Exposure variables
Primary exposure variable: Day-case or overnight stay. Day-case was defined as any procedure where the admission day and discharge day were the same.
Secondary exposure variable: In secondary analysis the exposure variable was defined at a trust level; trusts operating on ≥ 80% of patients as day-case and trusts operating on < 50% of patients as day-case were compared. Eighty percent was chosen to reflect trusts where day-case surgery appeared to be the default and 50% was chosen to reflect trusts where surgery with overnight stay was dominant. This analysis was conducted to minimise bias, when considering the data at a patient level due to patients with more severe illness (and so potentially poorer outcomes) being more likely to stay in overnight. For this analysis, trusts performing fewer than 50 procedures over the study period (average of ten per financial year) were excluded.