Harvard Catalyst Awards Four SHRINE Pilot Grants
Advancing their research from funding received through the Harvard Catalyst Shared Information Research Network (SHRINE) Query Prizes awarded in June 2012, four translational researchers were recently awarded $50,000 each for their proposals to utilize the web-based query tool in their innovative studies. Health conditions include autism, Alzheimer’s, and precocious puberty, with a fourth focusing on patient “problem” data. The awardees were chosen based on the project’s potential to have a demonstrable impact on human health.
For researchers needing aggregate data on patients, SHRINE is an indispensable tool yielding data previously unattainable. Launched in 2009, the network currently holds data on six million patients seen at the five participating Harvard-affiliated hospitals, including Boston Children’s Hospital (BCH), Beth Israel Deaconess Medical Center (BIDMC), Brigham and Women’s Hospital (BWH), Dana-Farber Cancer Institute (DFCI), and Massachusetts General Hospital (MGH), yielding more than 10 billion medical facts. Additionally, a pilot network is underway, connecting seven medical and clinical research centers in California (University of California, San Francisco), Michigan, North Carolina, Ohio, Texas, and Washington State.
SHRINE pilot funding aids researchers who require a large sample size of patients — a cohort that can only be culled from multiple institutions. “We are pleased to support these talented researchers with funding to conduct research using SHRINE,” said Isaac “Zak” Kohane, MD, PhD, professor of medicine and pediatrics at BCH, and the principal investigator of SHRINE. “Each of these proposals exemplifies innovative ways in which the tool can be applied and we look forward to following their studies.” SHRINE’s foundational technology is called Informatics for Integrating Biology and the Bedside (i2b2). This NIH-funded informatics framework establishes a common language and standards for securely exchanging clinical information across institutions.
Awardee Lindsay Oberman, PhD at BIDMC turned to SHRINE to retrieve preliminary data for her work on autism. Hypothesizing that autism might serve as a ‘protector’ from developing Alzheimer’s Disease (AD) or related dementias later in a patient’s life, Oberman relies on information found in patients’ medical records – data that SHRINE will help her validate. In her early research, Oberman hypothesized that the cortexes of individuals identified as “hyperplastic” (as seen in ASD) would be protected from AD, which she and her co-investigator, Alvaro Pascual-Leone, MD, PhD at BIDMC, suggest is a disease related to having a hypoplastic cortex. The two investigators used SHRINE to explore the number of patients who hold comorbid diagnoses of ASD and AD. Preliminary results are consistent with their hypothesis that individuals with autism have lower frequencies of AD than those without.
“Recent data suggests that patients with ASD have an excessive amount of plasticity,” says Oberman. “Plasticity is essential to the establishment and maintenance of brain circuitry. However, too much plasticity may lead to instability of structural connections and compromising of functional systems necessary for cognition and behavior.” Working with evidence that suggests that plasticity may underlie age-related cognitive decline, Oberman’s project will entail evaluating the presence of behavioral and neuroimaging markers for dementia in patients over 65 with ASD as compared to both patients with schizophrenia and a healthy control group.
Autism is also at the center of a study proposed by Roy Perlis, MD, MSc at MGH, which examines the connection between in utero exposure to antidepressants and an increased risk for autism. Perlis and co-investigators Roscoe Brady, MD, PhD at BIDMC, Sek Won Kong, MD at BCH, and Jordan Smoller, MD, ScD at MGH will utilize patient electronic health record (EHR) data based on SHRINE and i2b2 queries at their respective institutions.
Lisa Topor MD, MMSc at BCH will utilize her funding to examine the rise in precocious puberty affecting both genders. SHRINE inquiries she conducted have revealed a fivefold increase in annual visits for precocious puberty in network hospitals from 2001-2009. Working with co-investigators Amy Fleischman, MD, MMSc at BCH, Natalie Shaw, MD MMSc at MGH, Topor will continue to explore factors associated with the rise of precocious puberty and risk factors associated with comorbidities.
Focusing on the importance of maintaining accurate patient problem lists, Adam Wright, PhD at BWH, will work with a team of researchers from five Harvard-affiliated hospitals to build algorithms that will detect diagnoses missing from patient’s medical records. While patient problem lists remain essential for ensuring care, research has shown that problem lists are often inaccurate, incomplete, and poorly maintained. Following an initial study at BWH, Wright and his co-investigators Michael Hassett, MD, MPH at DFCI, Kenneth Mandl, MD, MPH at BCH, Charles Safran, MD at BIDMC, and David Ting, MD at MGH will work with their respective hospitals to study problem list usage, focusing on list completeness, validation of their existing problem inference rules, and the improvement and expansion of their rules. “We’ve developed these algorithms using data from the BWH and the initial results have been very positive,” says Wright. But even with a hospital as big as BWH, we can’t always get all the data we need – SHRINE lets us identify other potential sources of data to expand our analysis and improve the accuracy of our algorithms.”
Five conditions — diabetes, acute lymphoblastic and chronic myelogenous leukemias (ALL and CML), sickle cell disease and myasthenia gravis – will be examined. “For many conditions such as diabetes or hypertension, we have enough patients to make accurate algorithms here at BWH, and our algorithms are very accurate for these diseases,” he says. “But for rarer conditions, such as sickle cell disease and myasthenia gravis, we’ve had a harder time getting enough data to build accurate algorithms. We don’t see a lot of sickle cell patients at BWH, for example, but DFCI and BCH have a specialized sickle cell program. By collaborating with researchers there – which SHRINE enables – we can get the data we need to fill in these gaps.”