Researchers at the Rollins School of Public Health at Emory University received an award from the Patient-Centered Outcomes Research Institute (PCORI) to study statistical methods for missing data in large observational studies. The study, which was motivated by the Paul Coverdell National Acute Stroke Registry, is part of a portfolio of projects awarded by PCORI that will advance the field of patient-centered, comparative effectiveness research and provide patients with information that will help them make better-informed decisions about their care.

Qi Long, PhD, associate professor in the Department of Biostatistics and Bioinformatics at Rollins, will lead the research project that will focus on the development, dissemination and assessment of methods to account for missing data in large observational studies. The team anticipates that the proposed methods will enable investigators to obtain robust and valid results on patient outcomes while educating and training stakeholders and students on the application of statistical methods for missing data.

"Missing data are frequently encountered and prevalent in large observational studies including registries," explains Long. "Many of the variables that have missing values are necessary for relevant and robust findings. Our proposed methods, applied to the Coverdell Registry data, will identify important factors, steps, and gaps in patient care that would otherwise be missed."

"Long's study is one of 71 projects totaling more than $114 million approved for funding by PCORI's Board of Governors. The awards were a mix of projects that included the first studies specifically targeting improvement of research methods.

This project was selected for PCORI funding on the basis of its potential to advance the relatively new field of patient-centered comparative effectiveness research, which focuses on measuring outcomes that matter most to patients and those who care for them," says PCORI Executive Director Joe Selby, MD, MPH. 

"By enhancing the methods for conducting this new approach to health research, this study will provide information that will ultimately help people make more fully informed health and healthcare decisions. We look forward to following the study's progress and working with Emory's Rollins School of Public Health to share the results."

“Large observational studies, including registries, play an important role in patient-centered outcomes research,” explains Long. “Missing data can seriously compromise statistical analysis and the validity of their findings. Existing methods are not appropriate for handling missing data in these settings and likely lead to biased results that could produce harmful effects on patient outcomes wrong allocations and waste of resources. Our proposed methods are designed to overcome the limitations of the existing methods and ultimately improve patient outcomes.”