A Bayesian borrowing-by-parts method integrated with propensity score stratification for borrowing adult data for pediatric patients

In medical research, it's essential to find the most effective treatments for every patient group, including children. However, because children are less often included in clinical trials, it can be challenging to gather enough data to determine the best treatments for them. This project aims to address this challenge by using advanced statistical methods to better estimate how well treatments work for children by leveraging adult studies' data.

To explain our approach, let's break down some key terms and concepts: Bayesian methods are like making an educated guess that you continually improve as you gather more information. Bayesian borrowing is when we use information from one group (adults) to help make better guesses for another group (children). Propensity score stratification helps us make fair comparisons by ensuring that the groups we're comparing are similar in important ways. By combining these methods, we can make the most of the available data to estimate treatment effects more accurately for pediatric patients.

This research is crucial because it can lead to better-informed medical decisions and improve care for children, who often have fewer treatment options supported by strong evidence. The Belimumab treatment being studied, which is a drug that is used to treat certain autoimmune diseases, such as lupus, could impact thousands of children suffering from Systemic lupus erythematosus (SLE) disease, a chronic autoimmune disease where the immune system attacks the body's own tissues (causing inflammation and damage to various organs). By using data from adult studies to enhance our understanding of how these treatments work in children, we can improve healthcare outcomes for a large and vulnerable segment of the population. This research has the potential to directly benefit millions of children by providing them with more effective and safer treatment options.