Contents lists available at ScienceDirect [610012]

Contents lists available at ScienceDirect
Journal of A ffective Disorders
journal homepage: www.elsevier.com/locate/jad
Research paper
Associations between grip strength and generalized anxiety disorder in older
adults: Results from the Irish longitudinal study on ageing
Brett R. Gordona,b,1,⁎, Cillian P. McDowella,b, Mark Lyonsa, Matthew P. Herringa,b
aDepartment of Physical Education and Sport Sciences, University of Limerick, Limerick, Ireland
bPhysical Activity for Health Research Cluster, Health Research Institute, University of Limerick, Limerick, Ireland
ARTICLE INFO
Keywords:
Prospective cohortMuscular weaknessCross-sectional
Worry
ElderlyABSTRACT
Background: Associations between grip strength and mental health disorders have been established; however,
associations between grip strength and Generalized Anxiety Disorder (GAD) remain unstudied. Therefore, this
study investigates associations between grip strength and prevalent and incident GAD.Methods: A prospective cohort design was utilized. At baseline, participants aged ≥50 years ( N= 3,952)
completed a hand grip strength assessment and abbreviated Penn State Worry Questionnaire (PSWQ) and weredivided into sex-speci fic tertiles based on strength. A score of ≥23 on the PSWQ de fined caseness of GAD. At
two-year follow-up, GAD was assessed with the Composite International Diagnostic Interview —Short Form.
Results: Prevalence and incidence of GAD were 18.2% ( N= 718) and 0.9% ( N= 29), respectively. Adjusting for
age, sex, waist circumference, social class, smoking status, and physical activity, a one-standard-deviation (1-SD)increase in strength was associated (OR, 95%CI) with 12.1% (OR = 0.88, 0.80 –0.96; p< 0.01) lower odds of
prevalent GAD, and middle and high strength tertiles were associated with 27.3% (OR = 0.73, 0.59 –0.89;
p< 0.01) and 23.1% (OR = 0.77, 0.62 –0.95; p< 0.05) lower odds, respectively. A 1-SD increase in strength
was non-signi ficantly associated with 24.2% (OR = 0.76, 0.50 –1.14) lower odds of incident GAD, and middle
and high strength tertiles were non-signi ficantly associated with 31.4% (OR = 0.69, 0.30 –1.58) and 66.5%
(OR = 0.34, 0.11 –1.00) lower odds, respectively (all p> 0.05). There was no signi ficant interaction between
strength tertiles and sex.Limitations: The observational nature of the study limits inferring causality.
Conclusions: Increased hand grip strength may be associated with lower odds of developing GAD in older adults.
Larger investigations of prospective associations are needed.
1. Introduction
Anxiety disorders are the most prevalent mental disorders, with
lifetime prevalence of ∼30% ( Bandelow and Michaelis, 2015), and the
sixth leading cause of disability ( WHO, 2017 ). Generalized anxiety
disorder (GAD) is a highly prevalent anxiety disorder, with a 12-month
prevalence of approximately 3.4% in adults aged >65 years and 1.7%
in adults aged <66 years in Europe in 2010 ( Wittchen et al., 2011 ). The
economic burden of GAD has been reported at €1684 per case in Ger-
many ( Jacobi et al., 2002), and in Australia, total care costs for GAD are
$112.3 million (AUD) ( Issakidis et al., 2004 ). Frontline treatments such
as cognitive behavioral therapy, and pharmacotherapy, have demon-
strated e fficacy for the treatment of GAD ( Mitte, 2005 ); however, many
patients maintain residual symptoms, and treatment resistance occursdue to multiple factors ( Bystritsky, 2006 ). Thus, there is a need to
identify risk factors for GAD that can be targeted for prevention.
There is a growing interest in the importance of objective physical
fitness tests as indicators of physical and mental outcomes in older
adults. One such test, the maximal isometric handgrip strength test, is
the most widely used and practical measure of muscular function for
clinical settings ( Roberts et al., 2011 ;Leong et al., 2015). It has in-
creasingly been identi fied as an important physical health marker in
older adults ( Peterson et al., 2017 ), and has been cross-sectionally as-
sociated with mental health outcomes such as anxiety in people aged50+ years ( Cheung et al., 2013 ), and cognition in samples of 476,559
UK adults ( J. Firth et al., 2018b) and 24,174 patients with major de-
pression and bipolar disorder ( J. Firth et al., 2018a ). . Prospectively, it
has been associated with onset of depression ( Kim, 2019;
https://doi.org/10.1016/j.jad.2019.05.043
Received 5 February 2019; Received in revised form 14 May 2019; Accepted 24 May 2019⁎Corresponding author. Department of Physical Education and Sport Sciences, University of Limerick, PESS 1039, Limerick, Ireland.
E-mail addresses: brett.gordon@ul.ie (B.R. Gordon), cillian.mcdowell@ul.ie (C.P. McDowell), mark.lyons@ul.ie (M. Lyons),
matthew.herring@ul.ie (M.P. Herring).
