Lead/Presenter: Melissa Knox, COIN - Houston
All Authors: Knox MK (Center for Innovations in Quality, Effectiveness, and Safety, Baylor College of Medicine, Houston), Eck CS (Center for Innovations in Quality, Effectiveness, and Safety, Baylor College of Medicine, Houston) Dorsey LE (Michael E DeBakey VAMC, Houston) Mehta PD (University of Houston) Wong JJ (Center for Innovations in Quality, Effectiveness, and Safety, Baylor College of Medicine, Houston) Yang C (Center for Innovations in Quality, Effectiveness, and Safety, Baylor College of Medicine, Houston) Petersen LA (Center for Innovations in Quality, Effectiveness, and Safety, Baylor College of Medicine, Houston)
Objectives:
The COVID-19 pandemic resulted in an initial shock and sustained changes to the health care system. Pre-pandemic evidence associated inadequate nurse staffing with mortality, and pandemic-related labor shortages have heightened concerns about aligning staffing and workload. Objective comparisons of inpatient nurse staffing and workload prior to and during the pandemic have been limited. We tested the feasibility of using bar code medication administration (BCMA) data, now available at 98% of US hospitals, for assessing changes in nurse staffing and workload. Medication pass (medpass) analysis uses BCMA data to generate daily point-in-time staff-level metrics, allowing for comparisons across units and over time.
Methods:
We conducted medpass analysis for the highest frequency scheduled medication time (peak-time) on each Veterans Health Administration (VHA) acute-care nursing unit, assessing trends and comparing 2019 and 2020 data. Daily point-in-time staff-level metrics generated using BCMA data included: count of staff days, patients per staff (PPS), and medpass duration. Patient conditions vary by unit and patient acuity drives how many patients a nurse is assigned, so we examined differences by unit type and PPS. Population Studied: All staff (n = 36,534) administering medications scheduled for the peak time on 582 VHA acute-care nursing units during 2019 and 2020.
Results:
The number of staff days at peak-time medpass was similar between 2019 (923,787) and 2020 (929,057). Share of RNs and LPNs among all staff administering peak-time medications increased 1.4 percentage points (87.8% to 89.2%) and 0.3 percentage points (2.7% to 3.0%), respectively. Share of non-nursing staff and other nursing roles decreased. For RNs the number of peak-time medications per patient increased from 2019 (6.13) to 2020 (6.26). For all unit types except surgical units, RN PPS decreased from 2019 to 2020, even after accounting for the initial shock to patient census in March 2020. For RNs with the same PPS, medpass duration was longer in 2020 than in 2019, with variation across unit types and PPS. Year-over-year medpass duration increase was greatest on medical and medical-surgical units.
Implications:
RNs in VHA acute-care units, except for surgical units, administered medications to fewer patients per medpass in 2020 than in 2019. Patient assignments based on patient acuity result in lower PPS when patient needs are higher, and higher acuity among hospitalized patients in 2020 has been reported. Even as PPS decreased, our data suggest that RNs may have taken on greater workload in 2020 for the same number of patients compared to 2019. Fewer non-nursing staff were administering peak-time medications in 2020, the number of medications per patient administered by RNs increased, and it took more time for RNs to complete the medpass in 2020 at the same level of PPS.
Impacts:
Nursing workload is associated with patient and staff outcomes. However, few tools are available for assessing nursing workload at the individual, unit, and shift level and in a timely manner. Medpass analysis is a feasible tool for assessing near real-time and granular changes in nurse workload. Such analyses can support investigation of pandemic-related disrupted care.