Robust Respondents and Lost Limitations: The Implications of Nonrandom Missingness for the Estimation of Health Trajectories

Heide Jackson, Michal Engelman, Karen J Bandeen Roche

Research output: Contribution to journalArticle

Abstract

Objective: We offer a strategy for quantifying the impact of mortality and attrition on inferences from later-life health trajectory models. Method: Using latent class growth analysis (LCGA), we identify functional limitation trajectory classes in the Health and Retirement Study. We compare results from complete case and full information maximum likelihood (FIML) analyses, and demonstrate a method for producing upper- and lower-bound estimates of the impact of attrition on results. Results: LCGA inferences vary substantially depending on the handling of missing data. For older adults who die during the follow-up period, the widely used FIML approach may underestimate functional limitations by up to 20%. Discussion: The most commonly used approaches to handling missing data likely underestimate the extent of poor health in aging populations. Although there is no single solution for nonrandom missingness, we show that bounding estimates can help analysts to better characterize patterns of health in later life.

Original languageEnglish (US)
Pages (from-to)685-708
Number of pages24
JournalJournal of Aging and Health
Volume31
Issue number4
DOIs
StatePublished - Apr 1 2019

Fingerprint

Health
health
Retirement
Growth
retirement
mortality
Mortality
Surveys and Questionnaires
Population

Keywords

  • latent class growth analysis
  • longitudinal analysis
  • mortality and attrition bias

ASJC Scopus subject areas

  • Health(social science)
  • Sociology and Political Science
  • Gerontology
  • Community and Home Care
  • Geriatrics and Gerontology
  • Life-span and Life-course Studies

Cite this

Robust Respondents and Lost Limitations : The Implications of Nonrandom Missingness for the Estimation of Health Trajectories. / Jackson, Heide; Engelman, Michal; Bandeen Roche, Karen J.

In: Journal of Aging and Health, Vol. 31, No. 4, 01.04.2019, p. 685-708.

Research output: Contribution to journalArticle

@article{c0e3f1be7bea42b186b11808c5fc5410,
title = "Robust Respondents and Lost Limitations: The Implications of Nonrandom Missingness for the Estimation of Health Trajectories",
abstract = "Objective: We offer a strategy for quantifying the impact of mortality and attrition on inferences from later-life health trajectory models. Method: Using latent class growth analysis (LCGA), we identify functional limitation trajectory classes in the Health and Retirement Study. We compare results from complete case and full information maximum likelihood (FIML) analyses, and demonstrate a method for producing upper- and lower-bound estimates of the impact of attrition on results. Results: LCGA inferences vary substantially depending on the handling of missing data. For older adults who die during the follow-up period, the widely used FIML approach may underestimate functional limitations by up to 20{\%}. Discussion: The most commonly used approaches to handling missing data likely underestimate the extent of poor health in aging populations. Although there is no single solution for nonrandom missingness, we show that bounding estimates can help analysts to better characterize patterns of health in later life.",
keywords = "latent class growth analysis, longitudinal analysis, mortality and attrition bias",
author = "Heide Jackson and Michal Engelman and {Bandeen Roche}, {Karen J}",
year = "2019",
month = "4",
day = "1",
doi = "10.1177/0898264317747079",
language = "English (US)",
volume = "31",
pages = "685--708",
journal = "Journal of Aging and Health",
issn = "0898-2643",
publisher = "SAGE Publications Inc.",
number = "4",

}

TY - JOUR

T1 - Robust Respondents and Lost Limitations

T2 - The Implications of Nonrandom Missingness for the Estimation of Health Trajectories

AU - Jackson, Heide

AU - Engelman, Michal

AU - Bandeen Roche, Karen J

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Objective: We offer a strategy for quantifying the impact of mortality and attrition on inferences from later-life health trajectory models. Method: Using latent class growth analysis (LCGA), we identify functional limitation trajectory classes in the Health and Retirement Study. We compare results from complete case and full information maximum likelihood (FIML) analyses, and demonstrate a method for producing upper- and lower-bound estimates of the impact of attrition on results. Results: LCGA inferences vary substantially depending on the handling of missing data. For older adults who die during the follow-up period, the widely used FIML approach may underestimate functional limitations by up to 20%. Discussion: The most commonly used approaches to handling missing data likely underestimate the extent of poor health in aging populations. Although there is no single solution for nonrandom missingness, we show that bounding estimates can help analysts to better characterize patterns of health in later life.

AB - Objective: We offer a strategy for quantifying the impact of mortality and attrition on inferences from later-life health trajectory models. Method: Using latent class growth analysis (LCGA), we identify functional limitation trajectory classes in the Health and Retirement Study. We compare results from complete case and full information maximum likelihood (FIML) analyses, and demonstrate a method for producing upper- and lower-bound estimates of the impact of attrition on results. Results: LCGA inferences vary substantially depending on the handling of missing data. For older adults who die during the follow-up period, the widely used FIML approach may underestimate functional limitations by up to 20%. Discussion: The most commonly used approaches to handling missing data likely underestimate the extent of poor health in aging populations. Although there is no single solution for nonrandom missingness, we show that bounding estimates can help analysts to better characterize patterns of health in later life.

KW - latent class growth analysis

KW - longitudinal analysis

KW - mortality and attrition bias

UR - http://www.scopus.com/inward/record.url?scp=85063202672&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063202672&partnerID=8YFLogxK

U2 - 10.1177/0898264317747079

DO - 10.1177/0898264317747079

M3 - Article

C2 - 29254422

AN - SCOPUS:85063202672

VL - 31

SP - 685

EP - 708

JO - Journal of Aging and Health

JF - Journal of Aging and Health

SN - 0898-2643

IS - 4

ER -