This is the Continuous Integration Build of FHIR (will be incorrect/inconsistent at times).
See the Directory of published versions
Clinical Decision Support Work Group | Maturity Level: N/A | Standards Status: Informative | Compartments: No defined compartments |
This is the narrative for the resource. See also the XML, JSON or Turtle format. This example conforms to the profile Evidence.
Generated Narrative: Evidence example-statistic-model-include-if
version: 9; Last updated: 2025-03-10T14:59:32.227Z
Profile: http://hl7.org/fhir/uv/ebm/StructureDefinition/statistic-model
StructureDefinition Work Group: cds
url: https://fevir.net/resources/Evidence/346428
identifier: FEvIR Object Identifier/https://fevir.net/FOI/346428
title: Example of statistical analysis plan with different statistical test if non-normal distribution
status: active
author: Brian S. Alper, MD, MSPH:
publisher: HL7 International / Clinical Decision Support
contact: http://www.hl7.org/Special/committees/dss
copyright:
https://creativecommons.org/licenses/by-nc-sa/4.0/
description:
This example uses the StatisticModelIncludeIf extension to represent the model characteristic of Welch's 2-sample t-test if the data is consistent with a normal distribution (based on Shapiro-Wilk test with p-value >= 0.05) and a Wilcoxon Mann-Whitney test if not.
variableDefinition
description:
Adults with hypertension
variableRole: population
variableDefinition
description:
Reversitol vs. Placebo
variableRole: exposure
variableDefinition
description:
Change in systolic blood pressure from baseline to 3 months
variableRole: outcome
studyDesign: randomized assignment
statistic
statisticType: difference in means
modelCharacteristic
Statistic Model Include If
- attribute: data distribution
- value: normal distribution (Shapiro-Wilk test with p-value >= 0.05)
code: Welch's 2-sample t-test
intended: true
modelCharacteristic
Statistic Model Include If
- attribute: data distribution
- value: not consistent with normal distribution (Shapiro-Wilk test with p-value < 0.05)
code: Wilcoxon Mann-Whitney test
intended: true
Usage note: every effort has been made to ensure that the examples are correct and useful, but they are not a normative part of the specification.