When would you use a network meta-analysis?
where you have >2 drugs to treat a condition (e.g. multiple drugs for PPH) you can use network meta-analysis to rank studies in order of effectiveness.
What is a cohort study?
type of observational studies in which a cohort, or a group of individuals sharing some characteristic, are followed up over time, and outcomes are measured at one or more time points.
What is intention to treat analysis?
Where pts analysed in groups to which they were randomised. E.g. analysing pt to group receiving nifedipine even if they ended up receiving atosiban
How do you calculate risk ratio or relative risk?
How do you know which figure goes on bottom of division?
Divide one risk by another.
Placebo usually goes at bottom. If comparing two treatments, put established treatment at bottom of division and new treatment at top
How do you calculate absolute risk difference?
Difference between risk of two treatments. Treatment group risk – control group risk
How do you calculate NNT?
What do you do with your result if you don’t get a whole number?
1/absolute risk difference
Round it up (even 6.2 gets rounded up to 7)
Equation for sensitivity
TP/TP+FN
Equation for specificity
TN/TN+FP
Equation for positive predictive value
TP/TP+FP
Equation for negative predictive value
TN/TN +FN
Equation for Likelihood ratio +
sensitivity/(1-specificity)
Equation for Likelihood ratio -
(1-sensitivity)/specificity
What does it mean if heterogeneity is present in a study?
Where different studies looking at similar thing give you different answers.
What do the whiskers represent in forest plot?
95% CI
When diamond shape used to summarise forest plt, what do corners of diamond represent on diamond?
95% CI
Equation for standard error of the mean (SEM)
SEM = SD of sample/square root of n
(n = population)
What type of statistical test should be used for data that is normally distributed?
Parametric or non-parametric?
Parametric
(non-parametric = non-normal)
When does Odds Ratio approximate Risk Ratio?
When disease being studied is very RRRare
