Capturing surgical site infection after hip fracture surgery : insights from linked routine data.

Aims: To evaluate the incidence, timing, and capture of surgical site infections (SSIs) following hip fracture surgery using routinely collected primary and secondary care datasets, and to assess the limitations of using such resources to identify SSIs. Methods: We conducted a retrospective cohort study using linked Clinical Practice Research Datalink (CPRD) and Hospital Episode Statistics (HES) data from 1999 to 2013. Patients with a hip fracture were identified and compared with age- and sex-matched controls. SSIs were defined using diagnostic codes and categorized as 'definite' or 'possible'. Timing of diagnosis was evaluated, and risk factors for developing infection assessed, using Cox regression. Results: Among 13,920 hip fracture patients, 192 developed definite a SSI (n = 192, 2.4%) or possible SSIs (n = 15, 0.1%). Most infections were recorded in CPRD, with limited overlap between CPRD and HES. Median time to diagnosis was 130 days, with 43% of cases identified more than 90 days postoperatively. Charlson Comorbidity Index ≥ 3 was significantly associated with increased risk. Reoperation for infection occurred in 1.2% of cases (n = 37). Conclusion: Routinely coded data sources likely under-capture SSIs after hip fracture surgery. Poor agreement between primary and secondary care records, combined with delayed identification, limits the utility of routinely collected administrative datasets to accurately identify postoperative infection.