gender-specific associations between physical functioning, bone quality, and fracture risk in older...

ORIGINAL RESEARCH

Gender-Specific Associations Between Physical Functioning, BoneQuality, and Fracture Risk in Older People

Regula Furrer • Natasja M. van Schoor •

Arnold de Haan • Paul Lips • Renate T. de Jongh

Received: 26 June 2013 / Accepted: 15 December 2013

� Springer Science+Business Media New York 2014

Abstract The aim of this study was to investigate which

parameters of physical functioning are associated with

bone quality and fracture risk and whether gender-specific

differences exist within these associations. We studied

1,486 participants of the Longitudinal Aging Study

Amsterdam. As measures of physical functioning, handgrip

strength, physical performance, and level of physical

activity were assessed. To assess bone quality, broadband

ultrasound attenuation (BUA) and speed of sound (SOS)

were measured at baseline using quantitative ultrasound

and bone mineral density (BMD) at baseline and after

3 years by dual-energy X-ray absorptiometry. In addition,

fracture incidence over 6 years was assessed. After

adjustment for confounders (age, serum 25[OH]D, smok-

ing, and body weight), in men, physical performance was

positively related to BUA, SOS, and BMD cross-section-

ally and to BMD longitudinally. Using Cox proportional

hazards model, in men higher handgrip strength and

physical performance were associated with reduced

fracture risk after adjustment for confounders (hazard ratio

[HR] 0.96, 95 % confidence interval [CI] 0.92–0.99, and

HR 0.89, 95 % CI 0.80–0.98, respectively). In women, a

moderate level of physical activity was related to reduced

fracture risk (HR 0.57, 95 % CI 0.33–0.99). In conclusion,

in men, higher handgrip strength and physical performance

are related to higher bone quality and reduced fracture risk,

whereas in women, a moderate to high level of physical

activity is associated with reduced fracture risk. These

measurements may contribute to the identification of

individuals at high fracture risk. Both the causality of and

explanations for gender-specific differences in these rela-

tionships remain subject to further studies.

Keywords Osteoporosis � Fracture � Risk factor �Handgrip strength � Physical performance � Physical

activity

Introduction

With increasing age, the prevalence of osteoporosis, char-

acterized by low bone mineral density (BMD), increases.

Low BMD in combination with increased fall incidence

among older people results in higher fracture risk [1]. After

the age of 60, approximately 44 % of women and 25 % of

men will experience at least one fracture [2]. The high

morbidity and mortality related to fractures at old age [3, 4]

underline the necessity to search for indicators to identify

individuals at high fracture risk. These indicators may be

used to select individuals who may benefit from additional

interventions to prevent fractures.

Mechanical loading is an important contributor to bone

quality. It seems likely that measures of physical func-

tioning, such as muscle strength, physical performance, and

The authors have stated that they have no conflict of interest.

R. Furrer � A. de Haan

Faculty of Human Movement Sciences, MOVE Research

Institute Amsterdam, VU University Amsterdam, Amsterdam,

The Netherlands

N. M. van Schoor

Department of Epidemiology and Biostatistics, EMGO Institute

for Health and Care Research, VU University Medical Center

Amsterdam, Amsterdam, The Netherlands

P. Lips � R. T. de Jongh (&)

Department of Internal Medicine and Endocrinology, MOVE

Research Institute Amsterdam, VU University Medical Center

Amsterdam, Amsterdam, The Netherlands

e-mail: [email protected]

123

Calcif Tissue Int

DOI 10.1007/s00223-013-9836-1

physical activity, are important determinants for BMD.

Low handgrip strength and physical performance have

been shown to be associated with lower BMD and higher

fracture risk [5–11]. In addition, differences in handgrip

strength and walking tests were able to discriminate

between fallers with fractures, fallers without fractures, and

nonfallers in older men [12]. These results suggest an

association between parameters of physical functioning,

BMD, and fracture risk in older people. However, it

remains unclear which measures of physical functioning

are related to bone quality both cross-sectionally and lon-

gitudinally and whether this translates into increased frac-

ture risk.

Since postmenopausal women are at highest risk for

osteoporosis and fractures, most studies focus on women

[5, 7, 8, 11, 13]. Since men and women differ in body

composition, hormonal status, and patterns of physical

activity, relationships between measures of physical func-

tioning and bone quality may be gender-specific [14, 15].

Therefore, it is of interest to study gender-specific differ-

ences in these associations.

The aim of the present study was to investigate (1)

which parameters of physical functioning, i.e., handgrip

strength, physical performance tests, or physical activity,

are associated with bone quality as estimated by calcaneal

quantitative ultrasound and X-ray absorptiometry; (2)

which parameters of physical functioning are related to

longitudinal changes in BMD and 6-year fracture risk; and

(3) whether gender-specific differences exist in the afore-

mentioned relationships.

Materials and Methods

Study Sample

For this study, data from the second (1995–1996) and third

(1998–1999) cycles of the Longitudinal Aging Study

Amsterdam (LASA) were analyzed. The LASA is an

ongoing interdisciplinary cohort study on predictors and

consequences of changes in physical, cognitive, emotional,

and social functioning in older people. The sampling and

data collection have been described in detail elsewhere

[16]. Briefly, a random sex- and age-stratified sample of

older men and women was drawn from population registers

of 11 municipalities in three different regions of the

Netherlands. The study was approved by the Medical

Ethics Committee of the VU University Medical Center

Amsterdam, and all participants gave informed consent.

