Description
Bound an overall mean under supplied missing-outcome limits.
Missing outcomes can leave a mean only partly identified. If every unknown outcome is confined to a defensible range, placing all unknown values at its lower or upper endpoint gives deterministic worst-case limits; this range is different from sampling uncertainty.
Bound an overall mean under supplied missing-outcome limits.
When to use this tool
Use Clinical Research Missing Outcome Bound when checking the stated clinical-research record, allocation, timing, analysis or evidence-basis convention.
How to use the tool and how it works
- Enter observedSum, observedCount, missingCount, minimum, maximum. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read lowerMean, upperMean, width. Use the output convention below.
Bound an overall mean under supplied missing-outcome limits. Require nonnegative whole counts with positive total, maximum >= minimum and observedSum within observedCount times those bounds; lower/upper means substitute each missing value with minimum/maximum, width = missing*(max-min)/total.1
An estimand specifies the treatment-effect question, including intercurrent-event handling. Denominator choices and missing-data assumptions must align with that question.2
| Input field | Meaning and units | Accepted input |
|---|---|---|
| observedSum | Observed Sum | Required. Finite number. |
| observedCount | Observed Count | Required. Safe integer; minimum 0. |
| missingCount | Missing Count | Required. Safe integer; minimum 0. |
| minimum | Minimum | Required. Finite number. |
| maximum | Maximum | Required. Finite number. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| lowerMean | Worst-case mean with every missing outcome at the supplied minimum. | Required. Finite number. |
| upperMean | Worst-case mean with every missing outcome at the supplied maximum. | Required. Finite number. |
| width | missingCount*(maximum-minimum)/(observedCount+missingCount); not a confidence-interval width. | Required. Finite number. |
{
"input": {
"observedSum": 30,
"observedCount": 10,
"missingCount": 2,
"minimum": 0,
"maximum": 10
},
"output": {
"lowerMean": 2.5,
"upperMean": 4.166666666666667,
"width": 1.6666666666666667
}
}Limitations and assumptions
- Bounds are deterministic worst-case limits, not confidence intervals or a justified imputation model; outcome bounds must be defensible.
- Inputs must satisfy both the field types and the cross-field conditions above. Invalid input causes an error; a successful false flag or documented placeholder is a different result.
- Numeric calculations use binary floating point unless exact decimal-integer arithmetic is explicitly stated. No hidden measurement tolerance or decimal rounding is added; nonfinite or out-of-contract results are rejected. Numeric signed-zero results are exported as zero.
Alternative or Complementary methods
Use the complete approved protocol, authenticated source records and qualified clinical/statistical oversight for actual trial design, conduct and interpretation.
References
Similar or alternative tools
- Trial Randomized Treated Denominator Choice
Return a success proportion using randomized or treated subjects.
- Endpoint Missing Visit Denominator
Return evaluable-visit share under a scheduled/attended denominator policy.