Completed Projects

Learn more about our past projects here.


The Impact of COVID-19 on Primary Care Practice

Investigators: Monisa Aijaz, MD, MPH, Valerie Lewis, PhD, Genevra Murray, PhD

COVID19 has caused dramatic changes to primary care access and delivery that present significant issues to the primary care workforce. First, primary care practices have cancelled large swaths of in person visits while working to rapidly implement telehealth. This is coupled with dramatic declines in revenue that present organizations with difficult financial choices.  The intensity of care under COVID19 likely may be increasing burnout among clinicians. Currently, information on these changes is largely anecdotal. We propose to conduct qualitative interviews with approximately 40 primary care practices in several states, purposively sampled across key dimensions (e.g. size, rurality, ownership).


Interrupted Integrated Health Care: How Primary Care Practices Utilized Tele-Health and Coordinated Team-Based Care in Response to COVID-19

Investigators: Brianna Lombardi, PhD, MSW; Lisa de Saxe Zerden, PhD, MSW; Erica L. Richman, PhD, MSW

This study will investigate how integrated health care (IHC) teams adapted during COVID-19 with a specific focus on their use of tele-health and how they adapted components of IHC such as warm hand-offs between providers and patients, referral linkages, and communication during the pandemic. This study also aims to identify innovative ways IHC practices delivered primary care that addressed patient physical, social, and behavioral health needs during a pandemic and understand areas in which innovation can continue or be reinforced in the post-pandemic period.


Workplace Violence in Healthcare Settings

Investigators: Brianna Lombardi, Emily McCartha, Connor Sullivan, Erin Fraher

Overview: Workplace violence is described by the Occupational Safety and Health Administration (OSHA) as “any act or threat of physical violence, harassment, intimidation, or other threatening disruptive behavior that occurs at the work site. It ranges from threats and verbal abuse to physical assaults and even homicide.” Reports suggest workplace violence in healthcare settings has increased during the COVID-19 pandemic, but most research uses pre-pandemic data sources. This brief summarizes current research in healthcare settings in the COVID-19 pandemic context.


Addressing Burnout among the Frontline Workforce during COVID-19: A Scoping Review & Expert Interviews

Investigators: Lisa de Saxe Zerden, PhD, Brianna M. Lombardi, PhD, MSW, Erica L. Richman, PhD, Alexandria B. Forte, PhD Student, MSW, Meriel McCollum, PhD.

Objective: This study sought to identify the strategies and interventions that U.S. health systems implemented to reduce burnout and increase employee well-being during the first year of the COVID-19 pandemic.

Methods: Pairing a scoping review of published literature and expert interviews with healthcare administrators, the study identified burnout prevention and reduction patterns at the individual, organizational, and community system levels.

Findings: The scoping review identified eight types of interventions; spanning individual, organizational, and community systems. Qualitative data supported scoping review findings and uncovered themes related to communication, role shifting, and wellness initiatives as forms of burnout prevention and reduction.

Policy Implications: The results support the need for multi-pronged strategies that target multiple systems to address the healthcare workforce’s diverse needs at the individual, organizational, and community level. The eight intervention types identified to address burnout included: enhanced communication, wellness initiatives, individual counseling or support, group or unit counseling and support, provision of goods and services, peer support, access to adequate PPE and supplies, and expanded workplace flexibility. Importantly, the data also affirmed the presence and toll systemic racism takes on healthcare workers as well as the pressure healthcare workers feel as personal caregivers within their lives and balancing those needs. Finally, the study emphasized the shifting role of the behavioral health workforce in not only supporting patients but also colleagues.


Impact and responses of safety net practices in 20 states to the COVID-19 pandemic, as experienced by frontline clinicians.

Investigator: Donald Pathman, MD, MPH

Objective: This study examined how the pandemic has affected clinicians that serve low-income, racial and ethnic minority communities through safety-net practices, including clinicians participating in the National Health Services Corps (NHSC).

Methods: Using a national sample of safety-net health care practitioners (including physical, dental and behavioral health providers) across 20 states, the study conducted a survey to understand the experiences of these clinicians during the pandemic (October 2020 to January 2021).

Findings: Of the 1,518 participants, three-quarters (76.6%) of the safety-net clinicians scored in at-risk levels for burn-out. Physical and behavioral health providers reported more symptoms of burnout. Yet, dental providers were more likely to be furloughed or to have reduced hours.

Policy Implications: Like the patient population they serve, safety-net clinicians have experienced disrupted work, burn-out, and related mental health challenges. National programs and policies should ensure safety-net practices build cultures of support and prioritize clinicians’ work and their mental health needs.


Professional Structures and Physician Burnout

Investigator: Tania Jenkins, PhD

Objective: Burnout is a prominent and growing topic in the healthcare workforce literature. However, much of the research takes place at the individual level. This study used a novel socio-ecological framework adapted from the National Academies of Science, Engineering, and Medicine (NASEM) systems model to examine the interplay between individual, institutional, professional, and systems-level factors in shaping physicians’ responses to their work conditions across the career span of physicians.

Methods: The ethnographic study took place from December 2020 to August 2021 and examined pediatricians and trainees at an academic primary care pediatrics clinic affiliated with a University. In addition to observations, the study conducted interviews with 65 attendings, residents, medical students, and clinic staff or administrators.

Select Findings: Using the Maslach Burnout Inventory, 63% of those interviewed exhibited at least 1 symptom of burnout: high emotional exhaustion, high depersonalization or low personal accomplishment. Structural factors at the institutional, professional, and societal level shaped physician wellbeing & job satisfaction. Because structural factors exacerbate one another, interventions must be holistically designed to address multiple factors simultaneously

Policy Implications: Policy interventions at various structural levels should consider: 1) institutionally supporting interprofessional teams and spaces, 2) professionally (re)balancing service and learning in GME, and 3) societally empowering physicians of color.


Burnout in “Essential” Workers: Understanding the Unique Experiences of Low Wage Workers in Health Systems

Investigators: Lisa de Saxe Zerden PhD, MSW; Erica Richman PhD MSW; Brianna Lombardi PhD, MSW

Objective: Reports of burnout among the health workforce, particularly in relation to COVID-19, have expanded exponentially. However, burnout research often focuses on clinical providers such as nurses and doctors. Lower-paid, essential workers within health systems provide equally important, but often less lauded, services. This study aimed to understand the unique experiences of low-wage, essential, frontline hospital workers during COVID-19 to better identify strategies to support them during and beyond the pandemic.

Methods: To understand the unique experiences of environmental services workers at hospitals, including food service and house-keeping staff, during the COVID-19 pandemic this study conducted in-depth qualitative interviews with twenty workers from three states.

Results: Seven themes emerged from the interviews. Four themes focused on reported sources of burnout:

  1. Changes in duties and being understaffed;
  2. Fear of potentially contracting COVID-19 and risk in endangering others;
  3. Desire for recognition about risks and importance of their jobs;
  4. Lack of clarity around COVID-19 benefits and perceived unequal distribution of these resources.

Respondents also described protective factors that buffered their experiences:

  1. Employer provision of paid time off specifically designated for COVID-19;
  2. Organizational efforts to provide mental health support;
  3. Self-coping strategies and a sense of pride in their jobs.

Policy Implications: Clear communication on resources, sick leave, mental health supports, and self-coping strategies were beneficial for participants. Hospital systems need to deploy concrete interventions to support workers’ tangible needs, while simultaneously also creating a culture shift in how this type of work is valued across health systems.


Health Care Workforce Playbooks and the COVID-19 Pandemic

Investigators: Anna Dodson, BSPH; Nathan Nelson-Maney, BS; Thomas C. Ricketts, MPH, PhD; Jill Forcina PhD, RN, CNE, CNL, OCN

Objectives: To develop guidelines for training and information on accessing training resources to “surge” healthcare workforce professionals during the COVID-19 pandemic and beyond. These documents are called “Playbooks” and are meant to provide health care systems, states, the federal government and other stakeholders with updated, state-of-the-art resources to promote local flexibility to respond to workforce challenges and shortages.

Methods: The development of these Playbooks was informed by the principles of patient-centered care where the individual’s health needs and goals drive all health care decisions (NEJM Catalyst 2017). The team used expert advice from a range of informants: clinicians, policy makers, and administrators to develop the playbook guideline. In these playbooks, we describe how the targeted skillsets of the healthcare workforce can be matched to patients’ needs and goals no matter who delivers the care or support.