During the medical interview of the second cycle, 1,509

individuals participated. The total population used in the

present study included individuals who participated in the

medical interview of the second LASA cycle and who were

included in at least one of the performed analyses

(n = 1,486).

Measures of Physical Functioning

Handgrip Strength

Maximal handgrip strength was determined using a strain

gauged dynamometer (Takei TKK 5001; Takei Scientific

Instruments, Tokyo, Japan) and recorded to the nearest

kilogram. Participants performed two maximal trials with

the right and left hands. The highest value of the right or

left hand was used for analyses.

Physical Performance

Physical performance was determined using three different

tests: the walking test, as measured by the time taken to walk

3 m, turn around 180�, and walk back 3 m as fast as possible;

the chair-stand test, as measured by the time taken to rise

from a chair and sit down again five times as fast as possible

without using the hands; and the balance test, as measured by

the time one is able to maintain a tandem stance (one foot

directly in front of the other) with a maximum of 10 s. The

time of the walking test and chair stand test was divided into

quartiles of the population giving 1–4 points (4 for the fastest

performance, 1 for the slowest performance, and 0 if the test

could not be finished). The time of the balance test was

divided into three categories: unable (0 points), able to hold

position for 4–9 s (2 points), and able to hold position for

10 s (4 points). Total score for physical performance ranged

0–12 and was calculated by the sum of the scores of the three

different performance tests.

Physical Activity

To assess the level of physical activity, the validated LASA

Physical Activity Questionnaire was used [17]. The ques-

tionnaire addresses the following activities over the past

2 weeks: walking outdoors, bicycling, gardening, light and

heavy household activities, and sports activities. Taking

into account the frequency and duration of these activities,

average time (minutes) spent on physical activity per day

was calculated. Physical activity showed a skewed distri-

bution and was therefore divided into tertiles. As all data

were stratified by gender (due to interaction effects, see

‘‘Results’’), physical activity tertiles were gender-specific.

The cut-off points were for men 65 and 135 min/day and

for women 118 and 193 min/day.

R. Furrer et al.: Physical Functioning, Bone Quality, and Fracture Risk

123

Bone Quality Measurements

Total-hip BMD was measured by dual-energy X-ray

absorptiometry (DXA; Hologic QDR 2000; Hologic,

Waltham, MA, USA). Scans were analyzed using the

software version V4.71A:1. Broadband ultrasound attenu-

ation (BUA in decibels per megahertz) and speed of sound

(SOS in meters per second) were assessed by quantitative

ultrasound (QUS; CUBA Clinical Instrument; McCue

Utrasonics, Winchester, UK) through the calcaneus by

placing two transducers on either side of the heel with

coupling gel. Both heels were measured twice, and mean

values of BUA and SOS of all four measurements were

used for further analyses. All participants were invited to

undergo QUS in participating hospitals or smaller health

centers near the participant’s home. Participants living in

Amsterdam or its vicinity were invited to undergo DXA

measurements at the VU University Medical Center

Amsterdam (study population n = 522). All BMD mea-

surements were performed with the same DXA scanner.

All QUS measurements were performed with CUBA

Clinical Instruments, but as these measurements were

performed in three different regions of the Netherlands,

three different scanners were used. BUA values were not

different between these three regions (70.7 [± 20.1], 71.6

[± 21.0], and 69.6 [± 20.4] dB/MHz, respectively).

However, SOS values were significantly different between

these regions: 1,601.5 (± 39.6), 1,657.7 (± 51.9), and

1,617.2 (± 43.3) m/s. To test whether the scanner used for

the measurements influenced the associations with SOS,

we performed all analyses with additional adjustment for

the scanner used. None of the associations changed after

additional adjustment for scanner and are therefore pre-

sented without the additional adjustment.

Assessment of Fractures

Fractures occurring between 1995–1996 and 1998–1999

were collected prospectively with a fracture calendar.

Participants were asked to mail the calendar with recorded

fractures every 3 months. If the calendar was not returned,

not completed, or completed incorrectly, participants were

contacted by telephone. Proxies were contacted if partici-

pants were not able to respond. Between 1998–1999 and

2001–2002, fracture incidence was assessed during the

medical interview and telephone interview. Once a par-

ticipant reported a fracture, a general practitioner was

requested to verify the fracture and to identify the type,

date, and cause of the fracture. If fractures were not con-

firmed by the general practitioner, original X-rays were

requested and assessed by an experienced medical doctor.

If original X-rays were not available, the participant was

recontacted with the goal of verification.

Potential Effect Modifiers and Confounders

Gender was tested as a potential effect modifier. Variables

that were related to physical functioning as well as bone

quality in previous analyses in the LASA, which could

therefore influence associations between those measures,

were considered potential confounders. Potential con-

founders were age, anthropometric variables (weight, body

mass index, waist circumference, and lean body mass),

physical activity, number of chronic diseases, alcohol use

(number of units per week), smoking (yes/no), medication

use (yes/no), serum 25-hydroxyvitamin D (25[OH]D) lev-

els, and serum levels of inflammatory markers (C-reactive

protein [CRP] and interleukin 6 [IL-6]) [18–20]. Physical

activity, number of chronic diseases, alcohol use, smoking,

and medication use were assessed during an interview

based on self-report.

Prior to blood sample collection, participants were

allowed to have tea and toast but no dairy products. Blood

was centrifuged and serum was stored at -20 �C. Serum

25(OH)D was measured by a competitive protein binding

assay (Nichols Diagnostics, Capistrano, CA, USA). CRP

and IL-6 were determined using sensitive enzyme-linked

immunosorbent assays at Sanquin Research (Amsterdam,

The Netherlands). Lean body mass was calculated from

DXA whole-body scans using software version V5.70A.

Statistical Analysis

Baseline differences between the genders were tested by

independent-samples t-tests for normally distributed data,

Mann–Whitney U tests for skewed data, and Chi squared

tests for categorical variables. Since CRP, IL-6, and alco-

hol use were not normally distributed, data were log-

transformed. Effect modification was defined by a statis-

tically significant interaction term (p \ 0.10). All potential

confounders were tested in regression models and consid-

ered relevant and kept in the model if they induced a

change of the b coefficient of at least 10 %. Multiple linear

regression analyses were performed to study associations

of physical functioning and BMD, BUA, or SOS. In lon-

gitudinal analyses between physical functioning and BMD,

baseline BMD was included in the model. Assumptions of

linear regression analysis were checked by histograms and

normal probability plots of the standardized residuals.

Associations with fracture risk were studied using Cox

proportional hazards regression analyses. The assumption

of a constant hazard rate over time was checked with a log-

minus-log plot. IBM SPSS Statistics 20 (SPSS, Inc.,


123

Chicago, IL, USA) was used for all statistical analyses.

Significance was reported as p \ 0.05.

Results

Subject Characteristics

Table 1 shows baseline characteristics of the study popu-

lation. Men were heavier but leaner than women. In addi-

tion, compared to women, men had a better physical

performance, higher handgrip strength, BMD, BUA, and

SOS; but they were less active. The higher physical activity

level of women was mainly explained by the greater time

spent on light household tasks.

To measure BMD, participants had to come to the VU

Medical Center in Amsterdam, which may have caused a

selection bias of those able to come to the hospital and those

who were less mobile. We analyzed the characteristics of the

individuals with a BMD measurement at baseline (n = 522)

and those without a BMD measurement (n = 964). Of all

subject characteristics presented in Table 1, the following

characteristics were significantly different between the two

groups (BMD vs. no-BMD): height (167.2 ± 8.8 vs.

165.8 ± 9.5 cm), age (75.2 ± 6.4 vs. 76.2 ± 6.8 years), lean

body mass (45.0 ± 8.8 vs. 36.8 ± 4.7 kg), medication use (72

vs. 78 %), serum 25(OH)D (50 ± 22 vs. 55 ± 25 nmol/L),

serum IL-6 (2.5 [1.6–3.8] vs. 1.6 [0.9–2.6] pg/mL), alcohol use

(4 [0.5–12.5] vs. 2 [0–7] drinks/week), physical activity (129

[81–201] vs. 124 [60–93] min/day), physical performance

(7.8 ± 2.8 vs. 6.8 ± 3.5 points), and SOS (1,603 ± 38 vs.

1,635 ± 53 m/s). These data show that the group with a BMD

measurement were 1 year younger and more fit/active than the

group without a BMD measurement.

