DOCTORAL SEMINAR IN OBSERVATIONAL INFERENCE
Course Director: Timothy Church
Affiliate Faculty: George Maldonado, Bruce Alexander
PubH 8141
Fall Semester 2003
Friday 2:30-4:00
Mayo 1155
Course Credits: 2
Grading: S/N
Prerequisites: consent of instructor. Ideal preparation includes Epi I, II, III, (quarters) or Epi I and II (semesters) and Validity Concepts in Epidemiologic Research, or equivalent.
This seminar course in observational inference is designed for doctoral students in Public Health who are interested in practicing the fundamentals of epidemiologic inference, including methods for designing, analyzing and interpreting epidemiologic studies. Class time will be spent critically discussing methods papers, using tools such as bias assessment, sensitivity analysis, and meta-analysis to better understand applied papers, and designing studies or parts of studies related to various areas, including environmental and occupational health. Students will be required to prepare homework assignments, participate in the discussions, and to lead at least one discussion. Assignments will include reading, preparing examples, and leading discussions on a topic. Grade is dependent upon participation in all aspects of the seminar. This course will be offered every semester.
The topic for this semester will be:
“Screening, Early Detection, and Symptomatic Care:
Pervasive Sources of Ascertainment Bias in Epidemiologic Studies”
Researchers have long recognized threats to validity from lead-time and length bias in observational studies of mass screening for chronic disease. More recently attempts to quantify these potential biases have appeared in the literature, one pointing to the effects of differential timing of diagnosis. Two recently published commentaries by Weiss and by Joffe have pointed out that screening can bias studies of risk factors associated with screening behavior. The problem is deeper yet, in that any factor associated with the timing of disease ascertainment will be confounded. For example, people more likely to seek immediate rather than delayed care for colorectal cancer may also be less likely to smoke or eat meat and more likely to eat fiber and vegetables and to exercise regularly. The artifactual effect of this earlier diagnosis would then be transferred to eating vegetables and exercise. In this course, students will study the mechanisms by which screening-related lead-time, length-bias, and self-selection can affect the validity of studies of chronic disease diagnosis or mortality, how such bias can be extended to associated risk factors, and how other forms of early diagnosis can produce similar bias. Students will also review how well the bias can be ascertained in observational studies, and develop models of such bias, including probabilistic sensitivity analysis, to understand how different assumptions affect the results.
Reading material: Church TR. A novel form of ascertainment bias in case-control studies of cancer screening. Journal of Clinical Epidemiology 1999;52:837-47.
Weiss NS. Adjusting for Screening History in Epidemiologic Studies of Cancer: Why, When, and How to Do It. Am. J. Epidemiol. 2003;157:957-61.
Joffe MM. Invited Commentary: Screening as a Nuisance Variable in Cancer Epidemiology: Methodological Considerations. Am. J. Epidemiol. 2003;157:962-4.
Recommended background text: Modern Epidemiology, 2nd ed., by K. Rothman and S. Greenland. Little, Brown (Boston) 1998.
For more information, please contact Dr. Church (612-626-1494, trc@cccs.umn.edu, Dr. Maldonado (612-626-2104, gmphd@umn.edu, or Dr. Alexander (612-625-7934, balex@umn.edu.
Scholastic dishonesty is a violation of the student conduct code and is defined as "any act that violates the rights of another student in academic work or that involves misrepresentation of your own work. Scholastic dishonesty includes (but is not limited to): cheating on assignments or examinations; plagiarizing, which means misrepresenting as your own work any part of work done by another; submitting the same paper, or substantially similar papers, to meet the requirements of more than one course without the approval and consent of all instructors involved; depriving another student of necessary course materials; or interfering with another student’s work." Scholastic dishonesty in any portion of the academic work for a course shall be grounds for awarding a grade of "F" or "N" for the entire course. Please consult the student conduct code.
Page last updated: