Ambulatory Care Compact to Organize Risk and Decisionmaking (ACCORD)

Failure to followup on abnormal tests results and failure of care providers to adhere to recommended guidelines are common quality concerns in primary care. For many physicians, there is no reliable method to ensure that tests are completed and results adequately tracked. Prior research supports the importance of patient-clinician collaboration to agree on treatment goals to manage chronic health problems, communicate about the provision of preventive services, and promote medication adherence. Current systems lack comprehensive, practical approaches to translating a model of collaborative, patient-centered care that engages patients, clinicians, and practices in the explicit creation and execution of shared clinical care plans. In response, this project designed Ambulatory Care Compact to Organize Risk and Decision-making (ACCORD), primary care delivery system to transform care within the Massachusetts General Primary Care Practice-Based Research Network (MGPC-PBRN) by involving patients and families in the design of tools to foster shared decisionmaking, invigorate patient-clinician partnerships, and share care plans visibly. The project aims were to:

• Design a model for patient-centered primary care that facilitates patient-clinician partnerships and results in documented followup care plans that can be tracked reliably to reduce the risk of care plans being lost to followup in busy primary care networks. 
• Develop a health information technology architecture and software (ACCORD) to support the patient-centered care delivery model designed in the first aim. 
• Implement and evaluate ACCORD in a randomized controlled study within the MGPC-PBRN.

Two series of patient-provider focus groups were conducted as part of the iterative design effort. Implementation of the ACCORD systems included the definition of ACCORD temporal concepts; determination of representative use cases and business rules; definition of the ACCORD template and population of the ACCORD template library; development of the ACCORD authoring tool and definition of authoring guidelines; implementation of the ACCORD event detection engine and scheduler; and integration with provider and patient systems. A randomized controlled trial to evaluate the ACCORD system in the largest primary care practice at Massachusetts General Hospital was planned at the end of the project. 

The preliminary results from the focus groups showed the need for a model that would support a high variability in patient-provider collaborative decisionmaking styles, rather than imposing an ideal concept of shared decisionmaking. A comprehensive software system for authoring, proposing, and accepting the ACCORD system was implemented based on a usability and model evaluation.