Sites

Members of the IGs will implement the decision support intervention within the functioning electronic health record system at clinical sites in three phases. Selection of the phases and implementation sites has been performed to meet the following AHRQ expectations:

In the first phase, we will implement CDS for Asthma in one initial location . Yale Specialty clinic. Based on that, we will identify and address issues, then revise the plan for next phase. We will meet in person with the AHRQ Project Officer and key stakeholders to review progress and findings to date, and solicit feedback. Based on that feedback we will review, revise and seek PO approval of the plan for the second phase.

In the second phase, we expect to implement CDS for Obesity at Yale Primary Care and Delaware Primary Care, and for Asthma at Nemour.s Jacksonville, Orlando and Pensacola facilities. Following this implementation, we will also identify and address risks, and review status with the Project Officer and key stakeholders before finalizing the plan for the final phase of implementation.

In the third and final phase, we expect to implement CDS for Asthma at Yale Primary Care and Delaware Primary care.

This implementation locations and phasing are summarized in the table below:

Phase	Condition	Site	EHR System
1	Asthma	Yale Specialty	GE Centricity
2	Obesity	Yale Primary Care	GE Centricity
		Nemours Delaware PC	EpicCare
	Asthma	Nemours Orlando	EpicCare
		Nemours Jacksonville	EpicCare
		Nemours Pensacola	EpicCare
3	Asthma	Yale Primary Care	GE Centricity
		Nemours Delaware PC	EpicCare

Asthma interventions will be developed for both specialty and primary care clinicians, while obesity prevention will focus on primary care. Beginning at the Orlando and Yale Specialty Clinic sites with asthma interventions will offer the most controlled environments for testing the interventions. These sites also have a record of innovation and a high likelihood of initial success. Asthma (in Florida) and obesity interventions in New Haven and Delaware Valley will follow. We anticipate that the primary care clinics will require different asthma interventions from those planned at the specialty sites. In Phase 3, we will introduce asthma interventions in the primary care sites.

We believe this implementation plan will enable us to operationalize a replicable process for implementation at a wide variety of implementation sites that should demonstrate the external validity of the project.s findings. Relevant characteristics of the sites are:

·         Yale Primary Care Center is an academic, inner city, ambulatory care center that serves a low-income, multi-ethnic, Medicaid and uninsured population with generally low levels of healthcare literacy. Clinicians in training there (residents and nurse practitioners) will take skills in interaction with clinical decision support tools to geographically dispersed primary care and specialty practices when they finish their training.

·         The Pediatric Specialty Center at Yale Children.s Hospital sees children in referral from a wide range of socioeconomic segments. It is manned by academic pediatric subspecialists, postdoctoral fellows, and advanced practice RNs.

·         The Nemours multi-specialty centers in Orlando, Jacksonville, and Pensacola each has a unique culture and flavor. Community-based sub-specialists provide care to a wide spectrum of patients including both those with private insurance and Medicaid coverage.

·         The 41 pediatricians and 11 APRNs who practice in the 14 Delaware Valley Nemours-affiliated primary care practices cover a broad geographic area and their patients span a wide demographic range.

We believe that a clinician.s employer reflects on resource availability for acquisition of EHR systems. But the issues of effective implementation of decision support and adherence to guideline recommendations cross these lines. Incentive programs that require effective practice improvement (such as pay for performance) affect academic and multi-specialty groups as well as small privately owned practices. Therefore we anticipate applicability of our findings to a broad range of American healthcare providers.