Final Exam

Question 1

What are the three types of reviews with a description of each one

Answer 1

Narrative (non-systematic): An author’s interpretation of selected evidence on a general topic, Systematic: An author make a conclusion based on a (potentially blinded) review of data, meta-analysis: quantitative-systemic review that combines the outcomes from several studies into a single result while producing new data and providing supporting evidence for clinical decision making

Question 2

Which types of reviews have a focused clinical question, predifined literature search, blinding of reviews, and has no statistical analysis

Answer 2

Systemic and Meta-analysis

Question 3

What are the strengths of meta-analyses

Answer 3

There are patient/subjects included there is more power due to increased sample size, improved estimates of effect size, allows for indirect comparisons in absence of head-to-head trials (resolving discrepancies among smaller studies)

Question 4

What are limitations of meta-analyses

Answer 4

Limited by the quality of available studies, prone to inheriting the biases of included studies, degree of inconsistency between studies

Question 5

T/F: I2 statistics show the degree of inconsistency between studies and if greater than 40% it is significant

Answer 5

True

Question 6

What are the key elements of meta-analysis design

Answer 6

Introduction: explicit statement of question or questions being addressed and explaining the rationale for the MA
Methods: Inclusion/Exclusion criteria for the types of studies involved, literature search strategy
Results: show the study selection(via flow diagram) and the related study characteristics (in a table)
Discussion: Summarize the main findings but consider the relevance of the findings along with the implication for future research

Question 7

T/F: The methods section of a meta-analysis must explain the data collection process and their methods of measuring outcomes, assessing bias and heterogeneity. Often use a fixed-effects model or random-effects model to combine data

Answer 7

True

Question 8

T/F: The results section of a meta-analysis should individual study results and overall/combined results

Answer 8

True

Question 9

What are funnel plots and how are they conducted

Answer 9

Used to identify the existence of PUBLICATION BIAS using a scatter plot, X= treatment effect and Y= sample size, plot should look like a funnel to hint there is no publication bias

Question 10

What are forest plots and how are they conducted

Answer 10

Used to report the RESULTS of a meta-analysis using a visual display, provides individual and combined results with means and confidence intervals highlighting if statistical significance is present