Studied by 0 people
Looks like no one added any tags here yet for you.
what do we need systematic reviews for?
Pulls together papers on a particular subjects and give us a summary
what is cumulative meta analysis
a statistical procedure to calculate summary estimates from the results of similar trials every time the results of a further trial in the series had become available e.g each year
what are the methods of literature searching
identify PICO elements
identify search terms
use literature searching
review papers
summarise results
what is PICO
population, intervention, control, and outcomes
briefly describe the selection process for papers included in the meta-analysis
review titles to get rid of duplicates
review abstracts to exclude basic sciences
review full texts
what do papers need, to be included?
need to be experimental or observational
assess association b/w 2 things e.g BB’s and COPD mortality
clearly define terms being tested
what do you add when summarising papers?
Descriptive summary
describe studies bc combining results not usually sensible
meta-analysis if results CAN be combined
what should you be aware of when reviewing papers?
assessment of bias, confounding and design issues
how is the data actually summarised
you’ll have a table with all the studies, each study given one row of data
gives you a reasonable study
forest plot - useful for presenting results
what os a forest plot
graphically presents results
size of 95% confidence intervals compared to study sizes
the smaller the study, the larger the confidence interval
why is it difficult to give one overall result?
bc one paper could be more accurate than other studies
larger study size
more similar population to patients being treated.
what is meta-analysis
a statistical method for combining the results of several studies.
when is a fixed meta analysis used?
when you can assume that the treatment effect is the same in each study and variation is due to sampling variation
assumes true effect doesn’t differ b/w studies
when do you use random effect meta-analysis
if there is heterogeneity (you can test for it)
(basically, how consistent are these results?)
how do you test for heterogeneity
COCHRAN’S Q TEST
Q finds if there a significant difference b/w each study’s
which model has smaller weights? Random effects or fixed?
random effects! because they are more similar to each other
when are funnel plots used?
to determine whether there are any biases in the study
→ the bigger the error, probably the smaller the studies and at bottom of funnel
what are the difficulties with bias in studies? when do they arise
not all relevant studies are included
lots of info published but not enough detail to include in the meta-analysis
reporting of one major result instead of all results
what are Cochrane reviews?
A TYPE OF SYSTEMATIC REVIEW!
take in all papers on a particle subject and due a huge review of all those papers
to cover everything available
- incredibly reliable
