Modeling of ICU nursing workload to inform better staffing decisions
Author(s)Ma, Yiyin, M.B.A. Massachusetts Institute of Technology
Modeling of Intensive Care Unit nursing workload to inform better staffing decisions
Leaders for Global Operations Program.
Patrick Jaillet and Retsef Levi.
MetadataShow full item record
Beth Israel Deaconess Medical Center (BIDMC) has partnered with the Gordon and Betty Moore Foundation's to eliminate preventable harm in the Intensive Care Unit (ICU). Many medical publications suggest nursing workload as a major contributor to patient safety. However, BIDMC was not using any tool to measure nursing workload, and as a result, nurse staffing decisions were made solely based on the ad hoc judgment of senior nurses. The objective of this thesis is to create a prospective nursing workload measurement and ultimately use it to improve staffing decisions in ICUs. To create a nursing workload measurement, a wildly-adopted patient-based scoring system, the Therapeutic Intervention Score System (TISS), was modified to BIDMC's ICUs. With consultation from clinicians and nurses, changes were made to the TISS to reflect BIDMC's workflow, and a new nursing workload scoring system called the Nursing Intensity Score (NIS) was created. The NIS for each patient per shift was calculated over a two-year period to gain further insights to improve staffing decisions. After looking at the current state, there was no correlation between nursing staffing and overall patient workload in the unit. In addition, nurses with 1 patient (1:1 nurses) had significantly less workload than nurses with two patients (1:2 nurses) even though they were expected to be the same. Finally, there was one overworked nurse (150% of median nursing workload) in every three shifts in the ICU. A prospective approach to analyze patient workload was developed by dividing patients based on clinical conditions and categorizing the results on two axis: the nominal workload level and the variability around the nominal value of workload. This analysis suggests that, a majority of patients are predictable, including a few patients with high but predictable load. On the other hand, some patients are highly unpredictable. A nursing backup system was proposed to balance workload between 1:1 and 1:2 nurses. To test the proposal, a simulation was developed to model the ICU with the goal of minimizing the number of overworked nurses. The best backup system was a buddy pairing system based on predictive model of patient conditions, with the resource nurse as the ultimate backup.
Thesis: M.B.A., Massachusetts Institute of Technology, Sloan School of Management, 2015. In conjunction with the Leaders for Global Operations Program at MIT.Thesis: S.M., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2015. In conjunction with the Leaders for Global Operations Program at MIT.Cataloged from PDF version of thesis.Includes bibliographical references (pages 92-94).
DepartmentSloan School of Management.; Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.; Leaders for Global Operations Program.; Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science; Sloan School of Management
Massachusetts Institute of Technology
Sloan School of Management., Electrical Engineering and Computer Science., Leaders for Global Operations Program.