Clinical Trials Methods and Outcomes Lab

The Clinical Trials Methods and Outcomes Lab, led by Dr. Michael Harhay, aims to improve the analysis and interpretation of randomized trial data, both for critical care and related fields specifically, and trials more generally.

Randomized trials in critical care and related fields often conclude without finding a statistically significant difference between groups – said another way, the analysis demonstrates that the intervention being tested did not lead to a meaningful change. Dr. Harhay and his team interrogate trials further, examining the methodological choices that could have led to a null result. For instance, they look at sample sizes, analytic techniques, or the interpretation of results to assess if the true benefits of a given intervention were overlooked due to study design or analysis decisions.

More specifically, the Clinical Trials Methods and Outcomes Lab innovates statistical methods to analyze missing patient outcomes – a common occurrence due to gaps in administrative data or research participants who are unable to contribute responses – and variations in treatment responses. Missing data is particularly challenging in trials conducted by the PAIR Center; our trials often measure patient-reported outcomes, like quality of life or physical function, and include seriously ill patients who may die during a lengthy research study. Yet, without this key information, study analyses can become biased, and thus require innovative approaches informed by cutting-edge statistical theory and diverse stakeholders.

Dr. Harhay and his team also use mixed methods approaches to develop outcome measures that improve the efficiency and patient-centeredness of trials.

The work of the Clinical Trials Methods and Outcomes Lab directly informs the conduct and analysis of all PAIR Center trials and has broad applications for the research community at large.