10. Reviewing the Evidence Flashcards
What makes a systematic review a credible source of evidence?
It is explicit (clearly focused question), transparent and reproducible.
What is a systematic review?
An overview of primary studies that uses explicit and reproducible methods.
What is 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 the purpose of meta-analysis?
To facilitate the synthesis of a large number of study results, to systematically collate study results to reduce problems of interpretation due to variations in sampling. To quantify effect sizes and their uncertainty as a pooled estimate.
What is the quality criteria for meta-analyses?
A formal protocal specifying: compilation of complete set of studies, identification of common variable or category definition, standardised data extraction, analysis allowing for sources of variation.
What is used to give a pooled estimate odds ratio in meta-analyses?
The individual odds ratios and 95% CI for all studies.
How can a forest plot be interpreted?
The individual odds ratios are squares, with their 95% CI shown as lines. The size of the square is proportional to the weight given to that study. The diamond is the pooled estimate and the dotted line shows the pooled 95% CI. The solid line is the null hypothesis odds ratio.
What are three problems with meta-analysis?
Heterogeneity between studies (modelling for variation - fixed effect model vs random effects model and analysing the variation - sub-group analysis), variable quality of the studies, publication bias in selection of studies.
Ideally, how should studies in meta-analyses be similar?
Study design, participant profile, treatments or exposures, outcomes measured and statistical analysis used.
How does the fixed effect model differ from the random effects model?
The FEM assumes the studies are estimating exactly the same effect size but the REM assumes the studies are estimating similar, not the same, effect size.
How do the fixed effect model and random effect model typically differ in the following values?
a. Point estimate
b. 95% confidence interval
c. Weighting of the studies
a. Similar.
b. Wider in REM than FEM.
c. More equal weighting in REM than FEM, so more weighting on smaller studies in REM.
What are two types of sub-group analysis?
Stratification by study characteristics - subsets of while studies defined by study design (e.g. length of follow up) and participant profile (e.g. recruitment criteria).
Stratification by participant profile - data analysed by type of participants, e.g. age group, gender.
What can variable quality be due to?
Poor study design, poor design protocol, poor protocol implementation.
Order the following from least susceptible to bias to most susceptible.
Cohort studies, randomised controlled trials, case-control studies and non-randomised controlled trials.
Randomised controlled trials
Non-randomised controlled trials
Cohort studies
Case-control studies.
What are the two approaches that can be taken towards variable quality of studies in meta-analysis?
Define a basic quality standard and only include studies of that standard and above.
Score each study on its quality and incorporate the score into weighting of each study in analysis, use sub-group analyses to explore difference and meta-regression analysis.
