Nephrologist's Perceptions of Risk of Severe Chronic Kidney Disease and Outpatient Follow-up After Hospitalization with AKI: Multinational Randomized Survey Study

Background: Patients hospitalized with acute kidney injury (AKI) have variable risks for chronic kidney disease (CKD); however, there is limited knowledge about how this risk influences outpatient follow-up with nephrologists.
Objective: This survey study examined the likelihood that nephrologists would recommend outpatient follow-up of patients with varying risk profiles for CKD after hospitalization with AKI, and the effect of reporting the predicted risk of severe CKD on their decision-making.
Design: A randomized survey study examining the impact of providing predicted risks of severe CKD on nephrologists' follow-up recommendations for patients with AKI.
Setting: The study included nephrologists from the USA, UK, and Canada between September and December 2023.
Patients: Participants reviewed clinical vignettes of patients with AKI and varying risks of severe CKD (G4 or G5), using an externally validated prediction model. Measurements: The primary outcome was the likelihood of recommending nephrologist specialist follow-up for each case, scored on a seven-point Likert scale (1 = "definitely not," 7 = "definitely would").
Methods: Participants were randomized to receive a version of the survey either with or without the predicted risk of severe CKD included for each vignette. Responses were compared across categories of predicted risk (<10%, 10–49%, and ≥50%) using generalized estimating equations.
Results: Of the 203 nephrologists who participated, 73 (36%) were from the UK, 71 (35%) from Canada and 45 (22%) from the United States. Mean (95% CI) Likert scores increased from 4.01 (3.68, 4.34) for patients with a <10% predicted risk, to a 6.06 (5.76, 6.37) for those with a ≥ 50% predicted risk of severe CKD. Nephrologists were significantly less likely to recommend outpatient nephrology follow-up for patients with a <10% predicted risk of severe CKD when the risk was reported (mean difference -0.71 [95% CI -1.19, -0.23]), and significantly more likely to recommend follow-up for patients with a ≥50% predicted risk when the risk of severe CKD was reported (mean difference 0.49 [95% CI 0.04, 0.93]).
Limitations: This study focuses on nephrologists from high-income countries, and reliedance on hypothetical scenarios rather than real-world practices., and potential selection bias among survey respondents may not be representative of all nephrologists, though consistent findings across diverse subgroups strengthen its relevance findings.
Conclusions: When the predicted risk of severe CKD is reported, nephrologists are less likely to recommend follow-up for lower risk patients with AKI and more likely to recommend follow-up for higher risk patients, leading to better alignment of recommendations for outpatient follow-up with patient risk of severe CKD.