Critical Appraisal RCT Bias - Early Stop / Correct Group Flashcards
Intention to Treat Analysis
Analyzing patients in the group they were randomized into
On Treatment Analysis
Analyzing patients in the group that actually received the treatment
- Adds people from one arm to the other
Per Protocol Analysis
Analyzing only patients who are adherent to the assigned treatment
- Throws out those not adherent
Modified Intention to Treat
Analyzing only randomized patients who received one dose of the study medication
Effect of Patient Analysis
Not using an intention to treat analysis can bias trials
What Patient Analysis is best? Why?
Intention to Treat Analysis is best as it makes non-adherence and loss to follow up not an issue
- Preserves the strength of randomness
Effects of Per Protocol Analysis and On Treatment Analysis
Undermines randomization as it removes results from the group the patient was originally assigned to
- Removes non-adherence, contamination, and loss to follow up creating bias
What Analysis Method gives the smallest treatment effects
Intention to Treat Analysis
Reasons to stopping a trial early
Results were overwhelmingly positive
Results are harmful. Safety committee stops the study
Effects of stopping a trial early
Questions to ask when stopping a trial early
Were there preplanned stopping rules
Did those rules involve analysis of the data and a precise p value
Were there a large number of events
Affect of the Journal the study was published in
Just because the journal is reputable does not automatically make the study reliable
- Impact Factor
- Open Access Journal
Affect of the Author on the study
Check their publication history
Cite any conflicts of interest that could possibly introduce any bias
Affects of sponsorships
Leads to more favourable results and conclusions
- However, does not automatically make it biased
