Washington.
Patterson, D., Skillman, S. M. (2002). Health
professions education in Washington State: 1996-2000
program completion statistics. Working Paper #73.Seattle, WA: WWAMI Center for Health WorkforceStudies, University of Washington.
Patterson, D. G., Skillman, S. M., Hart, L. G. (2004).Washington State’s dental hygienist workforce through
2020: influential factors and available data. WorkingPaper #92. Seattle, WA: WWAMI Center for HealthWorkforce Studies, University of Washington.
Patterson, D. G., Skillman, S. M., Hart, L. G. (2004).Washington State’s radiographer workforce through
2020: influential factors and available data. WorkingPaper #89. Seattle, WA: WWAMI Center for HealthWorkforce Studies, University of Washington.
Skillman, S. M., Hutson, T., Andrilla, C. H. A. (2003).Washington State hospitals: results of 2002 workforce
survey. Working Paper #79. Seattle, WA: WWAMICenter for Health Workforce Studies, University ofWashington.
Skillman, S. M., Hutson, T., Andrilla, C. H. A.,Berkowitz, B., Hart, L. G. (2002). How are
Washington State hospitals affected by the nursing
shortage? Results of a 2001 survey. Working Paper#68. Seattle, WA: WWAMI Center for HealthWorkforce Studies, University of Washington.
Wright, G. E., Paschane, D. M., Baldwin, L. M.,Domoto, P., Cantrell, D., Hart, L. G. (2001).Distribution of the dental workforce in Washington
State: patterns and consequences. Working Paper #60.Seattle, WA: WWAMI Center for Health WorkforceStudies, University of Washington.
DATA SNAPSHOTSWWAMI Center for Health Workforce Studies,University of Washington (2000). Data snapshot: race
and ethnicity of Washington State health professionals
compared with state population. Seattle, WA: Author.
WWAMI Center for Health Workforce Studies,University of Washington (2001). Data snapshot:
licensed practical nurses (LPNs) in Washington:
demographics and employment characteristics.Seattle, WA: Author.
WWAMI Center for Health Workforce Studies,University of Washington (2001). Data snapshot:
registered nurses (RNs) in Washington: demographics
and employment characteristics. Seattle, WA: Author.
WWAMI Center for Health Workforce Studies,University of Washington (2002). Data snapshot:
pharmacists in Washington: demographic and
employment characteristics. Seattle, WA: Author.
For a complete list of publications by the Centerfor Health Workforce Studies, visithttp://www.fammed.washington.edu/CHWS/.