Life expectancy with and without cognitive impairment by diabetes status among older Americans

Carlos Díaz-Venegas, Daniel C. Schneider, Mikko Myrskylä, Neil K. Mehta

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Diabetes affects mortality and cognitive functioning. It is not known how diabetes influences life expectancy (LE) with and without cognitive impairment. We seek to examine age at onset of cognitive impairment and life expectancy (LE) with and without cognitive impairment by diabetes status among middle- and older-aged Americans. Data come from the U.S. Health and Retirement Study 2000–2012 linked to the National Death Index (N = 13,687). We use multinomial regression models stratified by gender and controlling for age, education and race/ethnicity to estimate diabetes-status specific transition probabilities, then use a Markov chain matrix population model to calculate age at onset of cognitive impairment and LE with and without cognitive impairment by diabetes status at age 50. LE at age 50 was 27.6 (men) and 32.1 (women). From age 50, those with diabetes had a first incidence of cognitive impairment 3 (men) and 4 (women) years earlier than those without. Diabetes reduced total LE by 5–7 years and cognitively healthy LE by 4–6 years. Those with diabetes lived one year less in a cognitively impaired state than those without. Over 80% of the lower LE associated with diabetes is attributable to the loss of cognitively-healthy years. Those with diabetes have a shorter LE with cognitive impairment because of higher mortality. In analyses by educational attainment, education was strongly protective of cognitive health, yet diabetes was associated with lower age of cognitive impairment onset and fewer cognitive healthy years lived in all educational groups. The excess mortality because of diabetes may be decreasing. If the mortality decline is not coupled with a comparable decline in the risk of cognitive impairment, the population level burden of impaired cognition may grow larger.

Original languageEnglish (US)
Article numbere0190488
JournalPLoS One
Volume12
Issue number12
DOIs
StatePublished - Dec 1 2017

Fingerprint

Medical problems
Life Expectancy
diabetes
Mortality
Age of Onset
Education
Cognitive Dysfunction
education
Markov Chains
Retirement
Health
Cognition
Population
nationalities and ethnic groups
cognition
Markov processes
Incidence
death
incidence

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Díaz-Venegas, C., Schneider, D. C., Myrskylä, M., & Mehta, N. K. (2017). Life expectancy with and without cognitive impairment by diabetes status among older Americans. PLoS One, 12(12), [e0190488]. https://doi.org/10.1371/journal.pone.0190488

Life expectancy with and without cognitive impairment by diabetes status among older Americans. / Díaz-Venegas, Carlos; Schneider, Daniel C.; Myrskylä, Mikko; Mehta, Neil K.

In: PLoS One, Vol. 12, No. 12, e0190488, 01.12.2017.

Research output: Contribution to journalArticle

Díaz-Venegas, Carlos ; Schneider, Daniel C. ; Myrskylä, Mikko ; Mehta, Neil K. / Life expectancy with and without cognitive impairment by diabetes status among older Americans. In: PLoS One. 2017 ; Vol. 12, No. 12.
@article{e4d3617306e145ddbeb6541739948e3a,
title = "Life expectancy with and without cognitive impairment by diabetes status among older Americans",
abstract = "Diabetes affects mortality and cognitive functioning. It is not known how diabetes influences life expectancy (LE) with and without cognitive impairment. We seek to examine age at onset of cognitive impairment and life expectancy (LE) with and without cognitive impairment by diabetes status among middle- and older-aged Americans. Data come from the U.S. Health and Retirement Study 2000–2012 linked to the National Death Index (N = 13,687). We use multinomial regression models stratified by gender and controlling for age, education and race/ethnicity to estimate diabetes-status specific transition probabilities, then use a Markov chain matrix population model to calculate age at onset of cognitive impairment and LE with and without cognitive impairment by diabetes status at age 50. LE at age 50 was 27.6 (men) and 32.1 (women). From age 50, those with diabetes had a first incidence of cognitive impairment 3 (men) and 4 (women) years earlier than those without. Diabetes reduced total LE by 5–7 years and cognitively healthy LE by 4–6 years. Those with diabetes lived one year less in a cognitively impaired state than those without. Over 80{\%} of the lower LE associated with diabetes is attributable to the loss of cognitively-healthy years. Those with diabetes have a shorter LE with cognitive impairment because of higher mortality. In analyses by educational attainment, education was strongly protective of cognitive health, yet diabetes was associated with lower age of cognitive impairment onset and fewer cognitive healthy years lived in all educational groups. The excess mortality because of diabetes may be decreasing. If the mortality decline is not coupled with a comparable decline in the risk of cognitive impairment, the population level burden of impaired cognition may grow larger.",
author = "Carlos D{\'i}az-Venegas and Schneider, {Daniel C.} and Mikko Myrskyl{\"a} and Mehta, {Neil K.}",
year = "2017",
month = "12",
day = "1",
doi = "10.1371/journal.pone.0190488",
language = "English (US)",
volume = "12",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "12",

