Abstract
Evaluations of medical information systems (MIS) often compare one or more dependent variables between experimental patients having the MIS available and control patients. Typically, physicians who are exposed to and therefore are potentially influenced by the MIS during visits with experimental patients also see control patients, thus possibly weakening the internal validity of subsequent comparisons between the MIS and the control conditions. The present study examines the possible bias due to carry-over of MIS influence to control patients in the context of a prospective, randomized design.
Get full access to this article
View all access options for this article.
