Key result
A predictive risk model using dynamically evolving clinical data outperformed a model using static variables for predicting 30-day readmission in kidney transplant recipients (AUC 0.73 vs 0.63; P=0.009).
Why the study?
Does a predictive risk model using dynamic clinical data improve the prediction of 30-day readmissions compared to fixed variables in adult kidney transplant recipients?
Observational (n=1,147)
Does a predictive risk model using dynamic clinical data improve the prediction of 30-day readmissions compared to fixed variables in adult kidney transplant recipients?
Absolute Event Rate: 0.73% vs 0.63%
p-value: p=0.009
Incorporating dynamic clinical data into risk models significantly improves the prediction of 30-day readmissions in kidney transplant recipients compared to using only static administrative data.
May support dynamic readmission models in kidney transplant recipients; hypothesis-generating and requires prospective validation.
BACKGROUND: Thirty-day readmissions (30DRA) are a highly scrutinized measure of healthcare quality and relatively frequent among kidney transplants (KTX). Development of predictive risk models is critical to reducing 30DRA and improving outcomes. Current approaches rely on fixed variables derived from administrative data. These models may not capture clinical evolution that is critical to predicting outcomes. METHODS: We directed a retrospective analysis toward: (1) developing parsimonious risk models for 30DRA and (2) comparing efficiency of models based on the use of immutable versus dynamic data. Baseline and in-hospital clinical and outcomes data were collected from adult KTX recipients between 2005 and 2012. Risk models were developed using backward logistic regression and compared for predictive efficacy using receiver operating characteristic curves. RESULTS: Of 1147 KTX patients, 123 had 30DRA. Risk factors for 30DRA included recipient comorbidities, transplant factors, and index hospitalization patient level clinical data. The initial fixed variable model included 9 risk factors and was modestly predictive (area under the curve, 0.64; 95% confidence interval [95% CI], 0.58-0.69). The model was parsimoniously reduced to 6 risks, which remained modestly predictive (area under the curve, 0.63; 95% CI, 0.58-0.69). The initial predictive model using 13 fixed and dynamic variables was significantly predictive (AUC, 0.73; 95% CI, 0.67-0.80), with parsimonious reduction to 9 variables maintaining predictive efficacy (AUC, 0.73; 95% CI, 0.67-0.79). The final model using dynamically evolving clinical data outperformed the model using static variables (P=0.009). Internal validation demonstrated that the final model was stable with minimal bias. CONCLUSIONS: We demonstrate that modeling dynamic clinical data outperformed models using immutable data in predicting 30DRA.
No takes yet. Share an insight, caveat, or question.
Taber et al. (2015) conducted an observational in Kidney transplantation (n=1,147). Dynamic clinical data risk model vs. Fixed variable risk model was evaluated on 30-day readmission predictive performance (AUC) (95% CI 0.67-0.79, p=0.009). A predictive risk model using dynamically evolving clinical data outperformed a model using static variables for predicting 30-day readmission in kidney transplant recipients (AUC 0.73 vs 0.63; P=0.009).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: