Personalized dynamic dose-finding for longitudinal observational data: an application to glucose intolerance in VLBW infants

The prescribed doses for many drugs are based on population norms or vary between physicians. For example, very low birth weight (VLBW) infants (weighing less than 1500 grams at birth) experience slower postnatal growth than in utero and receive glucose infusions to mitigate nutritional disruption. Effective treatment dosing is critical to ensure adequate neonatal weight gain while minimizing the risk of hyperglycemia. A single uniform dose is impractical due to the substantial heterogeneity in glucose metabolism among infants. Furthermore, little is known about each infant's glucose metabolism, which complicates the selection of the optimal dose value. We employ a flexible longitudinal mixed model to determine dynamic, patient-specific optimal doses for new infants. Empirical Bayes predictions of random effects are incorporated to capture individual dosage sensitivities that adapt as new data becomes available. To be viable in high-risk clinical environments, providing estimates of predictive uncertainty is crucial for supporting informed clinical decision-making that balances patient safety with treatment efficacy. We propose using the implicit value delta method to quantify the uncertainty of the predicted optimal doses. We validate the properties of our model through simulation studies and an analysis of VLBW infants in a neonatal intensive care unit (NICU) for the first 7 days of life, which demonstrates some key differences in optimal and prescribed dosing strategies.

Authors

Institutions

Publication Details

Journal
Journal of Biopharmaceutical Statistics
Published
2026-11-13
DOI
https://doi.org/10.1080/10543406.2026.2761999
Primary Topic
Statistical Methods in Clinical Trials
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Personalized dynamic dose-finding for longitudinal observational data: an application to glucose intolerance in VLBW infants

Stella Coker Watson Self, Md. Nasim Saba Nishat, Alexander C. McLain, Bryan L. Ohning et al.
Journal of Biopharmaceutical Statistics
Statistical Methods in Clinical Trials
article

Personalized dynamic dose-finding for longitudinal observational data: an application to glucose intolerance in VLBW infants

Stella Coker Watson Self, Md. Nasim Saba Nishat, Alexander C. McLain, Bryan L. Ohning, Rahul Ghosal, Christine Schammel, Mirinda Ann Gromley
article en

Abstract

The prescribed doses for many drugs are based on population norms or vary between physicians. For example, very low birth weight (VLBW) infants (weighing less than 1500 grams at birth) experience slower postnatal growth than in utero and receive glucose infusions to mitigate nutritional disruption. Effective treatment dosing is critical to ensure adequate neonatal weight gain while minimizing the risk of hyperglycemia. A single uniform dose is impractical due to the substantial heterogeneity in glucose metabolism among infants. Furthermore, little is known about each infant's glucose metabolism, which complicates the selection of the optimal dose value. We employ a flexible longitudinal mixed model to determine dynamic, patient-specific optimal doses for new infants. Empirical Bayes predictions of random effects are incorporated to capture individual dosage sensitivities that adapt as new data becomes available. To be viable in high-risk clinical environments, providing estimates of predictive uncertainty is crucial for supporting informed clinical decision-making that balances patient safety with treatment efficacy. We propose using the implicit value delta method to quantify the uncertainty of the predicted optimal doses. We validate the properties of our model through simulation studies and an analysis of VLBW infants in a neonatal intensive care unit (NICU) for the first 7 days of life, which demonstrates some key differences in optimal and prescribed dosing strategies.

Journal of Biopharmaceutical Statistics
Prisma Health (US), University of South Carolina (US)
Good health and well-being
Statistical Methods in Clinical Trials
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon