PURPOSE: The efficient management of intensive care unit patient turnover can significantly impact patient survival, patient medical expenses, overall patient satisfaction, and hospital operating expenses. Ultimately movement within a constrained health care delivery system is a dynamic and stochastic process that often eludes traditional analysis and prediction tools. For these reasons, we developed and validated a simulation model of movement through constrained hospitals systems, focusing on the ICU. Validation focused on the ability to faithfully represent patient movement throughout the hospital system.
METHODS: This pilot study consists of a retrospective analysis of a comprehensive sample of 5153 unique patients admitted to the VA Pittsburgh Health System (VAPHS) from April 27, 2010 to February 7, 2011. Data is extracted from the corresponding arrival and patient movement files to produce a cohort data set and time series analysis of patient in all inpatient units. Calculations of length-of-stay (LOS) and transition rates are produced. Blocking is defined as the difference between time of assignment and movement to specified location. Graphical simulation methodology and a commercial software packages, ARENA and ONMET, are used to create a preliminary computer simulation model of daily operations. Conceptual validation was achieved by demonstrating an animated model to clinical staff to establish face validity. Operational validation was achieved by producing Q-Q plots for comparison of two emergent characteristics produced by the model to actual recorded data.
RESULTS: This model graphically depicts (i) LOS rates, (ii) transition probabilities, (iii) location blocking rates, (iv) request realization rate. The greatest blocking time and the greatest assignment difference are recorded in the MM location. When considering the emergent characteristics, blocking and bed occupancy, we observed close linear relationships between the simulated model and recorded patient movement.
CONCLUSIONS: The described validated process modeling approach simulates and accurately describes patient movement within constrained health care delivery system.
CLINICAL IMPLICATIONS: Such a novel patient flow management tool will systematically and objectively aids managerial decision-making at both the unit and the hospital level.
DISCLOSURE: The following authors have nothing to disclose: Spencer Nabors, Theologos Bountourelis, Andrew Schaefer, Louis Luanghesorn, Jeffrey Kharoufeh, Lisa Maillart, Winston Yang, Gilles Clermont
No Product/Research Disclosure Information