Prudent Elective Surgery Scheduling: A Whole Systems Approach

The University Hospital of Wales performs approximately 25,000 surgical procedures per year. Due to high demand, many of these procedures are cancelled, often due to poor scheduling and the lack of beds for patients post surgery. In addition, the incorrect balance of surgical specialties within the timetable is often responsible for the lengthening of patient waiting lists.

Currently, the process of timetabling these operations is arduous and time-consuming, with eighteen different specialties having their own demand to satisfy in line with resource constraints and competing interests between departments. In the University Hospital of Wales, approximately 18% of operations are cancelled annually, with non-clinical hospital reasons, such as the lack of available beds post-surgery, accounting for 54% of these. In addition, after leaving the operating theatre, 31% of patients are also ultimately assigned to an inappropriate ward (that is, a ward that does not necessarily have the specialist nurses or equipment that is desirable for the patient) due to shortfalls in bed capacity in the intended ward. Such occurrences are known to be distressing for patients and frustrating for staff. Evidence suggests that they can also result in adverse patient outcomes and longer lengths of stay.

The process of theatre scheduling currently involves the periodic production of a so-called 'master surgery schedule'. This specifies a weekly timetable for each operating theatre, using morning and afternoon sessions, Monday to Friday. Currently, the number of sessions for each specialty is decided manually based on their perceived needs such as the current length of their waiting lists. Crucially, this master surgery schedule does not consider post-surgery resource requirements; consequently, bed requirements for certain specialties may occur in undesirable peaks at various points in the week.

Currently, such peaks are not picked-up by hospital staff until the day of or the day prior to surgery, with operations for some patients having to be cancelled if a suitable post-operative bed is not available.

Research conducted by a PhD student at our university under the supervision of the three applicants for this proposal has demonstrated through simulation that post-operative bed utilisation can be increased, and hence the number of cancellations can be decreased, through better planning and better resource management. This work is currently in its final stages and is being written up, with thesis submission due in the next six months.

The results have strongly indicated that the basis of this research is sound. Consequently, at this point we are now in a position to extend this research by also looking into the upstream effects, using innovative queueing and timetabling methods that take into account these features. In addition, we also wish to examine the issue of solution 'robustness' more closely, which involves assessing the ability of a particular surgery schedule to be able to cope with unexpected changes in patient demand. The outcome of this research will be a generalisable tool applicable to any healthcare setting which admits patients for elective surgery.