1Declaration of interest: none.Journal of Affective Disorders 255 (2019) 136–141
Available online 25 May 2019
0165-0327/ © 2019 Elsevier B.V. All rights reserved.
T

Fukumori et al., 2015 ;McDowell et al., 2018a ;Veronese et al., 2017 )
and cognitive decline ( Veronese et al., 2016 ) among older adults.
Further, randomized controlled trials have demonstrated that resistance
exercise training can signi ficantly reduce anxiety and depressive
symptoms among older adults ( Gordon et al., 2017; Gordon et al.,
2018 ), and improvements in anxiety and depressive symptoms were not
moderated by features of the resistance exercise stimulus, or by sig-nificant improvements in strength. Among young adults, resistance
exercise training can improve clinical severity ( Herring et al., 2012 )
and associated signs and symptoms of GAD ( Herring et al., 2011, 2015,
2016 ). However, prospective associations between grip strength and
GAD remain unknown. Thus, the objectives of this study were (1) to
investigate associations between grip strength and prevalent and in-
cident GAD, and (2) to investigate whether stronger associations were
evident with higher grip strength. The authors hypothesized that grip
strength would be associated with lower odds of GAD, with stronger
associations evident for higher grip strength.
2. Methods
This study used STROBE recommendations to guide reporting
(Von Elm et al., 2008 ).
2.1. Study population
The Irish Longitudinal Study on Ageing (TILDA) is a large pro-
spective cohort study that assesses social, economic, and health cir-
cumstances of community dwelling adults aged ≥50 years and their
partners of any age in Ireland. An initial multi-stage probability sampleof addresses was chosen by means of the RANSAM sampling procedure
(Whelan, 1979), with District Electoral Divisions selected at the first
stage and household addresses selected at the second stage. The re-
sponse rate was 62.0%. Participants gave full informed consent to
participate, and ethical approval was obtained from the Trinity College
Dublin Faculty of Health Sciences Research Ethics Committee.
The present study analysed data from TILDA Waves 1 (2009 –2010)
(Barrett et al., 2011 ) and 2 (2012 –2013) ( Nolan et al., 2014 ). Re-
spondents ( N= 4756) with complete grip strength and Penn State
Worry Questionnaire (PSWQ-A) data at Wave 1 and Composite Inter-
national Diagnostic Interview –Short Form (CIDI-SF) data at Wave 2
were included in the present analyses. Respondents ( N= 787) were
excluded for incomplete covariate data, resulting in a sample size of
3952 for cross-sectional analyses. Participants ( N= 3234) free of GAD
at baseline were included in the longitudinal analyses.
2.2. Grip strength
Grip strength was measured at baseline using a hydraulic hand
dynamometer (Fabrication Enterprises Inc, White Plains, NY).
Respondents were divided into sex-speci fic tertiles based on the mean
of their two scores (i.e., males: 0.0 –29.5, 30.0 –37.0, and 37.5 –65.0 kg,
respectively; females: 0– 17.5, 18.0 –22.0, and 22.5 –38.5
kg, respec-
tively). Standardized grip strength was derived by dividing a re-spondent's mean grip score by the sex-speci fic sample grip strength
standard deviation.
2.3. Generalized anxiety disorder
At Wave 1, worry symptom severity and GAD caseness were as-
sessed with the abbreviated Penn State Worry Questionnaire (PSWQ-A)
(Hopko et al., 2003 ). In the present sample, the PSWQ-A demonstrated
excellent internal consistency (Cronbach's α= 0.94) ( Stanley et al.,
2011 ). A score of ≥23 defi ned caseness of GAD. This cut-o ffscore has
been shown to correctly identify clinical status among older adults with
66.4% sensitivity and 92.5% speci ficity ( Wuthrich et al., 2014 ). At
Wave 2, the CIDI-SF ( Kessler et al., 1998) assessed whether respondentsfulfilled the Diagnostic and Statistical Manual of Mental Disorders-IV
criteria for GAD ( APA, 2000 ). The CIDI-SF has been shown to correctly
classify CIDI cases of GAD with an accuracy of 99.6% ( Kessler et al.,
1998 ).
2.4. Covariates
Covariates included age, sex, waist circumference, and social class.
Age was classi fied using four categories (i.e., 50– 59 years, 60 –69 years,
70–79 years, and 80+ years). Waist circumference was classi fied as
normal or above normal according to World Health Organization
guidelines (i.e., Males: >94 cm; Females >80 cm) ( WHO, 2008 ). Social
class was defi ned according to the European Socioeconomic Classi fi-
cation (EseC) Scheme ( Rose and Harrison, 2007 ). In order to improve
sample coverage, those who were not in paid employment were allo-cated to a ‘Not Working ’group ( N= 1034). Smoking status was self-
reported as current or non-current smoker. Physical activity (PA) wasmeasured using the International Physical Activity Ques-
tionnaire —Short Form ( Craig et al., 2003 ), for which respondents were
asked to report the number of days and duration of the vigorous,moderate, and walking activities they undertook during the last seven
days. Data were processed according to the IPAQ guidelines for data
processing and analysis. Respondents who reported PA greater than a
combined 16-hours per day were excluded ( N= 7).
