Manual Therapy 15 (2010) 100–104

Contents lists available at ScienceDirect

Manual Therapy
journal homepage: [Link]/math

Original Article

Effect of physical exercise interventions on musculoskeletal pain in all body

regions among office workers: A one-year randomized controlled trial
Lars L. Andersen a, *, Karl Bang Christensen b, Andreas Holtermann a, Otto M. Poulsen a, Gisela Sjøgaard c,
Mogens T. Pedersen d, Ernst A. Hansen e
a
National Research Centre for the Working Environment, Lersø Parkalle 105, DK 2100 Copenhagen Ø, Denmark
b
Faculty of Health Sciences, University of Copenhagen, Øster Farimagsgade 5, 1014 Copenhagen K, Denmark
c
Institute of Sports Science and Clinical Biomechanics, University of Southern Denmark, Campusvej 55, DK-5230 Odense M, Denmark
d
Institute of Exercise and Sport Sciences, University of Copenhagen, Nørre allé 51, 2200 Copenhagen N, Denmark
e
Norwegian School of Sport Sciences, P.O. Box 4014, Ullevål Stadion, NO-0806 Oslo, Norway

a r t i c l e i n f o a b s t r a c t

Article history: This study investigated effects of physical exercise on musculoskeletal pain symptoms in all regions of
Received 21 February 2009 the body, as well as on other musculoskeletal pain in association with neck pain. A single blind
Received in revised form randomized controlled trial testing a one-year exercise intervention was performed among 549 office
22 July 2009
workers; specific neck/shoulder resistance training, all-round physical exercise, or a reference inter-
Accepted 6 August 2009
vention. Pain symptoms were determined by questionnaire screening of twelve selected body regions.
Case individuals were identified for each body region as those reporting pain intensities at baseline of 3
Keywords:
or more (scale of 0–9) during the last three months. For neck cases specifically, the additional number of
Musculoskeletal diseases
Rehabilitation pain regions was counted. Intensity of pain decreased significantly more in the neck, low back, right
Occupational health elbow and right hand in cases of the two exercise groups compared with the reference group
Occupational diseases (P < 0.0001–0.05). The additional number of pain regions in neck cases decreased in the two exercise
groups only (P < 0.01–0.05). In individuals with no or minor pain at baseline, development of pain was
minor in all three groups. In conclusion, both specific resistance training and all-round physical exercise
for office workers caused better effects than a reference intervention in relieving musculoskeletal pain
symptoms in exposed regions of the upper body.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction Ylinen, 2007; Andersen et al., 2008b; Blangsted et al., 2008).

However, there is lacking knowledge about effects of physical
Musculoskeletal pain symptoms in the back, neck and extrem- exercise on musculoskeletal complaints in other regions of the
ities are common subjective health complaints (Ferrari and Russell, body, as well as the preventative effect of exercise on development
2003; Ihlebaek et al., 2007; Sjøgren et al., 2009), with high socio- of pain symptoms in individuals without complaints.
economic consequences in terms of health expenses and lost Although neck pain is the most prevalent musculoskeletal
working days (Henderson et al., 2005). In occupational groups with complaint in office workers (Blangsted et al., 2008), pain symptoms
monotonous and repetitive work tasks, e.g. computer users, neck in other body regions are reported as well (Juul-Kristensen et al.,
pain is the most prevalent musculoskeletal complaint (Juul- 2006). Thus, it is relevant to assess the effect of different exercise
Kristensen et al., 2006). While the aetiology of musculoskeletal interventions on pain in all regions of the body. Furthermore, people
pain symptoms is multifactorial with several physical and with neck pain are more likely to experience other musculoskeletal
psychosocial risk factors at work as well as during leisure (National pain in association with neck pain (Hagen et al., 2006; Juul-
Research Council, Institute of Medicine, 2001; Punnett and Weg- Kristensen et al., 2006) likely due to increased pain sensitivity
man, 2004; Andersen et al., 2007), there is general consensus about (Arendt-Nielsen and Graven-Nielsen, 2008). Therefore, it is also
the beneficial effect of physical exercise. Both specific muscle relevant to assess the effect of different exercise interventions on
training and all-round physical exercise have shown beneficial other musculoskeletal pain specifically in association with neck
effects on neck pain as well as low back pain (Hayden et al., 2005; pain. Previous randomized controlled trials on rehabilitation of
neck/shoulder pain have considered neck/shoulder pain symptoms
* Corresponding author. Tel.: þ45 39 16 53 19; fax: þ45 39 16 52 01. alone (Waling et al., 2000; Ylinen et al., 2003; Andersen et al.,
E-mail address: lla@[Link] (L.L. Andersen). 2008b). In the present group of participants we have recently

