2015 Keeneland Conference Session 3D

Workforce Enumeration & Development

Room: Bluegrass Room I & II
Wednesday, April 22, 2015, 10:45 AM to 12:15 PM

Moderator: Angela J. Beck, PhD, MPH


Jiali Ye, PhD

Changes in the Number of Public Health Nurses in Local Health Departments: 2005-2013

Co-Investigator(s): Nathalie Robin, MPH & Carolyn Leep, MS, MSPH

Background: Public health nurses are an important component of workforces in local health departments (LHDs). They work with individuals and communities to offer expertise in clinical care and health promotion.  Recent evidence suggests a shortage of registered nurses for public health practice, which is threatening the ability to meet the health needs of communities. Research Objective: This study aims to examine changes in the number of public health nurses in LHDs and rural-urban variation in the changes and to assess how expenditures and provision of clinical services are related to the size of nurse workforce in 2005-2013.   Data Sets and Sources: Four National Profile of Local Health Departments study datasets (2005, 2008, 2010, and 2013) were linked. NACCHO’s Profile survey core questionnaire contains questions on LHD workforce, funding, activities and services. The 2013 Health County Ranking data were used to obtain community characteristics, including poverty rate, unemployment rate, uninured rate, and ratio of population to primary care physician. Study Design: This study employed a longitudinal cohort research design. A total of 1231 LHDs that completed all four surveys and provided answers on the number of public health nurses in these surveys were included in the analysis. Analysis: The mean number of nurses per 100,000 population was computed by survey year and by rural-urban classification. Trends over time were assessed overall and by rural/urban subgroups. Fixed effects model was used to examine how changes in total count of clinical services and expenditure per capita may affect the number of nurses, with community-level socioeconomic factors and LHD infrastructure characteristics as covariates.   Principle Findings: On average, there was a significant decrease in the number of LHD nurses per 100,000 population over year (p<0.001). The mean number of nurse per 100,000 people dropped from 21.8 in 2005 to 16.9 in 2013. There was a reduction in the supply of nurses in LHDs in both rural and urban areas, but LHDs in rural areas demonstrated larger nurse supply losses (p<0.05). The fixed-effects model shows that the number of clinical services was positively associated with growth in the number of nurses (Estimate=. 0.51, p<0.001) after controlling for covariates. The variation in expenditures over time also affected the change in the number of nurses, but the impact was much smaller (Estimate=0.08, p<0.001). Conclusions: A longitudinal analysis demonstrates a decline in the number of public health nurses in LHDs. The shift of public health practice priority, particularly the reduction of clinical services, may be a major forces for reducing nursing positions.  Budget constrains may also be a barrier to adequate nursing staffing. Implications for Public Health practice and Policy: Current reforms in health delivery system and decreases in public health funding at local level contributed to a reduction of public health nurses workforce. Further research is needed to assess how such a reduction may affect the delivery of essential public health services and health outcomes of the population.


Craig Ziegler

The Relationship of Medical School Students’ Debt on Choice of Primary Care Specialty

Co-Investigator(s): Robert Steiner, MD; Doug Lorenz, PhD; Barry Wainscottt, MD; & Robert Esterhay, MD

