Comparison of Scottish Morbidity Record (SMR)01 data with clinician-based case note reviews in tetermining co-morbidities of patients with chronic kidney disease

Abstract

Background: Scotland is an international leader in using health information for health surveillance, planning and research. The Scottish Morbidity Record (SMR)01 has collected hospital discharge information for more than 40 years and is co-ordinated nationally by the Information Services Division. We aim to assess the validity of using SMR01 data to determine co-morbidities in patients with chronic kidney disease (CKD) by comparing
SMR01 data with clinician-based case note reviews (CNR).

Methods: A validation study nested in the Grampian Laboratory Outcomes Mortality and Morbidity Study-I (GLOMMS-I) cohort. GLOMMS-Iconsisted of 3426 individuals with CKD who had CNR. SMR01 data were linked to GLOMMS-I data using the patients’ Community Health Index number. To assess agreement between SMR01 and CNR data, we estimated prevalence based on SMR01 and CNR, sensitivity, specificity, positive and negative predictive values and the kappa statistic. We also performed sub-group analyses to quantify the limitations of recording for specific clinically relevant
groups