1356-689X/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/[Link].2009.08.004
 L.L. Andersen et al. / Manual Therapy 15 (2010) 100–104 101

reported a beneficial effect of specific resistance training and all- repetitions, combined with static neck exercises (neck flexion, neck
round physical exercise on neck/shoulder pain symptoms (Andersen extension, lateral flexion) for repetitions of 5 s duration at 70–80%
et al., 2008a; Blangsted et al., 2008). of MVC (Ylinen and Ruuska, 1994). The volume of training in SRT
The objective of the present study was to investigate effects of was set according to the recommendations by the ACSM for effi-
two different physical exercise interventions on (1) musculoskel- cient gains of muscle strength in untrained individuals (Kraemer
etal pain in all regions of the body, and (2) other musculoskeletal et al., 2002). The participants were encouraged to add weight when
pain in association with neck pain specifically. We hypothesized they were able to perform more than 15 repetitions per exercise.
that both specific training and all-round physical exercise are more The training sessions ended with a high-speed dynamic power
efficient than a reference intervention for reducing and preventing exercise, performed as 15 s all-out kayaking (Dansprint, Vanløse,
musculoskeletal pain symptoms in different regions of the body Denmark) or ergometer rowing (Concept2, Inc., Morrisville, USA).
among office workers. Training was performed three times a week and each session lasted
20 min. Two of the three weekly sessions were supervised by
2. Methods experienced instructors.
APE had a primary goal of inspiring the participant to inte-
2.1. Study design grate physical activity into their daily lifestyle in a motivating
way. The participants were encouraged to increase their level of
A single blind randomized controlled trial testing a one-year physical activity both during leisure time and at work, and
exercise intervention was performed. The procedure of participant received information about location and opening hours of local
recruitment has been described in detail previously (Andersen swimming baths, fitness clubs, etc. Experienced instructors
et al., 2008a; Blangsted et al., 2008). Briefly, participants were office introduced different forms of physical activities for all-round
workers recruited from 12 geographically different units of strength and aerobic fitness during 1–4 monthly visits. Walking
a national Danish public administration authority. Of the 2163 group sessions were organized, and some participants were
employees invited for the study, 1397 replied to the invitation, and supplied with step counters. In addition, an 8-min CD-based
841 were willing to participate. Exclusion criteria were hyperten- exercise program for aerobic fitness and general strength – but
sion, disc prolapse, severe spinal disorders, history of severe trauma not specifically for the neck and shoulder area – was offered to
or other factors including pregnancy. In total 616 participants the participants. The activities offered were varied throughout
participated in the baseline test (397 females of age 44.6 yrs, body the year according to an activity program developed by
mass 68.2 kg, height 1.68 m; and 219 males of age 45.7 yrs, body a professional company, Dansk Firmaidrætsforbund (DFIF) in
mass 83.1 kg, height 1.81 m) of which further 24 were excluded due Denmark. As part of the motivation, each participant filled in
to the above mentioned criteria and 43 withdrew, leaving a total of a ‘‘contract’’ stating the ways that more physical activity could be
549 participants for the randomization. The participants were included in the daily live, e.g. riding the bicycle for work instead
randomized at the cluster level into one of three intervention of driving the car, joining the local fitness centre, climbing the
groups: specific resistance training (SRT, n ¼ 180), all-round phys- stairs instead of using the elevator, etc.
ical exercise (APE, n ¼ 187), or a reference intervention without Reference group (REF) had a main purpose of ensuring that
physical activity (REF, n ¼ 182). Clusters of participants working in the participants received attention similar to the two other
on the same floor or in the same building participated in the same groups. Participants were encouraged to form groups which
type intervention to avoid contamination of the intervention and to should try to improve health and working conditions, e.g.
enhance compliance. A total of 79 clusters were identified and through improved workplace ergonomics, stress management,
cluster sizes ranged from 1 to 25 participants; only 9 clusters organization of work, and cafeteria food quality. The participants
contained a single participant. At one-year follow-up 440 partici- themselves were responsible for organizing presentations about
pants (w80%) replied to the questionnaire. health-promoting activities that they found interesting, e.g. diet,
The study protocol was approved by the local ethics committee stress management, weight loss, meditation, relaxation, and
(KF 01-201/04), and qualified for registration in the International indoor climate. Staff from our departments supported the group
Standard Randomised Controlled Trial Number Register on http:// by helping to organize the presentations. The participants in REF
[Link], and has been assigned a unique trial identification received the same amount of attention as the participants in SRT
number: ISRCTN31187106. All participants were informed about and APE, however no actual changes were implemented at the
the purpose and content of the project, and gave written informed worksites.
consent to participate.
We have previously reported the main results on this trial, 2.3. Musculoskeletal pain symptoms
showing approximately 30% reduction of neck/shoulder pain, 9–
10% increase of neck/shoulder muscle strength, and improvements The participants replied to an internet-based questionnaire with
in metabolic syndrome related risk factors with the SRT and APE regard to musculoskeletal pain symptoms in the neck, shoulders,
intervention (Andersen et al., 2008a; Blangsted et al., 2008; elbows, hands, upper back, low back, hips, knees and feet. The
Pedersen et al., 2009). regions were defined according to the Nordic Questionnaire
(Kuorinka et al., 1987). Both the right and left side were included for
2.2. Interventions the shoulders, elbows, and hands. Thus, a total of 12 body regions
were included. Intensity of pain during the last three months was
Intervention took place for a one-year period from February rated on a 10-point ordinal scale ranging from 0 to 9, where the
2005 until January 2006. All participants were allowed a total of 1 h following question was answered: ‘‘On average, how intense was
per week during working hours for the intervention activities, your pain in [body part] during the last three months on a 0–9
which have been described in detail previously (Andersen et al., scale?’’ (where 0 corresponded to ‘‘no complaints’’ and 9 corre-
2008a; Blangsted et al., 2008) and are also described below. sponded to ‘‘pain as bad as it could be’’) (Brauer et al., 2003).
SRT was performed for the neck and shoulder muscles. The core Subsequently, ‘control’ and ‘case’ individuals were defined for
of the program was dynamic strengthening exercises (front raise, each region of the body as those reporting pain intensities at
lateral raise, shoulder shrugs) performed for 2–3 sets of 10–15 baseline of 0–2 (controls, i.e. no or minor pain) and 3 or more
102 L.L. Andersen et al. / Manual Therapy 15 (2010) 100–104

