Clinical Epidemiology-Evidence Based Medicine Flashcards
What are the two options for summarizing clinical research evidence?
systematic (cochran collaboration)
narrative (linus pauling)
In regards to the two options for summarizing clinical research evidence which option is the strongest?
systematic
What are some differences in narrative and systematic reviews?
Systematic —> question: focused on clinical; sources: explicit search strategy; selection: criteria set up; appraisal: rigorous and critical; synthesis: quantitative ; Inferences: usually EBM
Narrative –> question: broad in scope; sources: not usually specified; selection: not usually specified; appraisal: variable; synthesis: qualitative; inferences: sometimes EBM
In general what is a traditional narrative review?
Expert summary of research evidence to make recommendations.
(take a broad based topic and review it but does have to be done by an expert)
What are some advantages to a traditional narrative review?
Can address broad-based topics by authors with experience
What are some disadvantages to a traditional narrative review?
Lack of structure, potential for value judgements, biases.
therefore may not be the most valid method
What is a systematic review?
Rigorous scientific review of evidence bearing on a specific clinical question with specific criteria and procedures set in advance.
what are some advantages of a systematic review?
Useful for single, focused questions; finds all quality studies; increases power; improves generalizability
(we can take small studies combine them and therefore we can have more statistical power, allows a better picture of whats going on)
what are some disadvantages of a systematic review?
expensive and laborious and time consuming
What are the steps in doing a systematic review?
- Define question
- Find relevant studies
- select strongest studies
- determine scientific strength and quality
- summarize in tables and forest plots
- determine if pooling (meta analysis) is okay
- calculate summary effect size
- identify if heterogeneity is present
Does systematic reviews involve published or non published works?
both :)
In regards to defining a question, what does the acronym PICO mean?
P= patients I= Intervention C= Comparison O= outcomes T= time
What are the types of studies for a systematic review?
Randomized trials
Cohort
Case Control
Diagnostic Tests
What are some examples of types of systematic reviews that are experimental studies?
- antiplatetlet drugs for cardiovascular prevention
2. Chrondoitan vs. Placebo for relief of pain from osteoarthritis
What are some examples of types of systematic reviews of observational and diagnostic studies?
- Smoking as as risk factor for breast cancer
- Leg raising test to detect lumbar disk herniation in patients with low back pain and leg pain
- Anti-TB IgG test for diagnosis of pulmonary pneumonia
What are some potential biases for publication?
Why studies are not reported? * Non-significant findings * Unpublished data * disagreement with findings Is there a remedy? * Cochran collaborators track down leads for systematic reviews * Encourage publication of all results * Registration of all clinical trials
One of your patients is an active 73 y.o with constant A-Fib and a history of hypertension. His meds are: aspirin; digoxin; hydrochlorothiazide, enalapril. You wonder whether you should change from aspirin to warfarin to prevent strokes and other clotting problems. What is an appropriate PICO question?
P= patient (Elderly with A-fib)
I= intervention (warfarin)
C= comparison ( aspirin or placebo)
O= outcomes (strokes or clotting problems)
Ex. In elderly persons with A-fib does warfarin perform better then placebo to prevent strokes and clotting problems
A systematic review finds how many studies that meet stringent criteria?
Five (in this case comparing warfarin with aspirin)
In regards to epi studies, for a 95% CI, if there is a 1 in the CI, what does this mean?
If one in the confidence interval this means that the study is not significant
(aka the null is in the value)
What is a meta-analysis?
Use of statistical methods to combine results of individual studies, usually from systematic reviews
What are the advantages of a meta analysis?
can provide an adequate sample size and power to evaluate small treatment effects; especially good if the analysis can be done with the data from individual patients
What are the disadvantages of a meta analysis?
Quality is dependent on studies, may be too heterogeneous to combine; patients are variable
What is a fixed meta analysis?
Assumes each study answers the same questions, and has same effect size, so results differ only by chance.
What is a random effect model?
Assumes studies address different but related questions, a random sample of all studies on that question; takes heterogeneity into account, less likely to overestimate precision, so wide CI considered more realistic.