Results & discussion: The Playbooks for the five domains of care are available online. They cover: 1. Home-Based Care, 2. Ambulatory Care, 3. Inpatient medical care, 4. Long-Term Care., and 5. Critical Care. Two of the Playbooks are supplemented by a free-standing “checklist” to further assist in planning. These playbooks are available via the NC AHEC website (and below) and it is our intention to modify them as needed. The uptake of the playbooks will be assessed in follow-up communications with users.


Emergent COVID-19 Research


State-Based Approaches to Leveraging Medicaid-Funded Graduate Medical Education

Principal Investigator: Erin Fraher, PhD, MPP; Jacob A. Rains MPH; Thomas J. Bacon DrPH; Julie Spero MSPH; Emily Hawes PharmD, BCPS, CPP

Total Medicaid funds invested in graduate medical education (GME) increased from $3.78 billion in 2009 to $7.39 billion in 2022. States have considerable flexibility in designing Medicaid GME payments to address population health needs. The goal of this study was to use states as “policy laboratories” to understand states’ impetus for using Medicaid funds for GME, the structure of their investments, the composition and charge of advisory bodies that guide these investments, and the degree of transparency and accountability in place to track whether Medicaid GME investments achieved desired workforce outcomes. Two rounds of structured interviews were conducted in 2015-2016 and 2019-2020 with key stakeholders from 10 states. Interview transcripts were analyzed and coded in six thematic areas: impetus for change, payment, advisory bodies, transparency, accountability, and challenges. Findings from this study can inform individual state efforts and guide federal policy makers interested in convening learning collaboratives to share best practices and strategies to address common challenges encountered by states in designing Medicaid GME to meet their workforce needs.


Developing a GME Policy Toolkit that States Can Use to Evaluate Return on Investment for Public Funds Invested in Training

        Investigators: Erin Fraher, PhD, MPP; Tom Ricketts, PhD, MPH; Ryan Kandrack, PhD

This study will develop and refine a methodology that states can use to evaluate the return on investment for public funds spent on Graduate Medical Education (GME). Armed with data about the outcomes of GME funds invested—in terms of yield of physicians in needed specialties and geographies–states can make more informed decisions about where to target future GME funds to produce the physician workforce needed to meet population health needs and ensure value for their investment.


Association Between Title VII Funding for Medical Schools and Physician Workforce Outcomes

Investigators: Ryan Kandrack, Erin P. Fraher, Thomas C. Ricketts

Objective: To examine the association between attending a Title VII-funded medical school and probability of practicing in primary care or underserved communities.

Data/Setting: We used data from the 2016 AMA Physician Masterfile, the Area Health Resource File, the American Community Survey, the Integrated Postsecondary Education Data System, Title VII grants data from the HRSA Data Warehouse, and NHSC participant data obtained from HRSA. The sample included physicians graduating from medical school between 2004 and 2010.

Design/Methods: Cross-sectional multivariable regression analyses of specialty and practice location as of 2016, comparing physicians exposed to Title VII funding during medical school relative and those who were not exposed.

Results: Exposure to both pre-doctoral and departmental Title VII funding was statistically significantly and negatively associated with practicing in a whole county HPSA. The amount of Title VII funding was statistically significantly and negatively associated with practicing in a whole county HPSA. These results were extremely small in magnitude. Exposure to Title VII funding was not statistically significantly associated with any other outcomes.

Conclusions: A longitudinal analysis examining trends before and after receiving Title VII funds and more years of data are needed to more systematically examine the outcomes of Title VII funding.


Investment in Graduate Medical Education Outcomes: An Analysis of Pediatric GME at the Residency Program Level

Investigators: Thomas C. Ricketts, PhD, Erin P. Fraher, PhD, MPP, Ryan Kandrack, PhD(cand)

Objectives: To assess the value of pediatric graduate medical education (GME) programs in producing the workforce needed at the state and national level.

Data/Setting: Physicians identified as graduating from an ACGME-approved general pediatrics residency training program in the United States.

Design/Methods: Cross-sectional analysis examining the practice characteristics in 2016 of physicians who graduated from pediatric residency training programs on or before 2005. Analyses used the 2005 and 2016 American Medical Association Masterfiles, the 2017 Area Health Resources File (HRSA) and residency data from the ACGME.

Results: On average across programs, more than half of pediatricians moved from their state of residency but retention varied by state/territory (range 7.6% to 71.2%) and training program (range 4.4% to 93.0%). There was also variation across programs in the tendency of graduates to remain in general pediatrics (range 46.2% to 96.8%) and practice in a non-metro location (range 1.6% to 14.0%) or HPSA (range 2.3% to 10.6%).

Conclusions: Workforce outcomes vary significantly by training program. Future research could combine data in this analysis with Children’s Hospital GME (CHGME) data reported to HRSA to further develop and refine metrics for evaluating program outcomes.


DocFlows: Mapping GME Location and Practice Location by Specialty

Investigators: Erin Fraher PhD, MPP, Erica Richman, PhD, MSW, Evan Galloway, MPS, Andy Knapton, MSc

Issue: Federal efforts to reform graduate medical education (GME) have stalled and states are actively exploring ways to use Medicaid and state appropriations to produce the workforce needed to meet population health needs. In our previous work, we documented these efforts and noted that states have voiced the need for better data. To address this need, we developed the DocFlows web app (, which allows users to query, download and share maps/graphs of interstate moves by residents and actively practicing physicians in 35 specialties.

Methods: The 2009 and 2015 American Medical Association Masterfiles® were merged and physicians’ practice locations in each year were compared. These data are reported separately for the 672,606 physicians who were in active practice in both years and for the 116,799 residents who were in training in 2009 and in active practice in 2016.

Findings: California, Florida, and Texas were the largest net importers of physicians and had the highest retention rates for actively practicing physicians, keeping over 90% of their actively practicing physicians in-state between 2009 and 2015. Wyoming had the lowest retention rate, keeping just 18% of residents and 76% of practicing physicians.

Discussion: The DocFlows app gives states information about where their physician workforce trained and the out-migration of their workforce to other states. There is significant migration between states of newly trained and actively practicing physicians, and many of these moves are between regions. Any change a state implements to expand GME or increase retention will affect other states.


The Future of Medicaid-Funded Graduate Medical Education as the Health Care System Shifts

Investigators: Erin P. Fraher, PhD, MPP; Julie C. Spero, MSPH; Tom Bacon, DrPH

Objective: In the absence of federal Graduate Medical Education (GME) reform, states are increasingly exploring ways to leverage Medicaid funds to shape the size, specialty mix and geographic distribution of their workforce.

Data/Setting: Ten states that had implemented, or planned to implement, GME reform were studied.

Design/Methods: Two structured interviews were conducted with key informants in each state to assess the structure of Medicaid GME payments, identify tracking and accountability metrics, and determine if GME funds were targeted to needed specialties or geographies. Interviews were coded and analyzed using directed content analysis.

Results: More states were in the planning stages of GME reform than had actually implemented changes. States engaged in reforming GME:

• developed alternative funding mechanisms;

• created oversight structures to drive GME reform; and

• noted there was little or no transparency/accountability of funds spent on GME.

Most states had little appetite for redistributing existing funds, instead seeking new funds to expand GME.

Conclusions/ Implications for Policy: More data, increased analytical capabilities and agreement on GME workforce metrics are needed. Uncertainty about the future of Medicaid and payment policy at the federal level may slow reform, but could accelerate it if states perceive that proposed Medicaid block grants provide another stimulus to change the way they invest in Medicaid GME.


Making Use of Workforce Projections to Inform the Graduate Medical Education Policy Debate in the United States

Author: Erin Fraher, PhD, MPP, Andy Knapton, MS, & Mark Holmes, PhD

Background: There is intense policy debate over whether the nation should expand graduate medical education (GME) or redistribute existing slots to high-need specialties and geographies. The recently released Institute of Medicine (IOM) report recommends creating a GME Policy Council to use data and workforce projections to more strategically target the $13 billion public dollars spent on GME annually. At the same time the IOM report was released, a new physician projection model—The FutureDocs Forecasting Tool—was launched. The model findings suggest that there will be significant shortages of physicians to meet the demand for health care services in some geographic areas while there will be surplus capacity in other geographies.

