Leveraging Electronic Health Record Data for Inpatient Nurse Workforce Management: A Mixed Methods Study

Background: To ensure adequate staffing levels, the Staffing Methodology for VHA Nursing Personnel (VHA Directive 1351) requires facilities to monitor daily inpatient nurse staffing levels and set staffing targets for each unit at least every two years. Despite efforts by the Office of Nursing Services (ONS) to consolidate data for decision-making, pilot data confirms that facility nurse leaders do not have centralized access to reliable, timely assessments of unit-level data needed to manage staffing. This gap exists for three reasons: facilities use non- interoperable staffing software systems, nationwide accounting tools are time-lagged, and electronic health records (EHR) have not been designed with analysis at the nursing-unit level in mind. Workload, an important aspect of staffing needs, is affected by patient census as well as factors such as patient turnover and acuity but is not accounted for in current staffing models. Nurse turnover is another critical dimension of staffing avail- ability. Human resources (HR) data undercount staff leaving inpatient service due to within-facility transfers to other units or roles, which we call latent turnover. To address these gaps, we developed a novel method using routinely collected data from the EHR (bar code medication administration, text integrated utility notes, and vital signs) to estimate daily nurse staffing and latent turnover for inpatient units. Significance: Nurse shortages have impacted inpatient staffing at VHA, the country’s largest employer of nurses. VHA estimates that an additional 77,500 nurses are needed in the next five years to achieve adequate staffing. A key element of workforce planning is knowing the appropriate staffing levels to safely manage inpatients. Low nurse staffing is associated with higher rates of mortality, adverse events, and nurse turnover. Innovation & Impact: Our EHR-based measures could allow VHA to identify changes in nurse staffing, work- load, and turnover across time, nursing units, and facilities. These EHR data are consistently collected across sites and can provide a VHA-wide view of unit-level inpatient direct-care nurse staffing and workload over time, an innovative use of EHR data. Our methods are designed with interoperability in mind. We have confirmed that syndicated Cerner data elements in CDWWork3 contain what is needed for our methods. Specific Aims: Aim 1: Use EHR-derived measures to establish VHA-specific benchmarks for inpatient direct- care registered nurse staffing levels at similar levels of workload nationally across units and time. Aim 2: Compare EHR and HR-derived measures of turnover and predict direct-care nursing turnover using EHR- derived staffing, workload, and turnover measures. Aim 3: In a formative evaluation, develop a roadmap for integrating EHR-derived nurse workforce measures into monitoring tools using the non-adoption, abandonment, scale-up, spread, and sustainability (NASSS) framework. Methodology: In Aim 1, we will use principal component analysis to provide unit-level, model-based empirical benchmarks, establishing current staffing levels for units with comparable workloads. In Aim 2, we will calculate monthly conventional and latent turnover rates for each unit and use mixed-effects models to identify unit-level predictors of each turnover rate. Aim 3 will use the NASSS framework to guide qualitative interviews with key stakeholders and to outline requirements and challenges in developing feasible monitoring tools that facilitate actionable insights. Quarterly input from a Subject-Matter Expert Panel, [including representatives from ONS] will be incorporated to inform research processes, findings, and outputs for all aims. Next Steps/Implementation: ONS commissioned our team to prepare static reports of unit-level staffing. We will build on this partnership to: 1) Study the relationship between staffing, workload, and turnover; 2) Establish system-wide benchmarking, and 3) Leverage our quali Project Number: 1I01HX003868-01A2 | Fiscal Year: 2025 | NIH Institute/Center: Veterans Affairs (VA) | Principal Investigator: LAURA PETERSEN | Institution: MICHAEL E DEBAKEY VA MEDICAL CENTER, HOUSTON, TX | Activity Code: I01 | Study Section: HSR-3 Healthcare Informatics & Access to Care[HSR3] View on NIH RePORTER: https://reporter.nih.gov/project-details/11111188