Associations Between Physical Functioning and Bone

Quality

Gender was an effect modifier of associations between

handgrip strength and BUA and SOS measures cross-

Table 1 Characteristics of the study population

Men N Women N p value

Study population (%) 48 % 718 52 % 768

Age (years) 75.8 ± 6.6 718 75.8 ± 6.6 768 0.973

Weight (kg) 78.0 ± 11.9 709 71.0 ± 13.0 758 \0.001

Height (cm) 173.0 ± 6.8 707 160.0 ± 6.4 757 \0.001

BMI (kg/m2) 26.0 ± 3.4 706 27.7 ± 4.9 756 \0.001

Waist circumference (cm) 99.3 ± 10.3 685 92.6 ± 11.6 717 \0.001

Lean body mass (kg) 52.1 ± 5.7 255 37.9 ± 4.6 263 \0.001

Number of chronic diseases 2 (1–2) 718 2 (1–3) 768 0.001

Medication use (%) 74 % 532/717 78 % 596/768 0.125

Serum 25(OH)D (nmol/L) 57.9 ± 24.5 643 48.9 ± 24.0 677 \0.001

Serum IL-6 (pg/mL) 1.9 (1.2–3.2) 628 2.0 (1.1–3.15) 657 0.416

Serum CRP (lg/mL) 3.3 (1.6–6.7) 629 3.1 (1.4–6.5) 658 0.393

Current smoker (%) 25 % 183/718 13 % 99/768 \0.001

Alcohol use (drinks/week) 6 (1–21) 717 1 (0–6) 766 \0.001

Physical activity (min/day) 96.4 (51.3–162.9) 689 152.1 (98.6–218.5) 736 \0.001

Total walking time 15.0 (4.3–34.6) 12.9 (1.7–30.0) 0.002

Total bicycle time 0.0 (0.0–14.3) 0.0 (0.0–7.1) \0.001

Light household tasks 42.9 (12.9–75.0) 120.0 (60.0–180.0) \0.001

Heavy household tasks 2.8 (0.0–12.9) 4.3 (0.0–17.1) 0.039

Sports activities 0.0 (0.0–8.6) 0.0 (0.0–8.6) 0.589

Physical performance (0–12) 7.8 ± 3.0 686 6.6 ± 3.4 726 \0.001

Handgrip strength (kg) 36.7 ± 8.5 715 21.8 ± 5.2 763 \0.001

Total hip BMD (g/cm2) 0.916 ± 0.15 259 0.784 ± 0.13 263 \0.001

BUA (dB/MHz) 81.0 ± 18.5 653 61.1 ± 17.2 693 \0.001

SOS (m/s) 1,638 ± 51 653 1,608 ± 46 693 \0.001

Fractures 6-year follow-up (%) 7 % 51/715 10 % 81/762 0.019

Values mean ± SD or median (interquartile range)

BUA broadband ultrasound attenuation, SOS speed of sound, BMD bone mineral density, BMI body mass index, CRP C-reactive protein


123

sectionally (p = 0.001 and p = 0.034, respectively) as

well as BMD longitudinally (p = 0.043). For comparabil-

ity reasons, data were stratified by gender in all further

analyses. Multiple regression analyses demonstrated that

age, serum 25(OH)D, smoking, and several anthropometric

variables were relevant confounders. Of anthropometric

variables, body weight had the largest effect and was

therefore used as a confounder in regression analyses. For

relationships of physical functioning with fracture risk, of

all potential confounders including baseline BMD, only

age was observed to be a relevant confounder.

Table 2 shows associations between measures of phys-

ical functioning and estimates of bone quality. After

adjustment for age, serum 25(OH)D, and smoking, higher

handgrip strength was associated with higher BUA and hip

BMD in both genders, although the latter association was

borderline significant in women (p = 0.062). These asso-

ciations were no longer significant after additional adjust-

ment for weight (Table 2, model 3). Higher physical

performance was associated with higher BUA and SOS

after adjustment for age, serum 25(OH)D, and smoking in

men but not in women (Table 2, model 2). After adjust-

ment for all relevant confounders, higher physical perfor-

mance was associated in men with higher BUA, SOS, and

BMD and in women with higher BUA (Table 2, model 3).

In men, for a 1 point higher score on physical performance

tests, BMD, BUA, and SOS were 0.01 g/cm2, 0.75 dB/MHz,

and 2.27 m/s higher, respectively. In women, for a 1 point

higher score on the physical performance tests, BUA score

was 0.76 dB/MHz higher. The individual physical perfor-

mance tests were not associated with BUA, SOS, or BMD.

In neither men nor women, level of physical activity was

associated with BUA, SOS, or BMD.

Associations Between Physical Functioning

and Total-Hip BMD After 3-Year Follow-Up

In men, but not women, higher physical performance was

significantly associated with higher follow-up BMD over

3 years (Table 2). This association remained significant

after correction for all relevant confounders. In men, for a 1

point higher score on the physical performance test, follow-

up BMD was 0.01 g/cm2 higher. In neither men nor

women, handgrip strength and level of physical activity

were associated with follow-up BMD.

Associations Between Physical Functioning and Time

to First Fracture during 6 Years of Follow-Up

In men, but not women, higher handgrip strength and phys-

ical performance were associated with reduced 6-year frac-

ture incidence independently of age. With 1 kg higher

handgrip strength or 1 point higher score during physical

performance test, the risk of having a fracture in 6 years was

approximately 5 and 10 % lower, respectively (Table 3).

However, in women, but not men, a higher level of physical

activity was associated with reduced 6-year fracture inci-

dence (Table 3). Women in the lowest tertile had a higher

fracture risk compared to women in the second (p = 0.045)

and third (p = 0.080) tertiles. In women, the risk of having a

fracture in 6 years was reduced by approximately 40 %

when activity was more than 118 min/day. To analyze

whether the gender-specific differences observed in the

association between physical activity and fracture risk were

due to the fact that women performed more light household

activities (see Table 1), we performed the same analyses

using a physical activity score that excluded light household

tasks. In contrast to gender-specific tertiles of total physical

activity, gender-specific tertiles of physical activity without

light household tasks were very similar in men and women.

The cut-off points for the tertiles were, for men, 25.8 and

66.0 min/day and, for women, 20.0 and 57.5 min/day. In

both men and women, no association was observed between

level of physical activity (without light household tasks) and

risk of having a fracture in 6 years. This suggests that the

observed association in women between level of total

physical activity and risk of having a fracture can be

explained by a higher level of light household activities.

Discussion

The main findings of this study are (1) in men higher

physical performance score is associated with higher bone

quality cross-sectionally as well as longitudinally, (2) in

men higher handgrip strength and physical performance

score are associated with reduced 6-year fracture risk, and

(3) in women a moderate to high level of physical activity

was associated with reduced 6-year fracture risk.

