Trends in Hospital Admissions Associated With an Acute Kidney Injury in England 1998-2020: A Repeated Cross-Sectional Study

Background: A rapid rise in diagnosis of Acute Kidney Injury (AKI) has been observed in the United Kingdom and worldwide. It is unclear whether this reflects true growth in incidence or improved detection. Methods: Aggregate data from Hospital Episode Statistics database on all admissions to National Health Service (NHS) hospitals in England 1998/99 - 2019/20 were used to extract the number of AKI diagnoses in different age groups and compare them to diseases of similar population incidence (pneumonia, heart failure, dementia) based on ICD-10 coding. Haemofiltration procedures were used as proxy for acute kidney replacement therapy (KRT). Temporal trends in usage of AKI-specific ICD-10 codes were examined.  Findings: A total of 356 million inpatient episodes over 22 years were analysed. Between 1998/99 and 2019/20, primary AKI diagnoses increased eight-fold from 14,041 to 110,960. From 2012/13, when first recorded, secondary AKI diagnoses doubled from 404,019 to 902,671; and the total proportion of hospital episodes associated with AKI doubled from 26·9 to 48·5 per thousand episodes. Average annual growth in AKI of 11·6% exceeded that of pneumonia, heart failure, and dementia. The proportion of acute KRT to total AKI cases decreased. In 2019/20, 96% of AKI cases were encoded as N17·9 (Acute renal failure, unspecified). Interpretation: A substantial increase in AKI cases observed within the NHS in the last two decades is likely a result of improved detection and coding, particularly of mild cases. Adoption of clinically orientated coding could facilitate further improvements in detection with benefits for clinical practice and research. Funding Statement: The authors received no funding for this study. Declaration of Interests: Dr John Prowle has received consultancy fees from MediBeacon, Nikkiso Europe GmbH, and Quark Pharmaceuticals; speaker honoraria from Baxter, Fresenius Medical Care, and Nikkiso Europe GmbH; and research support from bioMerieux. The other authors have declared no competing interests. Ethics Approval Statement: This study used anonymized, publicly available national data published by NHS Digital available under an Open Government License v3·0 and as such it did not require an ethics committee approval.