Reviewing the Evidence Flashcards
Outline the basis of a systematic review
Literature is searched to find primary studies that used explicit and reproducible methods, avoiding bias.
Outline the basis of a meta-analysis
A quantitative synthesis of the results of two or more primary studies that addressed the same hypothesis in the same way.
What is a ‘pooled estimate’?
The odds ratio and 95% CIs are calculated for all studies in a meta-analysis. They are then given a weighting (depending on the size of the study) and combined to give a pooled estimate.
On a forest plot, what do the small squares represent?
Individual odds ratios
On a forest plot, what do the horizontal lines represent?
The individual confidence intervals
On a forest plot, what does the size of the small squares represent?
The weighting given to the study
On a forest plot, what does the dotted, vertical line represent?
The pooled odds ratio
On a forest plot, what does the diamond represent?
The pooled estimate
Give 2 problems with meta-analysis
Publication bias may be present
The studies may be of varying qualities
Heterogeneity between studies
What is ‘heterogeneity’?
Ideally all studies in a meta-analysis should be similar in terms of study design, treatments or exposures, outcomes measured, participant profile and the statistical analysis used.
Describe the ‘fixed effect model’
Assumes that the studies in a meta-analysis are estimating exactly the same effect size
Describe the ‘random effects model’
Assumes that the studies in a meta-analysis are estimating similar, not the same, effect size
List the two ways sub-groups can be analysed
Stratification by study characteristics
Stratification by participant profile
What type of study is most prone to bias and confounding factors?
Case-control study
What is publication bias?
Studies with statistically significant or favourable results are more likely to be published than those studies with non-statistically significant or unfavourable results.