Objective/Aims: This project uses the projections released in The FutureDocs Forecasting Tool to develop scenarios about how GME slots might be redistributed from higher-capacity places to areas where there are not enough physicians to meet the population’s need for healthcare services. The project will show the outcome of redistributing GME slots according to three different scenarios:

A modest redistribution of slots that will bring the states and specialties most in shortage up to a lesser level of shortage;

A more significant redistribution of slots that will draw from overcapacity states and specialties to address workforce shortages in undercapacity states and specialties; and

An expansion of slots by 30,000 to be distributed according to the methodology described in the Resident Physician Shortage Reduction Act of 2013.

Methods: The number of visits in shortage/surplus for each type of health care service will be derived using the FutureDocs Forecasting Tool. The model’s plasticity matrix will be used to translate shortage/surplus visits into the specialty types of physicians who provide these visits in different states.

Policy Relevance: The purpose of this project is to demonstrate how workforce projections can be used to inform policy decisions regarding GME investments. The project will identify the geographies and specialties that should be targeted for GME expansions and show how this expansion could be accomplished by redistributing existing GME positions rather than increasing the total number of positions.

Understanding Registered Nurse Turnover in the U.S.

Investigators:  Cheryl B. Jones, PhD, RN; George Knafl, PhD; Meriel McCollum, BSN, RN, PhD student

Overview: Although nurse workforce shortages are not new, concerns about a shortfall of nurses have
increasingly intensified across healthcare settings during the COVID-19 pandemic. The purpose of this study was to provide a baseline understanding of RN turnover through a cross-sectional analysis of the most recent data (2020 release) from the National Sample Survey of Registered Nurses.

Methods: Data from the 2018 National Sample Survey of Registered Nurses (NSSRN), a publicly available and nationally representative data source produced by the National Center for Health Workforce Analysis in collaboration with the US Census Bureau, was used to model RN turnover that occurred between December 2016 and December 2017. Descriptive analyses and logistic regressions using sampling weights helped the study examine how RN turnover varied by the following factors: sociodemographic, professional, and economic.

Findings: Forthcoming when manuscript is published and publicly available.

Policy Implications: These data shed light on the areas where health care leaders can target interventions to retain nurses most at risk of turnover while policymakers can tailor programs to meet the needs of different sociodemographic, professional, employment, and economic groups.

Supporting the Dynamic Careers of Licensed Practical Nurses: A Strategy to Bolster the Long-Term Care Nurse Workforce

Investigators: Cheryl B. Jones, PhD, RN, FAAN, Meriel McCollum, PhD, RN, Alberta K. Tran, PhD, RN, CCRN, Mark TOles, PhD, RN, FAAN, George J. Knafl, PhD

Objective: The nature of careers has changed as employees gain greater ability to move between organizations, vocations, and modes of employment, such as full-time, part-time, per diem, at-home, web-based, and self-employment. Traditional models of work are being replaced by dynamic patterns of employment that are based on the “boundaryless” career, in which employees are more mobile and able to work across organizational and vocational boundaries (Arthur & Rousseau, 1996). With the rise in demand for LPNs, particularly in long-care settings, understanding the characteristics of LPN participation in the workforce is crucial to better address potential shortages based on this growth.

Data and Methods: Drawing on the theory of boundaryless careers, the investigators examined longitudinal employment data from LPNs in North Carolina (NC) and described patterns in LPN licensure and career transitions.
Results: Two career patterns were identified: (a) the continuous career, in which LPNs were licensed in 75% or more of the years they were eligible to be licensed and (b) the intermittent career, in which lapses in licensure occurred.
Conclusions: Findings indicated that LPNs who made job transitions were more likely to demonstrate continuous careers, as were Black LPNs. These findings suggest the importance of organizational support for LPN career transitions and support for diversity in the LPN workforce.


Barriers and facilitators for the LPN-to-RN transition: Perspectives from practicing LPNs

Investigators: Cheryl B. Jones, PhD, RN, Mark Toles, PhD, RN, Anna Beeber, PhD, RN, Meriel McCollum, RN BSN, Madeline Neal, BS

Objectives: Examine barriers and facilitators of LPN-to-RN transitions.

Data/Setting: Secondary analysis of data gathered in 2015 by the North Carolina Foundation for Nursing Excellence (NCFFNE). Over 3,400 LPNs (18% of licensed LPNs) responded to the survey.

Design/Methods: Qualitative content analysis of open-ended questions provided information on LPNs’ interest in pursuing an LPN-to-RN transition, the barriers to pursuing a transition, and information about educational goals and transition opportunities. Responses were used to characterize barriers and facilitators.

Results: 70% of survey respondents were over the age of 40; 42% worked in home health, hospice, or long-term care/assisted living. About 75% were interested in pursuing a LPN-to-RN transition. The nature of transitions were characterized by LPNs’ life stage and their health; barriers reflected LPNs’ views of academic programs, employers, and their personal perspectives about the transition and the RN role. “Modifiable” facilitators of LPN-to-RN transitions were identified in these areas.

Conclusions: LPNs who become RNs possess human capital from previous work experience, understand the unique pressures that accompany patient care delivery, and are more diverse than RNs. Institutions must recognize the value of LPNs on the care team. Efforts to improve and value the work of both LPNs and RNs may foster LPN-to-RN transitions.


Predicting Role Transitions for the LPN-to-RN Workforce in North Carolina

Investigators: Cheryl B. Jones, PhD, RN, FAAN; Mark Toles, PhD, RN; George J. Knafl, PhD; Anna S. Beeber, PhD, RN

Objective: Licensed practical nurses (LPNs) are an untapped resource in the health care workforce because they represent a group that can transition to become registered nurses (RNs), and also bring diversity to the RN workforce (IOM, 2011). We know very little about the factors that affect LPN-to-RN transitions, or the policies needed to foster LPN-to-RN transitions. This study examined the career trajectories of LPNs in North Carolina (NC) to describe predictors of their transitions to the RN role.

Data/Setting: Data on LPNs, including those who transitioned to become RNs, were drawn from the NC Health Professions Data System (HPDS). The HPDS is comprised of annual licensure files from various licensing boards in NC, including the NC Board of Nursing.

Design/Methods: A retrospective cohort design was used to study LPN to RN professional transitions. This project developed a concatenated dataset, using data from 2000 to 2013, to examine LPN-to-RN transitions that occurred between 2001 and 2013.

Results: Using logistic regression, several characteristics of LPNs at the time of their licensure were associated with a greater likelihood of a LPN to RN transition: obtaining LPN licensure between 1996-2013, and particularly between 1996 and 2004; obtaining licensure at younger ages; obtaining nursing education in the US; being employed in a hospital setting; being employed part-time; and specializing in medical/surgical nursing.

Conclusions: While very little is known about LPNs who transition to become RNs, they are a valuable resource for expanding the RN workforce. Strategies are needed to support LPN-to-RN transitions, including strategies that focus on individuals enrolled in LPN programs, on LPN education programs, and on LPN employers. Incentives are needed to foster LPN career development and bring diversity to the RN workforce.


Pathways in Nursing Careers: Transitions from the Role of Licensed Practical Nurse to Registered Nurse in North Carolina

Investigators: Cheryl B. Jones, PhD, RN, FAAN Mark Toles, PhD, RN, George J. Knafl, PhD, Anna S. Beeber, PhD, RN

Background: A more diverse registered nurse (RN) workforce is needed to provide healthcare in North Carolina (NC) and nationally. Studies describing licensed practical nurse (LPN) career transitions to RNs are lacking. This study explored the feasibility of using NC Health Professions Data System (HPDS) data to describe LPN-to-RN transitions in NC and, to the extent possible with the data, to characterize the occurrence of LPN-to-RN professional transitions.

Methods: A retrospective cohort design was conducted using licensure data on LPNs from 2001 to 2013. Cohorts were constructed based on year of graduation.

Results: It was partially feasible to use the HPDS data to describe LPN-to-RN professional transitions. Out of 39,398 LPNs in NC between 2001 and 2013, there were 3,162 LPNs (8.0%) that had a LPN-to-RN career transition between 2001 and 2013. LPNs were more likely to transition to RN if they were male; from Asian, American Indian, or other racial/ethnic groups; held a baccalaureate, associate or doctoral degree in their last year as an LPN; worked in a hospital in-patient setting; worked in the medical/surgical nursing specialty; and were from a rural area.

