Topic 9: Meta Analysis

Question 1

What is Meta Analysis

Answer 1

The statistical combination of results from two or more separate studies. A systematic review.

Question 2

Why might we want to do meta analysis

Answer 2

Individual studies may be too small, studies may address the same question but the trials differ in some ways.

Question 3

Why might individual trials be too small

Answer 3

Failed to recruit their target, they overestimated the MCID.

Question 4

In what ways can trials that address the same research question differ

Answer 4

Details of treatment, outcome, patient samples

Question 5

Why is meta-analysis needed in grant applications

Answer 5

To demonstrate a new trial is justified, and the question cannot be answered using existing data.

Question 6

What might make a trial difficult to include in meta analysis

Answer 6

Difficult to locate, trial might not have made it into literature, the trial is currently ongoing.

Question 7

What are the advantages of meta analysis

Answer 7

Increase the power to detect treatment differences, increased precision for estimates of treatment effect. Allows results from small inconclusive studies to contribute. Gain the ability to answer questions not posed by individual studies (eg subgroup effects). It is reassuring to see consistency across a broad range of patients.

Question 8

What do we need to consider for meta analysis to make sure it isn’t misleading

Answer 8

Specific study designs, within study biases, variation across studies, reporting bias.

Question 9

What are the preliminary steps for a meta analysis

Answer 9

Formulate review question, specify eligibility criteria, identify, select and criticially appraise studies, collect appropriate data, decide what would be meaningful to analyse.

Question 10

What does an assessment of quality of the study mainly look at

Answer 10

The risk of bias in the results: for example, over or under estimating a treatment effect.

Question 11

Give an example of a tool for assessing bias

Answer 11

Cochranes

Question 12

Give reasons a trial may not be considered of high enough quality to be included in a meta analysis

Answer 12

Small sample size, poor measurement of characteristics, poor intervention implications, non-random treatment assignment.

Question 13

How do we decide which studies to exclude

Answer 13

A study is given a quality score, and we exclude studies based on a quality score cut off.

Question 14

What are 2 other ways to assess the quality of trials being considered for the meta analysis

Answer 14

Weighting studies, down weighting those of a poorer quality. Or doing a sensitivity analysis where we compare results across pre-specified analyses.

Question 15

What quantity do we usually investigate when it is binary data

Answer 15

Risk ratio, odds ratio

Question 16

What quantity do we usually investigate when it is continuous data

Answer 16

Mean difference, standardised mean difference

Question 17

What quantity do we usually investigate when it is ordinal data

Answer 17

Proportional odds ratio

Question 18

What quantity do we usually investigate when it is counts and rates data

Answer 18

rate ratio

Question 19

What quantity do we usually investigate when it is time to event data

Answer 19

Hazard ratio

Question 20

What analysis method is used for meta analysis

Answer 20

Meta regression analysis

Question 21

How does the fixed effects model work

Answer 21

The underlying true treatment effect is assumed to be equal for all trials, and then the treatment effect observed in each trial is this plus a sampling error.

Question 22

What distribution does the fixed effects model assume for the sampling errors

Answer 22

normally distributed with mean 0 and variance.

Question 23

What is an alternative to the fixed effects model

Answer 23

The random effects model

Question 24

What does the random effects model say

Answer 24

allows the population treatment effects to vary across trials, since the idea that each observed treatment effect estimates that same underlying quality is not necessarily plausible.

Question 25

What does meta regression analysis do

Answer 25

Models the variation between trials investigating why results from different trials might have differred.

Question 26

What is heterogeneity in meta analysis

Answer 26

The variation in study outcomes between studies

Question 27

What is I-squared

Answer 27

The percentage of total variation across studies due to heterogeneity and not chance.

Question 28

What can we do before deciding whether to carry out a fixed or random effects model

Answer 28

Test for homogeneity

Question 29

What is I-squared used as an alternative for and why

Answer 29

Homogeneity tests. Homogeneity tests have low power.