Meta Analysis Flashcards
What is Meta analysis?
Process of using statistical methods to combine the results of different studies
- integrate the findings
- pool data
- identify overall trends
It is an optional part of a systematic review
Advantages of Meta analysis?
Increase in power
Improvement in precision
Answer questions not posed by individual studies
When should meta analysis only be undertaken?
Minimal differences in characteristics across studies
Same outcome measure
Data in each study is fully available
- otherwise potential ton mislead
Types of data in meta analysis?
Dichotomous data
- died or not
- born or not
- illness or not
Continuous data
- blood pressure
- weight
- amount of pain
What is OR?
ODDS RATIO
- measure of how strongly an event is associated with exposure
E.g. smoking and lung cancer
- 17 smokers / 100 have lung cancer, 83 / 100 do not
- 1 non smoker / 100 has lung cancer, 99 / 100 do not
17/83 = 0.205
1/99 = 0.01
Odds in exposed group / odds in non exposed group
0.205 / 0.01 = 20.5
Smokers 20x more likely according to this odds ratio
What is RR?
Risk ratio / relative risk
Determines risk of incidence of something in one group vs another group of data.
E.g. wound infection after incidental appendectomy
Had appendectomy
- yes, 7 infections and 124 non infection. 7/131 = 5.34%
- no, 1 infection and 78 non. 1/79 = 1.27%
RR = 5.34 - 1.27 = 4.2
4.2x the risk of infection compared to those who did not have surgery
If RR is <1 then shows there is a decrease in risk
What is RRR?
Relative risk reduction
- shows how much an intervention / treatment has reduced the risk of bad outcomes
1 - RR
What is ARD?
Risk of developing cancer in smokers = 0.48 e.g.
Risk non smokers = 0.2 e.g.
Risk difference = 0.28
- smoking adds this 0.28 risk or 28%
How calculate confidence interval?
CI = mean value plus/minus Z-value for CI
Multiplied by
S/root N