Limitations: This study’s findings are limited by the use of secondary data, from only one state, and with missing data elements, and the use of descriptive analyses.

Conclusion: This study fills an important gap in our knowledge of LPN-to-RN career transitions. Our findings indicate that the odds of an LPN-to-RN transition were greater if LPNs were: male; from all other racial groups than white; of a younger age at first LPN licensure; working in a hospital setting or in the specialty of medical-surgical nursing; employed part-time or working in a rural setting during the last year as an LPN.

Are Behavioral Health Providers Co-located with DEA Waivered Buprenorphine Prescribers?

Investigators: Lisa de Saxe Zerden, PhD, MSW; Brianna Lombardi, PhD, MSW; Erica Richman, PhD, MSW; Alex Forte, MSW; Evan Galloway, MPS

Objective: Targeted planning for co-location of DEA-waivered buprenorphine prescribers and BH clinicians could increase the use of medication for opioid use disorder (MOUD). Presently, less than half of all waivered prescribers, outside of hospitals, are co-located with BH clinicians. This study sought to provide a national benchmark to assess the degree to which DEA-waivered providers are co-located with behavioral health providers, and thus, implementing best practice models for MOUD.

Methods: To understand the rate of co-location of DEA-waivered providers and behavioral health providers, we conducted spatial analysis using two publicly available data sources. We identified the physical location of DEA waivered prescribers through the Substance Abuse and Mental Health Services Administrator Buprenorphine Practitioner Locator and the physical location of social workers and psychologists using the CMS National Plan and Provider Enumeration System (NPPES) downloadable file.

Findings: 1) Waiver size predicted colocation. Prescribers with the smallest waivers (>30) were more likely to be co-located with BH providers 2) There were significant geographic difference in co-locations. Prescribers in urban or metro areas were more likely to be co-located with BH providers 3) Co-location occurred differently at the individual, county, and state level. States with expanded Nurse Practitioner scope of practice regulations were more likely to be colocated, across all provider types

Where is Behavioral Health Integration Occurring? Using NPI Data to Map National Trends

Investigators: Erica L. Richman, PhD, MSW, Brianna Lombardi, PhD, MSW, Lisa de Saxe Zerden, PhD, MSW, Randy Randolph, MRP

Issue: Because most primary care visits include a behavioral health component, providing integrated services is now considered a priority for health systems to meet patient needs. Integrated care is a model where behavioral healthcare workers work on teams with primary care physicians (PCPs) to help address behavioral and social determinants of health alongside physical health. Co-location, where both types of providers work in the same physical space, is one key element of integration. Little is known about the rate of expansion of co-located services in the U.S. since the passing of legislation and increased federal funding incentivizing co-location.

Methods: This study was undertaken in collaboration with the Behavioral Health Workforce Research Center at the University of Michigan. National Plan and Provider Enumeration System data were used for analyses. PCPs reporting at least one of five specialties were included (family medicine, general practice, internal medicine, pediatrics, and obstetrics/gynecology). Behavioral health providers were limited to social workers and clinical psychologists. Provider practice addresses were geocoded to latitude and longitude coordinates using the Environmental Systems Research Institute StreetMap database and ArcGIS software. The geocoding system provided information on the quality of each geocoding result and an algorithm was applied to choose the best address. Straight-line distances between practice locations of behavioral health providers and PCPs were summarized and distances <0.01 miles were considered co-located.

Findings: Of the 380,690 PCPs in the sample, >44% were co-located with a behavioral health provider. Providers in urban settings were much more likely to be co-located than providers in rural locations. Smaller practices were significantly less likely to be co-located than larger ones and were more likely to be located in rural settings. Rate of co-location varied by PCP specialty; pediatricians, obstetricians/gynecologists, and internal medicine physicians were most likely to be co-located, whereas family medicine and general practitioners were least likely to be co-located.

Discussion: Co-location is occurring less frequently in rural settings and in smaller practices; these are the practices that will need greater assistance achieving integrated healthcare. Increasing rates of co-location by incentivizing behavioral healthcare providers to work alongside physicians could reduce physician burnout by minimizing the rural physician’s scope of practice and potentially reduce feelings of seclusion. The health workforce must be trained to work in integrated settings and understand how practice can incorporate both physical and behavioral health needs concurrently.


Social Work in Integrated Primary Care: A Systematic Review

Investigators: Mark Fraser, PhD, MSW; Brianna Lombardi, MSW; Shiyou Wu, MSW; Lisa de Saxe Zerden, PhD, MSW; Erica Richman, PhD, MSW; Erin Fraher, PhD, MPP

Objective: To describe the roles and functions of social workers in integrated healthcare settings.

Design/Methods: A systematic review was undertaken of randomized controlled trials (RCTs) of routine versus integrated care, where social workers provided patient care on interprofessional teams. A 5-phase review process was used to search eight electronic databases and the grey literature. Searches were conducted by two independent researchers who screened citations and extracted data from studies published between 2000 and 2016.

Results: The searches recovered 502 citations. After screening the abstracts, 107 reports were retained for a full-text review. Of these, 32 reports from 26 RCTs met study criteria.

Conclusions: The 26 RCTs revealed that three often overlapping core functions define the scope of practice for social workers in integrated care: 1) providing clinical interventions for patients with behavioral health problems; 2) managing care plans for patients with chronic conditions; and 3) engaging community resources on behalf of patients. Social workers on interprofessional healthcare teams are improving patient health outcomes by helping mitigate barriers caused by negative social determinants of health—without increasing costs. Findings from these 26 studies reveal that team based health care inclusive of social work is one of the building blocks that can transform our healthcare system to better address population health.


Toward a Better Understanding of Social Workers on Integrated Care Delivery Teams

Investigators: Brianna M. Lombardi, MSW, Lisa de Saxe Zerden, MSW, PhD, Erica L. Richman, MSW, PhD

Background: Social workers’ training and knowledge of psychosocial risk factors, behavioral health screening, assessment and intervention, and focus on the adaption of services to be culturally inclusive makes the profession uniquely positioned to assist in the treatment of the “whole person” in integrated care settings. However, there is a limited understanding of what social workers are currently doing in integrated care settings. Until now, work exploring social worker roles in integrated settings has been theoretical in nature and limited by sample size or geographical reach.

Objective: To address these gaps, this study used a convenience sample of Masters of Social Work (MSW) students throughout the US in integrated field placement settings and their MSW field instructors (N=395) to clarify how this workforce, not traditionally captured in workforce research contributes to integrated healthcare.

Methods: An electronic survey was developed using Qualtrics and administered to HRSA-funded Behavioral Health Workforce Education and Training (BHWET) MSW students and their field instructors. The survey focused on understanding the roles, tasks, and interventions of social workers in integrated health care.

Key findings:

  • The most commonly used skills were: team-based care; motivational interviewing; psychoeducation; using the social determinants of health; and adapting services to be culturally inclusive. The least used skills were medication management; SBIRT (screening, brief intervention, and referral to treatment); warm hand-offs; functional assessment of daily living skills; and behavioral activation.
  • Respondents had knowledge of or education related to most core competencies of integrated practice. However, many indicated they had not learned about SBIRT (34%); behavioral activation (25%); problem-solving therapy (19%); huddles (18%); or warm hand-offs (18%).
  • Most were co-located with the rest of the integrated care team (62%). About 80% talked with the team in person at least weekly, with more than 42% doing so daily. Participants who were co-located or worked in inpatient settings were more likely to communicate with team members in person
  • Over 53% reported that team members always have access to the same electronic health record (EHR), but 15% indicated team members never use the same EHR.

Conclusion: Findings suggest the importance of programs to train and deploy social workers in integrated settings, such as the BHWET federal funding mechanism, as MSW students appear to be learning the necessary skills needed to work in integrated care. However, social workers currently in practice still require retooling and training. The majority of field instructors indicated learning tasks and skills “on the job.” This education gap provides an opportunity for MSW educators to develop continuing education curriculums to support and re-tool the current social work workforce.


