A groundbreaking multicenter clinical trial has unveiled an innovative, data-driven approach to identifying liver transplant recipients at imminent risk for organ rejection due to medication nonadherence. Pioneered through collaboration between the Icahn School of Medicine at Mount Sinai and Texas Children’s Hospital, the study leverages the Medication Level Variability Index (MLVI)—a novel, electronic health record (EHR)-based biomarker—to transform how transplant patients’ adherence is monitored, particularly in the vulnerable adolescent and young adult population.
Medication nonadherence remains a formidable clinical challenge in transplantation medicine, directly correlating with the incidence of allograft rejection and subsequent morbidity. Adolescents and young adults are disproportionately affected, often struggling with complex immunosuppressive regimens required to prevent rejection. Traditional detection methods rely heavily on patient self-reporting or sporadic clinical assessments, both of which are prone to bias and oversight. The MLVI surmounts these limitations by quantifying fluctuations in immunosuppressant blood concentrations extracted from routine lab draws, thus providing an objective, continuously updated marker of adherence consistency.
The study encompassed 13 leading pediatric transplant centers across the United States and Canada, screening over 3,000 liver transplant recipients’ medical records. A subset of 148 high-risk patients, identified through elevated MLVI scores indicating inconsistent medication intake, were randomized to receive either standard care or a comprehensive two-year telehealth-based behavioral intervention. This innovative intervention utilized regular, remote engagement with behavioral specialists trained to reinforce adherence strategies and support patient self-management remotely, a particularly pivotal adaptation during the COVID-19 pandemic’s disruption of traditional in-person care.
Although the trial did not achieve statistical significance on the composite primary endpoint—which included rejection episodes, re-transplantation, and patient withdrawal—this was attributed to unexpectedly low rejection rates in both the intervention and control arms. Notably, recipients undergoing the remote behavioral intervention experienced roughly half as many rejection-related events and instances of re-transplantation relative to those in the standard care group. This signals a clinically meaningful impact akin to significant risk reduction, underscoring the utility of integrating MLVI-guided interventions into routine post-transplant care.
Beyond the study’s interventional efficacy, a compelling secondary finding was the noticeable reduction in overall rejection incidences associated with systematic MLVI implementation itself. Deployment of this risk marker as part of routine clinical workflows appeared to recalibrate clinician vigilance and patient management strategies effectively, driving digital medicine toward a more proactive, preemptive paradigm. This represents a paradigm shift from traditional reactive post-rejection treatment models to anticipatory, preventive care grounded in real-time data analytics.
In elucidating the clinical implications, Dr. Eyal Shemesh, lead investigator and behavioral health chief at Mount Sinai Kravis Children’s Hospital, emphasized the profound potential of harnessing extant EHR data in identifying nonadherence before it culminates in irreversible graft damage. This methodology enables focused allocation of clinical resources toward those patients most at risk, potentially mitigating adverse outcomes and associated healthcare costs through early behavioral interventions.
The underlying mechanistic rationale of the MLVI involves assessing intra-patient variability in immunosuppressant blood levels, primarily tacrolimus—a calcineurin inhibitor with a narrow therapeutic index vital for transplant survival. High MLVI values reflect erratic drug intake patterns, which compromise steady-state pharmacokinetics essential for consistent immunosuppression. This bioinformatics approach quantifies variability statistics, facilitating clinician recognition of subtle, yet clinically consequential lapses in adherence previously undetectable through conventional means.
Remote intervention strategies adopted in the study capitalized on telemedicine’s scalability and patient-centered design, utilizing frequent virtual check-ins to monitor, educate, and motivate adolescent transplant recipients. This framework not only fostered sustained engagement over two years but also offered a flexible alternative amidst pandemic-induced restrictions, showcasing telehealth’s integral role in contemporary chronic disease management and its potential to enhance access and adherence among geographically diverse populations.
Dr. Benjamin L. Shneider of Texas Children’s, senior author and leading gastroenterologist, highlighted the broader applicability of these findings beyond transplantation, suggesting that integrating biomarkers like MLVI could revolutionize care paradigms for myriad chronic conditions where adherence is paramount. The study advocates for embedding such objective adherence metrics within EHR platforms, enabling clinicians to harness precision medicine tools capable of individualized risk stratification and tailored intervention deployment.
Importantly, the study also delineates future avenues for investigation to optimize resource utilization and cost-effectiveness of MLVI-guided behavioral programs. While efficacy signals are positive, defining precise thresholds for intervention initiation, intervention intensity, and long-term sustainability require further study. Additionally, expanding the MLVI concept to encompass other organ transplant types and immunosuppressive agents could broaden the tool’s impact across transplant medicine.
The research was conducted within the multidisciplinary environment of the Texas Children’s Research Institute, leveraging cross-specialty collaboration to translate laboratory insights into tangible clinical strategies. Funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) under the National Institutes of Health (NIH), this project exemplifies how integrating data science with clinical expertise catalyzes breakthroughs in patient care.
In conclusion, this study marks a significant advance in transplant medicine’s ongoing quest to mitigate rejection through early detection of medication nonadherence. By employing an EHR-based biomarker coupled with remote, behaviorally informed interventions, clinicians can pivot toward a model that prioritizes prevention over treatment of rejection episodes. This shift harbors profound implications for improving long-term transplant survival, reducing hospitalizations, and enhancing quality of life for adolescent and young adult transplant recipients worldwide.
Subject of Research: Medication nonadherence detection and intervention in adolescent liver transplant recipients using an EHR-based biomarker (MLVI).
Article Title: A remote intervention to improve medication nonadherence guided by a marker of risk derived from the electronic health records of adolescent transplant recipients.
News Publication Date: June 15, 2026
Web References:
American Journal of Transplantation article
Texas Children’s Hospital
Mount Sinai Health System Facebook
Mount Sinai Health System Instagram
Mount Sinai Health System LinkedIn
Mount Sinai Health System Twitter (X)
Mount Sinai Health System YouTube
References: American Journal of Transplantation, AJT1253, May 27, 2026
Keywords: Medication Level Variability Index, MLVI, medication nonadherence, liver transplantation, organ rejection, adolescent transplant recipients, electronic health records, telehealth intervention, immunosuppressant monitoring, behavioral health, pediatric transplantation, remote patient monitoring
Tags: adolescent transplant patient adherencedata-driven transplant patient managementelectronic health record biomarkersimmunosuppressant blood concentration trackinginnovative transplant adherence toolsliver transplant patient monitoringmedication level variability indexmulticenter transplant clinical trialorgan rejection predictionpediatric liver transplant outcomesreducing organ rejection ratestransplant medication nonadherence
