Skip to main content
. 2021 Sep 7;31(7):515–525. doi: 10.1136/bmjqs-2021-013522

Figure 3.

Figure 3

Cumulative incidences of all in-hospital and postdischarge events. Part A shows data for endoscopic resection of prostate NEC; Part B shows data for repair of recurrent inguinal hernia using insert of prosthetic material; Part C shows data for primary total prosthetic replacement of hip joint NEC. Left-hand side: cumulative incidence of all in-hospital events after operation, split by provider type (NHS hospital=blue; ISHP=red) and event type (between hospital transfers=dotted; in-hospital death=solid; discharge=dashed). Events are considered as competing risks, formulated as a multistate model. Right-hand side: similar plots for postdischarge events (death=long dashed; within-specialty emergency readmission=dot dashed). The three operation types shown here were chosen so as to best represent together all 18 sets of results (the full set is shown in online supplemental figure 3). IH, inguinal hernia; ISHP, Independent Sector Healthcare Provider; NEC, not elsewhere classified; NHS, National Health Service; THR, total hip replacement.