Cross-national health comparisons using the Rasch model

findings from the 2012 US Health and Retirement Study and the 2012 Mexican Health and Aging Study

Ickpyo Hong, Timothy Reistetter, Carlos Díaz-Venegas, Alejandra Michaels-Obregon, Rebeca Wong

Research output: Contribution to journalArticle

Abstract

Purpose: Cross-national comparisons of patterns of population aging have emerged as comparable national micro-data have become available. This study creates a metric using Rasch analysis and determines the health of American and Mexican older adult populations. Methods: Secondary data analysis using representative samples aged 50 and older from 2012 U.S. Health and Retirement Study (n = 20,554); 2012 Mexican Health and Aging Study (n = 14,448). We developed a function measurement scale using Rasch analysis of 22 daily tasks and physical function questions. We tested psychometrics of the scale including factor analysis, fit statistics, internal consistency, and item difficulty. We investigated differences in function using multiple linear regression controlling for demographics. Lastly, we conducted subgroup analyses for chronic conditions. Results: The created common metric demonstrated a unidimensional structure with good item fit, an acceptable precision (person reliability = 0.78), and an item difficulty hierarchy. The American adults appeared less functional than adults in Mexico (β = − 0.26, p < 0.0001) and across two chronic conditions (arthritis, β = − 0.36; lung problems, β = − 0.62; all p < 0.05). However, American adults with stroke were more functional than Mexican adults (β = 0.46, p = 0.047). Conclusions: The Rasch model indicates that Mexican adults were more functional than Americans at the population level and across two chronic conditions (arthritis and lung problems). Future studies would need to elucidate other factors affecting the function differences between the two countries.

Original languageEnglish (US)
Pages (from-to)1-11
Number of pages11
JournalQuality of Life Research
DOIs
StateAccepted/In press - May 10 2018

Fingerprint

Retirement
Health
Arthritis
Population
Lung
Mexico
Psychometrics
Statistical Factor Analysis
Linear Models
Stroke
Demography

Keywords

  • Cross-cultural comparison
  • Disability
  • Health
  • Outcome measure
  • Rasch model

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health

Cite this

Cross-national health comparisons using the Rasch model : findings from the 2012 US Health and Retirement Study and the 2012 Mexican Health and Aging Study. / Hong, Ickpyo; Reistetter, Timothy; Díaz-Venegas, Carlos; Michaels-Obregon, Alejandra; Wong, Rebeca.

In: Quality of Life Research, 10.05.2018, p. 1-11.

Research output: Contribution to journalArticle

@article{bcaa5817210840b18797bbe11d9640dc,
title = "Cross-national health comparisons using the Rasch model: findings from the 2012 US Health and Retirement Study and the 2012 Mexican Health and Aging Study",
abstract = "Purpose: Cross-national comparisons of patterns of population aging have emerged as comparable national micro-data have become available. This study creates a metric using Rasch analysis and determines the health of American and Mexican older adult populations. Methods: Secondary data analysis using representative samples aged 50 and older from 2012 U.S. Health and Retirement Study (n = 20,554); 2012 Mexican Health and Aging Study (n = 14,448). We developed a function measurement scale using Rasch analysis of 22 daily tasks and physical function questions. We tested psychometrics of the scale including factor analysis, fit statistics, internal consistency, and item difficulty. We investigated differences in function using multiple linear regression controlling for demographics. Lastly, we conducted subgroup analyses for chronic conditions. Results: The created common metric demonstrated a unidimensional structure with good item fit, an acceptable precision (person reliability = 0.78), and an item difficulty hierarchy. The American adults appeared less functional than adults in Mexico (β = − 0.26, p < 0.0001) and across two chronic conditions (arthritis, β = − 0.36; lung problems, β = − 0.62; all p < 0.05). However, American adults with stroke were more functional than Mexican adults (β = 0.46, p = 0.047). Conclusions: The Rasch model indicates that Mexican adults were more functional than Americans at the population level and across two chronic conditions (arthritis and lung problems). Future studies would need to elucidate other factors affecting the function differences between the two countries.",
keywords = "Cross-cultural comparison, Disability, Health, Outcome measure, Rasch model",
author = "Ickpyo Hong and Timothy Reistetter and Carlos D{\'i}az-Venegas and Alejandra Michaels-Obregon and Rebeca Wong",
year = "2018",
month = "5",
day = "10",
doi = "10.1007/s11136-018-1878-4",
language = "English (US)",
pages = "1--11",
journal = "Quality of Life Research",
issn = "0962-9343",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Cross-national health comparisons using the Rasch model

