Categories
Nevin Manimala Statistics

Clinician Prediction of Early Readmission Among Kidney and Liver Transplant Recipients

Prog Transplant. 2024 Oct 30:15269248241288556. doi: 10.1177/15269248241288556. Online ahead of print.

ABSTRACT

Introduction: Patients are at risk of hospital readmission after kidney and liver transplantation due to the complexity of posttransplant care. Currently, clinical practice relies on providers’ prediction since there is a lack of specific strategies. However, the accuracy of clinicians’ ability to predict readmissions using clinical judgment alone is unknown. Research Question: What is the accuracy of clinicians’ ability to predict readmissions after transplantation using clinical judgment alone? Design: In 2019, clinical providers at a large, urban transplant center were electronically surveyed. Primary surgeons, nephrologists, transplant pharmacists, hepatologists, and nurses were asked, within 24 h of any kidney or liver transplant recipient discharge, to predict whether a patient would be readmitted within 30 days, and the suspected causes of readmission. Prediction accuracy was assessed by sensitivity, specificity, positive and negative predictive value, and F-score. Kappa scores were calculated to assess agreement between transplant surgeons and other providers. Results: Overall, N = 34 unique providers were surveyed about 148 kidney and 63 liver transplant recipients, and 27.0% of kidney recipients and 25.4% of liver recipients were readmitted within 30 days. The positive predictive values were low among clinical providers, ranging from 0.25 to 0.55. Agreements between providers were weak, but higher among kidney transplant providers (range: 0.42-0.44) than for liver transplant providers (range: -0.02-0.26). Conclusion: Clinical judgment alone to predict readmission among transplant recipients may not be sufficient and a combination of clinicians’ predictions, multitiered discharge surveillance strategies and data-based predictive models may better identify high-risk patients and guide interventions to reduce readmission.

PMID:39474702 | DOI:10.1177/15269248241288556

By Nevin Manimala

Portfolio Website for Nevin Manimala