Endogenous testosterone levels, physical performance, and fall risk in older men

Eric Orwoll, Lori C. Lambert, Lynn Marshall, Janet Blank, Elizabeth Barrett-Connor, Jane Cauley, Kris Ensrud, Steven R. Cummings

Research output: Contribution to journalArticle

130 Citations (Scopus)

Abstract

Background: Gonadal steroid levels decline with age in men. Whether low testosterone levels affect the development of common age-related disorders, including physical functioning and falling, is unclear. Methods: This longitudinal, observational follow-up study sought to determine whether low testosterone levels are associated with physical performance and fall risk in older men. A total of 2587 community-based men aged 65 to 99 years were selected using a stratified random sampling scheme from a study cohort of 5995 volunteers. Bioavailable testosterone and estradiol levels and physical performance measures were determined from baseline. Incident falls were ascertained every 4 months during 4 years of follow-up. Generalized estimating equations were used to estimate risk ratios for the relation of sex steroids to falls. Results: Fifty-six percent of the men reported at least 1 fall; many fell frequently. Lower bioavailable testosterone levels were associated with increased fall risk. Men with testosterone levels in the lowest quartile had a 40% higher fall risk than those in the highest quartile. The effect of low testosterone levels was most apparent in younger men (65-69 years) (relative risk, 1.8; 95% confidence interval, 1.2-2.7); testosterone level was not associated with falls in the oldest men (≥80 years). Lower testosterone concentrations were associated with reduced physical performance. However, the association between low testosterone levels and fall risk persisted despite adjustment for performance. Conclusions: Falls were common among older men. Fall risk was higher in men with lower bioavailable testosterone levels. The effect of testosterone level was independent of poorer physical performance, suggesting that the effect of testosterone on fall risk may be mediated by other androgen actions.

Original languageEnglish (US)
Pages (from-to)2124-2131
Number of pages8
JournalArchives of Internal Medicine
Volume166
Issue number19
DOIs
StatePublished - Oct 23 2006

Fingerprint

Testosterone
Accidental Falls
Steroids
Androgens
Volunteers
Estradiol
Cohort Studies
Odds Ratio
Confidence Intervals

ASJC Scopus subject areas

  • Internal Medicine

Cite this

Endogenous testosterone levels, physical performance, and fall risk in older men. / Orwoll, Eric; Lambert, Lori C.; Marshall, Lynn; Blank, Janet; Barrett-Connor, Elizabeth; Cauley, Jane; Ensrud, Kris; Cummings, Steven R.

In: Archives of Internal Medicine, Vol. 166, No. 19, 23.10.2006, p. 2124-2131.

Research output: Contribution to journalArticle

Orwoll, E, Lambert, LC, Marshall, L, Blank, J, Barrett-Connor, E, Cauley, J, Ensrud, K & Cummings, SR 2006, 'Endogenous testosterone levels, physical performance, and fall risk in older men', Archives of Internal Medicine, vol. 166, no. 19, pp. 2124-2131. https://doi.org/10.1001/archinte.166.19.2124
Orwoll, Eric ; Lambert, Lori C. ; Marshall, Lynn ; Blank, Janet ; Barrett-Connor, Elizabeth ; Cauley, Jane ; Ensrud, Kris ; Cummings, Steven R. / Endogenous testosterone levels, physical performance, and fall risk in older men. In: Archives of Internal Medicine. 2006 ; Vol. 166, No. 19. pp. 2124-2131.
@article{5929d32b0f624f68b83239ea9e37aa7a,
title = "Endogenous testosterone levels, physical performance, and fall risk in older men",
abstract = "Background: Gonadal steroid levels decline with age in men. Whether low testosterone levels affect the development of common age-related disorders, including physical functioning and falling, is unclear. Methods: This longitudinal, observational follow-up study sought to determine whether low testosterone levels are associated with physical performance and fall risk in older men. A total of 2587 community-based men aged 65 to 99 years were selected using a stratified random sampling scheme from a study cohort of 5995 volunteers. Bioavailable testosterone and estradiol levels and physical performance measures were determined from baseline. Incident falls were ascertained every 4 months during 4 years of follow-up. Generalized estimating equations were used to estimate risk ratios for the relation of sex steroids to falls. Results: Fifty-six percent of the men reported at least 1 fall; many fell frequently. Lower bioavailable testosterone levels were associated with increased fall risk. Men with testosterone levels in the lowest quartile had a 40{\%} higher fall risk than those in the highest quartile. The effect of low testosterone levels was most apparent in younger men (65-69 years) (relative risk, 1.8; 95{\%} confidence interval, 1.2-2.7); testosterone level was not associated with falls in the oldest men (≥80 years). Lower testosterone concentrations were associated with reduced physical performance. However, the association between low testosterone levels and fall risk persisted despite adjustment for performance. Conclusions: Falls were common among older men. Fall risk was higher in men with lower bioavailable testosterone levels. The effect of testosterone level was independent of poorer physical performance, suggesting that the effect of testosterone on fall risk may be mediated by other androgen actions.",
author = "Eric Orwoll and Lambert, {Lori C.} and Lynn Marshall and Janet Blank and Elizabeth Barrett-Connor and Jane Cauley and Kris Ensrud and Cummings, {Steven R.}",
year = "2006",
month = "10",
day = "23",
doi = "10.1001/archinte.166.19.2124",
language = "English (US)",
volume = "166",
pages = "2124--2131",
journal = "JAMA Internal Medicine",
issn = "2168-6106",
publisher = "American Medical Association",
number = "19",

