A study published in the Journal of Pain has shown that applying covariate adjustment can increase the precision of clinical trials measuring outcomes such as pain, mood, and fatigue, offering researchers a regulator-backed way to reduce variability without increasing patient numbers.

Measuring subjective endpoints such as pain, mood, and fatigue is one of the greatest challenges in clinical research. Variability between patients often obscures true treatment effects, making it harder to interpret trial results. The new peer-reviewed paper provides a practical framework for covariate adjustment, offering guidance on how to build composite baseline measures that improve statistical power and comply with regulatory expectations.

The authors applied the method to a Phase 3 acute lumbar pain trial. By selecting and building prognostic covariates based on patient characteristics, they were able to improve precision. Further gains were achieved by incorporating composite psychological predictors generated by Cognivia’s Placebell platform, which automates the process of creating these measures. According to the study, this approach improved results by up to 23.4%, without requiring additional patients or extended timelines.

Dominique Demolle, chief executive officer and co-founder of Cognivia, said: “Trials too often fail, not because therapies are ineffective, but because the signals get lost in noise. This study shows a clear, validated path for tackling that noise, without additional patients, delays or cost.”

Covariate adjustment is a method that accounts for differences between patients, such as psychological traits or baseline conditions, in order to reduce noise in trial outcomes. Although supported by regulatory guidance, including from the FDA, the approach remains underused in practice. Cognivia is the first technology company to publish a practical roadmap for implementing covariate adjustment at scale, supported by validation across three separate studies.

The researchers noted that while their case study focused on pain, the principles extend across therapeutic areas where subjective measures play a role. This includes central nervous system conditions, fatigue-related endpoints, and trials involving symptoms linked to emotional or mental health.

Samuel Branders, director of data science at Cognivia and co-author of the paper, said: “This approach is a game changer for trials with subjective endpoints displaying a high variability. It helps produce clear, more trustworthy results and makes better use of patient resources by increasing precision without inflating sample size.”

The article, From theory to practice: Simple rules for improving clinical trial confidence with covariate adjustment, was published in the September 2025 edition of the Journal of Pain.