Solving the hospital crisis with mathematical, statistical and software tools

Jun 6 2005

Melbourne researchers are hoping to ease the strain of bed shortages and limited resources in Australian hospitals in a project to develop mathematical, statistical and software tools that will enable hospital managers to see where queues or available resources exist within the hospital system.

The University of Melbourne's Professor of Operations Research Peter Taylor says, "Australian hospitals have difficulty managing increasing demand and need to adopt new approaches to service provision and patient management if they hope to maintain an acceptable quality of service."

"The problem is, understanding the dynamic relationship between the range of factors such as available beds, waiting lists, equipment, and so on has become complex to the point where it exceeds human ability to understand how an individual decision will affect the system as a whole."

"The result is that queues can form, ambulances can be forced to bypass hospitals and waiting times for surgery can grow."

"Put simply, hospitals have to find more efficient ways of managing their load."

Professor Taylor and Dr Graham Byrnes at the University of Melbourne, in collaboration with the Royal Melbourne Hospital and the Monash Institute of Health Services Research are now developing mathematical models that they hope will make hospitals easier to manage. They also intend to draw on experience and problems from other hospitals to ensure their solutions have widespread applicability.

Mathematical models are tools that have long been used in a range of other complex systems, such as manufacturing, the military and telecommunications, in order to enhance and optimize control. They have not yet been tested in the hospital system.

Research Fellow, Leon Au, who recently commenced work on the project says, "It is not a matter of simply transferring existing techniques, from telecommunications for example, over to hospitals because patients with life-threatening conditions need to be given priority over other patients and this has a flow on effect through the whole system."

"The route different patients take through the hospital system from diagnosis to rehabilitation can vary substantially, so consistent patterns can be difficult to find."

Funded by the Australian Research Council and Melbourne Health, the project draws researchers from mathematics, statistics and health. They will construct models for how patients arrive and are served by different parts of the hospital, from the emergency department triage through to discharge planning.

The individual models will then be linked to form a whole picture of the hospital, which will make it possible to determine how a delay in one department affects the rest of the system. For example, a blockage of the emergency department may have its root cause in the difficulty of finding a nursing home to discharge a long-staying ward patient.

The model will also follow individual patients as they move through the system from one department to another.

The researchers hope to eventually have unique software tools that are of direct practical use to health service planners, hospital strategic planners and operational managers by enabling them to monitor performance across the system and make informed decisions about where resources are most needed.