A new approach to over-bedding in American healthcare: Statistical analysis and computer simulation of patient admission in a West Virginia state psychiatric hospital

Abstract

Objective: To identify the factors that have a substantial impact on a West Virginia state psychiatric hospital’s bed occupancy by investigating historical admission data, and developing a computer simulation system to give insight into modifiable variables that reduce admission numbers, therefore to provide solutions to the over-bedding problem.
Methods: Quantitative review of hospital admission data from January 2007 to November 2013 allowed for the construction of a simulation model to estimate the inpatient flow. The system’s performance was evaluated after alteration of selected parameters and variables.
Results: The study revealed significant regional differences in admission numbers. The civil commitments and psychiatric hospitalizations do not directly correlate with county coverage populations. Some counties sent disproportionately more patients. Patients’ length of stay also varied among geographical areas. Re-admission was not uncommon. Using the percentage of diversion as the outcome measurement, the computer simulation model reconstructed the admission scenario multiple times, predicting that the diversion rate can be significantly reduced if certain variables (hospital capacity, patient arrivals from top referring counties, and patient length of stay) are changed.
Conclusions: Involuntary admissions were unevenly distributed according to geography and population in the studied American state psychiatric hospital. Using historical data, computer simulations can model hospital admission systems to evaluate performance and predict needs for change.