Evidence Based Medicine Flashcards
Describe features of parallel group designs
• Randomises allocation to treatment or control group
• controlled: one group is placebo controlled and the other for standard treatment
• blinding:
– Single blind - either patients or observers do not know
to which group patient has been allocated
– Double blind - neither patients nor observers know to which group the patient has been allocated
When do you use parallel vs cross over trials?
A parallel group design is used when the effect of treatment is not reversible e.g. → antibiotics for infection / chemotherapy for cancer.
A cross-over design is used when the effect of the treatment IS reversible e.g. → drug to lower serum cholesterol / inhaler for asthma
What is the null vs alternative hypothesis?
A null hypothesis: no statistical significance between the two variables in the hypothesis. It is the hypothesis that the researcher is trying to disprove.
An alternative hypothesis: opposite of null hypothesis. So, it would be that there IS indeed a statistically-significant relationship between the two variables.
What is the p value?
P Value: a probability (values from 0-1)
it measures the probability of getting the observed results if the null hypothesis is actually correct.
A P-value BELOW the α level (usually 0.05) means you can reject the null hypothesis
• A P-value GREATER or EQUAL to the α level means we have failed to find evidence against the null hypothesis and we cannot reject it
Describe type I and II errors
What is intention to treat analysis?
ITT analysis: all participants in a trial are analysed according to the intervention to which they were allocated and not whether they received it and did it properly or not.
This approach maintains the original randomised groups and thus the baseline similarity between the groups is maintained.
What is on treatment analysis?
(OTA; i.e. per protocol analysis in drug trials)
→ OTA means limiting the analysis of the data to patients that complied with treatment. The major problem with this is that those that comply with treatment are often different in nature to those who do not.
Hence, it is difficult to decipher whether any treatment effect may be due to intervention or due to difference in the make-up of participants
So there is a need to be cautious when interpreting OTA analysis, especially when ITT analysis shows no difference between treatment groups, and justification for the OTA approach must be made.