2.5. Statistical analysis
Statistical analyses were conducted using SPSS Version 22.0
(Armonk, NY: IBM Corp.). Chi-square tests examined di fferences in sex,
age, waist circumference, and social class categories between strengthtertiles. For signi ficant Chi-square tests, Ztests
were calculated for
column proportions for each row in the Chi-square contingency tableand Bonferroni-corrected for multiple comparisons ( Sharpe, 2015 ).
One-way ANOVAs quanti fied differences in Wave 1 worry symp-
toms between strength tertiles. Two-way ANOVAs examined variation
in Wave 1 worry symptoms by strength tertiles, sex, and their inter-
action. Binomial logistic regression quanti fied associations (i.e., odds
ratios (ORs)) between strength tertiles and standardized grip strength(treated as a continuous variable) and GAD status at Waves 1 and 2.
Covariates in adjusted models were age, sex, waist circumference, so-
cial class, smoking status, and weekly hours of MVPA. Due to the known
association of MVPA with both grip-strength ( McDowell et al., 2019 )
and GAD ( McDowell et al., 2018c ), sensitivity analyses excluding it as a
covariate were run. The Hosmer-Lemeshow test was performed and
Nagelkerke R
2calculated to assess the goodness-of- fit of the models.
Likelihood ratio tests examined covariate signi ficance.
3. Results
3.1. Participant characteristics
Compared to the analytic sample, a signi ficantly greater proportion
of excluded respondents (i.e., those with full grip strength, PSWQ-A,
and CIDI data, but missing covariate data) were aged 50 –59 years
(58.7% vs 41.1%) and male (50.9% vs 43.9%; both p< 0.05). Excluded
participants engaged in more weekly MVPA than included participants
(1.70 h; 95%CI: 0.69 –2.71). Excluded males were stronger than in-
cluded males (Mean di fference, 95%CI: 2.36 kg, 1.46 –3.27), and ex-
cluded females were stronger than included females (Mean di fference,
95%CI: 2.49 kg, 1.90 –3.08). There were no signi ficant di fferences in
waist circumference or GAD status.
Baseline participant characteristics are presented in Table 1. Table 2
presents the proportion of people with GAD at Waves 1 and 2 by levelsof strength and signi ficant covariates. Age ( p< 0.001) signi ficantly
differed between strength tertiles, but sex, waist circumference status,
smoking status, and social class did not (all p> 0.05). Results fromB.R. Gordon, et al. Journal of Affective Disorders 255 (2019) 136–141
137

follow-up tests are shown in Table 1. Brie fly, a signi ficantly greater
proportion of respondents aged 70 –79 and 80+ years were in the
lowest strength tertile ( p< 0.05).
3.2. Cross-sectional analyses
Worry symptoms did not signi ficantly di ffer by strength tertile. Theprevalence of GAD was 18.2% ( N= 718). Table 3 presents ORs for the
associations between strength and GAD status. Model 2 results are
summarized here: Compared to the low strength tertile, the middle and
high strength tertiles were signi ficantly associated with 27.3%
(p≤0.002) and 23.1% ( p< 0.02) lower odds of GAD, respectively. The
model fit the data well ( χ2(8) = 3.93, p≥0.86; R2= 0.09). Age, sex,
social class, and smoking status were signi ficant covariates (all
p< 0.01).
A one-standard-deviation increase in strength was associated with a
12.1% ( p≤0.004) lower odds of GAD. The model fit the data well
(χ2(8) = 5.63, p≥0.69; R2= 0.08). Age, sex, social class, and
smoking status were signi ficant covariates (all p≤0.01). Results did
not meaningfully di ffer in sensitivity analyses (see Supplementary
Table 1).
3.3. Longitudinal analyses
The incidence of GAD was 0.9% ( N= 29). Table 4 presents ORs for
the associations between strength and GAD status. Model 2 results are
summarized here: compared to the low strength tertile, the middle and
high strength tertiles were associated with 31.4% ( p≥0.38) and 66.5%
(p= 0.050045) reduced odds of GAD. The model fit the data well
(χ2(8) = 3.76, p≥0.88; R2= 0.10). Sex and smoking were signi ficant
covariates (both p< 0.01).
A one-standard-deviation increase in strength was associated with a
24.2% ( p> 0.18) reduced odds of GAD. The model fit the data well
(χ2(8) = 3.31, p≥0.91; R2= 0.10). Sex and smoking status were
significant covariates (both p< 0.01). [ Table 4 here]. Results did not
meaningfully di ffer in sensitivity analyses (see Supplementary Table 2).
3.4. Potential sex-related di fferences
Wave 1 worry symptoms were signi ficantly higher among females
(16.9 ± 8.0) than males (14.1 ± 6.8; F(1,3970) = 142.37, p< 0.001;
g= 0.38; 95%CI: 0.32 –0.45). Females were signi ficantly more likely to
report scores indicative of GAD at baseline (23.4% vs 11.5%,
p< 0.001). Males (33.6 ± 8.5 kg) were signi ficantly stronger than
females (19.8 ± 5.4 kg; F (1,3970) =3878.34, p< 0.001; g= 1.99; Mean
difference, 95%CI: 1.91 –2.07 kg).
