Epidemiological study design 2 Flashcards
first phase of clinical trials
1) first use in a small group to look at safety and dosage
second phase of clinical trials
Early studies e.g. efficacy in which there may/may not be a control group
third phase of clinical trials
Prove efficacy of treatment (large+ have control group)
Fourth phase of clinical trials
Side effect studies. No control group. Large
How to avoid confounding in clinical trials
- randomisation
- Randomisation will prevent confounding from factors we know could confound and unknown confounders
What are the two errors that are made by chance
Type 1 error (incorrectly rejecting Ho) (alpha)
Type 2 error (incorrectly accepting H0) (beta)
What is a type 1 error
incorrectly rejecting Ho
-believing the treatment works when it doesn’t
What is a type 2 error
Believing treatment doesn’t work when it does
How ro reduce type 2 error
Make study large
When does bias occur
IN allocation
When can double blind trials be hard to achieve
Surgery
One type of bias from the doctor
Can believe that one treatment is effective and the other is not so allocates one group of patients the one you think will work
What is generalization/applicability
Some drugs can’t be tested on certain groups of people e.g. pregnant women so are tested on other people
How to overcome the problem of generalisation
Use placebos
When is a trial not appropriate
When it intentionally causes harm
What is equipoise
Need there to be uncertainty as to the outcome. If a clinician believes one treatment is better then they would be obliged to offer it to all
What type of logistical difficulties may occur in clinical trials
Long delay between treatment and outcome which may be costly to drug companies
What to do when you have three different treatments
If there are 3 treatments, make them into 2 groups i.e. group 1 vs 2+3
Whats the advantage of grouping three groups into 2
Means you have a higher power
What are cross-over trials
Where patients get both treatments (one after each other)
What is an equivalence study
Trials to establish that a new treatment is equivalent
In subgroup studies, list the examples of when to use different studies
- subgroups may have different responses
- They can be analyzed separately
- If this is planned before trial begins, then we can test hypotheses in sub group
What type of trial would you use to see if a drug prevents myocardial infarction
Use randomized control trial
What type of trial would you use if you want to quickly see if a rare condition is caused by a drug
Case control study
What are rare outcomes best measured using
cohort studies
