Randomized Controlled Trials Flashcards
Who was the first person thought to conduct an RCT?
Sir Austin Bradford Hill
What is an RCT?
test whether an intervention works by comparing it to a control condition
How are subjects assigned to groups in RCTs?
randomized
-subjects assigned to study groups randomly
-equal probability of being assigned to either group (intervention or control)
True or false: RCTs can only contain two groups
false
can be >2 groups
What are the three types of RCT designs?
parallel:
-intervention and control group and then follow both over time and compare results
cluster:
-clusters of individuals are randomized instead of individuals
-often used in studying patient care
cross over:
-two groups, follow them and then swap the groups (control becomes intervention, vice versa)
-uses a washout period
-compares self to self
What are the advantages and disadvantages of cross-over RCTs?
advantages:
-less people needed
-minimizes confounding
disadvantages:
-complex
-sometimes a longer washout period
True or false: you can conduct an RCT before designing the trial
false
must design before it begins
Why do you need to make sure a study is designed before it begins?
to limit bias
What should be considered when choosing a population for an RCT?
start with a defined population
-who we want to study
-where do we get them from
specific inclusion and exclusion criteria
-much more strict in RCT
-can limit generalizability
sample size (n)
-how many subjects do we need?
-optimal number is determined statistically
What is the study power?
ability to show a difference between groups if a difference really exists
-n too low=underpowered
-n too high=unnecessary exposure, more resources
When should sample size for an RCT be determined?
before the study starts
Give examples of interventions that are specific and well done.
drug: dose, regimen, delivery method, follow-ups, length of therapy, etc
procedure: whats involved, timing, follow-up, length of intervention
What are some examples of what we can compare our intervention to?
nothing (placebo/usual care)
different dose
a different drug/procedure
standard treatment
What is the key takeaway regarding the treatment of the intervention and control groups?
they should be treated the same, the only difference is the intervention
Differentiate between an objective endpoint and a subjective endpoint.
objective: measurable (ex: BP, LDL)
subjective: subjects interpretation (ex: back pain)
Differentiate between hard endpoints and surrogate endpoints.
hard:
-major and non-disputable
-well defined, no subjectivity
surrogate:
-proxy or estimate
-can shorten duration of study
Differentiate between primary endpoint and secondary endpoint.
primary endpoint:
-main result that is measured at the end of the study to see the effect of the intervention
secondary endpoint:
-additional results of interest but not the main focus
-caution when interpreting as study wasnt designed around them
-never make a decision based on a secondary endpoint
What is a composite endpoint?
a primary endpoint that contains several events
What are the benefits and disadvantages of composite endpoints?
benefits:
-less subjects needed
disadvantages:
-hard to determine the true effect of the intervention on each of the event types
What are the results of doing randomization correctly?
ensures groups are similar in all aspects
-similar in things we can measure (age, sex, etc)
-similar in things we cant measure (genetics)
What is an important aspect of randomization?
should not be able to tell which group will be assigned next
What are the three methods of randomization?
complete
block
stratified
What is complete randomization?
no limitations-complete randomization
wont necessarily get equal numbers in each group
What is block randomization?
used to force balance in the number of subjects in each group
ex: block of 6=for every 6 subjects randomized,3 in intervention and 3 in control
What is stratified randomization?
used to achieve similarities in certain baseline characteristics
usually randomization is centralized and offsite (allocation concealment)
Differentiate between ITT and per protocol.
ITT: analyzing the data for ALL patients and according to the group they were originally randomized to
-preserves the value of randomization
per protocol: analyzing data from subjects who completed the study or followed protocol exactly
What is blinding?
unaware of what study group a subject is in
What can occur if group assignment is known?
subjects may report outcomes or AE differently or behave differently
investigators may report outcomes differently or treat subjects differently
researches may collect data, classify outcomes, and interpret responses differently
True or false: the more subjective an outcome is, the less critical blinding is
false
the more OBJECTIVE the outcome is, the less critical blinding is
What is the placebo effect?
perceived or actual effect from an ineffectual or inactive treatment
-conscious belief about drug/treatment
-subconscious association between being treated and recovery
-can be positive or negative
-sugar pill, saline, sham procedure
