Intention-to-Treat
Definition
Analysing every participant in the arm they were randomised to, whether or not they finished. It preserves the randomisation and usually gives a more conservative answer than analysing completers only.
What this means for research sourcing
Understanding evidence quality helps researchers prioritize compounds. Volta's research library summarizes the literature behind stocked compounds.
Related Terms in Evidence & Research
Endpoint
A defined, measurable outcome a trial uses to judge an intervention. Which endpoints were chosen, and when, is as informative as the results reported against them.
Primary Endpoint
The single outcome a trial was designed and powered to answer, fixed before data collection. Secondary endpoints are informative and are not what the trial succeeded or failed on, a distinction that press releases regularly blur.
Biomarker
A measurable indicator standing in for a process, a disease state or a response. Serum IGF-1 is the usual biomarker in growth hormone secretagogue work, and a biomarker moving is not the same as an outcome improving.
HbA1c (Glycated Hemoglobin)
A blood measure reflecting average glucose over roughly the preceding two to three months, reported as a percentage. It is the standard primary endpoint in metabolic trials, which is why a result quoted in points of HbA1c appears in almost every incretin paper.
Effect Size
How large the difference an intervention produced actually was, in real units or as a standardised measure. Once significance is established this is the only question left that matters.
Statistical Significance
A judgement that a result is unlikely to have arisen by chance under the null hypothesis, conventionally at p below 0.05. It says nothing whatever about whether the difference is big enough to care about.
p-Value
The probability of seeing a result at least as extreme as the one observed if the null hypothesis were true. Below 0.05 is the convention for significance, and it is not the probability that the hypothesis is wrong.
Confidence Interval
The range of values compatible with the data at a stated level, usually 95%. A wide interval means an imprecise estimate however striking the central figure looks, and it is the more honest number to read first.