Study Design (Segars) Flashcards
In interventional study designs, list in order of increasing evidence:
Phase 0 –> Phase 1 –> Phase 2 –> Phase 3 –> Phase 4
In observational study designs, list order of increasing evidence:
Cross sectional –> Case control –> Cohort
__=outcome is NOT yet known at start of study
__=outcome IS already known at the start of the study
prospective
retrospective
if don’t know the outcome at the beginning, its a prospective study
which studies prospective only?
Phase 0-4 (interventional)
Cohort (Observational)
Is researcher forcing group allocation?
Yes=__
No=__
Yes –> Interventional (Buzz word=Randomization) (sample size/focus/duration)
No –> Observational (outcome/exposure)
If you see “randomization” in a question stem, its what study design?
Interventional
For observational studies, if groups organized by DISEASE STATUS, it is a __ study
Case-control/Nested case-control
For observational studies, if groups organized by EXPOSURE STATUS, it is a __ study
Cohort
For observational studies, if “together d/t a common factor”, it is a __ study
cohort
For observational studies, if data collected across large population, it is a __ study
cross-sectional
A case-control design is useful when studying a rare disease. It commonly generates an __ for each and an __ as measure of association
Odds of exposure
Odds ratio (OR)
Case control studies are customarily conducted in a __ fashion.
Retrospective
A __ study derived from w/in or out of a cohort or interventional study
Nested case control
disease/outcome subjects from ‘other’ study become the cases of a subsequent (different) study
a __ design is useful when studying a rare exposure. Commonly generates the Risk of disease/outcome for each and a RR as measure of association
cohort
Cohort studies can be conducted in a __ fashion
retrospective or prospectve (or both)
In Case control, which boxes?
In Cohorts, which boxes?
Disease presence –> Column totals (A+C and B+D)
Exposure –> row totals (A+B and C+D)
A Group with something in common, i.e., Birth/inception/exposure/etc. is characteristic of __
BIG STAR
cohort study
Framingham heart study, nurses health study
Think of a __ study when the info gathered represents what is going on with disease AND exposures (at the same time) across the entire (a large) study population
x-sectional
“snap-shot” in time; examine relationships between disease AND exposure
Any survey that says “National” is a __ study
X-sectional
This interventional phase study is exploratory, investigational new drug, healthy volunteers, very small (<20), very short duration
0
This interventional phase study is for investigational new drug, assess safety/tolerance and PK of or more dosages, can have healthy or disease volunteers (depends on disease), small N (20-80), short duration (few weeks)
1
This interventional phase study is for investigation new drug in the indicated population. Assess effectiveness, diseased volunteers (narrow inclusion criteria for isolation of effects), larger N (100-300), short-medium duration (few wks to months)
2
This interventional phase study is for investigational new drug, indicated population. Assess effectiveness, diseased volunteers (may expand inclusion criteria and comparison groups for delineation of effects; Various stat-perspectives can be taken in studies such as superiority, non superiority, and equivalency), larger N (500-3000), longer durations (few months to years)
3
this interventional phase study is post FDA-approval. Registries/surverys also used in observational design
4
this interventional study format has subjects randomly allocated ONCE into a single tx group
Simple
this interventional study format has subjects randomly allocated into an initial group, then further randomly subdivided into a subgroup (multiple randomizations)
Factorial
useful in testing multiple hypotheses at the same time
This interventional study format has subjects exclusively managed in a single tx group. No switching groups after initial randomization
Parallel
This interventional study format has subjects forcibly switched to other tx group after initial tx assignment
X-over
Between & within group comparisons possible
which group allocation procedure most commonly utilized?
Random –> subjects have equal probability of being assigned to each of the pre-defined intervention groups
this form of randomization ensures balance within known confounding variables
Stratified
This form of blinding is neither investigators nor study subjects are informed which intervention (tx) group subjects are in
Double blind
Most conservative method of managing drop outs/lost to follow up by including them
intent to tx
- preserves randomization process
- maintains Power
__ role is to protect human subjects from undue risk. All human subject studies MUST be reviewed by them PRIOR TO study starting
IRB
__ role is as a semi-independent committee not involved with conduct of study but charged with reviewing data AS STUDY PROGRESSES to assess for undue risk of benefit
DSMB
can stop study early
