Smart Forms face twin integration challenges: to consolidate rules engine
output and data from multiple sources into a single view, and to be
perceived by users as a seamless extension of the LMR. The architectural
and technical approaches to meet these challenges are presented.
Service Oriented Architecture
Smart Form data requirements are satisfied via a series of Web services
made available by various sources of clinical information at Partners. Data
from services are consumed to dynamically generate individual Form
elements. Data generated by Form users flows back into repositories
also by way of services. Using existing services allows for high degree
of reusability, faster development and seamless integration.
Data Access Layer
The data access layer normalizes various data types into a common format. When
considering encapsulation of data from different sources, such
an intermediate layer is necessary to capture data, convert them to a
standard representation, and aggregate for ease of use.
Pertinent Patient Data
Access to disparate data is required to manage a particular condition. Vital
signs, lab results, medication lists, etc. are presented. Users
are able to review as well as enter relevant data.
A rules engine (an example of artificial intelligence technology) allows
for externalization of knowledge resources and provides the user with
a set of evidence-based, context-specific recommendations that are generated
at run time and can be acted upon.
Interactive Form components allow for flexible capture of coded data. Smart
Form output can then be integrated into a structured or unstructured
As a standard Web application, the Smart Form code is easy to maintain
and reuse. All aspects of the Smart Form are dynamically driven by Extensible
Markup Language (XML) and the Extensible Stylesheet Language Transformation (XSLT) technologies.