Background: US physician workforce requirements cannot meet current or future healthcare demands. The physician shortage is most notable among primary care physicians (PCPs) where an additional 52,000 are projected to be needed by 2025 (Petterson et al., 2012). The AAMC directed that medical school enrolment be increased 30%. However, more medical school matriculates does not necessarily equal having a sufficient PCP workforce. These statistics are pertinent because PCP functions differ from specialist.  PCPs holistically focused on the patient and are the patient’s first contact to the healthcare system. PCPs are also vital to controlling costs, usage, and distribution of healthcare, often arranging and overseeing patient care with specialist, particularly when patients have chronic diseases and comorbidities. They also serve as leaders for new healthcare delivery modalities, i.e., accountable care organizations and patient centered medical homes. PCPs’ median income ($186,044) dwarf specialists’ median income ($339,738) and specialist lifelong income is 3.5 million greater than PCPs (Youngclaus, et al., 2013). Further, the 2002 US graduating medical students debt burden exceeded $100,000 (Rosenblatt et al., 2005); and in 2013, medical school student median debt was $175,000 (AAMC Stat News, 2013), as medical schools’ tuition has increased approximately 6% annually since 2001 (Weinstein et al., 2010).   Research Objectives: Recognizing the PCP shortage and increasing student debt, does a link between the two exist? If so, what policies can resolve the student debt issue? Accordingly, this study’s objective is to assess debt’s influence on students’ choice of a PCP career and to impart PCP shortage and student debt linked strategies to address this issue.     Study Design: This study is a retrospective correlational study using a database of all University of Louisville 2001 to 2010 graduating medical students (n=1391).   Data Sets and Sources: Data collected came from the AMCAS medical school application, the American Medical Association Physician Masterfile, and University of Louisville Office of Medical Financial Aid.   Analysis: Multiple logistic regression was used to address if a relationship exists between medical school debt and matching to a primary care residency. Comparisons between lower debt quintiles and the highest quintile were made. Covariates in the model included gender, race, age, social economic status of parents, USMLE Step 1 scores, and year student graduated.   Principal Findings: The results found modest associations between debt levels and choosing primary care. Among medical students, the lowest quintile ($165,000: OR = 1.68, 95% CI: 1.04-2.70), the second quintile ($50,001-$100,000 versus >$165,000: OR = 1.68, 95% CI: 1.04-2.70), and the fourth quintile ($135,000-$164,999 versus >$165,000: OR = 1.61, 95% CI: 1.05-2.47) was significantly associated with an increased likelihood of choosing primary care.   Conclusion: The findings suggest debt is having some influence on specialty choice. If increased debt discourages students from becoming PCPs then policymakers must take steps to eliminate the rising debt burden including freezing tuition costs, prorating debt to students’ specialty choice earning potential, and/or shortening training, among other strategies.     Implications: PCP shortages disrupt the nation’s healthcare system and student debt relief may ease the problem.


Michele Issel, PhD, RN

Competing for Registered Nurses as Public Health Nurses: The Salary Gap between LHDs and Hospitals

Co-Investigator(s): Christine Fitzpatrick, BA, RN & Betty Bekemeier, PhD, MSN, RN

Background:  Recruitment of registered nurses (RNs) has been studied extensively, specifically into hospital employment.  The periodic shortage of RNs compared to hospital RN vacancies has contributed to long-term salary gains for RNs working in the acute care sector. Recruitment of RNs into public health nursing positions has been an enigma, with NACCHO studies indicating that RN positions are among the hardest to fill.   Research Objectives:  To describe the extent to which local health departments’ (LHDs’) RN wages are competitive with local hospital wages. In a sample markets, from 2010 to 2014: (1) what percent of LHDs had RN salary changes that addressed salary compression? (2) What percent of LHDs had increased RN salary at entry into the public health nursing position? (3) Did the RN salary difference between that for LHD and hospital positions change?   Data Sets and Sources: Primary data were collected via questionnaire. Respondents were either the director of public health nursing or from the human resources department. Although hospital wage data were often publically accessible due to the union contracts, comparable data could not be publicly found for LHD county employees.   Study Design: A longitudinal, repeated measures design was used. Each county was treated as a labor market. Six county LHDs were surveyed in 2010 and 2014, using the same questionnaire in both years.  All hospitals located in the six counties were invited to participate. Given the small sample size, this study constitutes a pilot study.   Analysis: Descriptive statistics were used to answer each research question. Tests of significance were not appropriate, given the small number of counties in our sample.   Principal Findings:  Question (1): Of the 4 LHDs for which we had both 2010 and 2014 data, 50% reduced the RN salary compression, as seen by small yet new wage increases in 2014 for RNS with 5, 10 and 18 years of work experience. Question (2):  Similarly, 50% of the LHDs had increased the hourly wage for beginning public health RNs, from an average $21.59 compared to $28.44 for the hospitals.  Question (3): In 3 of the county markets for which we had both LHD and hospital data in 2010 and 2014, 2 showed an increase in the gap between LHD and hospital hourly wages, with LHD wages being noticeably lower at all years of RN experience. In one county market, the gap between LHD and hospital wages decreased from $20.90 to $16.92 in 2014 relative to what it had been in 2010, but only at 18 years of experience.   Conclusions: Overall, RN hourly wages for public health nurses employed by LHDs are far less attractive compared to what RNs can earn in hospitals within the same county (and sometimes literally across the street).   Implications: To recruit motivated and highly competent RNs with the community and collaboration skills needed in the evolving public health environment, LHDs must offer truly competitive wages. LHDs are in the paradoxical position of paying lower wages for RNs that must have more complex and professional decision making and engagement skills than their hospital counterparts.


