ABSTRACT
OBJECTIVE: Clinical practice guidelines (CPGs) could have a more consistent and meaningful impact on clinician behavior if they were delivered as electronic algorithms that provide patient-specific advice during patient-physician encounters. We developed a computer-interpretable algorithm for U.S. and European users for the purpose of diagnosis and management of thyroid nodules that is based on the "AACE, AME, ETA Medical Guidelines for Clinical Practice for the Diagnosis and Management of Thyroid Nodules," a narrative, evidence-based CPG. METHODS: We initially employed the guideline-modeling language GuideLine Interchange Format, version 3, known as GLIF3, which emphasizes the organization of a care algorithm into a flowchart. The flowchart specified the sequence of tasks required to evaluate a patient with a thyroid nodule. PROforma, a second guideline-modeling language, was then employed to work with data that are not necessarily obtained in a rigid flowchart sequence. Tallis-a user-friendly web-based "enactment tool"- was then used as the "execution engine" (computer program). This tool records and displays tasks that are done and prompts users to perform the next indicated steps. The development process was iteratively performed by clinical experts and knowledge engineers. RESULTS: We developed an interactive web-based electronic algorithm that is based on a narrative CPG. This algorithm can be used in a variety of regions, countries, and resource-specific settings. CONCLUSION: Electronic guidelines provide patient-specific decision support that could standardize care and potentially improve the quality of care. The "demonstrator" electronic thyroid nodule guideline that we describe in this report is available at http://demos.deontics.com/trace-review-app (username: reviewer; password: tnodule1). The demonstrator must be more extensively "trialed" before it is recommended for routine use.
