Perioperative mortality in New Zealand related to hip and knee replacement surgery: comparing administrative and registry data.

INTRODUCTION: Perioperative mortality is of considerable importance, but few national assessments are available. New Zealand has a clinical registry and an administrative dataset that both capture national information about hip and knee arthroplasties. National perioperative mortality rates were compared between the two data sources.

METHOD: Data related to all patients undergoing an elective hip or knee replacement procedure (primary or revision) between 1 January 2007 and 31 December 2011 were separately extracted from the New Zealand Joint Registry and the National Minimum Dataset. The procedure date was used to define the occurrence of an event and dates were compared between datasets plus or minus 3 days. Date of death information was obtained from the National Mortality Collection and used to estimate 30 day mortality rates.

RESULTS: No statistically significant differences in perioperative mortality were evident between comparisons from the two data sources although more deaths were recorded among Registry-only procedures.

CONCLUSIONS: Estimates of 30 day perioperative mortality related to hip and knee arthroplasty procedures in New Zealand 2007-2011 are very similar regardless of data source. These data, coupled with perioperative mortality review using structured reports obtained from clinicians, could be used to develop a surveillance system to promote surgical safety.