The purpose of this study is to improve patient care and safety while decreasing ED visit rates by sending specific information about care transitions related to hospital admission and discharge and emergency department and specialty care visits to primary care practices, care managers and patients with the use of health information technology (HIT) shared across a community-based network of providers.

Cycle 1 focuses on the impact of notices about ED encounters and hospitalizations derived from billing data that are sent to care managers for all 47,000 patients in the Northern Piedmont Community Care Network (NPCCN). Cycle 2 explores the impact of letters sent to patients, and care event reports sent to a patient's medical home in addition to notices sent to care managers about ED encounters, hospitalization and specialty care based on ADT (Admission Discharge Transfer) and billing data on 4,600 patients with complex health needs.

This three-year project seeks to improve outcomes, quality and coordination of care for patients with complex healthcare needs by facilitating the availability of information following three types of care transitions into the ambulatory care setting. Specific information regarding care transitions will be made available to patients, primary care practitioners and care managers following hospitalizations, emergency department (ED) encounters, and specialty clinic evaluations.

This project will build upon a regional Health Information Exchange (HIE) network created to connect providers serving 47,000 Medicaid beneficiaries across traditional institutional boundaries from both rural and urban settings in a 6-county region in the Northern Piedmont of North Carolina. This network includes 25 ambulatory care practices, 3 federally qualified health centers, 4 rural health clinics, 3 urgent care facilities, 11 government agencies, 5 hospitals and 2 cross-disciplinary care management teams. Within this HIE network, 4,600 patients with complex healthcare needs have been identified.

For this project, a standards-based clinical decision support tool will be utilized in order to ensure that the proposed approach is generalized, portable, and scalable; and routinely available claims and scheduling data will be used as the primary data source. This approach will support both traditional clinic-based models of care as well as new care models including population health management and the use of cross-disciplinary teams.

Under Aim 1, the existing HIE network and decision support tool will be enhanced to enable detection of transitions in care and delivery of timely, patient-specific information regarding these care transitions to patients, primary care clinicians and multidisciplinary care management team members. Under Aim 2, the impact of the proposed approach will be evaluated in a two-cycle randomized controlled trial primarily involving approximately 47000 Medicaid beneficiaries with a special focus on 4600 patients with complex health needs, 309 primary care clinicians, and 31 care management workers. Cycle 1 will assess only daily notices sent to care managers and will use only billing data. Cycle 2 will evaluate all components of the proposed intervention and us both billing and ADT data (see below). For Cycle 1, patients will be randomly assigned by family unit to either receive or not receive email notices sent to their care managers. For Cycle 2 patients will be randomly assigned to one of three groups: 1) information on care transitions sent to patients and their clinic-based caregivers; 2) information sent to patients, their clinic-based caregivers and their care managers; and 3) no information sent. The primary outcome measure will be the overall rate of ED utilization for each study group. Under Aim 3, the economic attractiveness of the proposed approach will be determined. Under Aim 4, the technology and results of this study will be disseminated through public media, publications and presentations. Information-augmented care transitions between sites should result in improved care coordination, higher quality of care, and more appropriate care.

This trial will be deployed in two cycles in order to support the needs of the care management network while the full intervention is developed. Cycle 1 will run from December, 2009 through December, 2010. It will assess the impact of notices about hospital admissions and ED encounters derived from billing data and sent daily to care managers for the 47,000 patients enrolled in NPCCN on the study outcomes. Cycle 2 will run from December, 2010 through December, 2011 and will address AIM 2 of the original grant proposal. For Cycle 2, events detected from ADT and billing data will be generated daily. The events will include hospital admissions, hospital discharges, ED encounters, and specialty care visits. The responses to events will include event summary reports sent to patients' assigned medical homes, letters sent to patients or their guardians, and release of information requests on behalf of a patient's medical home. The response will be generated for 4,600 patients identified as having complex health needs. In addition, notices will be sent to care managers for detected hospital and ED events for all 47,000 patients enrolled in NPCCN. Special priority will be given to patients with complex heath needs.

Inclusion Criteria:

• North Carolina Medicaid beneficiary enrolled in the Northern Piedmont Community Care Network (NPCCN)
• Has complex healthcare needs as defined by having two or more IOM (Institute of Medicine) priority conditions (hypertension, coronary artery disease, congestive heart failure, stroke, asthma, diabetes) OR one of the following: moderate to severe mental health diagnosis (schizophrenic disorder, episodic mood disorder, delusional disorder, non-organic psychosis, anxiety, dissociative-somatoform disorder, personality disorder), end-stage renal disease, sickle cell disease
• Continuous enrollment in NPCCN for 10 of the previous 12 months