Continuous Variable Relationship Analysis — continuousrelationship • ClinicoPath

Analyzes relationships between continuous variables in clinical research using evidence-based approaches from BMJ best practices. Avoids common pitfalls like arbitrary categorization and linear assumptions.

Usage

continuousrelationship(
  data,
  outcome,
  predictor,
  covariates = NULL,
  modelType = "spline",
  nKnots = "4",
  knotPositions = "quantile",
  plotType = "doseresponse",
  referenceValue,
  showCI = TRUE,
  showRug = TRUE,
  showGuidance = TRUE,
  compareWithLinear = TRUE,
  showCategorizedPitfall = FALSE,
  testLinearity = TRUE,
  showModelFit = TRUE,
  showModel = FALSE
)

Arguments

data: The dataset containing clinical variables
outcome: The outcome/dependent variable (can be continuous or binary)
predictor: The continuous predictor variable to analyze
covariates: Variables to adjust for in the analysis
modelType: Statistical approach for modeling the relationship. Splines are recommended for most clinical applications.
nKnots: Number of knots for spline models. More knots = more flexibility
knotPositions: How to position knots across the predictor range
plotType: Type of visualization to create
referenceValue: Reference value for the predictor (e.g., BMI=25). If not specified, uses median value.
showCI: Display 95 percent confidence intervals on plots
showRug: Show rug plot of actual data points
showGuidance: Display educational guidance based on BMJ best practices to help interpret results and avoid common mistakes
compareWithLinear: Show comparison with simple linear model to demonstrate the importance of flexible modeling
showCategorizedPitfall: Educational: Show what happens when continuous variables are inappropriately categorized (with warning)
testLinearity: Perform statistical test for non-linear relationship
showModelFit: Display AIC, BIC, and other fit statistics
showModel: Display the model formula as HTML in the results

Value

A results object containing:

`results$guidance`					a html
`results$warningMessage`					a html
`results$modelFormula`					a html
`results$linearityTest`					a table
`results$modelFit`					a table
`results$doseResponsePlot`					an image
`results$partialEffectPlot`					an image
`results$comparisonPlot`					an image
`results$categorizedPlot`					an image
`results$interpretationGuide`					a preformatted

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$linearityTest$asDF

as.data.frame(results$linearityTest)

Examples

# \donttest{
# Example: Analyzing BMI vs mortality risk
continuousrelationship(
    data = clinical_data,
    outcome = "mortality",
    predictor = "bmi",
    modelType = "spline",
    nKnots = 4,
    showGuidance = TRUE
)
#> Error in continuousrelationship(data = clinical_data, outcome = "mortality",     predictor = "bmi", modelType = "spline", nKnots = 4, showGuidance = TRUE): argument "referenceValue" is missing, with no default
# }