Introduction to EBM Flashcards
What is a good strategy to help determine a good research question?
PICO (Population, Intervention, Comparison, Observation)
use this acronym to help identify a question that can lead to novel findings
refer to page 10
How can we acquire evidence for research questions?
Medline & PubMed are preferred
Cochrane Database
Google Scholar & Google (important, but do not be overeliant)
What is a common study design for drug therapy related research?
Randomized Control Trial (RCT)
What are the two main categories of all research studies?
Descriptive
Analytic (focus of PHAR 271)
What are the characteristics of a descriptive study?
They describe the situation at a given point in time
No intervention by the researcher
ex. survey
What are the characteristics of a analytic study?
They describe change in the situation over a given amount of time
Researcher applies an intervention or observes a sample over time
What does it mean to appraise a research study?
It is the process of carefully and systematically examining research to judge it validity, results and relevance
Why is critical appraisal of research studies important?
Sometimes study results are not always…
- Valid (differences are not caused by the intervention alone)
- Safe (invalid results can be harmful/wasteful)
- Useful (not meaningful or applicable to patient)
What are the two main types of validity?
Internal Validity
External Validity
What is internal validity?
Are the results due to the intervention (or exposure) or is there something else creating a discrepancy
What are the three major threats to internal validity?
- Chance (random error)
- Confounding (confusion of effects, correlation is not always causation)
- Bias (systematic errors that cause incorrect estimates of the association between intervention and outcome OR issues with study design)
What is external validity?
Are the results applicable (generalizable) to other populations (patients), settings, and time
What are some challenges for acheiving external validity?
Children, sick people, very elderly, and pregnant women are patient groups that are rarely included in trial studies
Therfore, application of trial findings to these patient groups is not straightforward