Social Work and Electronic Health Records: A New Frontier for Health Workforce Research

Investigators: Lisa de Saxe Zerden, MSW, PhD, Erica L. Richman, MSW, PhD, Brianna M. Lombardi, MSW, PhD, Kim Shoenbill, MD, PhD, Erin Fraher, MPP, PhD.

Introduction: As one of the largest groups of clinically trained mental health pro­viders in the United States, social workers are increasingly deployed on integrated health teams to address patients’ social determinants of health and provide behavioral health interventions. However, information about the specific content of social work practice in new models of healthcare is limited, and further exploration is needed to better define social worker roles and functions in integrated settings. This study aimed to demonstrate how EHR data can be used as a workforce research tool to assess the scope, contributions and value of social work, a profession still in the process of establishing its return on investment in health care.

Methods: To create a preliminary lexicon we conducted expert interviews and focus groups with 30 practicing social workers representing more than 20 primary care clinics, medical providers and informatics experts. These interviews, in conjunction with academic literature, formed the basis of the lexicon that was updated iteratively throughout the study. EHR data were obtained from the North Carolina Translational and Clinical Sciences Institute (TraCS), a broker of the Carolina Data Warehouse for Health (CDW-H), the data repository for the UNC Health Care System. A random sample of 60 patients who had clinical contact with a social worker between September 1, 2016 and August 31, 2017 were selected and 647 notes were reviewed and analyzed to evaluate social workers’ documentations.

Results: There were sizeable challenges to using EHR documentation to help clarify the role of social workers in integrated primary care settings. Because social workers do not bill for their services, they do not create their own clinical encounters and often append documentation to already existing provider notes. Furthermore, social workers were not consistently referred to as social workers but as case managers, care manager or other titles. However, in analyzing notes, we found social workers played broad roles in integrated primary care settings, both in direct patient care and management, and as part of a dynamic health care team.

Discussion: As this study helps demonstrate, EHR documentation shows social workers are regularly contributing to patient care and working as members of health care teams. Yet, if EHRs are to become a robust resource for workforce and health care researchers, informed and purposeful changes to the interface and usability of EHR systems are required in order to more fully understand the scope of practice and contributions of the social work health workforce.

  • Applicability and Usability of the EHR. This study helped identify limitations of EHR structures that impact the way social workers can document their work. Allowing clinical encounters to be created by providers regardless of billing permissions could increase the data abstraction process and general usability of data found within the EHR. Further, social workers consistently referring to themselves in documentation as their profession (i.e. social worker) as opposed to their job title (i.e. care manager) could also help improve the quality and ease of obtaining data.
  • Valuing Contributions of Social Work Interventions. One way to offset the challenge of understanding the complexity of social work interventions as documented in the EHR is to develop and expand how value is calculated.
  • Social Work Education Needs to Include EHR Documentation Strategies. As the EHR becomes a key source for helping define the social work workforce in healthcare, social workers should be trained to document in clear and consistent ways that help alleviate common barriers to EHR research.

Learn more:


Office Based Opioid Treatment (OBOT)—the Workforce Treating Opioid Use Disorder

Investigators: Lisa de Saxe Zerden, PhD, Brianna Lombardi, MSW, Erica Richman, PhD, Anjalee Sharma, MSW

Background: The current opioid crisis in the United States is a recognized national health emergency. The number of opioid-related deaths has more than quadrupled since 1999. Over 42,000 individuals died from opioid use in 2016 alone. To combat this epidemic, primary care providers are expanding clients’ access to care, particularly to medication-assisted treatment (MAT) programs, also referred to as office-based opioid treatment (OBOT). As primary care office-based treatment expands, understanding the workforce needed to effectively deploy this model of care is critical.

Objective: This exploratory analysis had three primary research questions:

  1. Which professionals comprise the workforce that provides MAT in primary care?
  2. How do OBOT teams communicate about patient care?
  3. What are the behavioral components of MAT provided in primary care settings?

Methods: We conducted 20-minute interviews with professionals working in expert OBOT teams across the United States and Health and Human Services regions. Twelve experts from 11 outpatient primary care clinics in several Northeast, mid-Atlantic, Southeast and Midwest states agreed to interviews.

Results & discussion: While every OBOT team must include a DEA-waivered medical provider and there were a few consistent trends in roles, we also found significant variation in team composition and workforce. The primary team components we found were prescriber, behavioral health provider, MAT registry coordinator, other team members and operation staff. Every team had a team member serving as behavioral health provider, most often a social worker (MSW/ LCSW). The most commonly employed psychosocial interventions echoed Fraser and colleagues’ (2018) identification of three primary roles performed by social workers on integrated behavioral health teams: individual behavioral health treatment (i.e., motivational interviewing, CBT), case management and referral services. For OBOT teams to function most effectively, purposeful, structured communication and defined meeting times can help ensure well-coordinated team-based MAT from the various professions included on the OBOT team.

Policy implications: As the treatment needs of people with OUD continue to warrant national attention, efforts will need to be targeted toward developing the diverse, team-based workforce needed to address the complexities of collaborative OUD treatment. Improving MAT across primary care settings will require workforce researchers, health systems and educators to recognize how the services provided by different behavioral health providers contribute individually and collectively to comprehensive OBOT practices. Effectively incorporating behavioral health providers’ skill sets will require greater understanding of the unique contributions of various types of behavioral health providers, from peer-support specialists to LCSWs.


What EHRs Tell Us about How We Deploy Health Professionals to Address the Social Determinants of Health

Investigators: Erica Richman, PhD, MSW, Brianna Lombardi, PhD, MSW, Lisa de Saxe Zerden, PhD, MSW

Background: Increasing aware­ness of the Social Determinants of Health (SDOH) has prompted health systems to im­plement strat­egies to screen for and address pa­tient SDOH. These resources are valuable for health systems but do not speak to the workforce needed to implement SDOH screening and re­lated inter­ventions. Little research has exam­ined which health care workers screen for SDOH, who actually addresses needs related to patient SDOH, and how to incorporate screening and intervention into the clinic work­flows.

Objective: This feasibility study used EHR data from a large health care system located in the south­east­ern United States to describe the mix of profession­als identifying and addressing pa­tient SDOH. The following research questions were considered:

  1. How well does EHR documentation identify which health professionals act to address patient SDOH?
  2. What actions regarding patient SDOH are documented in the EHR?

Methods: Two domains of SDOH, food insecurity, and housing insecurity, were studied. Researchers used the Electronic Medical Record Search Engine (EMERSE), a program that sys­tematically searches EHR notes using key words or terms to identify a patient pool. Limits were set to include only patients over age 17 and who had a note docu­mented between September 1, 2017 and August 31, 2018. After the final pool of notes was identi­fied, analysts at the Carolina Data Warehouse ran­domly extracted 480 notes (240 from each SDOH: food and housing insecurity) for researchers to an­alyze. Using a random number generator, research­ers randomly sampled 60 patient’s notes under each SDOH category, all of which contained one or more of the selected terms.

Results: The mean age of patients was 52 years of age (SD=16), were slightly more likely to be female (53%, n=63), and most were either white (49%, n=59) or black (40%, n=48) (see table 1). Out of 120 notes that contained a reference to food or housing insecurity, 72% (n=86) also contained information on an intervention or action that was taken to help address a patient’s need related to the specific SDOH. When an intervention was taken to help ad­dress food insecurity, 62% (n=29) of providers were social workers, 15% (n=7) were dieticians, and the remaining 11 providers were a mix of registered nurses (RNs), community health work­ers, medical assistants, physicians, and others. When an inter­vention to help address housing inse­curity was documented, 64% (n=25) of providers were social workers, 10% (n=7) were physicians, and the re­maining 10 providers were a mix of chap­lains, RNs, care manager assistants, and others.

Conclusion: This study identified that health care team members are screening and addressing patient SDOH. The analysis revealed that with dieticians, chaplains, and other health care professionals, so­cial work is the pri­mary workforce addressing pa­tient needs related to SDOH. Preliminary evidence supports social care as an effective intervention to improve patient health and well-being behavioral health providers, from peer-support specialists to LCSWs. Future work is needed to identify the most effective mix of health team members to address SDOH, as well as evaluate sustainable financial models to meet patient social needs.

