Interpreting Study Results Flashcards
How are results typically presented?
means
proportions
-simple percentages
-risk ratio
-odds ratio
time to an outcome
-survival curve
-hazard ratio
What is the mean?
continuous outcomes
compare the average of the outcome between groups or change from baseline
What are proportions?
the fraction of the total that possesses the outcome
compare the proportion that have the outcome between groups
What is the absolute risk reduction (ARR) or absolute risk difference (ARD)?
the absolute difference between the probability of the event in the control group and probability of the event in the intervention group
What is the relative risk (RR)?
dichotomous outcomes
shows what the “risk” of the outcome in the intervention group is compared to the risk in the control group
probability of event (intervention)/probability of event (control)
What does the relative risk number tell us?
RR=1
-no difference in risk between groups
RR<1.0
-less risk out outcome in intervention group
RR>1.0
-higher risk of outcome in intervention group
What is the relative risk reduction (RRR)?
the degree to which baseline risk is reduced (or increased) by the intervention
RRR=1-RR
What is the odds ratio (OR)?
shows the odds of the outcome occurring in the intervention group compared to the control group
odds (intervention)/odds (control)
What does the OR tell us?
OR=1.0
-no difference in odds between groups
OR<1.0
-less odds of outcome in intervention group
OR>1.0
-higher odds of outcome in intervention group
What is the number needed to treat (NNT)?
number of subjects who would have to be treated (receive the intervention) in order for one additional subject to “benefit” in comparison to the control
NNT=100/ARD
must take into account the duration of the study
What is the number needed to harm (NNH)?
number of subjects who would be treated before you see one additional subject with an adverse effect compared to control
NNH=100/ARD
What are Kaplan-Meier survival curves?
compares how long it takes subjects in each group to reach the outcome
often reported as “median survival time”
-time for half the subjects to reach the outcome
-p value will tell you if theres a statistically significant difference between the groups
What does the hazard ratio tell us?
HR=1.0
-no difference in hazard between groups
HR<1.0
-less hazard of outcome in intervention group
HR>1.0
-higher hazard of outcome in intervention group
What are confidence intervals?
range where the true effect of the intervention (treatment) lies
the narrower the CI, the more precise the results
increase sample size=increase precision
What can CI show us?
magnitude of the effect
-best case-worse-case scenario
if there is really is a difference
-if the CI crosses the threshold for “no difference”, then the results are not statistically significant
-for means/proportions: no difference=0
-for RR or OR: no difference=1
Which is better, the 95% CI or p-value?
reporting CI is becoming more common/required
CI show us the range in the actual data measurements
CI shows us the direction of effect, p-value just indicates if a difference exists
CI can help distinguish between clinical and statistical significance
Differentiate between superiority, non-inferiority and equivalence.
superiority:
-want to show an intervention is better than control (active/placebo)
-is there statically significant/clinically relevant difference between groups?
non-inferiority:
-is the new intervention not substantially worse than an established intervention?
equivalence:
-is the new intervention neither worse nor better than an established intervention?
What are non-inferiority trials?
clinical trial to establish that an intervention is not clinically worse than a comparison by more than a pre-determined margin
When should the non-inferiority margin be determined?
a priori
-prevents bias
-based on a combination of statistical reasoning and clinical judgement
What kind of analyses should be done and reported for non-inferiority trials?
ITT
per protocol
confidence intervals are key
What would happen if only ITT was reported for a non-inferiority trial?
it can bias towards non-inferiority if lots of participants are lost to follow-up or have protocol violations
-dont get to observe the maximal effect of the drug
What would happen if only per protocol was reported for a non-inferiority trial?
problems with confounding, power, etc
What can be done once non-inferiority is established?
an analysis for potential superiority can be conducted
True or false: lack of superiority in an superiority trial design means non-inferiority
false
lack of superiority in a superiority trial design does not mean non-inferiority and cannot claim non-inferiority
What is a propensity score?
the probability that a subject would be in a particular treatment group based on their observed baseline characteristics
What is the goal of propensity scores?
mimic RCTs by balancing observed characteristics between study groups
What do propensity scores help to reduce?
selection bias and confounding in observational studies
True or false: propensity scores replace the value of randomization
false
What is an adaptive clinical trial design?
clinical trial design that allows for prospectively planned modifications to one or more aspects of the design based on accumulating data from subjects in that trial
What are examples of pre-planned changes for adaptive clinical trial designs? What are the benefits of these changes?
adjusting sample size
-prevent underpowered studies
-prevent unnecessary exposure to subjects
stopping treatments arms/groups
-prevent unnecessary exposure to useless/harmful treatment
-prevent wasting resources and time
changing allocation of subjects to study groups
-fewer subjects randomized to less beneficial/useless treatment
-more subjects receive beneficial treatment
identifying subjects who are likely to benefit and focus recruitment efforts on them
-identify subjects likely to benefit
stopping study early because of obvious benefit or harm
-prevent unnecessary exposure
-prevent wasting resources and time
-quicker dissemination of positive findings
What do interim data analyses create the potential for?
operational bias