There was no signi ficant interaction between strength tertiles and
sex in predicting Wave 1 worry symptoms ( F(2,3946) = 0.32, p> 0.72).
There was no signi ficant interaction between sex and strength tertiles or
continuous strength in cross-sectional or longitudinal logistic regression
models (all p> 0.20)Table 1
Baseline characteristics of included sample.
Tertile 1
(N= 1284) n(%)Tertile 2(N= 1393) n(%)Tertile 3(N= 1275) n(%)
PSWQ-A score
(8–40)
(mean ± SD)15.93 ± 7.86 15.51 ± 7.46 15.55 ± 7.54
Grip strength (kg)
(mean ± SD)18.35 ± 6.32 25.74 ± 6.80 33.51 ± 9.23
MVPA (hours)
(median (IQR))0.21 (7.0) 2.00 (10.0) 4.0 (14.0)
Sex
Male 550 (42.8) 603 (43.3) 582 (45.6)Female 734 (57.2) 790 (56.7) 693 (54.4)
Age category (years)
50–59 326 (25.4)
a563 (40.4)b735 (57.6)c
60–69 457 (35.6)a,b514 (39.6)b410 (32.2)a
70–79 369 (28.7)a274 (19.7)b121 (9.5)c
80+ 132 (10.3)a42 (3.0)b9 (0.7)c
Waist circumference category
Normal 629 (49.0) 687 (49.3) 602 (47.2)Above normal 655 (51.0) 706 (50.7) 673 (52.8)
Social class
Professional
workers45 (3.5) 66 (4.7) 50 (3.9)
Managerial and
technical301 (23.4) 355 (25.5) 383 (30.0)
Non-manual 204 (15.9) 224 (16.1) 186 (14.6)Skilled manual 129 (10.0) 128 (9.2) 116 (9.1)Semi-skilled 137 (10.7) 125 (9.0) 126 (9.9)Unskilled 40 (3.1) 40 (2.9) 30 (2.4)Farmers 71 (5.5) 84 (6.0) 78 (6.1)Not working 357 (27.8) 371 (26.6) 306 (24.0)
Current smoker
Yes 169 (13.2) 216 (15.5) 192 (15.1)No 1115 (86.8) 1177 (84.5) 1083 (84.9)
Each superscript letter denotes a subset of PA Guidelines categories whose
column proportions do not di ffer signi ficantly from each other at the 0.05 level
(ap< 0.05,bp< 0.01,cp< 0.001). Abbreviations: PSWQ-A: Penn State Worry
Questionnaire; BMI: Body Mass Index; IQR: interquartile range; kg: kilogram;
m: meter; MVPA: moderate-to-vigorous physical activity; wk: week; yrs: years.
Table 2Proportion of people with GAD at Waves 1 and 2 by levels of strength and signi ficant covariates.
Covariate Wave 1 Wave 2
Strength tertile (cases/persons at risk) Total Strength tertile (cases/persons at risk) Total
Low Moderate High Low Moderate High
Age (years)
50–59 100/326 119/563 152/735 371/1624 –– – –
60–69 92/457 75/514 60/410 227/1381 –– – –
70–79 47/369 35/274 19/121 101/764 –– – –
80+ 15/132 3/42 1/9 19/183 –– – –
Waist circumference
Normal 133/629 126/687 118/602 377/1918 –– – –
At risk 121/655 106/706 114/673 341/2034 –– – –
Sex
Male 70/550 64/603 65/582 199/1735 2/480 1/539 2/517 5/1536
Female 184/734 168/790 167/693 519/2217 11/550 10/622 3/526 24/1698
Smoking status
Current 46/169 45/216 52/192 575/3375 8/907 8/990 3/903 29/2800
Past/Never 208/1115 187/1177 180/1083 143/577 5/123 3/171 2/140 10/434
Abbreviations: GAD: generalized anxiety disorder.B.R. Gordon, et al. Journal of Affective Disorders 255 (2019) 136–141
138

4. Discussion
To the authors ’knowledge, this is the first study to examine pro-
spective associations of grip strength with diagnosed generalized an-
xiety disorder. The findings demonstrated large, signi ficant, inverse,
cross-sectional associations between grip strength and GAD amongolder adults, with stronger associations found among those with greater
strength. Large, inverse prospective associations were also observed;
however, findings were statistically non-signi ficant, likely as the in-
cidence rate of GAD was too small to provide good precision of thepoint estimate. Although worry symptoms were signi ficantly higher
among females compared to males, there was no signi ficant interaction
between strength and sex.
