Study Design Flashcards
Understand the principles and key features of common study designs: cross‐sectional, case control, cohort and RCT. Know how a research question and study project are developed
What are case series?
Individual based investigations, lot of individual positive outcomes. Hence, no comparison group. Selective sampling and limited information. E.g. Cases of birth defects with rubella around the world.
What are correlational (ecological) studies?
Start with a population defined geographically or temporally and investigate them on exposure and outcome. E.g. John Snow cholera outbreak in London. Still no comparison group. Large scale, cheap and easy since can extract data from existing databases. Many confounding factors.
What are cross sectional studies?
Obtain exposure and outcome data at the same time. Takes place at a single point in time and so only offer a snapshot. Difficulty in recalling past events may also contribute bias. But can calculate the prevalence. We don’t know cause or effect.
What are case control studies?
Start with the outcomes and move backwards to see who was exposed or not. E.g. The cases would be persons with lung cancer, the controls would be persons without lung cancer. Then you look at both groups and find those who smoke (exposed) and those who aren’t and find the correlation. Good for investigating cause and making hypotheses and determining risk. Faster and cheaper than cohort studies. Cannot calculate prevalence.
What is a cohort study?
Exposure first then outcomes. Such a study would recruit a group of smokers and a group of non-smokers (the unexposed group) and follow them for a set period of time and note differences in the incidence of lung cancer between the groups at the end of this time. Not good for long induction times. Can investigate the nature of the disease. Requires large populations, expensive and time consuming.
What is a retrospective cohort study?
Start with exposed and look at the outcomes.
What is a randomized control trial?
Start with a population and randomly expose some and not others and observe the outcomes. Best for evaluating treatment, ethical considerations, expensive and time consuming.
