Prediction Model of Immunosuppressive Medication Non-adherence for Renal Transplant Patients Based on Machine Learning Technology

Abstract

ObjectivesPredicting adherence to immunosuppressive medication (IM) is important to improve and design future prospective, personalized interventions in Chinese renal transplant patients (RTPs).MethodsA retrospective, multicenter, cross-sectional study was performed in 1,191 RTPs from October 2020 to February 2021 in China. The BAASIS was used as the standard to determine the adherence of the patients. Variables of the combined theory, including the general data, the HBM, the TPB, the BMQ, the PSSS and the GSES, were used to build the models. The machine learning (ML) models included LR, RF, MLP, SVM, and XG Boost. The SHAP method was used to evaluate the contribution of predictors to predicting the risk of IM non-adherence in RTPs.ResultsThe IM non-adherence rate in the derivation cohort was 38.5%. Ten predictors were screened to build the model based on the database. The SVM model performed better among the five models, with sensitivity of 0.59, specificity of 0.73, and average AUC of 0.75. The SHAP analysis showed that age, marital status, HBM-perceived barriers, use pill box after transplantation, and PSSS-family support were the most important predictors in the prediction model. All of the models had good performance validated by external data.ConclusionsThe IM non-adherence rate of RTPs was high, and it is important to improve IM adherence. The model developed by ML technology could identify high-risk patients and provide a basis for the development of relevant improvement measures.

Development and validation of a LASSO-based prediction model for immunosuppressive medication nonadherence in kidney transplant recipients

“…The inclusion criteria were as follows: (1) age ≥18 years; (2) no graft failure after kidney transplant; (3) no exercise taboo (such as paralysis, osteoarthropathy, and cardiopulmonary disease); KTRs who were not Mandarin-speaking, had multiple transplantations, with cognition/mental disorders, and with postoperative time <3 months were excluded. A total of 1011 KTRs were enrolled in the derivation set and 180 in the validation set [ 12 ]. The flowchart of enrollment is illustrated in Figure 2 .…”

“…Zhu et al. [ 12 ] developed a prediction model based on machine learning technology and presented the predicted results by formulation, which was complex to apply to clinical practice. So far, few models were practical and well-suited to predict IM nonadherence in KTRs, and an innovative model that is theory-based, comprehensive, and easy to apply is needed.…”

Development and Internal Validation of a Questionnaire Assessing Predisposition to Nonadherence to Immunosuppressive Medication in Kidney Pretransplant Patients

“…However, the proposed model is applicable only after transplant. More recently, Zhu et al, 56 using support vector machine technology, a machine learning algorithm used for classification problems, developed a predictive tool based on 10 variables and reported an AUC of 0.75. Again, the tool is applicable only on posttransplant patients.…”

Revolutionizing Kidney Transplantation: Connecting Machine Learning and Artificial Intelligence with Next-Generation Healthcare—From Algorithms to Allografts

“…Sridharan et al [98] identified significant predictors useful in optimizing tacrolimus and cyclosporine dosing regimens using various ML algorithms. Zhu et al's [99] study on non-adherence found the SVM model to be the most effective, with a sensitivity of 0.59, a specificity of 0.73, and an average AUC of 0.75. Monitoring and managing disease recurrence in a transplanted organ is another vital aspect of post-transplant care.…”