Association of Reduced Physical Functioning

with Increased Fracture Risk in Men

The importance of physical performance with respect to

BMD and fracture risk has been shown previously [8, 10–

12, 21]. Within the LASA population, physical perfor-

mance was also associated with 3-year fracture risk [10].

Our data suggest that in men, but not women, low physical

performance is a risk factor for lower bone quality and

increased fracture incidence during 6 years. In both gen-

ders, associations of handgrip strength with bone measures

were largely explained by body weight, a proxy for body

size. Within the literature, the relationship between muscle

strength and bone quality is subject to controversy.

Although handgrip strength and leg extension strength have

been reported to be associated with femoral neck BMD in


123

postmenopausal women [5], several studies in older men

and women did not show associations between leg extensor

strength and BMD or QUS measures after adjusting for

anthropometry [22–24]. Body mass and composition are

well-known determinants of BMD [15, 18, 25, 26], and

since bone adapts to mechanical loading, higher forces on

bone during activities in heavy people may at least in part

explain their higher BMD. These findings suggest that

anthropometry, in particular body size and weight, plays an

important role in associations between muscle strength and

BMD and should be taken into account when studying

these relationships.

In men, higher handgrip strength was related to reduced

fracture risk despite the absence of a relationship with bone

quality. In older people, handgrip strength was shown to be

positively correlated with contraction velocity of ankle

plantar flexors and knee extensors [27]. Maximal contrac-

tion velocity of ankle plantar flexors and knee extensors

was shown to be substantially reduced in older fallers

compared to nonfallers [27]. This suggests that in men

handgrip strength may reflect muscle characteristics that

contribute to a decreased fall incidence and consequently

reduced fracture risk.

Training intervention studies in older people showed

contradictory results regarding the reduction of fall risk

[28–32]. A meta-analysis of the Frailty and Injuries:

Cooperative Studies of Intervention Techniques trials

already demonstrated that exercise could reduce the risk of

falls [33]. Later trials were not always conclusive.

Although exercise training increased maximal force and

physical performance, fall risk was not significantly

reduced [28, 29]. In contrast, other studies reported a

reduced fall risk after training intervention, which even-

tually may result in a reduced number of fractures [30, 31].

In general, a combination of strength and balance training

appears most effective in reducing fall risk, but it remains

to be determined whether this is the most effective training

strategy to reduce the number of fractures. Meanwhile, our

study indicates that physical performance tests and mea-

surement of handgrip strength may contribute to the iden-

tification of men with a high fracture risk.

Association of Low Level of Physical Activity

with Increased Fracture Risk in Women

In women, of all parameters of physical functioning, only

level of physical activity was related to reduced fracture

incidence. Women being active for less than 120 min/day

had a 40 % higher fracture risk compared to more active

women. Since level of physical activity was not associated

with bone quality, it is suggested that other parameters may

play an important role in the reduction of fracture risk in

these women. Physical activity, mainly consisting of lightTa

ble

2A

sso

ciat

ion

sb

etw

een

ph

ysi

cal

fun

ctio

nin

gan

db

on

em

easu

res

Model

BU

A(n

=1,2

31)

SO

S(n

=1,3

01)

Tota

l-hip

BM

D(n

=492)

Foll

ow

-up

BM

Da

(n=

327)

Men

Wom

enM

enW

om

enM

enW

om

enM

enW

om

en

B(9

5%

CI)

B(9

5%

CI)

B(9

5%

CI)

B(9

5%

CI)

B(9

5%

CI)

9100

B(9

5%

CI)

9100

B(9

5%

CI)

9100

B(9

5%

CI)

9100

Han

dgri

pst

rength

(kg)

Model

10.2

7**

(0.0

7–0.4

7)

0.4

1**

(0.1

2–0.7

0)

0.3

9(-

0.1

6to

0.9

3)

0.4

3(-

0.3

6to

1.2

2)

0.3

0*

(0.0

3–0.5

7)

0.3

5�

(-0.0

1to

0.7

1)

0.1

7(-

0.1

4to

0.4

9)

0.1

0(-

0.1

5to

0.3

5)

Model

20.2

7**

(0.0

6–0.4

7)

0.4

0**

(0.1

1–0.6

9)

0.2

7(-

0.2

8to

0.8

2)

0.3

5(-

0.4

4to

1.1

4)

0.3

2*

(0.0

5–0.5

8)

0.3

5�

(-0.0

2to

0.7

1)

0.1

7(-

0.1

5to

0.4

8)

0.1

1(-

0.1

4to

0.3

5)

Model

30.1

3(-

0.0

8to

0.3

4)

0.0

8(-

0.2

0to

0.3

5)

0.5

3(-

0.0

5to

1.1

1)

0.1

6(-

0.6

5to

0.9

7)

0.0

5(-

0.2

1to

0.3

0)

0.1

5(-

0.1

8to

0.4

7)

0.0

4(-

0.2

8to

0.3

7)

0.1

5(-

0.1

0to

0.4

0)

Physi

cal

per

form

ance

(0–12

poin

ts)

Model

10.8

1**

(0.2

5–1.3

6)

0.3

8(-

0.0

7to

0.8

4)

