The optimal management of neonates with jaundice in an era of reduced post partum length of stay requires the availability of clinical information across sites of care as well as adherence to guidelines for diagnosis and management. This project will establish a computer based information system that will link key clinical data across institutions and provide computer based decision support (CDSS) to enhance adherence to guidelines for the management of hyperbilirubinemia.
The specific aims of the project are to demonstrate the feasibility of linking clinical information using an architecture that permits the sharing of patient data from heterogeneous electronic medical record systems while maintaining a high level of security; and of providing CDSS to clinicians seeing infants with jaundice. The project will determine whether the use of this program results in enhanced adherence to the guideline, changes in resource use, and enhanced knowledge concerning management. We will implement this program at two major obstetric hospitals in Boston, Brigham and Women's Hospital and Beth Israel Hospital, which together provide 60% of all newborn deliveries in Boston; The birth data will be linked to primary care and emergency department sites at Children's Hospital, Boston, as well as to a private community based practice. The impact of this program will be assessed with a quasi- experimental design, using before and after measurements at both intervention and matched non-intervention sites. Data will be derived from medical review criteria based chart review, as well as surveys of clinicians at study sites. The guideline used will be the American Academy of Pediatrics guideline for the management of hyperbilirubinemia, with local adaptation through a formal guideline process at Children's Hospital, Boston. The project addresses an important clinical entity, neonatal jaundice; it will demonstrate a broadly applicable mechanism to link data across sites of care in order to enhance communication and decrease duplication; and will give insight into the effectiveness of computer based decision support as a mechanism to enhance the implementation of practice guidelines in order to reduce variability in care, improve quality, and reduce cost.
| Sun, Y; van Wingerde, F J; Kohane, I S et al. (1999) The challenges of automating a real-time clinical practice guideline. Clin Perform Qual Health Care 7:28-35 |
| van Wingerde, F J; Sun, Y; Harary, O et al. (1998) Linking multiple heterogeneous data sources to practice guidelines. Proc AMIA Symp :391-5 |
