Premium
Statistical modelling of the differences between successive R–R intervals
Author(s) -
Mandrekar Sumithra J.,
Nagaraja Haikady N.,
Berntson Gary G.
Publication year - 2005
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.1948
Subject(s) - statistics , mathematics , weibull distribution , normality , confidence interval
Understanding the behaviour of R–R interval data and its successive differences is critical to the dynamics of cardiac control. Several time domain measures that quantify R–R interval variability have important clinical significance in terms of risk stratification and evaluating the effectiveness of treatment procedures. The present approach at examining the distributions of successive beat‐to‐beat differences of R–R interval data from different populations and under different conditions (baseline and reaction times) provides a valuable insight into their previously unexplored distributional properties. In particular, our analysis reveals that the successive differences have non‐normal statistical distributions (a contradiction to the commonly used assumption of normality), and the absolute successive R–R interval differences approximately follows a Weibull distribution. As an illustration of the utility of this approach, we explore the statistical properties of the time domain measure: root mean square successive difference, study the association between the Weibull scale parameter estimate and respiratory sinus arrhythmia, and propose improvements in artifact detection algorithms. Copyright © 2004 John Wiley & Sons, Ltd.