A key component of a learning health care system is the ability to collect and analyze routinely collected clinical data in order to quickly generate new clinical evidence, and to monitor the quality of the care provided. To do this, clinical data must be easy to extract and stored in computer readable formats. CERTAIN’s Automation Project intended to assess the availability of such data specifically for comparative effectiveness research (CER) and quality improvement (QI) on surgical procedures.
The study was conducted over three years across multiple hospitals which participate in the Surgical Care Outcomes and Assessment Program (SCOAP). Hospitals participate in SCOAP by submitting information regarding patients undergoing surgical and interventional procedures at their site. In return, hospitals receive reports benchmarking their performance in care delivery and outcomes. Through this mechanism, SCOAP has helped hospitals significantly reduce surgical and interventional care complications and re-operations, as well as associated costs. However, SCOAP hospital data is extracted from medical records by hospital staff, which limits the ability to expand data gathering for broader quality improvement purposes.
In partnership with five SCOAP hospitals and Caradigm, CERTAIN’s Automation Project assessed how Caradigm’s Amalga clinical data repository software, combined with novel CERTAIN tools for text-mining, or Natural Language Processing (NLP) could enhance the ability to monitor and advance QI and, in the future, be utilized for CER. Specifically, the partnership with hospitals was to evaluate whether CERTAIN Amalga could be used to replicate some or all of the manual medical chart review process for SCOAP data collection, increase the timeliness of hospital quality initiatives, and promote opportunities for hospitals to participate in more SCOAP modules with less staff effort. More broadly, CERTAIN also assessed the feasibility of using existing electronic health record (EHR) data for secondary use in CER and as a model of a Learning Healthcare System via a clinical data repository pulling information across multiple institutions.
As the Automation Project comes to a close in 2014, CERTAIN will be sharing results and lessons-learned from this work. Please review the related Automation Project publications on the right for more information about the study and its conclusions.
The Automation Project is supported by an award received through the Agency for Healthcare Research and Quality.
Automation Project Publications: