New York State Psychiatric Institute/Columbia University, Mayo Clinic, Icahn School of Medicine at Mount Sinai, and Weill Cornell Medicine have been awarded a $7.6 million five-year collaborative grants (R01 MH121921, R01 MH121922, R01 MH121923, and R01 MH121924) from the National Institute of Health (NIH) to leverage hospital electronic health records (EHRs) to study mental illness, its causes and clinical outcomes, and its genetics. The team will predict psychiatric diagnoses including depression, anxiety and substance use disorder, using clinical and social determinants of health data and genetic risk scores.
Major depressive disorders (MDD), anxiety disorder, and substance use disorders (SUD) are highly prevalent in the United States and contribute to significant public health issues, including suicide. The National Alliance on Mental Health estimates an 11-year average delay between the onset of mental illness symptoms and treatment. During those gap years, these three conditions can lead to increased healthcare utilization and cost, and higher morbidity and mortality, and decreased quality of life.
Recent studies have shown that MDD, anxiety, and SUD frequently co-occur due to shared genetic and environmental risk factors. As a result, there is an increased interest in using polygenic risk scores (PRSs), an aggregate of the impact of many DNA variants, and social determinants of health (e.g., education level, employment), as predictors and as biomarkers to identify and stratify patients.
Dr. Jyotishman Pathak
Principal investigators on this grant include Dr. Joanna Biernacka, professor of biostatistics and psychiatry at Mayo Clinic and Director of Mayo’s Psychiatric Genomics and Pharmacogenomics Research Program; Dr. J. John Mann, Paul Janssen Professor of Translational Neuroscience at Columbia University; Dr. Jyotishman Pathak, Frances & John L. Loeb Professor of Medical Informatics and Psychiatry, chief of the Division of Health Informatics, and vice-chair of the Department of Healthcare Policy & Research at Weill Cornell Medicine; Dr. Eli Stahl, assistant professor of Genetics and Genomic Sciences and of Psychiatry, and Associate Director of the Pamela Sklar Division of Psychiatric Genomics at Icahn School of Medicine at Mount Sinai; and Dr. Priya Wickramaratne, associate professor of clinical biostatistics (in psychiatry) at Columbia University and research scientist at the New York State Psychiatric Institute.
Together, researchers from all four medical centers will analyze EHR data from over 30 million patients linked to biobank data from almost 60,000 patients. The clinical derive from INSIGHT Clinical Research Network, a research collaboration across six health systems in New York City, and the biobank data include the Mount Sinai BioMe biobank and the Mayo Clinic Biobank. A large, demographically and geographically diverse data set is critical in understanding the association between psychiatric phenotypes and clinical outcomes, including its biological underpinnings. While PRSs have exhibited potential in risk prediction, their clinical utility for mental health disorders has not yet been well explored.
“There’s no “single gene” that predicts your mental health disease susceptibility, and multiple genetic and environmental factors can affect your emotional and physical wellbeing,” said Dr. Pathak. “By using real patient data derived from large-scale electronic health records linked to institutional biobanks, we can develop methods that improve our approach to mental health disorder research and hopefully clinical practice.”
The project has three aims: use structured data and unstructured clinical data to develop deep phenotyping for MDD, anxiety, SUD, and related outcomes; evaluate the performance of PRSs in predicting these EHR-derived phenotypes; and assess the association of PRSs and other predictors with outcomes, including treatment resistance and healthcare utilization.
Researchers hope that this study will substantially advance psychiatric disorders research using the vast data located in electronic health records and encourage the application of genetics and epidemiology in patient care.
- Highlights