Thefindings of lower odds of GAD for each 1-SD increase in grip
strength are consistent with previously reported cross-sectional findings
of increased odds of anxiety with each 1-SD decrease in grip strength
(Cheung et al., 2013 ). The magnitude of the longitudinal associations
between increased grip strength and odds of GAD is similar to pre-viously reported findings for depression among older adults
(Fukumori et al., 2015 ), as well as cognition in elderly women cancer
survivors ( Yang et al., 2018 ). Additionally, this finding is consistent
with previous investigations of the same TILDA sample, which foundthat each 1-SD increase in grip strength was associated with a sig-
nificant 16.4% lower odds of depression with ( McDowell et al., 2018a ),
and higher volumes of physical activity were associated with loweranxiety status and symptoms ( McDowell et al., 2018b ). These consistent
positive associations of grip strength with mental health highlight theimportance of maintaining strength in old age.
Smoking was a signi ficant covariate in all analyses, with current
smoking (vs past/never) associated with increased risk of developing
GAD (OR = 2.44, 95%CI:1.63 –3.65). Studies suggest that smoking in
early life may increase the risk of developing an anxiety disorder,people with an anxiety disorder may smoke to reduce anxiety, and
shared vulnerabilities may increase the risk of both ( Jiang et al.,
2014;Moylan, Jacka, Pasco, and Berk, 2012 ). Further, recent cross-
sectional findings from TILDA have demonstrated that older adults in
Ireland with mental health di fficulties, including anxiety, are morelikely to smoke ( Burns et al., 2017 ).
As resistance exercise training (RET) improves strength and func-
tional limitations ( Latham et al., 2004 ;Peterson et al., 2010), and
randomized controlled trials have shown that RET reduces anxietysymptoms in older adults ( Gordon et al., 2017 ), the current study fur-
ther support for RET for older adults to improve strength and protectagainst GAD. Declines in strength may be associated with increased
worry symptoms through decreases in physical function
(Wolitzky ‐Taylor et al., 2010 ).
RET addresses this potential underlying
mechanism by delaying the onset of physical disability in the elderly(Roth et al., 2000 ), or prolonged exposure to the symptoms of GAD may
contribute to greater decline in physical function in older adults
(Kim, 2019), further supporting the recommendation of RET for older
adults for physical and mental health bene fits.
Although the current study examines a population of older adults,
the bene fits of muscular strength is not limited to this population. For
example, muscle strengthening exercise (i.e., RET) is also anxiolytic
among young adults ( Gordon et al., 2017 ). Speci fically, in young adults
with GAD, RET can improve clinical severity ( Herring et al., 2012) and
associated signs and symptoms of GAD ( Herring et al., 2011, 2015,
2016 ). Adolescents with higher muscular strength are less likely to have
a psychiatric diagnosis and are at lower risk for mortality due to suicide
(Ortega et al., 2012 ). Unlike physical disorders, GAD typically arises in
early adulthood ( de Lijster et al., 2017), so finding e ffective treatments
and preventions for this age-group is crucial. Additional research on
physical activity and strength training interventions in these popula-
tions, particularly those with emerging signs of GAD, is needed.
Although strengths of the current study are the prospective design,
objective and standardized assessment of strength, adjustment for reg-
ular physical activity levels, and large sample size, the observational
nature of the study limit inferring causality. RCTs focused on RET are
needed to establish causality and to more rigorously explore potential
sex-related di fferences. Additionally, the assessment of GAD using dif-
ferent measures at Waves 1 and 2 was not ideal; however; the CIDI-SFwas only introduced in Wave 2 of TILDA. Further, the use of grip
strength alone as a proxy for strength in older adults is a limitation;
however, combining quadriceps strength and hand grip strength inTable 3
Odds ratios (OR) and 95% con fidence intervals (CI) derived from binominal logistic regression analyses as indicators of cross-sectional associations between grip
strength and GAD.
Grip strength Cases/Persons at risk Model 1 ORa(95% CI) Model 2 ORb(95% CI)
Grip-strength (per 1SD increase) 718/3952 0.86 (0.79 to 0.94)⁎⁎⁎0.88 (0.80 to 0.96)⁎⁎
Low 254/1284 REF REF
Moderate 232/1393 0.71 (0.58 to 0.87)⁎⁎0.73 (0.59 to 0.89)⁎⁎
High 232/1275 0.73 (0.59 to 0.90)⁎⁎0.77 (0.62 to 0.95) *
Abbreviations: REF: reference category, SD: standard deviation.
⁎p< 0.05.
⁎⁎p< 0.01.
⁎⁎⁎p< 0.001.
aAdjusted for age and sex.
bAdjusted for age, sex, waist circumference, social class, smoking status, & MVPA.
Table 4
Odds ratios (OR) and 95% con fidence intervals (CI) derived from binominal logistic regression analyses as indicators of prospective associations between grip
strength and GAD.
Grip strength Cases/Persons at risk Model 1 ORa(95% CI) Model 2 ORb(95% CI)
Grip-strength (per 1SD increase) 29/3234 0.77 (0.52 to 1.14) 0.76 (0.50 to 1.14)
Low 13/1029 REF REFModerate 11/1162 0.70 (0.30 to 1.59) 0.69 (0.30 to 1.58)High 5/1043 0.34 (0.12 to 1.01) 0.34 (0.11 to 1.00)
Abbreviations: REF: reference category, SD: standard deviation.
aAdjusted for age and sex.
bAdjusted for age, sex, waist circumference, social class, smoking status, & MVPA.B.R. Gordon, et al. Journal of Affective Disorders 255 (2019) 136–141
139

older adults have been reported to not provide additional value to
handgrip strength alone in the prediction of poor health outcomes
(Chan et al., 2014).
