PHIL 12 Lecture 11: Randomized Controlled Trials

3 Pages
Unlock Document

Kerry Mc Kenzie

PHIL 12 – Lecture 11 – Randomized Controlled Trials Reach ch 20 Going Further? • Is there a way to ensure we identify properties that cause other proportions, not merely correlate with them? o Randomized Controlled Trial: in ideal allows us to infer with near deductive certainty what was causally responsible for an effect; aka RCT ▪ Can’t always perform RCT’s • Often for ethical reasons ▪ Exporting inferences about what caused E in a sample to the general population is subject to inductive risks ▪ Most powerful method for identifying causes RCTs: The Gold Standard in Medicine • Gold standard in causal inference, foundation of Evidence-Based Medicine (EBM) • Ex: statins; did RCT and found that it reduced the risk of heart disease, stroke, and mortality in comparison with a placebo o So trusted and effective that there has been suggestions to put them in the water supply o Use of statins justified on basis of RCT evidence alone ▪ Don’t know why/how they work • Allow you to identify that something is causally effective without requiring you to know how Phases of Drug Testing • We want to see if lithium is effective at preventing the recurrence of suicidal thoughts in adults o Preclinical Phase: animal. Testing o Phase 0: ‘first in man’; sub-therapeutic dosages in small numbers of subjects to determine rates of metabolization (how fast it’s leaving the body) o Phase 1: small-scale trial (10-30 subjects) to determine safe dosage, administration, side effects, preliminary effectiveness o Phase 2: larger-scale RCT (100-300 subjects) against placebo to begin to evaluate effectiveness and further determine safety o Phase 3: large-scale RCT (300-3,000 subjects) to further determine efficacy, perhaps against existing treatments; success here gets the drug on the market o Phase 4: continued studies when on market, including feedback from doctors The RCT Phase: Step By Step • Step 1: obtain a sample o Need a representative sample ▪ How do we acquire a sample of people with X? ▪ How do we ensure that the sample is representative of the population? • Is sample matching feasible? o Sample Matching: figure out secondary properties will be relevant to X ▪ Hard enough to find people with X in the first place, let alone with secondary properties • How would we do random sampling? o Assign a number to each name, and use a random number generator ▪ Selection bias even in the best-case scenario • Ex: Gallop poll, where they called people randomly from the phone book, but the phone book was biased because rich people owned phones at the time • Step 2:
More Less

Related notes for PHIL 12

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.