Validity of the diagnosis of appendicitis in the Danish National Patient Register

Abstract

Aims: Appendicitis is a common disease. The nationwide Danish National Patient Register provides an important data source for epidemiological research. Data used in register-based epidemiological research needs to be validated. We aimed to validate the diagnosis of appendicitis in the Danish National Patient Register. Methods: From 1997 to 2015 nationwide data from the Danish National Patient Register, the Danish Pathology Register, and the Danish Civil Registration System were used to validate the diagnosis of appendicitis or the combination of the diagnosis for appendicitis and surgical removal of the appendix in the Danish National Patient Register. Sensitivity, specificity, and positive and negative predictive values were calculated using pathology reports as golden standard. Results: Diagnosis of appendicitis in the Danish National Patient Register had a sensitivity, specificity, positive predictive value, and negative predictive value of 0.928 (95% confidence interval (CI): 0.927; 0.930), 0.995 (95% CI: 0.995; 0.995), 0.769 (95% CI: 0.767; 0.771), and 0.999 (95% CI: 0.999; 0.999). A diagnosis of appendicitis and a procedure code for surgical removal of the appendix had a sensitivity, specificity, positive predictive value, and negative predictive value of 0.886 (95% CI: 0.885; 0.888), 0.998 (95% CI: 0.998; 0.998), 0.895 (95% CI: 0.894; 0.897), and 0.998 (95% CI: 0.998; 0.998). Conclusions: The diagnosis of appendicitis alone or in combination with the registered surgical removal of the appendix in the Danish National Patient Register showed acceptable validity. Whether to use the diagnosis for appendicitis only or in combination with procedure codes for the removal of the appendix depends on whether high sensitivity or high positive predictive values are warranted.