Assessing the Concordance of Coded Morbidity and Mortality Data for In-Hospital Trauma- Related Deaths

Abstract

To examine the effectiveness of trauma systems, in-hospital mortality rates are key indicators. However, to date there has been little research examining the congruence of morbidity and mortality data as recorded in hospital databases and mortality databases. A retrospective review of coded medical records from a trauma registry, a metropolitan teaching hospital and the National Death Index for 1747 patients admitted to hospital for ≥ 24 hours for acute treatment of injury was undertaken. This study was established to determine the number and cause of differences in coded data between hospital records and National Death Index (NDI) records for in-hospital deaths following admission for trauma. Of the 60 patients who died in hospital, 26% died from non-traumatic causes according to NDI coded data. Over half of the cases with trauma coded in the NDI contained external cause data which were less specific or did not match external cause data from hospital records. When reviewing trauma-related outcomes, where possible injury researchers and policy advisors should consider both mortality and morbidity datasets when reviewing in-hospital deaths, as this research identified that the morbidity dataset provides greater detail for injury and external causes of injury than mortality coded data for in-hospital deaths.