EPIDEMIOLOGYÐICS Flashcards
Describe a cohort study.
Examines a group of subjects exposed to a certain risk factor/situation. Can be prospective (exposed and then followed over time) or retrospective (examines exposed group in whom disease has already occurred.)
Conclusion of a cohort study
Relative risk
Describe a case-control study
Retrospectively compares a group with disease to healthy controls. Frequency of certain exposures in both groups considered.
Conclusion of a case-control study
Odds ratio
Disadvantages of a case-control study
Recall and selection bias
Describe a cross-sectional study
Estimates disease PREVALENCE following exposure
What the heck is a “length” bias?
Screening test detects SLOWLY progressive diseases and MISSES rapidly progressive diseases, thus overestimating the effectiveness of a screening test.
What is the publication bias?
The idea that studies that SHOW a difference are more likely to be published compared to studies that end up supporting the null hypothesis. The issue that comes in is meta-analyses will then only include these studies, and not include ALL studies, including the ones that supported the null hypothesis.
Self-selection bias
The idea that patients with a certain medical history (think: histrionic) are more likely to participate in a study. Thus, the study isn’t representative of the general population.
Relative risk equation
Disease in exposed/disease in unexposed.
RR=1
No relationship between exposure and disease
Odds ratio equation
Odds of exposure in diseased/odds of exposure in healthy.
Odds ratio is most accurate as an estimate of relative risk in cases of .. ?
Rare disease.
Attributable risk equation
Rate of disease in exposed population - rate of disease in unexposed population.
Positive likelihood ratio
Sensitivity/(1-specificity)