Christine Plepys, MSc

Recent Graduate Employment Reporting Methodology and Results

Background:  The Association of Schools and Programs of Public Health (ASPPH) assists members with data collection and analysis services for data-informed decision-making. With 98 institutional members, an important aspect of our role is to build cohesiveness around data collection efforts. ASPPH has collected data annually on numbers of graduates by various demographics. One area where we are working toward building an annual data collection strategy is on recent graduate employment information. ASPPH has historical answers from the 1970’s and 1990’s, but recent aggregate data collection efforts have struggled due to changes in higher education, graduate survey fatigue, and alumni tracking issues. In addition, while each ASPPH member collects graduate employment data for Council on Education for Public Health (CEPH) accreditation requirements, there are many methods used, survey questions and response options differ, and response rates can be low.  Research Objective: Our objective is to build a cohesive data collection methodology to use annually with member institutions in order to provide a reliable, aggregate answer to the question, “Where do recent graduates of ASPPH member institutions find employment?” Data Sets and Sources: The data set was developed specifically for this project. Study Design With funding from the Centers for Disease Control and Prevention and working with internal ASPPH committees, a project plan was developed on reporting annual graduate employment data to ASPPH.  The 2014 pilot project was supported centrally at ASPPH and implemented locally at member institutions. ASPPH developed a Common Questions Survey (see attachment) that includes the employment, salary and debt information ideally sought from graduates. The survey was intended to be inserted into the participating member’s current alumni survey or used as a stand-alone survey. Since many members already conduct their own graduate employment survey, they also provided their survey questions and results to ASPPH. Data collection was conducted in spring 2014, with results due to ASPPH in summer/fall. With the goal to match the CEPH accreditation requirement “within 12 months following award of the degree”, the spring 2014 pilot survey was sent to alumni that graduated in spring and summer 2013, and winter 2013/2014.  Analysis: ASPPH members submitted an Excel sheet with de-identified responses, copy of survey, response rate, and methodology. The individual member Excel spreadsheets were combined into a single database utilizing as much of the member-developed survey results as possible. Analysis is being conducted using Tableau 8.3. Principal Findings: In the pilot year, 60% of members participated providing 4,888 eligible responses. Response rates by institution ranged from 6.3% to 99.5%, with the survey results providing an overall response rate of approximately 47% (based on the ASPPH 2013 Graduate Report). For those respondents who were employed, the top employer types were university or colleges; hospital or other healthcare provider; associations, foundation or other non-profit, and the federal government. Further results will show the proportion of respondents working at state and local health departments, plus salary and debt information. Conclusions: This initial pilot was a success in terms of providing preliminary data on where graduates of CEPH-accredited schools and programs of public health find employment. A number of lessons learned will inform the project moving forward. Implications:  Graduate employment rates, employment settings, salary, and debt load provide important data for advocacy efforts, benchmarking, student recruitment, and workforce trends.