Problem focused knowledge navigation: implementing the problem focused medical record and the O-HEAP note.

The current organization of most Computerized Medical Records (CMR) is based on the Problem Oriented Medical Record (POMR) and the SOAP (Subjective, Objective, Assessment and Plan) note. The organizational structure of the POMR and especially the SOAP note, does not allow for optimal use of computer capabilities in the follow up note. Since follow up visits are the most common office visit by far, this is a major flaw in the CMR. The authors propose a Problem Focused Medical Record and the OHEAP (Orientation, History, Exam, Assessment and Plan) note to resolve this problem. OHEAP starts with a powerful orientation structure that brings forward the timeline, last Assessment and Plan, and Plan Results for each problem along with the patient's historical tables as the starting point of every follow up visit. The Assessment and Plan portion brings problem specific differential diagnoses and their workups along with other relevant tables such as expert systems, treatments, instructions, medical literature or pathways. This leads to Problem Focused Knowledge Navigation that brings powerful efficiencies to the CMR. By recognizing the true workflow in the longitudinal diagnosis and management of any medical problem, the efficiency of the CMR is maximized. OHEAP allows for optimal use of both personal and external data elements in the medical record. Its powerful orientation attributes minimize the time spent in analyzing the current status of the problem while its connections to problem specific databases help resolve the problem.

Education review: applied medical informatics--informatics in medical education.

The importance of informatics training within a health sciences program is well recognized and is being implemented on an increasing scale. At Chicago Medical School (CMS), the Informatics program incorporates information technology at every stage of medical education. First-year students are offered an elective in computer topics that concentrate on basic computer literacy. Second-year students learn information management such as entry and information retrieval skills. For example, during the Introduction to Clinical Medicine course, the student is exposed to the Intelligent Medical Record-Entry (IMR-E), allowing the student to enter and organize information gathered from patient encounters. In the third year, students in the Internal Medicine rotation at Norwalk Hospital use Macintosh power books to enter and manage their patients. Patient data gathered by the student are stored in a local server in Norwalk Hospital. In the final year, we teach students the role of informatics in clinical decision making. The present senior class at CMS has been exposed to the power of medical informatics tools for several years. The use of these informatics tools at the point of care is stressed.

Intelligent Medical Record--entry (IMR-E).

This paper describes an automated medical record designed to allow providers to enter patient data at the point of care. The system runs on PCs and Macintoshes and uses a graphical user interface and object-oriented programming to take advantage of current mouse and pen technologies. The provider acquires all relevant patient data by pointing and clicking at selections on input screens, many of which contain anatomical drawings to help the provider quickly and accurately describe patient findings. The system also generates a grammatically correct progress note using the problem-oriented structure. Furthermore, items identified in the assessment and plans portion of the program can be ported to expert systems for medical decisions assistance or to billing systems. The system allows the provider to obtain the necessary information on a focused patient visit in less than 5 min or to enter a complete history and physical.

Informatics Workup.

We introduce the concept of a Medical Informatics Workup performed by fourth year medical students working in a busy inner-city Emergency Room. These students use portable computers (Macintosh PowerBook 170s connected to a removable cartridge hard drive and CD-ROM drive) to do the patient workups. The PowerBook 170 contains the automated medical record entry software (IMR-E), five expert system software packages, and a program that allows the PowerBook to emulate a PC-compatible computer. With this configuration the student has a portable system that allows for the creation of a computerized medical record at the patient's bedside, along with the ability to analyze the data and generate a list of differential diagnoses.

A test selection module for MEDAS.

A test selection module to recommend proper tests to the physician during the diagnostic process has been added to MEDAS (the Medical Emergency Decision Assistance System). Test selection is an important part of the diagnostic process. This module can use either diagnostic hypotheses set up by the physician or derive its own diagnostic hypotheses from the differential diagnosis. This module recommends pertinent questions to be asked about the patient's history and physical examination and recommends laboratory and radiological tests. The recommended tests are presented in two different ways: as individual tests and as groups of sequential tests.

A feature dictionary supporting a multi-domain medical knowledge base.

Because different terminology is used by physicians of different specialties in different locations to refer to the same feature (signs, symptoms, test results), it is essential that our knowledge development tools provide a means to access a common pool of terms. This paper discusses the design of an online medical dictionary that provides a solution to this problem for developers of multi-domain knowledge bases for MEDAS (Medical Emergency Decision Assistance System). Our Feature Dictionary supports phrase equivalents for features, feature interactions, feature classifications, and translations to the binary features generated by the expert during knowledge creation. It is also used in the conversion of a domain knowledge to the database used by the MEDAS inference diagnostic sessions. The Feature Dictionary also provides capabilities for complex queries across multiple domains using the supported relations. The Feature Dictionary supports three methods for feature representation: (1) for binary features, (2) for continuous valued features, and (3) for derived features.