Population inference with mortality and attrition in longitudinal studies on aging

A two-stage multiple imputation method

Ofer Harel, Scott Hofer, Lesa Hoffman, Nancy L. Pedersen, Boo Johansson

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

A major challenge for inference regarding aging-related change in longitudinal studies is that of study attrition and population mortality. Inferences in longitudinal studies can account for attrition and mortality-related change as distinct processes, but this is made difficult when follow-up of all individuals (i.e., age at death) is not complete. This is a common problem because most longitudinal studies of aging either have incomplete follow-up or are still collecting data on subsequent outcomes, including time of death. A statistical approach is suggested for including time-to-death as a predictor in models with incomplete follow-up using a two-stage multiple-imputation procedure. An empirical example using data from the OCTO-Twin study is presented that shows the utility of his procedure for making inferences conditional on mortality when mortality data are incomplete.

Original languageEnglish (US)
Pages (from-to)187-203
Number of pages17
JournalExperimental Aging Research
Volume33
Issue number2
DOIs
StatePublished - Apr 2007
Externally publishedYes

Fingerprint

Longitudinal Studies
Mortality
Population
Twin Studies
Attrition
Inference
Longitudinal Study
Incomplete

ASJC Scopus subject areas

  • Geriatrics and Gerontology
  • Aging
  • Psychology(all)

Cite this

Population inference with mortality and attrition in longitudinal studies on aging : A two-stage multiple imputation method. / Harel, Ofer; Hofer, Scott; Hoffman, Lesa; Pedersen, Nancy L.; Johansson, Boo.

In: Experimental Aging Research, Vol. 33, No. 2, 04.2007, p. 187-203.

Research output: Contribution to journalArticle

Harel, Ofer ; Hofer, Scott ; Hoffman, Lesa ; Pedersen, Nancy L. ; Johansson, Boo. / Population inference with mortality and attrition in longitudinal studies on aging : A two-stage multiple imputation method. In: Experimental Aging Research. 2007 ; Vol. 33, No. 2. pp. 187-203.
@article{a3e9ab5ecb594fef8fc900017f609bf7,
title = "Population inference with mortality and attrition in longitudinal studies on aging: A two-stage multiple imputation method",
abstract = "A major challenge for inference regarding aging-related change in longitudinal studies is that of study attrition and population mortality. Inferences in longitudinal studies can account for attrition and mortality-related change as distinct processes, but this is made difficult when follow-up of all individuals (i.e., age at death) is not complete. This is a common problem because most longitudinal studies of aging either have incomplete follow-up or are still collecting data on subsequent outcomes, including time of death. A statistical approach is suggested for including time-to-death as a predictor in models with incomplete follow-up using a two-stage multiple-imputation procedure. An empirical example using data from the OCTO-Twin study is presented that shows the utility of his procedure for making inferences conditional on mortality when mortality data are incomplete.",
author = "Ofer Harel and Scott Hofer and Lesa Hoffman and Pedersen, {Nancy L.} and Boo Johansson",
year = "2007",
month = "4",
doi = "10.1080/03610730701239004",
language = "English (US)",
volume = "33",
pages = "187--203",
journal = "Experimental Aging Research",
issn = "0361-073X",
publisher = "Taylor and Francis Ltd.",
number = "2",

}

TY - JOUR

T1 - Population inference with mortality and attrition in longitudinal studies on aging

T2 - A two-stage multiple imputation method

AU - Harel, Ofer

AU - Hofer, Scott

AU - Hoffman, Lesa

AU - Pedersen, Nancy L.

AU - Johansson, Boo

PY - 2007/4

Y1 - 2007/4

N2 - A major challenge for inference regarding aging-related change in longitudinal studies is that of study attrition and population mortality. Inferences in longitudinal studies can account for attrition and mortality-related change as distinct processes, but this is made difficult when follow-up of all individuals (i.e., age at death) is not complete. This is a common problem because most longitudinal studies of aging either have incomplete follow-up or are still collecting data on subsequent outcomes, including time of death. A statistical approach is suggested for including time-to-death as a predictor in models with incomplete follow-up using a two-stage multiple-imputation procedure. An empirical example using data from the OCTO-Twin study is presented that shows the utility of his procedure for making inferences conditional on mortality when mortality data are incomplete.

AB - A major challenge for inference regarding aging-related change in longitudinal studies is that of study attrition and population mortality. Inferences in longitudinal studies can account for attrition and mortality-related change as distinct processes, but this is made difficult when follow-up of all individuals (i.e., age at death) is not complete. This is a common problem because most longitudinal studies of aging either have incomplete follow-up or are still collecting data on subsequent outcomes, including time of death. A statistical approach is suggested for including time-to-death as a predictor in models with incomplete follow-up using a two-stage multiple-imputation procedure. An empirical example using data from the OCTO-Twin study is presented that shows the utility of his procedure for making inferences conditional on mortality when mortality data are incomplete.

UR - http://www.scopus.com/inward/record.url?scp=33847696223&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33847696223&partnerID=8YFLogxK

U2 - 10.1080/03610730701239004

DO - 10.1080/03610730701239004

M3 - Article

VL - 33

SP - 187

EP - 203

JO - Experimental Aging Research

JF - Experimental Aging Research

SN - 0361-073X

IS - 2

ER -