Co-location of Pharmacists with Primary Care Providers: An Analysis of NPI vs Other Data Sources

Investigators: Emily M. Hawes, PharmD, BCPS, CPP; Brianna Lombardi, PhD, MSW; Evan Galloway, MPS; Hilary A. Campbell, PharmD, JD; Cristen P. Page, MD, MPH; Mary Roth McClurg, PharmD, MHS

Objective: Despite evidence supporting the integration of pharmacists in team-based primary care, little information exists on the co-location of pharmacists with primary care physicians in the United States and even less information exists on the factors associated with these models in primary care. This study sought to understand the degree to which pharmacists are co-located with primary care practices and characteristics associated with co-location.

Methods: This study used the National Plan and Provider Enumeration System’s (NPPES), National Provider Identifier (NPI) database, a national and publicly available data source that includes all health providers with an NPI, and CMS data to establish geographic co-location of pharmacists and primary care providers. Further, the study compared pharmacist-PCP co-location across a number of covariates. It assessed rurality, setting type, Medicaid expansion states, and expanded scope of work states to determine where co-location was more likely based on these factors.

Findings: The sample included 502,373 Physicians (51% Internal Medicine, 28% Family Medicine, 21% Other PCP sub-specialties) and 221,534 Pharmacists.

  • 23% percent of pharmacists in the (n=51,034) were co-located with a PCP.
  • Co-location varied by state, and within state by urban/rural geography
    • Pharmacists in states that have expanded Medicaid were more likely to be colocated.
    • Pharmacists in urban areas were more likely to be co-located with PCPs.
    • States with expanded pharmacist scope of practice regulations had higher rates of colocation.
  • Co-located pharmacists are most commonly working in larger physician practices

Policy Implications: The co-location of pharmacists and primary care physicians can increase access to integrated interprofessional care models, which are a critical component of high-quality primary care. This study provides a benchmark to understand the current state of this type of co-location. Understanding where co-location is occurring and predictors for pharmacist-PCP co-location can provide information to policymakers, payors, and healthcare organizations to bolster co-location.


Experience of Physician Assistants and Nurse Practitioners in Onboarding Programs

Investigators: Nathalie Ortiz Pate, MPH, MHS, Hilary Barnes, PhD, Lorraine Anglin, MHS, Mara Sanchez, MMS, Heather Batchelder, MA, LPA, Christine Everett, Ph.D., Perri Morgan, Ph.D.

Objective: Newly graduated nurse practitioners (NPs) and physician assistants (PAs) face challenges in adapting to their first jobs. Because of this, some organizations are instituting formal onboarding programs to support PAs and NPs as they transition from students to practicing clinicians. Yet, more research is needed to identify which components of NP and PA onboarding programs are successful and why. This study seeks to address that gap.

Methods: Between 2020-2021, the study conducted thirteen primary care NPs and PAs semi-structured interviews. Respondents answered questions about their onboarding programs, noting strengths and weaknesses of the programs and whether they felt their transition to clinical practice was successful after the program.

Findings: Findings were coded into six structural components and two psychosocial components of training programs. Structural components included:

  1. Improving competencies
  2. Providing mentorship
  3. Orientating to organizational dynamics
  4. tailoring ramp up of patient scheduling
  5. Clarifying expectations
  6. Providing clear organizational support.

Psychosocial components of training programs were described as:

  1. Creating a feeling of comfort
  2. Building professional self-confidence.

Policy Implications: Employers and organizational managers can use onboarding programs to increase retention and decrease employee turnover, which is expensive and disruptive to patient care. This study highlights six structural components and two psychosocial components that employers and managers could strive to incorporate into their onboarding programs.


Geographic Access Measure for Primary Care in Rural Areas

Investigators: Mark Holmes, PhD, Paul Delamater, PhD

Objective: A key issue in the urban-rural disparity policy debate is the use of different definitions of rural. This study sought to: (1) assess the degree to which rural definitions identify the same areas as rural; and (2) assess rural-urban disparities identified by each definition across socioeconomic, demographic, and health access and outcome measures.

Methods: The study determined the rural status of each census tract and calculated the rural-urban disparity resulting from each definition, as well as across the number of definitions in which tracts were designated as rural (rurality agreement). Using 8 federal rural definitions, the study examined the population in 72,506 census tracts. Population characteristics included percent with a bachelor’s degree, income below 200% poverty, population density, percent with health insurance and whether various health care services were within 30 minutes driving time of the tract centroid.

Findings: The rural population varied from slightly < 6.9 million people to >75.5 million across definitions. The largest rural-urban disparities emerged using Urban Influence Codes. Urbanized Area and Urbanized Cluster tended to generate smaller disparities. Population characteristics such as population density and percent white had notable differences across levels of rurality, while others such as percent with a bachelor’s degree and income below 200% poverty varied consistently.

Policy Implications: The study suggests that rural-urban populations and disparities are sensitive to the specific definition used. The relative strength of definitions varied across population characteristics. Researchers and policymakers should carefully consider the choice of outcome and region when deciding the most appropriate rural definition.


Descriptive Analysis of the IQVIA OneKey Database

Investigators: Ryan Kandrack PhD; Erin Fraher, PhD

In this study, we explored the IQVIA OneKey nursing, physician assistant, dentist, and Accountable Care Organization files to: document what data are in the files, numbers of health professionals in each profession compared to national estimates from other, established data sources, and identify the types of analyses the OneKey data would be well-suited for compared to other workforce datasets. Such an analysis serves as a resource for future health workforce research.

  • Brief developed and shared internally with HRSA.


Integration of Rehabilitation Care from the Acute to Community Setting: The Role of Physician Referral

Investigators: Janet K. Freburger, PT, PhD; Samannaaz S. Khoja, PT, PhD; Timothy S. Carey, MD, MPH

Objectives: To examine trends of primary care physician (PCP) referral to physical therapy (PT) and other physicians and to identify factors associated with PT referral.

Data/Setting: Office-based ambulatory care practices in the U.S.

Design/Methods: Analysis of data from the National Ambulatory Medical Care Survey (2003–2014). We calculated adjusted rates of referral by year and examined predictors of PT referral in multivariate models.

Results: PT referral rates declined from 19.9/1,000 visits in 2003-2004 to 11.4/1,000 visits in 2013-2014 (p=0.01), while physician referral rates increased from 88.7/1,000 visits to 115.1/1,000 visits (p<.001). 71.5 percent of visits referred to PT were for musculoskeletal conditions. Physician and practice characteristics associated with a decreased likelihood of PT referral for musculoskeletal conditions included narcotic prescription; being the patient’s PCP; and Medicare patient revenue>50%. Physician and practice characteristics associated with an increased likelihood of PT referral included being a doctor of osteopathy; practice ownership by a health plan/HMO; private insurance patient revenue>75%; and managed care patient revenue<25%.

Conclusions: PCP referral to PT declined, and varied by insurance coverage and physician and practice characteristics. A better understanding of the role of physical therapists in primary care is needed as care delivery moves to value-based payment.


The Role of Practice Facilitators in Meeting the HIT Needs of Rural Practices

Investigators: Ann Lefebvre, MSW, CPHQ, Mary McCaskill; Kristin Reiter, PhD, Jason Mose, MS, MBA, CHFP, PhD, Erin P. Fraher, PhD, MPP, Warren P. Newton, MD, MPH

Objective: To understand how practice facilitators enable rural and smaller practices to meet the evolving and increasing demands of health information technology (HIT).

Data/Setting/Design/Methods: This case study of North Carolina used bivariate and multivariate analyses to evaluate the skills, knowledge and effectiveness of practice facilitators in helping small, rural practices meet the requirements of Meaningful Use.

Results: Results suggest that deploying practice facilitators into smaller, rural practices in North Carolina may have helped to bridge skill gaps and support practices in achieving meaningful use performance measures. Findings suggest that effective practice facilitators have training and experience within a traditional healthcare profession, however, additional training is necessary to allow them to assist small, rural practices in meeting the evolving HIT demands of the health care system.

Conclusions: Small, rural ambulatory care practices often lack resources to hire new staff or develop new skills in their existing staff to meet the demands of increasing use of health information technology. A model that builds upon the practice facilitator’s experience in a healthcare profession, and allows them to access continuous learning opportunities enables practices to retrain members of their existing workforce who can assist in training other providers and staff in small, rural practices.


