DBP.RTM: Regression to the mean

In mcdr65/RegToMeanExample: Visualisation Regression to the Mean

Regression to the mean

Description

Regression to the mean

Usage

DBP.RTM(
  mu = 90,
  sigma = 8,
  r = 0.76,
  n = 1000,
  limit = 95,
  TrueChange = 0,
  show.plot = TRUE,
  show.out = FALSE
)

Arguments

mu	The mean DBP at baseline
sigma	The standard deviation at baseline and at follow-up
r	The pre-post correlation
n	sample size
limit	Limit for hypertension
TrueChange	True Change from Baseline. Default is zero.
show.plot	Add plot, default=TRUE
show.out	print results, default=FALSE

Value

A list with

• plot showing equality line and the linear regression slope for all observations

• plot showing extreme subjects only with regression to the mean

• A list with

  • summary I - Summary statistics for all observations

  • summary II - Summary statistics for extreme cases (with hypertension) only

  • correlation Pre-Post correlation or intraclass correlation

  • t-test all

  • t-test extreme

  • model Coefs from simple linear regression (that is, by "fixing" baseline), equals the correlation coefficient for standardized data

  • fraction of regression to the mean, RTM, equals (1-rho)

  • dataI all observations

  • dataII extreme cases

Author(s)

meichtry

References

Senn, S. (2009) Three things that every medical writer should know about statistics. The Write Stuff, 18 (3). pp. 159-162. ISSN 1854-8466

Examples

##Assume a true change of -10 (on diastolic blood presssure)
DBP.RTM(TrueChange=-10)

mcdr65/RegToMeanExample documentation built on April 27, 2024, 4 a.m.