15283-B-Blokker

83 Agreement between MIA and CA: a cross-sectional study 5 Data Analysis Agreement on Cause of Death COD determined respectively by MIA and CA were compared; in cases of disagreement between the two methods a consensus COD was reached for each case in three successive reviews by independent experts, as described in Appendix D. Diagnoses The MIA researcher (BMB) extracted and coded all different post - mortem diagnoses from the final MIA and CA reports using the International Classification of Diseases, Tenth Revision 138 with minor modifications. Diagnoses were sorted by 20 organ/tissue categories and one category of general diagnoses (Appendix E). Per case, a MIA researcher (IMW) extracted and coded all pre-mortem diagnoses, using the prepared ICD-10 list. The MIA pathologist and radiologist jointly scored all post - mortem diagnoses per MIA and two independent pathologists (KHL, VSM) together scored all post - mortem diagnoses per CA. Per case, independently for MIA and CA, they scored a post - mortem diagnosis as “certain” according to established radiological and/ or pathological criteria, or as “probable”, if there was any uncertainty. Also per case, for MIA and CA independently, a post - mortem diagnosis was classified as major if it was directly related to COD. In retrospect, these major diagnoses were, per organ/tissue category, grouped to combine related diagnoses (e.g. necrosis of the lungs, plus infection of the lungs, plus acute pneumonia). In the final step of our analysis GMD were per case included only if they were directly related to the consensus cause of death. Diagnostic Errors Retrospectively the diagnostic errors of MIA and CA were classified. A perceptual error was defined as an abnormality that, though present, was not reported or as an abnormality that, though absent, was reported. 139-141 A cognitive error was defined as an abnormality that, though reported correctly, was not correctly interpreted. An error was defined as a sampling error when a biopsy of a suspected radiological finding, confirmed with CA, was negative. Analyses All data were sealed in an OpenClinica electronic database. Agreement on COD between MIA and CAwas calculated. Furthermore COD established byMIA and CA, was compared to the consensus COD, and the percentage of cases in which COD was classified as