Determining the Value and Outcomes of the Doctor of Nursing Practice (DNP)

Investigators: Anna Song Beeber, PhD, RN; Cheryl Jones, PhD, RN, FAAN; Carrie Palmer DNP,
RN, ANP-BC; Julee Waldrop DNP, PNP-BC; Mary Lynn PhD, RN

Objective: The role of the DNP outside of academic settings has not been explored and the value that DNP-prepared nurses bring to their practice settings is largely unknown. This study aimed to survey existing DNP programs to identify the non-academic settings into which their DNP graduates are employed and conduct semi-structured interviews with employers to identify the role of the DNP prepared nurse in non-academic settings.

Data/Setting: Data were collected from December 2016-August 2016 in two parts: 1) an online survey of the DNP programs; and 2) qualitative semi-structured telephone interviews with employers.

Design/Methods: The online survey examined the types of non-academic institutions that hire DNP graduates. Employers were identified either by DNP program directors through the online survey (34 employers) or by a convenience sampling method (12 employers). Descriptive analyses and content analyses were performed to identify the roles and practice settings in which DNP-prepared nurses were employed.

Results: A total of 130 DNP program directors responded to the online survey (45% response rate). 23 employers participated in telephone interviews. Employers described the role of the DNP-prepared nurse as Advanced Practice Nurses (APN) providing patient care or as leaders, administrators, and managers. Employers identified that the DNPs practicing as APNs had less flexibility (in terms of time in their day and availability to fill in for other staff), while DNPs in leadership roles were flexible to fill in for administrators and managers.

Conclusions: The role of the DNP-prepared nurse in non-academic settings is not clearly defined. These nurses appear to typically function as APNs in clinical care or as a health care system leaders. However, the number of DNPs in these settings is still limited making it challenging to clearly articulate the role of DNPs in various settings.


Physician Assistant and Nurse Practitioner Roles in Patient-Centered Medical Homes

Investigators: Christine M. Everett, PhD, MPH, PA‐C; Brandi Leach, MS; Perri Morgan, PhD, PA‐C

Background: Team-based care involving physician assistants (PAs) and nurse practitioners (NPs) is one recommended strategy for improving access, quality, and cost of care in the patient-centered medical home (PCMH). The fact that PA/NPs can, and do, perform a variety of roles on primary care teams, suggesting that there is plasticity within the professions and between PA/NPs and physicians. While some studies have described PA/NP roles within a given PCMH, it is currently unclear the extent to which each potential role is being implemented in PCMHs nation-wide.

Key Findings/Policy Implications:

  1. The finding that PAs and NPs perform different patterns of clinical tasks in primary versus supplemental provider roles may be useful for workforce modeling of the task substitution potential of PAs and NPs.
  2. PCMH PA and NP roles seem to be well suited to current training, but since PAs and NPs report not maximally using their training up to 30% of the time, team modifications and training that promote task delegation by PAs and NPs to staff with less training might increase efficiency of care.
  3. Few PAs and NPs report performing tasks for which they are underqualified, but since these situations are potentially dangerous, provisions for back‐up should be available to these PAs and NPs.
  4. Few PAs and NPs are spending significant time performing some newer PCMH functions such as population health management and quality improvement. Increased training in these areas may improve the likelihood of performing some of these functions.
  5. Additional investigation is warranted to find potential causes and solutions to the dissatisfaction reported by the PAs and NPs who did experience role changes associated with PCMH implementation.


The Role of Physical and Occupational Therapists in the Acute to Post-Acute Care Transition

Investigators: Janet K. Freburger, PT, PhD, Dongmei Li, MS, Anna Johnson, PhD, Erin Fraher, PhD

Objectives: Little is known about the use of therapists in the acute to post-acute transition for patients discharged home following stroke. The objectives of this study were: 1) to describe the use of physical and occupational therapists in the acute and post-acute settings (i.e., home, outpatient department) and 2) to identify predictors of therapist use in these settings.

Design: Retrospective cohort design using Medicare claims data linked to American Hospital Association data and Area Health Resource File data.

Setting: Short-term acute care hospitals (N=3,128).

Sample: Patients 66 years or older admitted to the hospital for stroke who survived the inpatient stay, were discharged home, and survived the first 30 days at home (N=42,955).

Methods: Descriptive analyses of the use of physical and occupational therapists in the acute and postacute settings. Multivariate logistic regression analyses to identify sociodemographic, clinical, hospital-level, and county-level variables associated with physical and occupational therapist during the inpatient hospitalization and during the 30 days following discharge home. Analyses were conducted for any therapist use and for specific use of a physical or occupational therapist

Results: 70 percent of the sample saw a therapist in the inpatient setting, but only 33 percent saw a therapist in their home or outpatient setting. Several hospital and county-level factors were associated with therapist use. Of particular note, patients treated at hospitals with higher RN FTEs/admissions were more likely to receive care from therapists in the acute and post-acute setting and therapist supply was associated with the timeliness and intensity of therapist use. The association between physician supply and therapist use varied depending on physician type. We also found evidence of socioeconomic and geographic disparities in therapist use.

Conclusions: There is an under-utilization of therapists in the acute and post-acute settings for patients discharged home following stroke.


Comparison of Specialty Distribution of Nurse Practitioners and Physician Assistants in North Carolina, 1997-2013

Investigators: Perri Morgan, PhD, PA-C, Anna Johnson, PhD, MSPH, Erin Fraher, PhD, MPP

Background: As healthcare demand in the United States is expected to grow, increased use of nurse practitioners (NPs) and physician assistants (PAs) is seen as a partial solution to potential physician shortages. NPs and PAs can change specialties throughout their careers and may constitute a flexible reservoir of health professionals to meet emerging healthcare needs. This project studied changes in NP and PA specialty distribution in North Carolina between 1997 and 2013.

Key Findings: Results show that both professions more than tripled in size between 1997 and 2013, growing much faster than either the state’s population or the number of practicing physicians. Due to this rapid growth, absolute numbers of NPs and PAs increased in most medical specialties. Both NPs and PAs reported practice in a wide range of medical specialties, but there were patterns specific to each profession. Over the time period studied, the proportion of PAs –but not NPs– reporting practice in primary care dropped significantly. PAs were more likely than NPs to report practice in urgent care, emergency medicine, and surgical subspecialties. The proportion of NPs practicing in obstetrics/gynecology and pediatrics also fell, while the proportion practicing in adult medical subspecialties grew. Overall, the findings indicate that the NPs and PAs practice in a wide variety of specialties and their specialty distribution in the workforce can change relatively quickly.

Policy Relevance:

  1. Nurse practitioners and Physician Assistants are a flexible workforce that the nation can draw upon to meet emerging healthcare needs.
  2. There are important similarities and differences between NPs and PAs that should be considered in workforce policy design.
  3. Future research should advance understanding of mutable individual, educational, professional, and system level factors affecting specialty choice among NPs and PAs.
  4. Policies to attract NPs and PAs to specialties where they are most needed should be developed, tested, and implemented.

Methods: The study analyzed self-reported specialty for active, licensed NPs and PAs practicing in NC in 1997 and 2013. We used 1997 and 2013 licensure data for NPs and PAs from the NC Health Professions Data System, which were derived from the NC Board of Nursing and NC Medical Board.


Assessing Shifts in Outpatient Visits to Physicians of Other Specialties in Rural Areas with Shortages of Cardiologists and Gastroenterologists: A Preliminary Analysis

Investigators: Donald E. Pathman, MD MPH; G. Mark Holmes, PhD; Samuel Berchuck, BS; James W. Terry, Jr., BA

Executive Summary/Key Findings

Although physician workforce planning approaches the need for physicians of each specialty individually, in fact many services are provided by physicians of several specialties. Further, there is some evidence that physicians adjust the scope and balance of services they provide when there are too few physicians of other specialties in their communities, although when this “service shifting” happens and for which services and specialties is not known. This study uses Medicare data to assess changes in the number of outpatient visits made to various specialty groups for atrial fibrillation and for esophageal, gastric and duodenal disorders in rural areas that vary in their local availability of cardiologists and gastroenterologists, respectively. Analyses find evidence in one situation for visit shifting across specialties; specifically, where there are fewer local gastroenterologists, rural elderly make more visits for the selected gastrointestinal disorders to other, non‐primary care physicians. This partially offsets the loss of visits made to gastroenterologists. Visit numbers to primary care physicians did not change for either atrial fibrillation with fewer cardiologists or for these gastrointestinal disorders with fewer gastroenterologists. Similarly, visits for atrial fibrillation made to other, non‐primary physicians did not change with fewer cardiologists. The finding of shifting of visit numbers across some specialties for some medical conditions has implications for workforce planning and modeling. With shifting of outpatient visit numbers between specialties, physicians of various specialties can flexibly absorb local demand for services and reduce the service shortages otherwise anticipated when there are too few physicians of some specialties. Medical educators need to prepare physicians so they are able to provide services beyond the traditional focus of their specialty to help fill local service needs due to shortages in other specialties.