}

TY - JOUR

T1 - Life expectancy with and without cognitive impairment by diabetes status among older Americans

AU - Díaz-Venegas, Carlos

AU - Schneider, Daniel C.

AU - Myrskylä, Mikko

AU - Mehta, Neil K.

PY - 2017/12/1

Y1 - 2017/12/1

N2 - Diabetes affects mortality and cognitive functioning. It is not known how diabetes influences life expectancy (LE) with and without cognitive impairment. We seek to examine age at onset of cognitive impairment and life expectancy (LE) with and without cognitive impairment by diabetes status among middle- and older-aged Americans. Data come from the U.S. Health and Retirement Study 2000–2012 linked to the National Death Index (N = 13,687). We use multinomial regression models stratified by gender and controlling for age, education and race/ethnicity to estimate diabetes-status specific transition probabilities, then use a Markov chain matrix population model to calculate age at onset of cognitive impairment and LE with and without cognitive impairment by diabetes status at age 50. LE at age 50 was 27.6 (men) and 32.1 (women). From age 50, those with diabetes had a first incidence of cognitive impairment 3 (men) and 4 (women) years earlier than those without. Diabetes reduced total LE by 5–7 years and cognitively healthy LE by 4–6 years. Those with diabetes lived one year less in a cognitively impaired state than those without. Over 80% of the lower LE associated with diabetes is attributable to the loss of cognitively-healthy years. Those with diabetes have a shorter LE with cognitive impairment because of higher mortality. In analyses by educational attainment, education was strongly protective of cognitive health, yet diabetes was associated with lower age of cognitive impairment onset and fewer cognitive healthy years lived in all educational groups. The excess mortality because of diabetes may be decreasing. If the mortality decline is not coupled with a comparable decline in the risk of cognitive impairment, the population level burden of impaired cognition may grow larger.

AB - Diabetes affects mortality and cognitive functioning. It is not known how diabetes influences life expectancy (LE) with and without cognitive impairment. We seek to examine age at onset of cognitive impairment and life expectancy (LE) with and without cognitive impairment by diabetes status among middle- and older-aged Americans. Data come from the U.S. Health and Retirement Study 2000–2012 linked to the National Death Index (N = 13,687). We use multinomial regression models stratified by gender and controlling for age, education and race/ethnicity to estimate diabetes-status specific transition probabilities, then use a Markov chain matrix population model to calculate age at onset of cognitive impairment and LE with and without cognitive impairment by diabetes status at age 50. LE at age 50 was 27.6 (men) and 32.1 (women). From age 50, those with diabetes had a first incidence of cognitive impairment 3 (men) and 4 (women) years earlier than those without. Diabetes reduced total LE by 5–7 years and cognitively healthy LE by 4–6 years. Those with diabetes lived one year less in a cognitively impaired state than those without. Over 80% of the lower LE associated with diabetes is attributable to the loss of cognitively-healthy years. Those with diabetes have a shorter LE with cognitive impairment because of higher mortality. In analyses by educational attainment, education was strongly protective of cognitive health, yet diabetes was associated with lower age of cognitive impairment onset and fewer cognitive healthy years lived in all educational groups. The excess mortality because of diabetes may be decreasing. If the mortality decline is not coupled with a comparable decline in the risk of cognitive impairment, the population level burden of impaired cognition may grow larger.

UR - http://www.scopus.com/inward/record.url?scp=85039859348&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039859348&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0190488

DO - 10.1371/journal.pone.0190488

M3 - Article

VL - 12

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 12

M1 - e0190488

ER -