17 Strategic Revision for Exams Flashcards
Cohort study. Method with example involving mobile phone?
A: Starts from healthy people. E.g. year one med students at Imperial. Split into groups- those that are exposed to risk factors and those that are not. E.g. mobile phone use and neurodegenerative disease.- low exposure and high exposure- those that use phones a lot and those that do not. Watch for 10,20,30 years for those who get neurodegenerative disease- new prevalence of disease.
Q: What is crucial for a cohort study?
A: patience
Q: Cohort study calculations. Incidence? Relative risk?
A: INCIDENCE: NEW CASES/TOTAL CASES
DIVIDE INCIDENCE OF EXPOSED BY UNEXPOSED: GIVES RELATIVE RISK
Q: What’s the opposite of a cohort study?
A: case control study
Q: Case control study. Method including example involving mobile phones?
A: Start with disease- split by disease or no disease. E.g. Cases- those who have neurodegenerative disease. Control- those who do not have neurodegenerative disease.
Ask those in both groups. E.g. neurodegenerative- did you use a mobile phone a lot in the last 10/20/30 years?
Look at odds of exposure in disease group compared to control groups- ODDS RATIO. If odds ratio is higher in those with neurodegenerative disease and have used mobile phone- can prove there is a correlation.
Q: How do you prove a correlation in a case-control study?
A: . If odds ratio is higher in those with neurodegenerative disease and have used mobile phone- can prove there is a correlation.
Q: What makes a case control study better than a cohort study? (2)
A: faster
less expensive
Q: What makes a case control study worse than a cohort study? (5)
A: : Cannot get incidence of disease.
Need to rely on records, mixed up samples or what people say- cannot know that risk factor preceded the disease.
Cannot be 100% certain of timescale of events-
cannot be 100% sure of causation. Can be sure of association.
Recall bias
Q: What should be used to confirm causation? Needs?
A: Bradford Hill criteria- to be sure of causation
Need to know the temporal sequence of events
Q: What is attributable risk in epidemiology?
A: attributable risk or excess risk is the difference in rate of a condition between an exposed population and an unexposed population.
Q: How is attributable risk calculated?
A: Incidence of risk in exposed subtract incidence of risk in unexposed.
Q: Where is attributable risk mostly calculated?
A: mostly calculated in cohort studies, where individuals are assembled on exposure status and followed over a period of time.
