Study Design Basics
What is an epidemiological study?
A scientific method to observe, record, and analyze health and disease patterns in groups.
Who are the “cases” and “controls” in this type of study?
Cases = people with the disease.
Controls = similar people without the disease.
How is a cohort study structured?
Follows a group of people (exposed and unexposed) over time to see who develops disease.
What makes an experimental study different from an observational one?
Observational = just watching. Experimental = assigning treatment or exposure.
What is “recall bias”?
When participants don’t remember past events accurately.
What’s the main difference between observational and experimental studies?
Observational: no researcher control over exposure. Experimental: researcher assigns exposure/intervention.
Why is recall bias a problem in case-control studies?
Because people with the disease may remember exposures differently (over- or under-report).
What is one advantage of cohort studies in measuring disease risk?
Can directly measure incidence and calculate relative risk (RR).
What kind of study is a randomized control trial (RCT)?
A randomized control trial (RCT) - experimental
What is 'selection bias'?
When the way participants are selected leads to an unrepresentative sample.
Why is it important to choose the right study design before research begins?
Choosing the right study design ensures valid, ethical, and cost-effective data collection that truly answers the research question.
Why are case control studies useful for rare diseases?
Rare diseases - since it starts with cases already having the condition.
What are two disadvantages or challenges of cohort designs?
Costly, long-term, risk of loss to follow up.
Why are RCTs considered the 'gold standard' for testing treatments?
Because randomization minimizes bias and confounding, providing strong evidence for causation.
Which biases can affect cohort studies?
Selection bias, attrition (loss to follow up), misclassification.
How do you decide which study type (cohort, case-control, cross-sectional, ecological or RCT) fits your question?
Depends on the question:
Cohort - for best for studying causes and incidents.
Case control - rare diseases
Cross-sectional - prevalence
Ecological - group level data
RCT - interventions
What measure of association is used in case- control studies?
Odds Ratio (OR)
Which measure is used to compare risk between exposed and unexposed groups?
Relative risk (RR) or Risk ratio.
Why might it not always be ethical to use and experimental study?
When it's unethical (eg., you can't assign people to smoke or non-smoke).
Name one limitation common in observational studies.
How to prove causation, prone to confounding.
Which study design can best show cause-and- effect and why?
Randomized Controlled Trial (RCT) because randomization reduces bias and best shows causation.
Give one example of a case control study question.
'Is lung cancer linked to past cigarette smoking?'
Give one example of a cohort study question.
'Do smokers have a higher risk of developing heart disease over 10 years?'
Give one example of when you would use observational instead of experimental design.
Studying the link between pollution and asthma- you can't assign exposure, so you observe it.
Name one limitation common in randomized trials.
Expensive, time-consuming, and limited generalizability.