Journal of clinical computing最新文献

The results of a prospective controlled study designed to evaluate the performance of a computerized diagnostic aid in the clinical setting are reported herein. Two hundred and fifty patients referred to an outpatient clinic for gastrointestinal complaints were entered into the study. The diagnostic aid program was used routinely and the findings elicited from patients' medical histories and physical examinations were entered into the computer at the time of initial evaluation. The accuracy of the program's diagnoses was measured and compared with the accuracy of diagnoses made by a control group of physicians. The overall diagnostic accuracy of the program was found to be 69.2% (P much less than .0000001), a value significantly greater (P = .0005) than the overall accuracy of the control group of physicians, which was found to be 52%. A high agreement rate between program and physicians was also observed. These preliminary results seem to indicate a practical use for the program as a support in the initial screening of patients presenting with gastrointestinal complaints.

Computerized information systems that support medical decision-making are based upon formalized medical knowledge. Therefore, appropriate methods of knowledge engineering are of conspicuous importance. In order to arrive at a medical-history expert system, a target-oriented construction method of knowledge engineering was devised. Beginning with a framework of diagnostic categories, an optimal set of reliable and valid questions were formulated. In this paper, two independently developed general medicine questionnaires prepared by such a method are comparatively presented. It is shown that (a) inspite of a good deal of similarity between the two frames of reference considerable differences exist between the two sets of selected questions, and (b) the valid collection of medical information is limited to diagnostic categories that allow a clear description of medical symptoms.

Automated fact extraction from discharge summaries of the hospital chart yielded a useful set of facts for characterizing the clinical history and care of a hospitalized patient. The extracted facts depict the clinical events of a single patient, whereas the aggregate of the facts reflects the hospital's pattern of health care. Automated analysis of the discharge summaries provides good data for monitoring quality of care: Appropriateness and timeliness of hospitalization, effectiveness of diagnostic hypothesis generation, judicious use of resources, iatrogenic problems (surgical complications, other therapeutic mishaps), and inconsistencies (such as an abnormal finding without follow-up) are some of the issues recognizable when studying the care rendered. Fiscal studies are facilitated by the 50 to 100 facts on each patient considered pertinent and significant by the physician. These facts on each patient provide a much better definition of the illness and the resources used than the currently used discharge abstract data for DRG grouping. Medical audit: Ongoing clinical self-audit and cost study should stimulate a hospital's medical staff to constantly improve the quality of patient care management, and to be aware of the cost of the various resources. Alternative treatment patterns can be compared in the institutional audit reports, based on outcome, patient satisfaction, and/or cost. Clinical medicine could use a multi-institutional fact data-base as a retrievable source of yesterday's clinical experience, to guide clinical decisions of today.