Predicting the risks of kidney failure and death in adults with moderate to severe chronic kidney disease: multinational, longitudinal, population based, cohort study

Ping Liu,Simon Sawhney,Uffe Heide-Jørgensen,Robert Ross Quinn,Simon Kok Jensen,Andrew Mclean,Christian Fynbo Christiansen,Thomas Alexander Gerds,Pietro Ravani

DOI: https://doi.org/10.1136/bmj-2023-078063

2024-04-15

BMJ

Abstract:Abstract Objective To train and test a super learner strategy for risk prediction of kidney failure and mortality in people with incident moderate to severe chronic kidney disease (stage G3b to G4). Design Multinational, longitudinal, population based, cohort study. Settings Linked population health data from Canada (training and temporal testing), and Denmark and Scotland (geographical testing). Participants People with newly recorded chronic kidney disease at stage G3b-G4, estimated glomerular filtration rate (eGFR) 15-44 mL/min/1.73 m 2 . Modelling The super learner algorithm selected the best performing regression models or machine learning algorithms (learners) based on their ability to predict kidney failure and mortality with minimised cross-validated prediction error (Brier score, the lower the better). Prespecified learners included age, sex, eGFR, albuminuria, with or without diabetes, and cardiovascular disease. The index of prediction accuracy, a measure of calibration and discrimination calculated from the Brier score (the higher the better) was used to compare KDpredict with the benchmark, kidney failure risk equation, which does not account for the competing risk of death, and to evaluate the performance of KDpredict mortality models. Results 67 942 Canadians, 17 528 Danish, and 7740 Scottish residents with chronic kidney disease at stage G3b to G4 were included (median age 77-80 years; median eGFR 39 mL/min/1.73 m 2 ). Median follow-up times were five to six years in all cohorts. Rates were 0.8-1.1 per 100 person years for kidney failure and 10-12 per 100 person years for death. KDpredict was more accurate than kidney failure risk equation in prediction of kidney failure risk: five year index of prediction accuracy 27.8% (95% confidence interval 25.2% to 30.6%) versus 18.1% (15.7% to 20.4%) in Denmark and 30.5% (27.8% to 33.5%) versus 14.2% (12.0% to 16.5%) in Scotland. Predictions from kidney failure risk equation and KDpredict differed substantially, potentially leading to diverging treatment decisions. An 80-year-old man with an eGFR of 30 mL/min/1.73 m 2 and an albumin-to-creatinine ratio of 100 mg/g (11 mg/mmol) would receive a five year kidney failure risk prediction of 10% from kidney failure risk equation (above the current nephrology referral threshold of 5%). The same man would receive five year risk predictions of 2% for kidney failure and 57% for mortality from KDpredict. Individual risk predictions from KDpredict with four or six variables were accurate for both outcomes. The KDpredict models retrained using older data provided accurate predictions when tested in temporally distinct, more recent data. Conclusions KDpredict could be incorporated into electronic medical records or accessed online to accurately predict the risks of kidney failure and death in people with moderate to severe CKD. The KDpredict learning strategy is designed to be adapted to local needs and regularly revised over time to account for changes in the underlying health system and care processes.

medicine, general & internal

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the prediction of the risk of renal failure and death in adults with moderate - to - severe chronic kidney disease (CKD, stage G3b to G4). Specifically, the study aims to:

1. **Develop a tool**: This tool can simultaneously predict the risk of renal failure (taking into account the competing risk of death) and all - cause death, and is applicable to adults newly diagnosed with moderate - to - severe chronic kidney disease (stage G3b to G4). This tool is called KDpredict and can be used through electronic medical records or online access.

2. **Improve prediction accuracy**: Compared with existing benchmark models (such as the renal failure risk equation), KDpredict shows higher prediction accuracy in external validations in Denmark and Scotland, especially in predicting the risk of renal failure.

3. **Support clinical decision - making**: By providing simultaneous predictions of renal failure and death risks, KDpredict supports a more comprehensive discussion and a patient - centered decision - making process, avoiding the adverse consequences that may be brought about by only focusing on the risk of renal failure while ignoring the risk of death.

4. **Adaptability and update**: The design of KDpredict is flexible, can adapt to the needs of different regions, and is regularly revised to reflect the changes in the health system and care processes, thereby providing a more personalized tool.

In summary, the goal of this study is to develop a more accurate and comprehensive risk prediction tool to support the clinical management and personalized treatment decision - making of patients with moderate - to - severe chronic kidney disease.