Interventional Studies Flashcards
what methods are at the top of the pyramid?
systematic reviews
meta analyses
what method is at the very bottom of the pyramid?
in vitro research
list the evidence pyramid in order from lowest to highest
in vitro animal case report case series ecological cross-sectional case-control cohort interventional sys. reviews & meta
systematic reviews
compilation of multiple studies into one review
‘collective book report’
meta-analysis
compilation of the actual data of multiple studies
other names for interventional studies
clinical trial or study
experimental study
human study
investigational study
what are the key differences between observational and interventional studies?
forced allocation and ability to show causality in interventional studies
what 4 factors determine the phase of interventional study?
- purpose/focus
- study subjects
- sample size
- duration (disease specific)
as your phase number increases what also increases?
the studies duration and sample size
phase 0 and 1 are the only phases to include?
only logical places to include healthy subjects
in phase 4 you never see …..?
healthy subjects
describe ‘pre-clinical’ study
prior to human investigation
animal testing/study
what does phase 0 study?
exploratory or investigation of a new drug
phase 0 – list the 4 factors
- asses drug actions in a single/few doses
- typically healthy subjects (oncology)
- very small samples (typically <20)
- short duration (just long enough to monitor effects of the dose)
phase 1 – factors
- assess safety/how the body tolerates a drug
- healthy or diseased
- small 20 - 80 (<100)
- short duration - few weeks
what phases can represent first in human studies?
phase 0 or 1
or
phase 0 and 1 can be combined into one study
phase 2 – factors
- assess effectiveness of drug
- diseased
- large 100-300
- medium duration - weeks to months
in which phases do we assess effectiveness of a drug before it gains FDA approval?
phase 2 and 3
phase 3 – factors
- assess effectiveness
- diseased
- larger 500 - 3000
- long duration - months to year+
discuss phase 3 subjects
typically all diseased
may expand to have inclusion criteria to create different comparison groups
introduce different perspectives of null hypothesis
superior/equal/not worse
what is the final stage before FDA approval ?
phase 3
must do several phase 3 studies to show a greater ratio of positive to negative effects
generalize the purpose of phase 4
post-market study
FDA approval has already been gained and drug is on the market
study assesses the effect of the drug as it is used throughout the population
long term effects
co-morbidities
phase 4 – factors
- assess long-term safety and effectiveness
- diseased - those taking the drug
- entire population - hundreds to thousands
- varies - can be several years and ongoing
advantages of interventional studies?
can demonstrate causality
disadvantages of interventional studies
cost
time/complexity
ethics
validity
exploratory vs. explanatory studies
explore - looking at effectiveness, usefulness of a drug. non-real clinical environment. looking at how many doses or how big a dose to get effects
explain - interventional style
how to treat disease and patient
more real environment, flexibility to clinical, ability to change treatments
another term for explanatory studies
pragmatic studies
interventional study designs
simple or factorial
all studies are simple or factorial and parallel or cross-over
between simple and factorial, which requires more subjects?
factorial
describe simple design
divides subjects into x # of groups with only 1 round of randomization
used to test 1 hypothesis
no further division into groups
describe factorial design
divides subjects into >2 groups with at least 2 or more rounds of randomization to create subgroups
can test multiple hypotheses at once
major advantage of factorial design
-improves efficiency for answering clinical Q’s
characteristics of factorial design
- bigger study population
- increased complexity
- increased risk of drop outs
- possible restriction to validity
explain parallel studies
groups are simultaneously and exclusively studied
no switching between interventions is allowed
explain cross-over studies
subjects can serve as their own control by doing one intervention, wash-out phase
and then doing the other intervention
allowing between and w/in group comparisons
pictorials of cross-over studies
there is always an arrow present showing mvt of subjects
wash out
a period between cross-overs where the subjects go thru a time period to get the first intervention out of their system
this serves to eliminate contamination
lead in
a period before the study even starts to clear the subjects system of anything they might have been doing in their normal lives
what is the difference between washout and lead in phases?
lead in is always before the study begins and washout occurs when subjects are switching groups
what are the other benefits of utilizing a lead in period?
