Reducing Errors in the Diagnosis of Melanoma using an Intelligent Tutoring System

点击翻译按钮获取中文摘要

DESCRIPTION (PROVIDED BY APPLICANT): Our objective is to improve the diagnostic accuracy of community generalist pathologists for malignant melanoma through the use of a cognitive simulation system. Medical error research and clinical error reduction initiatives to date have primarily focused on therapeutic (e.g. medication) errors rather than on diagnostic errors. The diagnostic error rate in general surgical pathology is generally accepted to be approximately 1 %, although there is evidence that these errors are most likely underreported. Cancer of the skin is the most common of all cancers, and melanoma accounts for about 4% of skin cancer cases but causes a large majority of skin cancer deaths. The American Cancer Society estimates that the number of new melanomas diagnosed in the United States is increasing, and about 7,910 people are expected to die of melanomas during 2006. Since 1973, the mortality rate for melanoma has increased by 50%. False negative diagnostic errors for malignant melanoma are the most frequent diagnostic errors for which patients pursue litigation against pathologists and previous studies have estimated that approximately 10% of all pigmented lesions examined are associated with a clinically significant false negative error related to melanoma resulting in significant harm to patients. Previous studies have also shown that generalist pathologists in community private practices (the majority of pathologists examining pigmented lesions) tend to make more errors than pathologists who have received specialty dermatopathology training. This project represents a logical collaboration between two previously independent, successful, and productive investigators. The first is the proposed project PI; a significant portion of her work has been on a project showing how multi institutional voluntary sharing and root cause analysis of anatomic pathology diagnostic errors (including errors related to melanoma) may lead to successful process changes that reduce errors. The second is one of the proposed co-investigators (Crowley), whose major focus has been on the development, deployment, and evaluation of a highly novel cognitive simulation tool, an intelligent tutoring system, which has been shown to effectively improve diagnostic performance for melanomas and other melanocytic lesions. In order to achieve our objective, we aim to test the effectiveness of the intelligent tutoring system as a performance improvement and error reduction intervention in a prospective case control study, with volunteer community pathologists as study subjects. Pre- and post-interventional performance and error rates will be longitudinally measured and tracked. The result of this project will be a model diagnostic error reduction system for melanoma using cognitive simulation that we plan to use locally for proficiency testing and training and to enhance for use at national educational conferences and for national networking.