University of Bristol and UK Renal Registry

ILD102

The UK Renal Registry (UKRR) collects and reports information on UK children who receive long-term (>90 days) kidney replacement therapy (KRT) for kidney failure. This data is expected to be highly complete. Hospital Episode Statistics (HES) and the Patient Electronic Database for Wales (PEDW) are electronic hospital health records which record all inpatient and outpatient activity carried out in NHS hospitals in England and Wales. The ‘real world’ data collected by the above electronic health records offers an opportunity to learn more about rare diseases and their management through research. However, an assessment of data quality is required, to understand where gaps in coding for disease may be missing or inaccurate, which may compromise the validity of research findings. Currently, studies that evaluate the use of hospital records for chronic kidney disease surveillance in children are limited. This project aims to validate hospital record data against UKRR data for children with established kidney failure. We will compare information held in the UKRR dataset with that contained within the hospital record using linked HES/PEDW data. We will describe how children who receive long-term KRT for kidney failure are coded for kidney disease and KRT in the electronic hospital record, and determine the difference in timing between datasets for key events including date of first nephrology review and KRT start. We will use UKRR data to identify all children aged <16 years who started KRT (incident cohort) during the study period of 01/01/2000 and 31/08/2020. We anticipate over 1000 patients will be eligible. In doing so, we will understand how reliable HES/PEDW data is in identifying children with kidney failure who receive KRT. This will help to understand data quality issues for future research utilising hospital record data. We will also investigate whether there are differences in children with and without evidence of KRT/ kidney failure codes, such as age of child, sex, and socioeconomic deprivation factors. To our knowledge, this is the first study to use UKRR-data to validate hospital record data in children.