Bias in Observational Studies
Recall the case-control study of the relationship between estrogen use and endometrial cancer. The researchers later decided socioeconomic status might be a confounder, so they decide to independently explore the relationship between socioeconomic status and endometrial cancer. This is their study design: Cases consisted of 250 women with endometrial cancer who were referred to Massachusetts General Hospital (MGH) for treatment. Patients were referred to MGH from all over Massachusetts. Control subjects were recruited among residents who lived within 1 mile of MGH. Investigators recruited controls by going from door-to-door in the community during 9:00 a.m. to 5:00 p.m. on weekdays. Socioeconomic status was determined by neighborhood of residence at the time the patients were diagnosed. In your initial post, address the following: What type of study design is this? Why might the researchers have been concerned socioeconomic status was a confounder in the first case-control study? What types of bias are present in this study? Describe each element of the study methodology that contributed. Are there any additional confounders the investigators should be concerned about? Why? Imagine you are talking to the investigators. Suggest how they could modify the study to reduce the bias and address confounders, if you identified any. Feel free to also suggest another type of study design to explore this question. Your initial post should: Be between 150 and 300 words. Be well developed by providing clear answers with evidence of critical thinking. Add greater depth to the discussion by introducing new ideas. Demonstrate a high level of scholarship by connecting your ideas to course materials and peer-reviewed journal articles reporting on original research, as needed.