CLINICAL RESEARCH ARTICLE Evaluation of administrative case definitions for chronic kidney disease in children Allison Dart 1 , Mariette Chartier 1,2 , Paul Komenda 1 , Randy Walld 2 , Ina Koseva 2 , Charles Burchill 2 and Navdeep Tangri 1 INTRODUCTION: Administrative data is increasingly used for chronic disease surveillance; however, its validity to define cases of chronic kidney disease (CKD) in children is unknown. We sought to evaluate the performance of case definitions for CKD in children. METHODS: We utilized population-based administrative data from the Manitoba Center for Health Policy to evaluate the validity of algorithms based on a combination of hospital claims, outpatient physician visits, and pharmaceutical use over 1–3 years in children <18 years of age. Algorithms were compared with a laboratory-based definition (estimated glomerular filtration rate < 90 ml/min/1.73 m 2 and/or presence of proteinuria). RESULTS: All algorithms evaluated had very low sensitivity (0.20–0.39) and moderate positive predictive value (0.52–0.68). Algorithms had excellent specificity (0.98–0.99) and negative predictive value (0.96–0.97). Receiver operating characteristic (ROC) curves indicate fair accuracy (0.60–0.68). Sensitivity improved with increasing years of data. One or more physician claims and one or more prescriptions over 3 years had the highest sensitivity and ROC. CONCLUSIONS: The sensitivity of administrative data algorithms for CKD is unacceptably low for a screening test. Specificity is excellent; therefore, children without CKD are correctly identified. Alternate data sources are required for population-based surveillance of this important chronic disease. Pediatric Research (2020) 87:569–575; https://doi.org/10.1038/s41390-019-0595-1 INTRODUCTION Chronic kidney disease is a significant health problem in North America, affecting ~10–11% of the population. 1,2 The disease is associated with significant morbidity 3,4 and can ultimately progress to end-stage kidney disease (ESKD), requiring dialysis or kidney transplant to sustain life. 5 The impacts of ESKD are significant with respect to quality of life 6 as well as economic impacts. 7 In Canada, children make up <2% of the entire ESKD population. 8 However, when children develop CKD the conse- quences are dire, with significant impacts on quality of life, morbidity, and mortality. 9–12 In addition, over the past few decades there is a growing burden of early-onset risk factors for CKD in Canadian children, including hypertension, 13 obesity, 14 and diabetes. 15 Of concern is the lack of population-based data describing the prevalence of CKD, and therefore an incomplete understanding of the scope of this important health problem. Early identification is critical in order to implement therapy to slow the progression to ESKD. 16–18 As children have a longer life expectancy than adults, their lifetime risk of ESKD may be significant. Unfortunately, there is a paucity of research describing the population burden of CKD in children; therefore, the prevalence of children at risk for progression to ESKD is unknown. Administrative data are a valuable tool that can be used for chronic disease surveillance. 19 They are reliably utilized for some chronic diseases such as diabetes and inflammatory bowel disease. 20–22 Administrative case definitions for CKD have been evaluated in adult populations 23,24 with discouraging results. To our knowledge, no published studies have examined the validity of using administrative data in defining cases of CKD in children. In this study, we sought to evaluate the validity of administrative data definitions to identify children <18 years with CKD. MATERIALS AND METHODS In order to evaluate the performance of administrative case definitions for childhood CKD, we first identified a validation cohort utilizing laboratory data from Winnipeg, Manitoba, Canada. A gold standard cohort of children who reliably met the criteria for CKD (definitions described below) was identified, as was a cohort of children who had received testing for CKD and had normal results (no CKD cohort). Case definitions (described below) were then applied to the pediatric Winnipeg population and compared against this validation cohort of children with and without CKD. Approvals were obtained from the University of Manitoba Ethics Committee and the Manitoba Health Information Privacy Committee. DATA SOURCES The Manitoba Health Services Insurance Plan contains registration files, physician reimbursement claims (medical services data), hospital discharge abstracts, and records of outpatient prescrip- tions dispensed. These data are stored for research purposes in de-identified form in the Population Health Research Data Repository (herein referred to as the Repository) housed at the Manitoba Center for Health Policy. Canada’s health care system provides universal medical cover- age to all Canadian citizens. Funding is public and administered on a provincial basis, within guidelines set by the federal Received: 2 May 2019 Revised: 13 August 2019 Accepted: 5 September 2019 Published online: 2 October 2019 1 University of Manitoba, Winnipeg, MB R3T 2N2, Canada and 2 Manitoba Center for Health Policy, Winnipeg, MB, Canada Correspondence: Allison Dart (adart@hsc.mb.ca) www.nature.com/pr © International Pediatric Research Foundation, Inc. 2019 1234567890();,: