American Scientist

Have you ever watched a TV mystery where the key clue is just out of view? That’s a lot like what happens when scientists study illnesses such as Huntington disease. Researchers can see some factors that affect disease progression, but other factors are only partially visible because studies end before they can be fully observed, in what’s called right-censored data. It’s like trying to figure out the full story of a mystery show when the last few minutes get cut off. Typically, scientists handle this problem using methods designed for "missing data," but right-censored data contains partial information—hints about what the hidden values might be—unlike missing data, which gives researchers nothing. Tanya Garcia is an associate professor of Biostatistics at the University of North Carolina at Chapel Hill and leads a transdisciplinary research team of statisticians and neuroscientists toward designing robust statistical methods for studying neurodegenerative diseases. In this recorded talk, Garcia discusses her team’s methods as applied to Huntington disease data, looking at how cognitive function changes before diagnosis.