- gives subjects a practice period
- can observe the compliance and follow-up abilities of the subjects before the study actually starts
another term for lead in phase
run-in period/phase
disadvantages of cross-over design
-only good for long-term conditions that are not curable or the treatment only gives short-term relief
- subjects must stick w/ study for longer duration
- washout required
- complex data analyses
outcomes or endpoints in interventional studies
- primary
- secondary, tertiary, etc.
- composite
primary outcomes
the most important outcome
-answers the main research question
secondary outcomes
lesser importance than primary
-possible for future hypothesis generation
composite outcomes
combines multiple endpoints into one
can also be considered the primary outcome, with the individual endpoints being the secondary endpoints/outcomes
POEMs
patient oriented endpoints—most clinically relevant
ex. death, stroke, MI, hospitalization, preventing dialysis
DOEs
disease oriented endpoints—-elements used in place of evaluating POEMs
ex.
blood pressure for stroke risk
cholesterol for MI risk
group allocation
random or non-random
non-random group allocation
subjects do not have equal probability of being selected to each intervention group
random group allocation
subjects have equal chance of begin assigned to every intervention group
randomization - purpose
to make groups as equal as possible
based on known and unknown important factors/confounders
but group equality is never guaranteed
randomization is used…..
in interventional studies only
observational studies do not use randomization
examples of forms of randomization
simple
blocked
stratified
simple randomization
equal probability for allocation within one of the study groups
blocked randomization
ensures balance within each intervention group
best to ensure all groups are equal in size
stratified randomization
ensures balance with known confounding variables
can also pre-select levels to be balanced within each confounder
new term for confounder
interfering factors
masking of study subjects and investigators
single blind
double blind
open-label
single blind
masking
subjects are not informed of their intervention group
researchers do know
double blind
masking
subjects and researchers do not know who belongs to which group
open-label
unmasked/no blinding study
subjects and researchers know who is in what group
what can be used to assess adequacy of blinding
post-hoc’s survey
forms of blinding
placebo
placebo-effect
hawthorne-effect
placebo therapy
- -inert treatments made to look identical in all aspects to the actual treatment but has no real effects
ex. dosage form, frequency, monitoring, therapy requirements
double dummy
more than 1 placebo is used
placebo effect
improvement in condition by power of suggestion of begin treated, can be as powerful as 30-50%
the positive benefits you get from knowing you could be getting a treatment but you are actually getting the placebo
hawthorne-effect
study subjects change their behavior solely due to awareness of being studied/observed
post-hoc subgroup analysis
data-dredging or fishing
reduces power of data and increases risk of type 2 error
when is post-hoc accepted?f
not accepted when not planned
accepted when it is planned or performed for hypothesis generation
sample size determination
always have excess in sample size to account for possible drop-outs, deaths, move-away, etc.
managing drop-outs
either include them anyway or drop them
intention to treat method
a study that includes data from drop-outs, anything missed is considered null or no benefit
intention to treat results in
- preserves randomization process
- preserves baseline characteristics of groups
- maintains statistical power of original sample size
ignoring drop outs
include only compliant subjects in data analyses
must get a larger N to counteract but need drop out to be equal in groups
per-protocol or efficacy-analysis
as-treated method
ignores group allocation
subjects can switch groups
subjects are evaluated in the group they move to, end in, or spend most time in
per-protocol method
ignoring drop outs
efficacy-analysis
only including compliant subjects in data, ignoring drop outs
impact of per-protocol results
- biases can effect
- typically over estimates effects
- reduces generalizability
assessing adherence
ways in which the researchers know someone is complying with the rules
drug levels
pill counts
bottle tops
methods of improving adherence
ways to increase the compliance of subjects
frequent communications or follow-ups
treatment alarms
dosage containers
list the bottom 3 studies of the research evidence pyramid beginning with the least
in vitro
animal studies
case reports
what is the top of the evidence pyramid
systematic reviews and meta-analyses
then interventional exploratory and pragmatic studies
list the observational study types in order of high to low evidence
cohort case-control cross-sectional ecological case series case reports
animal research
in vitro
of the following which has the least amount of evidence?
ecological
pragmatic studies
case series
case reports
case reports
case series
ecological
pragmatic
