REVISION HIP COMPLEXITY CLASSIFICATION (RHCC): VALIDATION OF AN APP-BASED SCORING SYSTEM AND ANALYSIS OF INTER-OBSERVER INCONSISTENCY

The Revision Hip Complexity Classification (RHCC) provides a framework to allow discussion of complex cases as part of local or regional revision hip multidisciplinary team meetings. The British Hip Society (BHS) produced a RHCC mobile app to help simplify the classification process. We aimed to test the inter-observer and intra-observer reliability of the RHCC both with and without the use of the app. Two consultant orthopaedic surgeons, three orthopaedic registrars and two orthopaedic practitioners graded 50 consecutive revision hip cases without the app, on two occasions two weeks apart. Two weeks later a further 50 consecutive revision hip cases were graded using the app, then repeated two weeks later. Weighted Cohen's kappa was used to test for intra-observer reliability and Kendall's W was used to test for inter-observer reliability. We also identified reasons for ‘major disagreements’ by analysing cases whereby at least 3/7 reviewers disagreed. Without the app, intra-observer reliability of the RHCC classification was ‘substantial’, with an average kappa value 0.69 (range 0.37 – 0.85). When using the app, the intra-observer reliability improved for 6/7 reviewers, with an average kappa value of 0.78 (range 0.46 – 0.95). Without the app, inter-observer reliability of the RHCC classification was ‘moderate’ with Kendall's W score of 0.55. When using the app, the inter-observer reliability was similar with a Kendall's W score of 0.56. Without the app, 9/15 (60%) of major disagreements were due to discrepancies in assessing bone loss with Paprosky grading; with the app this figure rose to 15/20 (75%). The BHS RHCC app is a validated tool for grading the complexity of hip revisions. Use of the app improves intra-observer reliability of the RHCC classification across a range of health professionals, but not inter-observer reliability. The majority of disagreements were due to discrepancies in assessing bone loss with Paprosky grading.