Researchers at the University College London Hospital have developed an algorithm for predicting which patients are most likely to miss hospital appointments.
The mathematical modelling tool was developed using data from 22,318 MRI scan appointments across two NHS hospitals. The data included the time of day, the number of previous scans and the distance between residence and hospital.
The findings showed that the tool was able to precisely identify 90 per cent of patients who would miss appointments. The tool could reduce the number of patients to be called by the hospital for preventing one appointment being missed, from 11 to 5.3.
According to a team member, Amy Nelson, the team wanted to develop a modelling tool that accounted for complex human behaviours which induce different reasons for missed appointments. Thus, the accuracy of the model increased as more variables were introduced. Optimal predictive performance needed 81 variables.
It is estimated that the cost of not attending scheduled hospital appointments is £1 billion per year for the NHS. The successful introduction of the tool could result in substantial savings and shorter waiting times.