The glossary below provides definitions for several terms found in the structured abstracts of theEvidence-Based Orthopaedics section, many of which may not be familiar to orthopaedic surgeons.
The following terms are commonly used in abstracts describing trials of therapeutic interventions.
EER is the experimental (new treatment) event rate.
CER is the control (old treatment or nontreatment) event rate.
CI (confidence interval) quantifies the uncertainty in measurement. CI is usually reported as a 95% CI, which is the range of values within which we can be 95% sure that the true value for the whole population lies.
When the experimental treatment reducesthe risk for a bad event:
ARR (absolute risk reduction) is the absolute arithmetic difference in bad event rates between the experimental and control groups, calculated as |EER - CER|.
RRR (relative risk reduction) is the proportional reduction in bad event rates between the experimental and control groups, calculated as |EER - CER|/CER and accompanied by a 95% CI.
When trying to interpret the benefit of a new therapy, the NNT can be helpful to clinicians.
NNT (number needed to treat) is the number of patients who need to be treated to prevent 1 additional bad outcome, calculated as1/ARR, rounded up to the nearest whole number, and accompanied by a 95% CI.
When the experimental treatment increases the probability of a good event:
ABI (absolute benefit increase) is the absolute arithmetic difference in good event rates between the experimental and control groups, calculated as |EER -CER|.
RBI (relative benefit increase) is the proportional increase in good event rates between the experimental and control groups, calculated as |EER - CER|/CER and accompanied by a 95% CI.
NNT (number needed to treat) is thenumber of patients who need to be treatedto achieve 1 additional favorable outcome, calculated as 1/ARR, rounded up to thenearest whole number, and accompanied by a 95% CI.
When the experimental treatment increases the probability of a bad event:
ARI (absolute risk increase) is the absolute arithmetic difference in bad event ratesbetween groups, when the experimental treatment harms more patients thanthe control treatment, calculated as|EER - CER|.
RRI (relative risk increase) is the proportional increase in bad event rates between the experimental and control groups,calculated as |EER - CER|/CER and accompanied by a 95% CI.
NNH (number needed to harm) is thenumber of patients that, if they received the experimental treatment, would lead to 1additional person being harmed compared with the number of patients who receivedthe control treatment, calculated as 1/ARI, rounded up to the nearest whole number, and accompanied by a 95% CI.