Article

Mobile App Informs Decisions on Kidney Transplant, Dialysis

After developing a predictive model that compares a patient's individualized risk of mortality while undergoing dialysis to the risk of kidney transplant, a research team has created a free mobile app called iChoose Kidney that enables patients and providers to make informed decisions on treatment choices.

After developing a predictive model that compares a patient's individualized risk of mortality while undergoing dialysis to the risk of kidney transplant (KTx), a research team has created a free mobile app called iChoose Kidney that enables patients and providers to make informed decisions on treatment choices.

Rachel E. Patzer, PhD, MPH, of the Emory Transplant Center at Emory University, reported the findings of her group's research and development at Kidney Week 2013, the American Society of Nephrology's annual meeting, held November 5-10, 2013, in Atlanta, GA.

According to the researchers, patients with end-stage renal disease (ESRD) have 2 main treatment choices: dialysis or KTx. In order to compare the individualized risks of mortality with dialysis to that of KTx and make informed decisions about treatment choices, patients and providers need predictive models.

To develop their predictive model, the researchers examined mortality in a cohort of 721,380 ESRD patients over 18 years of age in the United States Renal Data System (USRDS) surveillance registry between 2005 and 2001 who received either dialysis or KTx. They used multivariable logistic regression to predict 1-year and 3-year mortality risks for patients on dialysis and patients who received a living or deceased donor transplant. The investigators then divided the data equally into derivation and validation datasets and evaluated the models using concordance statistics and measures for model discrimination and calibration for 3-year outcomes.

Over the 7-year study period, 47.4% of the 663,900 dialysis patients died, compared to 5.9% of the 57,671 transplant recipients. The factors that were strongly associated with increased odds of death for dialysis and KTx patients included older age; minority race or ethnicity; longer dialysis vintage; non-private insurance; delayed access to pre-ESRD nephrology care; lack of erythropoietin use prior to dialysis; smoking and drug use; and comorbidities that included cardiovascular disease and congestive heart failure. According to the authors, the discriminatory ability of the models for 1-year and 3-year mortality was high.

The characteristics of patients associated with mortality in the univariate analyses were added to a multivariable logistic regression model. Backwards elimination was carried out based on the statistical significance of variables, model discrimination, and goodness of fit. The final model was based on a combination of model fit and the practical and expedient aspects of collecting data in real-time for use as a decision-making tool in the form of the iChoose Kidney app.

iChoose Kidney uses validated mortality risk prediction models and a short questionnaire to incorporate demographic and clinical factors and provide individualized patient risks of mortality with different scenarios. Coefficients are applied to each questionnaire response and the outputs are risk estimates of patient-specific outcomes according to treatment modality. The risk estimates can be applied to national surveillance data of ESRD patients and kidney transplant recipients.

Risk prediction estimates for mortality on dialysis compared to transplant may help identify patients at high risk for poor outcomes; communicate knowledge about risks of morbidity and mortality to patients and providers; and encourage lifestyle or behavioral changes that promote health.

The iChoose Kidney app can be downloaded for free.

The researchers disclosed that their work was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health and the National Institute on Minority Health and Health Disparities.

Related Videos
Ashfaq Marghoob, MD: Artificial Intelligence, Smartphone Use for Pigmented Lesion Classification
Steve Nissen, MD | Credit: Cleveland Clinic
Major Diagnostic Challenges for Pigmented Lesions, with Ashfaq Marghoob, MD
Sherona Bau, NP | Credit: UCLA Health
Jessica Crimaldi, NP | Credit: Jessica Crimaldi on LinkedIn
Benjamin Scirica, MD | Credit: Brigham and Women's Hospital
© 2024 MJH Life Sciences

All rights reserved.