Biostatics 4 Flashcards
What study designs are commonly used in statistics?
Cross-sectional study, cohort study & randomized controlled trial (RCT), and case-control study.
What does a cross-sectional study sample?
A subset of the study population.
What measure of occurrence is used in cross-sectional studies?
Prevalence.
What measure of association is used in cross-sectional studies?
Prevalence ratio.
What samples are used in cohort studies and RCTs?
Sample of exposed and sample of unexposed.
What measures of occurrence are used in cohort studies and RCTs?
Incidence proportion and incidence rate.
What measures of association are used in cohort studies and RCTs?
Risk ratio and rate ratio.
What samples are used in case-control studies?
Sample of cases and sample of non-cases.
What measure of occurrence is used in case-control studies?
Odds
What measure of association is used in case-control studies?
Odds ratio.
What do we measure in a study involving smoking status and lung cancer status?
Smoking status (yes E+/no E-) and lung cancer status (yes D+/no D-).
What would we measure in a cross-sectional study?
The prevalence of lung cancer in smokers and non-smokers.
What measure of association would we report in a cross-sectional study?
Prevalence ratio.
How would we interpret the prevalence ratio if it was 5.44?
Those who smoke were 5.44 times as likely to have lung cancer compared to non-smokers.
What would we measure in a cohort study?
The incidence of lung cancer in smokers and non-smokers.
What measure of association would we report in a cohort study using risk?
Risk ratio (relative risk).
How would we interpret the risk ratio if it was 5.44?
Those who smoke had 5.44 times the risk of developing lung cancer compared to non-smokers.
What is another measure of occurrence we could use in a cohort study?
Incidence rate.
What measure of association would we report using the incidence rate?
Incidence rate ratio.
How would we interpret the incidence rate ratio if it was 7.5?
Those who smoke had 7.5 times the rate of developing lung cancer compared to non-smokers.
What important factors should we account for in a cohort study?
Loss to follow-up and death.
When can you calculate odds instead of prevalence or incidence of disease?
When you cannot calculate prevalence or incidence of disease, you can calculate odds.
How do you select participants in a case-control study?
You select participants as cases (with disease) and controls (without disease) and retrospectively assess exposure status.
Why is the prevalence of cancer in a case-control study not meaningful?
Because the study design selects an equal number of cases and controls, making the prevalence artificially 50%.
Can odds ratios (OR) be calculated in cohort or cross-sectional studies?
Yes, but they should not be interpreted as risk.
How does the OR compare to the relative risk (RR)?
The OR always overestimates the RR.
When does the OR approximate the RR?
The OR approximates the RR only when the disease is very rare (case-control study).
Format of research protocol
Abstract
Introduction
Methods
Data management & analysis
Ethics
Resources
Logistics
References
Appendices