Systematic Reviews and Meta-Analyses Flashcards
What are secondary research sources?
Systematic reviews and meta-analyses
They facilitate the process of identifying relevant articles and synthesizing (valid) findings into useful evidence
What are systematic reviews not?
Review papers
Thought pieces
Industry white papers
Extended background/introduction in primary research articles
Textbook chapters
*these do not follow an explicit, structured, and replicable methodology
They are written by “experts” who have already made up their mind
What is primary research vs secondary research?
Primary - written by researcher; patients as subjects
Secondary - systematic reviews (SR) and meta-analyses (MA); studies as subjects
How are secondary reviews similar to primary research?
Both are conducted according to explicit, transparent, structured methodologies
This allows other investigators to replicate and extend/update the review at a later time
The fundamental difference between primary research and secondary research does NOT lie in the scientific principles applied to the investigation, but in the focus of the investigation
What makes a systematic review different from other types of reviews?
The methods
Search strategy
Selection criteria
Assessment of study validity
Formal synthesis of evidence
How do you conduct a search strategy for a systematic review?
Should be clear and replicable
How did the investigators choose their articles?
When did the search window begin and end?
Which databases/search engines were used for the search?
Which search terms did they use?
These should be guided by the research question/objective
What is the selection criteria for systematic reviews?
Inclusion/exclusion criteria
Patient population
Intervention
Outcome
Study type
Why do we have to assess quality?
Ensures that high-quality (well-designed) studies are “weighted” more in the synthesis of findings
Typically, use critical appraisal sheets to evaluate study validity (e.g., random assignment, selection bias, blinding)
What is the synthesis of evidence?
How were the findings organized so that they can be summarized in a meaningful way
What is a meta-analysis?
A meta-analysis uses quantitative methods to summarize the current status of evidence by pooling results of similar studies asking the same question
Almost always based on a systematic review
Moves beyond reiterating/synthesizing the results of individual studies
Resolves issue of conflicting results, improves generalizability
Provides improved power and precision of effect size estimates by aggregating data from multiple studies to increase sample size
Can studies be combined?
Yes
If a clinical question is important, it probably has been addressed in one way or another in a series of clinical trials
These studies can sometimes be quantitatively combined
Dependent on a homogeneous (consistent) effect - careful consideration of similarities and differences of patient characteristics, contexts, and care processes is required before combining studies and claiming that one result fits all
Are statistical tests used to evaluate the degree of homogeneity across studies?
Yes
Indices of the extent of agreement among a set of studies include the Q test and I2 Index
The results of these tests determine whether the studies reflect a single (homogeneous) underlying effect and may be combined statistically
What is an overall effect size?
In a randomized controlled efficacy trial, this would be the average difference between groups divided by the pooled standard deviation
The contribution of each study is averaged through a weighting process by one of several methods
Generally, weighting is inversely related to variability (precision of the estimate) and proportionately related to sample size:
The studies with lower variability in the outcome and larger sample sizes are accorded more weight
