The Minuteful Kidney Test Evaluation: A Randomised Controlled Trial

The National CKD Audit (2017) (1) identified that 11 million people in the UK are at risk of chronic kidney disease (CKD), a long-term irreversible deterioration in kidney function with occurrence being more common in those with diabetes. NICE CG182 and NICE QS05 define testing Albumin:Creatinine ratio (ACR) in at risk populations as critical from a clinical and cost point of view. Despite its importance in the prevention of kidney disease, ACR testing is the worst performing of all 9 NICE diabetes care processes with low screening rates among patients, making this a critical area for improvement. Home testing strategies may therefore be useful in improving proteinuria screening adherence, which the National CKD Audit (2017) (1) finding that up to 65% of people with diabetes have relevant annual urinary ACR tests. The overarching aim of The Minuteful Kidney Test Evaluation * https://healthy.io/services/kidney * https://www.nhsx.nhs.uk/ai-lab/explore-all-resources/understand-ai/healthyio-smartphone-albuminuria-urine-self-testing/ is to increase the use of ACR testing to identify previously undiagnosed albuminuria in those with diabetes. The primary hypothesis is that there will be an increase in proportion of individuals screened in general practices utilising the Minuteful Kidney Test (plus usual care) compared with general practices delivering usual care alone; and that this will result in a higher proportion of previously undiagnosed CKD being diagnosed in individuals in the intervention practices. The suitability of the technology for widescale deployment will also be addressed by exploring differences in outcomes by baseline characteristics alongside the qualitative process evaluation which will also gain an understanding of the technology's application across the UK population through a mixed method evaluation. Using a Cluster RCT design, randomisation will operate on a 1:1 basis with a 50% chance of a cluster being allocated to the intervention. 348 practices (clusters) are required, with one cluster added to each arm to account for any whole cluster drop out. The average cluster size will consist of 470 patients with DM, determined as constant by how many patients with DM are eligible for annual review. 348 GP practices would provide 80% power at the 5% significance level to detect an increase in rate of CKD from 3% in the control arm to 4.3% in the intervention arm. This would provide over 90% power at the 5% significance level to detect an absolute increase of 7% in ACR test rate in the intervention arm compared to control (65% to 72%). The calculations assumed an ICC of 0.05, a coefficient of variation to allow for variation in in cluster size of 0.128 (cluster size range of 360-600).