Mathematical approach improves predictability of length of hospitalisation due to oral squamous cell carcinoma: a retrospective investigation of 153 patients

Oral squamous cell carcinoma (OSCC), a common cancer of the head and neck, is a major public health problem. The length of stay in hospital (LOS) of patients with OSCC, which can range from a few days to several months, has implications for the patient’s recovery. The aim of the study was to identify and evaluate risk factors that have an impact on the prolongation of inpatient hospital stay. A four-year retrospective study reviewed hospital records of 153 inpatients with OSCC. A statistical model for discrete time-to-event data, with the LOS in hospital measured in days for which the event of interest was discharge from hospital, was applied. The model utilises a tree-building algorithm to identify relevant risk factors for a prolonged LOS. Age, type of flap, and occurrence of complications turned out to be relevant variables. Before, and on day 12, the LOS was mainly dependent on flap type and age, whereas after day 12 it was influenced by the presence of early complications. Predicting the likelihood of discharge can improve the management and resource utilisation of the healthcare system among inpatients.