Sample size estimation for GEE method for comparing slopes in repeated measurements data

Sin Ho Jung, Chul Ahn

Research output: Contribution to journalArticle

70 Citations (Scopus)

Abstract

Sample size calculation is an important component at the design stage of clinical trials. Controlled clinical trials often use a repeated measurement design in which individuals are randomly assigned to treatment groups and followed-up for measurements at intervals across a treatment period of fixed duration. In studies with repeated measurements, one of the popular primary interests is the comparison of the rates of change in a response variable between groups. Statistical models for calculating sample sizes for repeated measurement designs often fail to take into account the impact of missing data correctly. In this paper we propose to use the generalized estimating equation (GEE) method in comparing the rates of change in repeated measurements and introduce closed form formulae for sample size and power that can be calculated, using a scientific calculator. Since the sample size formula is based on asymptotic theory, we investigate the performance of the estimated sample size in practical settings through simulations.

Original languageEnglish (US)
Pages (from-to)1305-1315
Number of pages11
JournalStatistics in Medicine
Volume22
Issue number8
DOIs
StatePublished - Apr 30 2003

Fingerprint

Repeated Measurements
Generalized Estimating Equations
Sample Size
Slope
Repeated Measurement Designs
Rate of change
Clinical Trials
Sample Size Calculation
Calculator
Asymptotic Theory
Missing Data
Statistical Model
Controlled Clinical Trials
Statistical Models
Closed-form
Interval
Simulation

Keywords

  • AR(1)
  • Compound symmetry
  • Independent missing
  • Independent working correlation
  • Missing completely at random
  • Monotone missing

ASJC Scopus subject areas

  • Epidemiology

Cite this

Sample size estimation for GEE method for comparing slopes in repeated measurements data. / Jung, Sin Ho; Ahn, Chul.

In: Statistics in Medicine, Vol. 22, No. 8, 30.04.2003, p. 1305-1315.

Research output: Contribution to journalArticle

@article{fb5ed743070c48f0b7e0d1bef68e109f,
title = "Sample size estimation for GEE method for comparing slopes in repeated measurements data",
abstract = "Sample size calculation is an important component at the design stage of clinical trials. Controlled clinical trials often use a repeated measurement design in which individuals are randomly assigned to treatment groups and followed-up for measurements at intervals across a treatment period of fixed duration. In studies with repeated measurements, one of the popular primary interests is the comparison of the rates of change in a response variable between groups. Statistical models for calculating sample sizes for repeated measurement designs often fail to take into account the impact of missing data correctly. In this paper we propose to use the generalized estimating equation (GEE) method in comparing the rates of change in repeated measurements and introduce closed form formulae for sample size and power that can be calculated, using a scientific calculator. Since the sample size formula is based on asymptotic theory, we investigate the performance of the estimated sample size in practical settings through simulations.",
keywords = "AR(1), Compound symmetry, Independent missing, Independent working correlation, Missing completely at random, Monotone missing",
author = "Jung, {Sin Ho} and Chul Ahn",
year = "2003",
month = "4",
day = "30",
doi = "10.1002/sim.1384",
language = "English (US)",
volume = "22",
pages = "1305--1315",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "8",

}

TY - JOUR

T1 - Sample size estimation for GEE method for comparing slopes in repeated measurements data

AU - Jung, Sin Ho

AU - Ahn, Chul

PY - 2003/4/30

Y1 - 2003/4/30

N2 - Sample size calculation is an important component at the design stage of clinical trials. Controlled clinical trials often use a repeated measurement design in which individuals are randomly assigned to treatment groups and followed-up for measurements at intervals across a treatment period of fixed duration. In studies with repeated measurements, one of the popular primary interests is the comparison of the rates of change in a response variable between groups. Statistical models for calculating sample sizes for repeated measurement designs often fail to take into account the impact of missing data correctly. In this paper we propose to use the generalized estimating equation (GEE) method in comparing the rates of change in repeated measurements and introduce closed form formulae for sample size and power that can be calculated, using a scientific calculator. Since the sample size formula is based on asymptotic theory, we investigate the performance of the estimated sample size in practical settings through simulations.

AB - Sample size calculation is an important component at the design stage of clinical trials. Controlled clinical trials often use a repeated measurement design in which individuals are randomly assigned to treatment groups and followed-up for measurements at intervals across a treatment period of fixed duration. In studies with repeated measurements, one of the popular primary interests is the comparison of the rates of change in a response variable between groups. Statistical models for calculating sample sizes for repeated measurement designs often fail to take into account the impact of missing data correctly. In this paper we propose to use the generalized estimating equation (GEE) method in comparing the rates of change in repeated measurements and introduce closed form formulae for sample size and power that can be calculated, using a scientific calculator. Since the sample size formula is based on asymptotic theory, we investigate the performance of the estimated sample size in practical settings through simulations.

KW - AR(1)

KW - Compound symmetry

KW - Independent missing

KW - Independent working correlation

KW - Missing completely at random

KW - Monotone missing

UR - http://www.scopus.com/inward/record.url?scp=0037739710&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037739710&partnerID=8YFLogxK

U2 - 10.1002/sim.1384

DO - 10.1002/sim.1384

M3 - Article

C2 - 12687656

AN - SCOPUS:0037739710

VL - 22

SP - 1305

EP - 1315

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 8

ER -