Our mission is to facilitate investigations drawing on large clinical databases, most notably large-scale electronic health records (EHR).
- Aim is to increase ability of clinical/translational investigators to generate and use emerging informatics technology for research by…
–Teaching about existing tools
- Courses in use of electronic health records for research
–Building new tools
- CQH Mart, Collaboration Acceleration Tool, …
–Providing consultations to investigators
- “How do I recruit patients admitted for MI for an intervention study?”
- “How do I capture adherence in outpatient pediatric clinics?”
- “How do I examine antibiotic exposure in a postoperative cohort?”
–Helping to generate pilot data and text to power grants applying these methods
- Referrals to/from: eCare, biostats cores, LCS, Center for Connected Health, CCDS, …
- “But don’t we already spend a lot of $$ supporting groups that do this kind of work?”
- YES – and they don’t consistently talk to each other, or to investigators.
Director: Roy Perlis, MD MSc
Research Director: Thomas McCoy, MD
- “From query to analysis in an hour or your money back.”
- 1.1 million MGH and BWH hospital admissions, from 2005-present
- Includes all standard RPDR data, plus…
–Outcomes including time-to-readmission, time-to-death, time to ER visit
–Curated medication exposure variables:
- “Find me everyone who received at least 60 days of antibiotic exposure after discharge.”
–New phenotypic variables derived from machine learning
–Links to Partners biobank and MGH NeuroBank
- Generates files ready for analysis – no pivoting, no data cleaning, no variable generation required
- Extends i2b2 platform – off the shelf, easy to maintain.
- First class 10/23 – course filled in first day
CAT: Collaboration acceleration tool
- Matchmaking tool for clinical/translational investigators
- “For the patients in my research cohort, how many also have data available through other studies?”
–How many patients in my cohort already have functional MRI data available?
–How many patients in my cohort already have iPS cell lines generated?
–How many participants in the biobank also participate in the MGH obstetric longitudinal cohort?
–Studies must include standard IRB-approved language allowing sharing
–Investigators decide who to collaborate with, and how
–No clinical/research data stored in the tool itself, only encrypted/hashed identifiers
- Helps transition investigators toward tissue banks and data repositories – no more ‘one-and-done’ studies.
- Consultation with investigators to plan and refine investigations making use of large clinical datasets;
- Assistance in preparation of materials describing investigations of large clinical data sets for submission to IRB or other regulatory review;
- Assistance in generating descriptions of methods for incorporation in funding applications;
- Assistance in generating pilot data to motivate funding applications, and for incorporation in publications;
- Development of tools to allow researchers to address questions which would otherwise be difficult to study in large data sets;
- Assistance in accessing appropriate large data sets and integration of multiple data sets – for example, use of birth and death certificate data;
- Assistance in development of decision support tools which make use of results derived from large clinical data sets.