Workforce Transformations Needed to Staff Value-Based Models of Care

Investigators: Erin Fraher, PhD, MPP; Rachel Machta, BS; Jacqueline Halladay, MD, MPH

Abstract: Secretary Burwell recently announced that by 2018, 50% of Medicare payments will be tied to value through alternative payment and care delivery models. What will this shift to value-based payment models mean for the workforce? This paper synthesizes the existing body of evidence on the workforce implications of new models of care. We identify: 1. task shifting that is occurring in the delivery of traditional health care services; 2. new staff roles that are emerging to provide enhanced care services; 3. how employers are “putting it all together”; and 4. the implications of these trends for health workforce research and policy.

Key Findings: Health care professionals are taking on new roles with medical assistants having one of the most rapidly evolving roles in new models of care. New roles are emerging that focus on: 1) coordinating and managing patients’ care within the health care system; and 2) “boundary spanning” functions that address the patient’s health care needs across health and community‐based settings. Employers are struggling to rewrite or create new job descriptions, reconfigure workflows, and develop training to support task shifting and new roles. Health workforce researchers and policy makers need to shift focus from “old school” to “new school” approaches. Specifically, they need to shift focus from: 1) workforce shortages to developing a better understanding of how the existing workforce could be redeployed and reconfigured to address the demand‐capacity mismatch; 2) provider type to provider role because different types of health care providers can take on the same roles; and 3) training new professionals to retooling the existing workforce since they will be the ones who will transform care.


The Evolving Role of Medical Assistants in Primary Care Practice: Divergent and Concordant Perspectives from MAs and Family Physicians

Investigators: Erin P Fraher, PhD, MPP, Allison Cummings, MD, Dana Neutze, MD PhD

Background: Medical assistants (MAs) are a flexible, low-cost and redeployable resource in primary care practices and their roles are swiftly transforming. A robust literature has identified a gap between the full potential of MA role expansion and its implementation in primary care practice. This gap has been attributed to role confusion, shortcomings in MA training, physician resistance to delegate tasks, and MA reluctance to take on new roles.

Methods: We surveyed medical assistants and family physicians in primary care practices in North Carolina about MA roles related to visit planning; direct patient care; documentation; patient education, coaching or counseling; quality improvement; population health and communication. MA surveys were delivered by practice facilitators employed by the NC Area Health Education Centers (AHEC) program. Family Physicians were reached through the NC Academy of Family Physicians newsletter. MAs and family physicians were surveyed to gauge whether they agreed on the roles MAs were currently performing and to identify their level of confidence in MAs’ ability to perform these tasks. For those tasks not being currently being performed, we assessed physician willingness to transition tasks to MAs with additional training and MA willingness to pursue that training.

Results: 118 medical assistants and 175 family physicians responded to the surveys. The majority (59%) of MAs in our sample reported experiencing a role change since beginning their current position. Findings suggest that many of the activities related to visit planning, direct patient care, and documentation that were once considered extended roles are now routinely performed by MAs on most days. MAs and physicians generally agreed on the tasks currently being performed by MAs, but they had diverging perceptions of MA roles on tasks related to patient education, coaching and counseling. While 44% of MAs reported performing motivational interviewing (MI), only 13% of physicians said MAs in their practices currently perform MI. 41% of MAs report assisting patients with chronic diseases to set goals compared to 18% of physicians who believe their MAs do this; 44% of MAs said they educated patients with chronic disease about preventive care while only 25% of physicians reported MAs took on this role.

Physicians and MAs agreed on the tasks that MAs perform least frequently with only 9% of physicians and 27% of MAs reported that MAs performed scribing. Tasks in the survey that physicians indicated that they would be most willing to delegate to MAs that MAs are currently not performing in their practice were population health activities such as identifying patients in need of preventive screening (mammography, colorectal screening, etc); finding patients with diabetes who are overdue for A1c tests and pending the order; and extracting data from the EHR to manage patient lists

Conclusion: Closing the gap between MA potential and actual roles in population health and panel management; patient education, coaching and counseling; and scribing will become even more important as our health care system moves toward value-based and risk-based payment models that emphasize addressing the upstream, preventive and chronic care needs of patients. While these results are promising, deploying MAs in the new roles will require increasing MA to physician staffing ratios, protected time for MAs to perform these functions, and redesigning workflows to accommodate these changes.


Primary care practice staffing characteristics

Investigators: Valerie Lewis PhD and Ryan Kandrack PhD(cand)

Contemporary health care policy and quality initiatives often explicitly or implicitly rely on primary care practices as a central locus for change. Despite this, no work has systematically examined the composition of the workforce of US primary care practices; how the workforce varies across key types of practices; or how team composition has changed over time.

This project will provide more comprehensive data on staffing configurations for virtually all primary care practices in the US and will examine how these staffing patterns vary in rural versus urban areas, by practice type and ownership structure.


Rural-urban differences within primary care scope of practice

Investigators: Ryan Kandrack, PhD, Erin Fraher, PhD, Mark Holmes, PhD

There is some evidence that primary care physicians in rural areas respond to the limited local supply of specialists by offering a broader array of clinical services than their urban colleagues. These studies suggest that there is “plasticity” in the scope of services provided by physicians. Plasticity describes the dynamic and real world nature of scope of practice; physicians within the same specialty may provide different types of services depending on their demographic characteristics and training, the density of other physicians who provide overlapping services in their local area, the needs of their local patient population, and their personal preferences. Previous studies have described primary care physicians’ plasticity between rural and urban locations but little work has been done investigating the factors that affect plasticity within primary care specialties. The goal of this project is to investigate how the breadth and volume of primary care services varies within family medicine and general internal medicine between rural and urban areas.

Diffusion of Physicians and Access to Primary Care: The Role of Person, Program, and Place

Investigators: Thomas C. Ricketts, PhD, MPH

Background: Federal and state policies try to influence physician location choices using a combination of programmatic or extrinsic factors (e.g. bonus payments, subsidized loans) and personal or intrinsic factors (e.g. promoting volunteerism, orienting trainees to underserved populations and selecting motivated students) to encourage physicians to practice in communities where the population has difficulty accessing medical care. Unfortunately, there has not been a combined evaluation of these policies, making it difficult to assess the actual effects. The central hypothesis of this project is that physician diffusion can be estimated based on the characteristics of the places physicians go to and come from. The focus of this report is on locations that are eligible for placement incentives from the federal government based on their designation as Health Profession Shortage Areas (HPSAs), but which may or may not have physicians recruited to them, as well as those places that could become HPSA

Conclusions and Policy Relevance:

1) It is not currently possible to generate models that predict who will move into or out of a rural HPSA or which rural HPSAs are more or less likely to attract physicians.

2) Federal programs and incentives that seek to promote practice in underserved communities are only one of a number of factors that influence a physician’s choice of practice location.

3) We have seen an overall steady and relatively well-distributed pattern of growth in physician supply. At the same time we have seen an overall increase in the number of areas designated as HPSAs despite growth in provider supply.

4) Analyses need to be conducted over a longer time period at a more fine-grained level to better understand the role of physicians serving in underserved places.

5) The cumulative effect of federal policies in reducing or eliminating geographic shortages is not known. A comprehensive evaluation is needed to judge the effects of current policies and factors that influence choice of location or the potential for emerging underservice.

Methods: This analysis makes use of multiple years of the American Medical Association’s (AMA) Physician Masterfile® that have been linked by the unique identifier for each physician in the files. The individual-level data sets included all physicians in the Masterfile at the end of calendar years 2006 and 2013.