Introduction: Routinely collected hospital discharge data are a valuable and widely used source of information when assessing the incidence of sports and other injuries in epidemiological studies. However, in order to ensure the usefulness of a hospital discharge register for scientific purposes, the coverage and accuracy of the database should be evaluated. In this study, cruciate ligament rupture was used to assess the coverage and accuracy of the Finnish National Hospital Discharge Register (NHDR).
Materials and methods: The coverage of the diagnosis of cruciate ligament injury in the NHDR was examined by comparing patients with a confirmed cruciate ligament injury in the arthroscopy register of the Finnish Central Military Hospital (N=262) to their register data in the NHDR. To assess the accuracy of this diagnosis in the NHDR, the original medical reports including magnetic resonance imaging (MRI) and arthroscopy files were reviewed.
Results: Of the above noted 262 conscripts with a confirmed cruciate ligament injury, 241 were found to be correctly coded in the NHDR (register coverage 92%). Concerning register accuracy, of the 217 patients having cruciate ligament injury as the main or secondary diagnosis in the NHDR, 194 were confirmed to have this injury as verified by MRI or arthroscopy of the knee (register accuracy 89%).
Conclusions: Our results showed that the coverage and accuracy of diagnosis registration of a cruciate ligament injury of the knee in the Finnish NHDR are very good. Further multi-centre studies are needed to confirm our findings.