Optimising cardiac services using routinely collected data and discrete event simulation

Abstract

Background: The current practice of managing hospital resources, including beds, is very much driven by measuring past or expected utilisation of resources. This practice, however, doesn’t reflect variability among patients. Consequently, managers and clinicians cannot make fully informed decisions based upon these measures which are considered inadequate in planning and managing complex systems. Aim: to analyse how variation related to patient conditions and adverse events affect resource utilisation and operational performance. Methods: Data pertaining to cardiac patients (cardiothoracic and cardiology, n=2241) were collected from two major hospitals in Oman. Factors influential to resource utilisation were assessed using logistic regressions. Other analysis related to classifying patients based on their resource utilisation was carried out using decision tree to assist in predicting hospital stay. Finally, discrete event simulation modelling was used to evaluate how patient factors and postoperative complications are affecting operational performance. Results: 26.5% of the patients experienced prolonged Length of Stay (LOS) in intensive care units and 30% in the ward. Patients with prolonged postoperative LOS had 60% of the total patient days. Some of the factors that explained the largest amount of variance in resource use following cardiac procedure included body mass index, type of surgery, Cardiopulmonary Bypass (CPB) use, non-elective surgery, number of complications, blood transfusion, chronic heart failure, and previous angioplasty. Allocating resources based on patient expected LOS has resulted in a reduction of surgery cancellations and waiting times while overall throughput has increased. Complications had a significant effect on perioperative operational performance such as surgery cancellations. The effect was profound when complications occurred in the intensive care unit where a limited capacity was observed. Based on the simulation model, eliminating some complications can enlarge patient population. Conclusion: Integrating influential factors into resource planning through simulation modelling is an effective way to estimate and manage hospital capacity.