Researchers Taking Advantage of PCORnet’s Scale to Study Rare Diseases
The Patient-Centered Outcomes Research Institute (PCORI) is funding four research studies on rare diseases that will take advantage of data gleaned from electronic health records and other health data sources. Another study is using computable phenotypes to identify people with rare diseases.
The four rare disease studies, approved for a cumulative total of $19 million, will use PCORnet, the National Patient-Centered Clinical Research Network, to provide new evidence about management of these rare diseases. With health records for 66 million patients available for observational studies, PCORnet provides vast scale to power research on conditions affecting even small numbers of people. PCORnet also makes it easier to identify individuals who may be good candidates to participate in research studies and invite them to take part.
One study aims to produce information to help families of children with a rare form of epilepsy decide between additional anti-seizure medication or surgery based on how often patients using each option seek emergency room care and how the options affect the children’s developmental and functional difficulties. Another study will assess how effective different strategies to monitor and treat high blood pressure are at preserving kidney function in children who have chronic kidney disease.
The CER-NET study aims to determine the optimal sequencing of treatments for people with neuroendocrine tumors, a rare form of cancer. A fourth study seeks to understand whether people with complex congenital heart disease developed in childhood have better outcomes if they continue to receive recommended care from a heart disease specialist when they transition to adulthood rather than primary care alone.
Separately, PCORI approved $1 million to fund the development of automated ways to more efficiently and accurately identify people with rare diseases or complex health problems via electronic health records, enabling treatments to start sooner and helping researchers offer individuals opportunities to participate in studies.
The study, being led by the University of Pennsylvania Perelman School of Medicine, includes this project summary: “In a clinical research network, if we can identify specific patient cohorts at scale, we can more efficiently trial new treatments. To locate these patients, investigators formulate a set of criteria, known as a computable phenotype, that can be used to search for patients in clinical data.
"Unfortunately, because health records are not designed for this purpose, constructing a computable phenotype that identifies most appropriate patients but does not identify many inappropriate patients requires considerable time, expertise, and troubleshooting. In this project, we propose to develop innovative machine learning methods, and incorporate them into software, which will largely automate the construction of such computable phenotypes. These methods will learn from a small set of target patients, but do not require a definitive set of nontarget patients, which should enable them to perform particularly well for diseases in which many affected patients are unaware that they have the disease. These methods will learn from many clinical practices’ data without the need to share individuals’ data—something that should enable them to learn how to find patients with rare conditions. Moreover, the methods will be configurable to ensure that resulting computable phenotypes are equitable across diverse patients.”
