HSCI 330 Lecture Notes - Lecture 6: Gastrointestinal Bleeding, Combined Oral Contraceptive Pill, Recall Bias
12 Mar 2017
School
Department
Course
Professor
Document Summary
Incorrect diagnosis: subject self-report, records incorrectly coded in data base. Misclassification table: column: true disease/exposure status, row: classified disease/exposure. 18 people in total. a+c= total true disease, b+d= total true non-disease. Can result from a poorly constructed questionnaire or survey process that doesn"t ask the right question. Can result from a faulty measuring device or observation technique: self-report by subjects. Typically, a problem in case-control studies: cases may be more motivated to recall past exposures than controls; also, occur in cohort studies. Reluctance of subjects to be truthful in reporting exposure relating to behavior considered socially unacceptable (food intake, sexual behavior illegal drug-use) Interviewer bias: an interviewer may probe more thoroughly about exposure for cases than for controls: data coded incorrectly in a database. Misclassification can sometimes occur for both exposure and disease in the same study: both exposure and illness information were obtained by subjects" recall (sydney beach)
