Article Text

PDF

GLOSSARY

Statistics from Altmetric.com

Terms used in therapeutics

Allocation concealed: the study adequately conceals study group allocation from those responsible for assessing participants at trial entry.

Allocation not concealed: study group allocation was known to those assessing participants for entry in the trial.

Unclear allocation concealment: the article does not contain a full description of how participants were allocated.

Blinded (or `masked'): clinicians, participants, outcome assessors or statisticians were not told who received the treatment or control interventions. In abstracts, those blinded are explicitly indicated.

Unblinded: clinicians, participants, outcome assessors and statisticians knew how interventions and controls were allocated.

Terms used in systematic reviews

Heterogeneity: occurs when there is more variation between the study results than would be expected by chance alone. In the presence of heterogeneous results, a random effects pooled odds ratio is usually preferred over a fixed effects pooled odds ratio.

Odds ratio (OR): the `odds' of an event is the ratio of the chance of it occurring to the chance of it not occurring. The odds ratio is a measure of the relative benefit of the experimental treatment, comparing the odds of an event in the intervention group to the odds of an event in the control group.

Terms used in diagnosis

Sensitivity: the sensitivity of a test is the proportion of people with the condition of interest who test positive (according to the diagnostic gold standard).

Specificity: the proportion of people who do not have the condition of interest who test negative (according to the diagnostic gold standard).

Likelihood ratio for a positive test result/positive predictive value: the likelihood that a positive test comes from a person with the condition of interest rather than one without the disorder.

Likelihood ratio for a negative test result/negative predictive value: the likelihood that a negative test comes from a person with the condition of interest rather than someone without the disorder.

Terms used when presenting data

ABI (absolute benefit increase)/ARR (absolute risk reduction) /ARI (absolute risk increase): the absolute arithmetic difference in rates of an outcome between the intervention and control groups.

NNT (number needed to treat)/NNH (number needed to harm): the number of people who need to receive the experimental treatment to create one additional improved outcome (NNT) or one additional harmful outcome (calculated as 1/ARR). The lower the NNT, the more effective the intervention.

RBI (relative benefit increase)/RRR (relative risk reduction)/ RRI (relative risk increase): the proportional increase or decrease in events between the intervention and control groups. RRI relates to negative or harmful events.

Confidence interval (CI): the range of values within which the true value for the population is expected to lie with a given degree of certainty (usually reported as 95% CI).

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.