}

TY - JOUR

T1 - Endogenous testosterone levels, physical performance, and fall risk in older men

AU - Orwoll, Eric

AU - Lambert, Lori C.

AU - Marshall, Lynn

AU - Blank, Janet

AU - Barrett-Connor, Elizabeth

AU - Cauley, Jane

AU - Ensrud, Kris

AU - Cummings, Steven R.

PY - 2006/10/23

Y1 - 2006/10/23

N2 - Background: Gonadal steroid levels decline with age in men. Whether low testosterone levels affect the development of common age-related disorders, including physical functioning and falling, is unclear. Methods: This longitudinal, observational follow-up study sought to determine whether low testosterone levels are associated with physical performance and fall risk in older men. A total of 2587 community-based men aged 65 to 99 years were selected using a stratified random sampling scheme from a study cohort of 5995 volunteers. Bioavailable testosterone and estradiol levels and physical performance measures were determined from baseline. Incident falls were ascertained every 4 months during 4 years of follow-up. Generalized estimating equations were used to estimate risk ratios for the relation of sex steroids to falls. Results: Fifty-six percent of the men reported at least 1 fall; many fell frequently. Lower bioavailable testosterone levels were associated with increased fall risk. Men with testosterone levels in the lowest quartile had a 40% higher fall risk than those in the highest quartile. The effect of low testosterone levels was most apparent in younger men (65-69 years) (relative risk, 1.8; 95% confidence interval, 1.2-2.7); testosterone level was not associated with falls in the oldest men (≥80 years). Lower testosterone concentrations were associated with reduced physical performance. However, the association between low testosterone levels and fall risk persisted despite adjustment for performance. Conclusions: Falls were common among older men. Fall risk was higher in men with lower bioavailable testosterone levels. The effect of testosterone level was independent of poorer physical performance, suggesting that the effect of testosterone on fall risk may be mediated by other androgen actions.

AB - Background: Gonadal steroid levels decline with age in men. Whether low testosterone levels affect the development of common age-related disorders, including physical functioning and falling, is unclear. Methods: This longitudinal, observational follow-up study sought to determine whether low testosterone levels are associated with physical performance and fall risk in older men. A total of 2587 community-based men aged 65 to 99 years were selected using a stratified random sampling scheme from a study cohort of 5995 volunteers. Bioavailable testosterone and estradiol levels and physical performance measures were determined from baseline. Incident falls were ascertained every 4 months during 4 years of follow-up. Generalized estimating equations were used to estimate risk ratios for the relation of sex steroids to falls. Results: Fifty-six percent of the men reported at least 1 fall; many fell frequently. Lower bioavailable testosterone levels were associated with increased fall risk. Men with testosterone levels in the lowest quartile had a 40% higher fall risk than those in the highest quartile. The effect of low testosterone levels was most apparent in younger men (65-69 years) (relative risk, 1.8; 95% confidence interval, 1.2-2.7); testosterone level was not associated with falls in the oldest men (≥80 years). Lower testosterone concentrations were associated with reduced physical performance. However, the association between low testosterone levels and fall risk persisted despite adjustment for performance. Conclusions: Falls were common among older men. Fall risk was higher in men with lower bioavailable testosterone levels. The effect of testosterone level was independent of poorer physical performance, suggesting that the effect of testosterone on fall risk may be mediated by other androgen actions.

UR - http://www.scopus.com/inward/record.url?scp=33750324635&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33750324635&partnerID=8YFLogxK

U2 - 10.1001/archinte.166.19.2124

DO - 10.1001/archinte.166.19.2124

M3 - Article

C2 - 17060543

AN - SCOPUS:33750324635

VL - 166

SP - 2124

EP - 2131

JO - JAMA Internal Medicine

JF - JAMA Internal Medicine

SN - 2168-6106

IS - 19

ER -