EBM Terminology Flashcards
P-value
This is an arbitrary number chosen by the investigators which determines the “level” below which we are sure that the results of the study were due to chance alone.
*this level is commonly chosen as p < 0.05, meaning there is less than a 5% chance that the results in this study were due to chance alone. Or these data are “statistically significant.”
Hazard Ratios
the hazard ratio (HR) is the ratio of the hazard rates corresponding to the conditions described by two levels of an explanatory variable.
Hazard Ratios vs. Relative Risks
Hazard ratios differ from relative risks in that the latter are cumulative over an entire study, using a defined endpoint, while the former represent instantaneous risk over the study time period, or some subset thereof.
Hazard Ratios vs. Relative Risks explained another way…
A relative risk considers the entire length of the study for a specific outcome to occur.
A hazard ratio is the risk of developing an outcome at some particular point during the study.
They are both ways of expressing the strength/weight of the outcome.
Control Event Rate (CER)
Control event rate (CER) –> experimental event rate/ control event rate == relative risk reduction
CER –EER |/CER = RRR
Absolute Risk Reduction (ARR)
CER-ERR | = ARR
Number Needed to Treat (NNT)
NNT = 1/ARR
Odds
A ratio of the number of people incurring an outcome event to the number of people who don’t have an event.
Odds Ratio
The ratio of the odds of having the target disorder in the experimental group relative to the odds in favor of having the target disorder in the control group. (Usually used in cohort studies or systematic reviews
Odds and Risks (Probabilities)
Odds = Probability / 1 - probability
*Example: When probability is 80%,
Odds = 0.80/1-0.80 or 0.80/0.20 = 4 (or odds of 4:1)
Probability = Odds / 1 + Odds
*Example: When odds are 4,
Probability = 4/1+4 = 0.80 or 80%
What is the odds ratio in a treatment trial?
Odds ratio = ad/cb
95% Confidence Interval
It basically says, if you repeat this study 100 times the results will fall between these limits 95% of the time.
When the 95% CI cross 1…
the study can be considered non-significant because you can’t tell whether the treatment helps or hurts the patient.
95% CI (3-32) …
means that the difference between the means of the groups being compared could be as low as 3 or as high as 32.
Non-inferiority trials
The investigators “arbitrarily” choose a % or margin that the drug being tested can be worse than the “standard” drug, and still be considered non-inferior.