T2 - findings from the 2012 US Health and Retirement Study and the 2012 Mexican Health and Aging Study

AU - Hong, Ickpyo

AU - Reistetter, Timothy

AU - Díaz-Venegas, Carlos

AU - Michaels-Obregon, Alejandra

AU - Wong, Rebeca

PY - 2018/5/10

Y1 - 2018/5/10

N2 - Purpose: Cross-national comparisons of patterns of population aging have emerged as comparable national micro-data have become available. This study creates a metric using Rasch analysis and determines the health of American and Mexican older adult populations. Methods: Secondary data analysis using representative samples aged 50 and older from 2012 U.S. Health and Retirement Study (n = 20,554); 2012 Mexican Health and Aging Study (n = 14,448). We developed a function measurement scale using Rasch analysis of 22 daily tasks and physical function questions. We tested psychometrics of the scale including factor analysis, fit statistics, internal consistency, and item difficulty. We investigated differences in function using multiple linear regression controlling for demographics. Lastly, we conducted subgroup analyses for chronic conditions. Results: The created common metric demonstrated a unidimensional structure with good item fit, an acceptable precision (person reliability = 0.78), and an item difficulty hierarchy. The American adults appeared less functional than adults in Mexico (β = − 0.26, p < 0.0001) and across two chronic conditions (arthritis, β = − 0.36; lung problems, β = − 0.62; all p < 0.05). However, American adults with stroke were more functional than Mexican adults (β = 0.46, p = 0.047). Conclusions: The Rasch model indicates that Mexican adults were more functional than Americans at the population level and across two chronic conditions (arthritis and lung problems). Future studies would need to elucidate other factors affecting the function differences between the two countries.

AB - Purpose: Cross-national comparisons of patterns of population aging have emerged as comparable national micro-data have become available. This study creates a metric using Rasch analysis and determines the health of American and Mexican older adult populations. Methods: Secondary data analysis using representative samples aged 50 and older from 2012 U.S. Health and Retirement Study (n = 20,554); 2012 Mexican Health and Aging Study (n = 14,448). We developed a function measurement scale using Rasch analysis of 22 daily tasks and physical function questions. We tested psychometrics of the scale including factor analysis, fit statistics, internal consistency, and item difficulty. We investigated differences in function using multiple linear regression controlling for demographics. Lastly, we conducted subgroup analyses for chronic conditions. Results: The created common metric demonstrated a unidimensional structure with good item fit, an acceptable precision (person reliability = 0.78), and an item difficulty hierarchy. The American adults appeared less functional than adults in Mexico (β = − 0.26, p < 0.0001) and across two chronic conditions (arthritis, β = − 0.36; lung problems, β = − 0.62; all p < 0.05). However, American adults with stroke were more functional than Mexican adults (β = 0.46, p = 0.047). Conclusions: The Rasch model indicates that Mexican adults were more functional than Americans at the population level and across two chronic conditions (arthritis and lung problems). Future studies would need to elucidate other factors affecting the function differences between the two countries.

KW - Cross-cultural comparison

KW - Disability

KW - Health

KW - Outcome measure

KW - Rasch model

UR - http://www.scopus.com/inward/record.url?scp=85046803191&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046803191&partnerID=8YFLogxK

U2 - 10.1007/s11136-018-1878-4

DO - 10.1007/s11136-018-1878-4

M3 - Article

SP - 1

EP - 11

JO - Quality of Life Research

JF - Quality of Life Research

SN - 0962-9343

ER -