5. Conclusions
The current findings support the protective e ffect of grip strength on
GAD in older adults. However, longitudinal findings were not statisti-
cally signi ficant after adjustment for potential confounders, likely due
to the relatively low incidence of GAD across two-year follow-up. Thesefindings support maintaining strength levels in later life for both phy-
sical and mental health bene fits.
Authors ’contributions
All authors made substantial contributions to conception and de-
sign, acquisition of data, and preparation of the manuscript.
Role of the funding source
The TILDA study was supported by the Irish Government, the
Atlantic Philanthropies, and Irish Life PLC. This work was also sup-
ported by the Health Research Board of Ireland (grant reference:
HRA_PHS/2012/30). The sponsors played no role in study design,
methods, subject recruitment, data collection, analysis or preparation of
the paper. Brett R. Gordon is funded by the Irish Research Council
under the Government of Ireland Postgraduate Programme
Acknowledgments
Declaration of interest statement: all authors have no con flicts of
interest.
Supplementary materials
Supplementary material associated with this article can be found, in
the online version, at doi: 10.1016/j.jad.2019.05.043 .
References
American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental
Disorders, Fourth ed. American Psychiatric Association Press, Washington DC .
Bandelow, B., Michaelis, S., 2015. Epidemiology of anxiety disorders in the 21st century.
Dialogues Clin. Neurosci. 17 (3), 327 PMID: 26487813 .
Barrett, A., Burke, H., Cronin, H., et al., 2011. Fifty Plus in Ireland 2011: First Results
from the Irish Longitudinal Study on Ageing (TILDA). Trinity College Dublin, Dublin.
http://epubs.rcsi.ie/psycholrep/45 .
Burns, A., Strawbridge, J.D., Clancy, L., et al., 2017. Exploring smoking, mental health
and smoking-related disease in a nationally representative sample of older adults in
Ireland –A retrospective secondary analysis. J. Psychosom. Res. 98, 78 –86.https://
doi.org/10.1016/j.jpsychores.2017.05.005 .
Bystritsky, A., 2006. Treatment-resistant anxiety disorders. Mol. Psychiatry 11 (9), 805.
https://doi.org/10.1038/sj.mp.4001852 .
Chan, O.Y.A., van Houwelingen, A.H., Bussekloo, J., et al., 2014. Comparison of quad-
riceps strength and handgrip strength in their association with health outcomes in
older adults in primary care. Age 36 (5), 9714. https://doi.org/10.1007/s11357-014-
9714-4 .
Cheung, C.L., Nguyen, U.S.D., Au, E., et al., 2013. Association of handgrip strength with
chronic diseases and multimorbidity. Age 35 (3), 929 –941. https://doi.org/10.1007/
s11357-012-9385-y .
Craig, C.L., Marshall, A.L., Sjöström, M., et al., 2003. International physical activity
questionnaire: 12-country reliability and validity. Med. Sci. Sports Exerc. 35 (8),1381 –1395. https://doi.org/10.1249/01.MSS.0000078924.61453.FB .
Firth, J., Firth, J.A., Stubbs, B., et al., 2018a. Association between muscular strength and
cognition in people with major depression or bipolar disorder and healthy controls.JAMA Psychiatry 75 (7), 740 –746. https://doi.org/10.1001/jamapsychiatry.2018.
0503 .
Firth, J., Stubbs, B., Vancampfort, D., et al., 2018b. Grip strength is associated with
cognitive performance in schizophrenia and the general population: a UK biobank
study of 476559 participants. Schizophr. Bull. 44 (4), 728 –736. https://doi.org/10.
1093/schbul/sby034 .
Fukumori, N., Yamamoto, Y., Takegami, M., et al., 2015. Association between hand-grip
strength and depressive symptoms: locomotive syndrome and health outcomes inAizu cohort study (LOHAS). Age Ageing 44 (4), 592 –598. https://doi.org/10.1093/
ageing/afv013 .
Gordon, B.R., McDowell, C.P., Lyons, M., et al., 2017. The e ffects of resistance exercise
training on anxiety: a meta-analysis and meta-regression analysis of randomized
controlled trials. Sports Med. 47 (12), 2521 –2532. https://doi.org/10.1007/s40279-
017-0769-0 .
Gordon, B.R., McDowell, C.P., Hallgren, M., et al., 2018. Association of e fficacy of re-
sistance exercise training with depressive symptoms; meta-analysis and meta-re-
gression analysis of randomized clinical trials. JAMA Psychiatry 75 (6), 566 –576.
https://doi.org/10.1001/jamapsychiatry.2018.0572 .
Herring, M.P., Jacob, M.L., Suveg, C., et al., 2012. Feasibility of exercise training for the
short-term treatment of generalized anxiety disorder: a randomized controlled trial.