2.5

8**

(1.0

6–4.1

0)

0.1

5(-

1.0

8to

1.3

9)

0.6

5(-

0.1

7to

1.4

6)

-0.2

8(-

0.8

6to

0.3

0)

1.0

0*

(0.0

6–1.9

4)

0.1

5(-

0.2

9to

0.5

9)

Model

20.7

7**

(0.2

1–1.3

3)

0.3

7(-

0.0

9to

0.8

3)

2.2

4**

(0.7

1–3.7

7)

0.0

0(-

1.2

5to

1.2

5)

0.6

3(-

0.1

8to

1.4

4)

-0.3

4(-

0.9

6to

0.2

7)

0.9

5*

(0.0

0–1.9

0)

0.1

3(-

0.3

1to

0.5

8)

Model

30.7

5**

(0.2

0–1.3

0)

0.7

6***

(0.3

4–1.1

9)

2.2

7**

(0.7

4–3.8

0)

0.2

3(-

1.0

4to

1.4

9)

0.7

4*

(0.0

1–1.4

7)

0.3

7(-

0.1

9to

0.9

3)

1.3

3**

(0.3

9–2.2

6)

0.0

6(-

0.4

1to

0.5

3)

Model

1,

adju

sted

for

age;

model

2,

adju

sted

for

age,

seru

m25-h

ydro

xyvit

amin

D,

smokin

g;

model

3,

adju

sted

for

age,

seru

m25-h

ydro

xyvit

amin

D,

smokin

g,

wei

ght;

CI

confi

den

cein

terv

al

*p\

0.0

5;

**

p\

0.0

1;

***

p\

0.0

01;

�p

=0.0

56;

�p

=0.0

62

aA

llm

odel

sw

ere

adju

sted

for

bas

elin

eB

MD


123

household tasks in the present study, may improve coor-

dination, endurance, and/or balance, which may not man-

ifest in better physical performance score but may

contribute to a reduced number of falls. A previous study in

the same LASA population showed that several variables

of physical functioning were associated with the risk of

recurrent falls [34]. This study demonstrated that an

increased level of physical activity was related to a reduced

risk of recurrent falls over 3 years [34]. In addition,

recurrent falls were reported to be reduced in women doing

light and heavy household activities [35]. This suggests

that a reduction in fall incidents in women may contribute

to the reduced fracture risk. Further intervention studies are

essential to study the potential causality of these relation-

ships. Nevertheless, for women, assessing the level of

physical activity may be useful to identify those with

increased fracture risk.

Gender-Specific Differences in Associations Between

Physical Functioning and Fracture Risk

Differences in hormonal status, body composition, level of

physical activity, muscle contractile characteristics, as well

as prevalence of chronic diseases may contribute to the

gender differences observed in our study [14, 15]. Previ-

ously, gender differences between fatigability were repor-

ted, showing that the muscles of men were more fatigable

than those of women, which may be explained by a higher

percentage of faster muscle fiber [36]. Impaired function-

ing due to aging and chronic diseases, which may result in

increased falls and eventually fracture risk, may be

reflected in poor physical performance and handgrip

strength more prominently in men than in women. Hand-

grip strength in men is approximately 1.7-fold higher than

that in women. The higher mean value of handgrip strength

and larger variation in men may discriminate better

between men with good and poor neuromuscular function

compared to women. Gender-specific differences in the

association between physical activity and fracture risk were

shown previously [37]. Comparable with our results,

moderate and high levels of physical activity were only

associated with decreased fracture risk in women [37].

Possibly, women who are able to attain at least 120 min of

daily physical activity (mainly consisting of light house-

hold tasks) may be generally healthier than women who are

not. This selection of women who are still able to attain a

high level of physical activity may not occur in men as they

perform substantially fewer household tasks. The under-

lying mechanisms for gender-specific differences in asso-

ciations between physical performance and bone quality

and fracture risk are unclear and of potential interest for

further studies.

Strength and Limitations

The strengths of the present study are its population-based

nature and the assessment of several measures of physical

functioning and bone quality together with fracture risk

within one cohort. In addition, the questionnaire on phys-

ical activity includes numerous daily activities such as light

and heavy household tasks and has been properly validated

[17]. A limitation of this study is the relatively small

number of individuals with available DXA measurements.

In summary, our results show that associations of

physical functioning with bone quality and fracture risk

differ between men and women. In men, low handgrip

strength and physical performance are risk factors for

fractures. In older women, low level of physical activity as

assessed with a questionnaire is a risk factor for fractures.

These findings may contribute to an improvement of the

identification of individuals at high fracture risk. The

gender-specific differences in relationships of measures of

physical functioning with bone quality and fracture risk

remain subject to further studies.

