Systematic Review Flashcards
A systematic review of observational studies of antioxidant vitamins to prevent cardiovascular disease combined the results of 12 studies to obtain a summary effect size and confidence interval. Which of the following would be the strongest rationale for combining studies results?
A You wish to obtain a more generalizable conclusion.
B. A statistical test shows that the studies are heterogeneous.
C. Most of the component studies are statistically significant.
D. The component studies have different and to some extent complement, biases.
E. A more precise estimate of effect size is needed than the individual studies provide.
E
When justified, pooling study results does provide a more precise estimate of effect size than any of the individual studies alone.
You are asked to critique a review of the literature on whether alcohol is a risk factor for breast cancer. You see that the reviewer has searched MEDLINE and found several observational studies of this question. All of the following are limitations of this search strategy except
A. Studies with negative results tend not to be published.
B. MEDLINE searches typically miss some articles even when they are included in the MEDLINE database.
C. Many of the world’s studies are not published in MEDLINE
D. Some articles are found by asking experts in this field, looking up recent review articles, and following up on citations in the articles you have found.
E. The studies you have found are not randomized controlled trials.
E
Because alcohol cannot be randomized, observational studies are the only feasible way of studying whether alcohol is a risk factor for breast cancer. All the other responses describe true limitations of a search strategy that includes only a MEDLINE search.
Systematic review of antiplatelet agents for cardio vascular disease prevention have combined individual patients, not trials. An advantage of this approach is:
A. It is more efficient for the investigator.
B. It is possible to obtain more precise estimates of effect in clinically important subgroups of patients.
C. Flaws in study design can be overcome.
D. It is not necessary to choose between fixed- and random-effects models when combining data
E. Publication bias is less likely.
B
The main advantage of patient-level meta-analyses is that it is possible to place each patient into clinically important subgroups and to have enough patients within those subgroups to provide relatively precise estimate of effects.
You are aware of growing evidence that aspirin is effective for the primary prevention of CVD. You find a cumulative meta-analysis of this question in the Cochrane Database of Systematic Review. The following are advantages of this graphic summary:
You can see when and how each new trial has changed the overall evidence of effectiveness. T/F
T
The cumulative meta-analysis shows how evidence is accumulated and the summary effect changes over time.
You are aware of growing evidence that aspirin is effective for the primary prevention of CVD. You find a cumulative meta-analysis of this question in the Cochrane Database of Systematic Review. The following are advantages of this graphic summary:
The summary effect size is the best available estimate of the true effect size for similar patients. T/F
T
If the underlying assumptions are met, the summary effect size is the best estimate of effect form all of the trials as a whole.
You are aware of growing evidence that aspirin is effective for the primary prevention of CVD. You find a cumulative meta-analysis of this question in the Cochrane Database of Systematic Review. The following are advantages of this graphic summary:
You can be confident that all of the trials studied exactly the same research question. T/F
F
One would need to look elsewhere in the report to decide how confident he or she is that the studies were all of the same question.
You are aware of growing evidence that aspirin is effective for the primary prevention of CVD. You find a cumulative meta-analysis of this question in the Cochrane Database of Systematic Review. The following are advantages of this graphic summary:
You can see how many trials of this question have been published. T/F
T
In the typical forest plot, one can easily form a visual impression or count the number of trials that met inclusion criteria.
You are aware of growing evidence that aspirin is effective for the primary prevention of CVD. You find a cumulative meta-analysis of this question in the Cochrane Database of Systematic Review. The following are advantages of this graphic summary:
Bias is a likely explanation for results, especially for the large trials. T/F
F
These graphics summarize studies without regard to whether or not they are scientifically sound. For that, the reader would need to look at the authors’ criteria for including studies.
You are aware of growing evidence that aspirin is effective for the primary prevention of CVD. You find a cumulative meta-analysis of this question in the Cochrane Database of Systematic Review. The following are advantages of this graphic summary:
You can see how precisely the effect size has been estimated. T/F
T
These graphics typically show the confidence intervals for all available studies at the time each new study is added to the rest.
Choose one or more. Systematic reviews are important because
a. The research has been critically appraised
b. Multiple studies provide information of greater statistical significance
c. They only identify positive outcomes
d. They provide incidence statistics
e. They are based on a comprehensive literature search
a, b, d, e
The strongest study design in weighing evidence is A. Systematic review B. Randomized clinical trial C. Cohort with control D. Case control
A
In the absence of bias, studies with small sample size are spread at the bottom of the figure
a. Publication bias
b. Funnel plot
c. Forrest plot
d. Meta-analysis
b
Advantages of a meta-analysis
a. Useful in resolving conflicting results among studies
b. Increases the sample size for the data analysis for the question at hand
c. Can give you one summary statistics for several studies
d. All of the above
e. A and C only
e
Representation of a running summary of all the studies up to each point in time
a . Forrest plot
b. Funnel plot
c. Meta-analysis
d. Cumulative meta-analysis
d
The first step in a systematic review is A. Literature search B. Plan methodology C. Gather expert opinions D. Assign a reviewer
A