Psychother. Psychosom. 81 (1), 21 –28.https://doi.org/10.1159/000327898 .
Herring, M.P., Jacob, M.L., Suveg, C., et al., 2011. E ffects of short-term exercise training
on signs and symptoms of generalized anxiety disorder. Ment. Health Phys. Act 4 (2),
71–77.https://doi.org/10.1159/000327898 .
Herring, M.P., Kline, C.E., O'Connor, P.J., 2015. E ffects
of exercise on sleep among young
women with generalized anxiety disorder. Ment Health Phys Act 9, 59 –66.https://
doi.org/10.1016/j.mhpa.2015.09.002 .
Herring, M.P., Johnson, K.E., O'Connor, P.J., 2016. Exercise training and health-related
quality of life in generalized anxiety disorder. Psychol. Sport Exerc. 27, 138 –141.
https://doi.org/10.1016/j.psychsport.2016.08.011 .
Hopko, D.R., Reas, D.L., Beck, J.G., et al., 2003. Assessing worry in older adults: con-
firmatory factor analysis of the Penn State Worry Questionnaire and psychometric
properties of an abbreviated model. Psychol. Assess 15 (2), 173. http://dx.doi.org/
10.1037/1040-3590.15.2.173 .
Issakidis, C., Sanderson, K., Corry, J., Andrews, G., Lapsley, H., 2004. Modelling the
population cost-e ffectiveness of current and evidence-based optimal treatment for
anxiety disorders. Psychol. Med. 34 (1), 19 –35.
Jacobi, F., Wittchen, H.U., Hölting, C., et al., 2002. Estimating the prevalence of mental
and somatic disorders in the community: aims and methods of the German National
Health Interview and Examination Survey. Int. J. Methods Psychiatr. Res. 11, 1 –18.
https://doi.org/10.1002/mpr.118 .
Jiang, F., Li, S., Pan, L., et al., 2014. Association of anxiety disorders with the risk of
smoking behaviors: a meta-analysis of prospective observational studies. DrugAlcohol Depend. 145, 69 –76.https://doi.org/10.1016/j.drugalcdep.2014.10.022 .
Kessler, R.C., Andrews, G., Mroczek, D., et al., 1998. The world health organization
composite international diagnostic interview short ‐form (cidi‐ sf). Int. J. Methods
Psychiatr. Res 7 (4), 171 –185. https://doi.org/10.1002/mpr.47 .
Kim, J.H., 2019. E ffect of grip strength on mental health. J. A ffect. Disord. 245, 371 –376.
https://doi.org/10.1016/j.jad.2018.11.017 .
Latham, N.K., Bennet, D.A., Stretton, C.M., et al., 2004. Systematic review of progressive
resistance strength training in older adults. J. Gerontol. A Biol. Sci. Med. Sci. 59 (1),48–61.https://doi.org/10.1093/gerona/59.1.m48 .
Leong, D.P., Teo, K.K., Rangarajan, S., et al., 2015. Prognostic value of grip strength:
findings from the prospective urban rural epidemiology (PURE) study. Lancet 386
(9990), 266 –273. https://doi.org/10.1016/S0140-6736(14)62000-6 .
de Lijster, J.M., Dierckx, B., Utens, E.M., et al., 2017. The age of onset of anxiety dis-
orders: a meta-analysis. Can. J. Psychiatry 62 (4), 237. https://doi.org/10.1177/
0706743716640757 .
McDowell, C.P., Gordon, B.R., Herring, M.P., 2018a. Sex-related di fferences in the asso-
ciation between grip strength and depression: results from The Irish LongitudinalStudy on Ageing. Exp. Gerontol. 104, 147 –152. https://doi.org/10.1016/j.exger.
2018.02.010 .
McDowell, C.P., Gordon, B.R., Andrews, K.L., et al., 2018b. Associations of physical ac-
tivity with anxiety symptoms and status: results from The Irish longitudinal study on
ageing. Epidemiol. Psychiatr. Sci. 1 –10.https://doi.org/10.1017/
S204579601800001X Epub ahead of print.
McDowell, C.P., Dishman, R.K., Vancampfort, D., et al., 2018c. Physical activity and
generalized anxiety disorder: results from The Irish longitudinal study on ageing
(TILDA).
Int. J. Epidemiol. 47 (5), 1443 –1453. https://doi.org/10.1093/ije/dyy141 .
McDowell, C.P., Gordon, B.R., MacDonncha, C., et al., 2019. Physical activity correlates
among older adults with probable-generalized anxiety disorder: results from the Irish
longitudinal study on ageing. Gen. Hosp. Psychiatry 59, 30 –36.https://doi.org/10.
1016/j.genhosppsych.2019.04.012 .
Mitte, K., 2005. Meta-analysis of cognitive-behavioral treatments for generalized anxiety
disorder: a comparison with pharmacotherapy. Psychol. Bull. 131 (5), 785. https://
doi.org/10.1037/0033-2909.131.5.785 .
