CBL-514 Results: Percentages and Denominators
A CBL-514 percentage is only interpretable when its numerator, denominator, endpoint and time point are known. Responder rates, relative changes and between-group differences answer different questions.
Three percentages that are easy to confuse
A responder rate is the fraction of people meeting a defined rule. A relative change compares a measurement with its baseline. A percentage-point difference subtracts one percentage from another. None can replace the others without changing the claim.
| Hypothetical example | Correct description |
|---|---|
| 12 of 20 participants meet a threshold | 60% responder rate |
| A measurement falls from 100 to 80 units | 20% relative reduction |
| Response rates are 60% and 20% | 40 percentage-point difference |
These numbers are invented for explanation. They are not CBL-514 trial results.
Keep enrollment separate from analysis
ClinicalTrials.gov organizes results into participant flow and outcome sections, including the people represented in an analysis.[1] Review these before using a trialβs total enrollment as the denominator.
For example, if a hypothetical trial enrolls 30 people but reports a measurement in 20 at a later visit, dividing a reported responder count by 30 may not reproduce the published analysis. The missing ten observations and their handling matter. Neither ignoring them nor guessing their outcomes is an acceptable shortcut.
A four-part annotation for every result
Write the endpoint, visit, analysis population and uncertainty beside the estimate. If a result is adjusted by a statistical model, preserve that fact. An adjusted mean need not equal the simple average a reader could calculate from a visible subset of observations.
A small p-value does not tell you the size or practical importance of a change. For interpretation, look at the estimated difference, its uncertainty and the outcomeβs meaning. Statistical evidence and clinical relevance are related questions, not identical labels.
This approach makes CBL-514 summaries easier to audit. A reader can trace a number to the table that generated it, rather than having to infer which definition was intended.
Frequently asked questions
Is a 40 percentage-point difference the same as a 40% relative increase?
No. Percentage points subtract rates; relative change divides a difference by a stated reference value.
Can I calculate a missing trial result from a headline?
Usually not reliably. Locate the underlying endpoint, denominator and analysis rules first.
CBL-514 product and documentation
Review the current format and pack choices, then match the supplied lot to its documentation. No batch-specific identity, purity or content result is asserted in this article.
