アブストラクト（34巻1号：The Bulletin of Kanagawa Dental College）

The Bulletin of Kanagawa Dental College

English

Title　： Good Studies of Prognosis
Subtitle　： BKDC CLINICAL AND RESEARCH TOPICS : Evidence-based Medicine and Dentistry
Authors　： Yoshiyuki Sasaki
Authors（kana）　：
Organization　： Center for Education and Research in Oral Health Care, Faculty of Dentistry, Tokyo Medical and Dental University
Journal　： The Bulletin of Kanagawa Dental College
Volume　： 34
Number　： 1
Page　： 41-45
Year/Month　： 2006 / 3
Article　： Report
Publisher　： Kanagawa Odontological Society
Abstract　： [Points to be appraised] 1. Is it considered that bias does not occur? 2. Is the considerable confounding factor measured and adjusted? 3. Does the result factor lead to a true outcome rather than a surrogate outcome? 4. In a process of causal inference, is a distinction being made as to whether the exposure factor is a factor or a marker? 5. Is the generalizability of an available result high? In an observational study, a cohort study is the most suitable study design to obtain evidence helpful in making prognoses. A cohort study designed adequately is performed using the following procedures. Exposure factors are measured after selecting subjects (cohort) from the population. The outcome factors are measured after a fixed time course. In other words, a cohort study is a longitudinal section study that prescribes a target population in the exposure of a factor, and exposure factors are measured before the occurrence of outcome factors. In a cohort study of a classic design, it is common to include a control group. However, a dental disease makes ends meet with a "web of causation" of multifactor. It is difficult to form groups using a combination of each factor in such a disease. Therefore a multivariate analysis is performed from data of plural exposure factors without making a control group.
Practice　： Dentistry
Keywords　：