Critical Appraisal Flashcards
What is critical appraisal?
It concerns assessing the validity of the research and the statistical techniques employed in studies and generating clinically useful information from them
What are the two types of research validity?
Internal- The degree to which the results of the study are likely to be true, believable and free from bias taking into consideration research methodology and design
External- Can the results obtained be generalised to the wider population?
What are the 6 main checks when doing a critical appraisal?
- Are aims, clinical question and hypothesis clearly defined?
- Is the research method appropriate to answer the clinical question?
- Is the sample size determined?
- Is the statistical analysis appropriate?
- Are the conclusions based on the results?
- Can I apply the results of this study to my patients?
What are 4 types of bias?
- Selection bias- Differences between baseline
characteristics of the groups being compared. Avoided by randomisation. - Performance bias- Difference between groups in the care provided. Avoided by Blinding
- Detection bias- Differences between groups in how outcomes are determined. Avoided by Blinding
- Attrition bias- Differences between groups in
withdrawals from the study. Avoided by statistic methods
Why does sample size matter?
As a general rule, the larger the sample size the more power the study has.
• Power should be at least 80%
• Power can be used to calculate the minimum sample size required to avoid type 2 error (failure to detect a difference when it actually exists)
How do you assess if the study design is appropriate?
The research question will determine which study design is suitable. One question can be answered by more than one study design, however one study design cannot answer all clinical questions.
How do you control confounding factors?
Restriction
Matching
Randomisation (stratified sampling)
Statistic methods using regression analysis
What does the p-value mean?
The probability that an observed difference could have occurred by chance. If p<0.05, this is the threshold for statistical significance.
What do confidence intervals mean?
Usually set at 95% limits, they represent the likelihood that the mean is located between these values. The wider the interval range, the less precise the results of the study are, and this is usually caused by small sample size
How do you determine if you can apply the findings of a study to your patients?
Think of the external validity!
Is the study population similar to mine?
Are the settings similar?
Are all the outcomes accounted for?
What about my patients values and believes?
