Coded diagnoses from general practice electronic health records are a feasible and valid alternative to self-report to define diabetes cases in research

Accurate information on the exposure, confounding factors and participants ’ health outcomes is crucial in epidemiological and clinical research. Methods such as questionnaires, medical registries and interviews can be used to define the case status. Self-report is a method that is often used for the definition of type 2 diabetes mellitus (DM) cases in research [1–3]. Previous studies have reported a high agreement between self-reported diagnoses of DM and medical records (kappa ranging between 0.71 and 0.92) [4–8].

Source: Primary Care Diabetes - August 31, 2020 Category: Primary Care Authors: A.W. de Boer, J.W. Blom, M.W.M. de Waal, R.C.A. Rippe, E.J.P. de Koning, I.M. Jazet, F.R. Rosendaal, M. den Heijer, M.E. Numans, R. de Mutsert Tags: Original research Source Type: research