Table 3 Associations between physical functioning and 6-year fracture risk

Fracture risk 6-year follow-up

Men HR (95 % CI) Number of fractures/N Women HR (95 % CI) Number of fractures/N

Handgrip strength (kg) 0.96* (0.92–0.99) 51/712 0.99 (0.94–1.04) 81/757

Physical performance (0–12 points) 0.89* (0.80–0.98) 46/684 0.95 (0.88–1.03) 75/720

Physical activity (min/day)

Tertile 1 Reference 17/209 Reference 35/202

Tertile 2 1.07 (0.56–2.05) 20/211 0.57* (0.33–0.99) 21/227

Tertile 3 0.51 (0.22–1.14) 9/221 0.60� (0.34–1.01) 21/224

Levels of physical activity: first tertile (men\65 min/day, women\118 min/day), second tertile (men 65–135 min/day, women 118–193 min/

day), third tertile (men [135 min/day, women [193 min/day)

Adjusted for age: * p \ 0.05; � p = 0.08

HR hazard ratio, CI confidence interval


123

Acknowledgments The Longitudinal Aging Study Amsterdam is

largely supported by a Grant from the Netherlands Ministry of Health

Welfare and Sports, Directorate of Nursing Care and Older Persons.

References

1. Johnell O, Kanis JA (2006) An estimate of the worldwide prev-

alence and disability associated with osteoporotic fractures. Os-

teoporos Int 17(12):1726–1733

2. Nguyen ND, Ahlborg HG, Center JR, Eisman JA, Nguyen TV

(2007) Residual lifetime risk of fractures in women and men.