Moylan, S., Jacka, F.N., Pasco, J.A., et al., 2012. Cigarette smoking, nicotine dependence
and anxiety disorders: a systematic review of population-based, epidemiologicalstudies. BMC Med. 10 (1), 123. https://doi.org/10.1186/1741-7015-10-123 .
Nolan, A., O'Regan, C., Dooley, C., et al., 2014. The Over 50s in a Changing Ireland:
Economic circumstances, Health and Well-Being. Trinity College Dublin (TILDA,Dublin .
Ortega, F.B., Silventoinen, K., Tynelius, P., et al., 2012. Muscular strength in male ado-
lescents and premature death: cohort study of one million participants. BMJ 345,
e7279 https://doi.org /10.1136/bmj.e7279 .
Peterson, M.D., Duchowny, K., Meng, Q., et al., 2017. Low normalized grip strength is a
biomarker for cardiometabolic disease and physical disabilities among US andChinese adults. J. Gerontol. A Biol. Sci. Med. Sci. 72 (11), 1525 –1531. https://doi.
org/10.1185/03007995.2013.770731 .
Peterson, M.D., Rhea, M.R., Sen, A., et al., 2010. Resistance exercise for muscular strength
in older adults: a meta-analysis. Ageing Res. Rev. 9 (3), 226 –237. https://doi.org/10.
1016/j.arr.2010.03.004 .
Roberts, H.C., Denison, H.J., Martin, H.J., et al., 2011. A review of the measurement of
grip strength in clinical and epidemiological studies: towards a standardisedB.R. Gordon, et al.
Journal of Affective Disorders 255 (2019) 136–141
140

approach. Age Ageing 40, 423 –429. https://doi.org/10.1093/ageing/afr051 .
Rose, D., Harrison, E., 2007. The European socio-economic classi fication: a new social
class schema for comparative European research. Eur. Soc. 9 (3), 459 –490. https://
doi.org/10.1080/14616690701336518 .
Roth, S., Ferrell, R., Hurley, B., 2000. Strength training for the prevention and treatment
of sarcopenia. J. Nutr. Health Aging 4 (3), 143 –155.
Sharpe, D., 2015. Your chi-square test is statistically signi ficant: now what. Pract Assess.
Res. Eval. 20 (8), 2 –10.http://pareonline.net/getvn.asp?v=20&n=8 .
Stanley, M.A., Diefenbach, G.J., Hopko, D.R., et al., 2011. Erratum to: the nature of
generalized anxiety in older primary care patients: preliminary findings. J.
Psychopathol. Behav. Assess. 33 (2), 298. http://dx.doi.org/10.1023/
A:1025903214019 .
Veronese, N., Stubbs, B., Trevisan, C., et al., 2017. Poor physical performance predicts
future onset of depression in elderly people: progetto veneto anziani longitudinal
study. Phys. Ther. 97, 659 –668. http://dx.doi.org/10.1093/ptj/pzx017 .
Veronese, N., Stubbs, B., Trevisan, C., et al., 2016. What physical performance measures
predict incident cognitive decline among intact older adults? A 4.4 year follow up
study. Exp. Gerontol. 80, 110 –118. http://dx.doi.org/10.1016/j.arr.2017.01.003 .
Von Elm, E., Altman, D.G., Egger, M., et al., 2008. The strengthening the reporting of
observational studies in epidemiology [STROBE] statement: guidelines for reportingobservational studies. Gaceta Sanitaria 22 (2), 144 –150. https://doi.org/10.1371/journal.pmed.0040296 .
Wittchen, H.U., Jacobi, F., Rehm, J., et al., 2011. The size and burden of mental disorders
and other disorders of the brain in Europe 2010. Eur. Neuropsychopharmacol. 21 (9),655–679. https://doi.org/10.1016/j.euroneuro.2011.07.018 .
Whelan, B.J., 1979. Ransam a random sample design for Ireland. Econ. Soc. Rev. 10 (2),
169. http://hdl.handle.net/2262/68820 .
Wolitzky ‐Taylor, K.B., Castriotta, N., Lenze, E.J., et al., 2010. Anxiety disorders in older
adults: a comprehensive review. Depress Anxiety 27 (2), 190 –211. https://doi.org/
10.1002/da.20653 .
World Health Organization, 2008. Waist circumference and waist-hip ratio: report of a
WHO expert consultation. pp. 8 –11 GenevaDecember .
World Health Organization (2017) Depression and other common mental disorders:
global health estimates. WHO/MSD/MER/2017.2.
Wuthrich, V.M., Johnco, C., Knight, A., 2014. Comparison of the penn state worry
questionnaire (PSWQ) and abbreviated version (PSWQ-A) in a clinical and non-clinical population of older adults. J. Anxiety Disord. 28 (7), 657 –663. https://doi.
org/10.1016/j.janxdis.2014.07.005 .
Yang, L., Koyanagi, A., Smith, L., et al., 2018. Hand grip strength and cognitive function
among elderly cancer survivors. PLoS One 13 (6), 1 –12.https://doi.org/10.1371/
journal.pone.0197909 .B.R. Gordon, et al.
Journal of Affective Disorders 255 (2019) 136–141
141