J Bone Miner Res 22(6):781–788

3. Johnell O, Kanis J (2005) Epidemiology of osteoporotic fractures.

Osteoporos Int 16(Suppl 2):S3–S7

4. Leboime A, Confavreux CB, Mehsen N, Paccou J, David C, Roux

C (2010) Osteoporosis and mortality. Joint Bone Spine 77(Suppl

2):S107–S112

5. Rikkonen T, Sirola J, Salovaara K, Tuppurainen M, Jurvelin JS,

Honkanen R, Kroger H (2012) Muscle strength and body com-

position are clinical indicators of osteoporosis. Calcif Tissue Int

91(2):131–138

6. Taaffe DR, Simonsick EM, Visser M, Volpato S, Nevitt MC,

Cauley JA, Tylavsky FA, Harris TB (2003) Lower extremity

physical performance and hip bone mineral density in elderly

black and white men and women: cross-sectional associations in

the Health ABC Study. J Gerontol 58(10):M934–M942

7. Karkkainen M, Rikkonen T, Kroger H, Sirola J, Tuppurainen M,

Salovaara K, Arokoski J, Jurvelin J, Honkanen R, Alhava E

(2009) Physical tests for patient selection for bone mineral den-

sity measurements in postmenopausal women. Bone

44(4):660–665

8. Khazzani H, Allali F, Bennani L, Ichchou L, El Mansouri L,

Abourazzak FE, Abouqal R, Hajjaj-Hassouni N (2009) The

relationship between physical performance measures, bone min-

eral density, falls, and the risk of peripheral fracture: a cross-

sectional analysis. BMC Public Health 9:297

9. Cheung CL, Tan KC, Bow CH, Soong CS, Loong CH, Kung AW

(2012) Low handgrip strength is a predictor of osteoporotic

fractures: cross-sectional and prospective evidence from the

Hong Kong Osteoporosis Study. Age 34(5):1239–1248

10. Stel VS, Pluijm SM, Deeg DJ, Smit JH, Bouter LM, Lips P

(2004) Functional limitations and poor physical performance as

independent risk factors for self-reported fractures in older per-

sons. Osteoporos Int 15(9):742–750

11. Zhu K, Devine A, Lewis JR, Dhaliwal SS, Prince RL (2011)

‘‘Timed up and go’’ test and bone mineral density measurement

for fracture prediction. Arch Intern Med 171(18):1655–1661

12. Rosengren BE, Ribom EL, Nilsson JA, Mallmin H, Ljunggren O,

Ohlsson C, Mellstrom D, Lorentzon M, Stefanick M, Lapidus J,

Leung PC, Kwok A, Barrett-Connor E, Orwoll E, Karlsson MK

(2012) Inferior physical performance test results of 10,998 men in

the MrOS Study is associated with high fracture risk. Age Ageing

41(3):339–344

13. Rouzi AA, Al-Sibiani SA, Al-Senani NS, Radaddi RM, Ardawi

MS (2012) Independent predictors of all osteoporosis-related

fractures among healthy Saudi postmenopausal women: the

CEOR Study. Bone 50(3):713–722

14. Murtagh KN, Hubert HB (2004) Gender differences in physical

disability among an elderly cohort. Am J Public Health

94(8):1406–1411

15. Makovey J, Naganathan V, Sambrook P (2005) Gender differ-

ences in relationships between body composition components,

their distribution and bone mineral density: a cross-sectional

opposite sex twin study. Osteoporos Int 16(12):1495–1505

16. Huisman M, Poppelaars J, van der Horst M, Beekman AT, Brug

J, van Tilburg TG, Deeg DJ (2011) Cohort profile: the Longitu-

dinal Aging Study Amsterdam. Int J Epidemiol 40(4):868–876

17. Stel VS, Smit JH, Pluijm SM, Visser M, Deeg DJ, Lips P (2004)

Comparison of the LASA Physical Activity Questionnaire with a

7-day diary and pedometer. J Clin Epidemiol 57(3):252–258

18. Pluijm SM, Visser M, Smit JH, Popp-Snijders C, Roos JC, Lips P

(2001) Determinants of bone mineral density in older men and

women: body composition as mediator. J Bone Miner Res

16(11):2142–2151

19. Sohl E, de Jongh RT, Heijboer AC, Swart KM, Brouwer-Brolsma

EM, Enneman AW, de Groot CP, van der Velde N, Dhonukshe-

Rutten RA, Lips P, van Schoor NM (2013) Vitamin D status is

associated with physical performance: the results of three inde-

pendent cohorts. Osteoporos Int 24(1):187–196

20. Kuchuk NO, Pluijm SM, van Schoor NM, Looman CW, Smit JH,

Lips P (2009) Relationships of serum 25-hydroxyvitamin D to

bone mineral density and serum parathyroid hormone and

markers of bone turnover in older persons. J Clin Endocrinol

Metab 94(4):1244–1250

21. Cawthon PM, Fullman RL, Marshall L, Mackey DC, Fink HA,

Cauley JA, Cummings SR, Orwoll ES, Ensrud KE (2008)

Physical performance and risk of hip fractures in older men.

J Bone Miner Res 23(7):1037–1044

22. Owings TM, Pavol MJ, Grabiner MD (2002) Lower extremity

muscle strength does not independently predict proximal femur

bone mineral density in healthy older adults. Bone 30(3):515–520

23. Ribom E, Olofsson H, Piehl-Aulin K, Mallmin H, Ljunghall S

(1999) Correlations between isometric quadriceps muscle

strength and bone mineral density. J Musculoskelet Res

3(4):275–284

24. Segal NA, Torner JC, Yang M, Curtis JR, Felson DT, Nevitt MC

(2008) Muscle mass is more strongly related to hip bone mineral

density than is quadriceps strength or lower activity level in

adults over age 50 year. J Clin Densitom 11(4):503–510

25. Stewart KJ, Deregis JR, Turner KL, Bacher AC, Sung J, Hees PS,

Tayback M, Ouyang P (2002) Fitness, fatness and activity as

predictors of bone mineral density in older persons. J Intern Med

252(5):381–388

26. Lim S, Joung H, Shin CS, Lee HK, Kim KS, Shin EK, Kim HY,

Lim MK, Cho SI (2004) Body composition changes with age

have gender-specific impacts on bone mineral density. Bone

35(3):792–798

27. Pijnappels M, van der Burg PJ, Reeves ND, van Dieen JH (2008)

Identification of elderly fallers by muscle strength measures. Eur

J Appl Physiol 102(5):585–592

28. Hauer K, Rost B, Rutschle K, Opitz H, Specht N, Bartsch P, Oster

P, Schlierf G (2001) Exercise training for rehabilitation and

secondary prevention of falls in geriatric patients with a history of

injurious falls. J Am Geriatr Soc 49(1):10–20

29. Freiberger E, Haberle L, Spirduso WW, Zijlstra GA (2012) Long-

term effects of three multicomponent exercise interventions on

physical performance and fall-related psychological outcomes in

community-dwelling older adults: a randomized controlled trial.

J Am Geriatr Soc 60(3):437–446

30. Barnett A, Smith B, Lord SR, Williams M, Baumand A (2003)

Community-based group exercise improves balance and reduces

falls in at-risk older people: a randomised controlled trial. Age

Ageing 32(4):407–414

31. Clemson L, Fiatarone Singh MA, Bundy A, Cumming RG, Ma-

nollaras K, O’Loughlin P, Black D (2012) Integration of balance

and strength training into daily life activity to reduce rate of falls

in older people (the LiFE study): randomised parallel trial. BMJ

345:e4547

32. Cadore EL, Rodriguez-Manas L, Sinclair A, Izquierdo M (2013)

Effects of different exercise interventions on risk of falls, gait


123

ability, and balance in physically frail older adults: a systematic

review. Rejuvenation Res 16(2):105–114

33. Province MA, Hadley EC, Hornbrook MC, Lipsitz LA, Miller JP,

Mulrow CD, Ory MG, Sattin RW, Tinetti ME, Wolf SL (1995)

The effects of exercise on falls in elderly patients. A preplanned

meta-analysis of the FICSIT Trials. Frailty and injuries: cooper-

ative studies of intervention techniques. JAMA 273(17):

1341–1347

34. Stel VS, Smit JH, Pluijm SM, Lips P (2003) Balance and mobility

performance as treatable risk factors for recurrent falling in older

persons. J Clin Epidemiol 56(7):659–668

35. Peeters GM, Verweij LM, van Schoor NM, Pijnappels M, Pluijm

SM, Visser M, Lips P (2010) Which types of activities are

associated with risk of recurrent falling in older persons? J Ger-

ontol 65(7):743–750

36. Wust RC, Morse CI, de Haan A, Jones DA, Degens H (2008) Sex

differences in contractile properties and fatigue resistance of

human skeletal muscle. Exp Physiol 93(7):843–850

37. Hoidrup S, Sorensen TI, Stroger U, Lauritzen JB, Schroll M,

Gronbaek M (2001) Leisure-time physical activity levels and

changes in relation to risk of hip fracture in men and women. Am

J Epidemiol 154(1):60–68


123

gender-specific associations between physical functioning, bone quality, and fracture risk in older...

Documents