(cases, i.e. more severe pain), respectively, on a scale of 0–9. 3.2. Post intervention
Thus, the frequency of cases and controls varies for each body
region according to the percentage of cases as shown in Table 1. In response to the intervention, there were main effects for
For neck pain cases specifically, the additional number of pain Region (F ¼ 3.04, P < 0.0005), Group (F ¼ 2.93, P ¼ 0.05), and Status
regions was counted (i.e. 0–11 regions). A pain region was defined (F ¼ 905, P < 0.0001). Subsequent ANOVAs for each body region
as a region with a reported pain intensity of 3 or more. For example, showed greater decrease in intensity of pain in the feet in APE
in a neck pain case individual reporting pain intensities of 3 or more compared with both SRT (P < 0.001) and REF (P < 0.05). For the
simultaneously in the low back, shoulder, elbow, and hand, remainder of body regions, no significant differences were
respectively, the additional number of pain regions were 4. observed between APE and SRT, and these were therefore collapsed
in the statistical analysis and compared with REF to yield higher
statistical power. Intensity of pain generally decreased over time
2.4. Statistics among cases in all three intervention groups. However, an overall
better pain relief was seen in SRT and APE compared with REF. Thus,
All data were analyzed according to the principle of intention- intensity of pain decreased more in the neck, low back, right elbow
to-treat. Baseline associations of pain intensity between different and right hand (P < 0.0001–0.05) (Fig. 1, left). In individuals with no
body regions were evaluated with non-parametric Spearman’s or minor pain symptoms of each specific body region (i.e.
correlation coefficient. Analysis of variance (ANOVA) was per- ‘controls’), APE compared with REF had a preventative effect on
formed in SAS version 9 using the mixed procedure. While the development of pain symptoms in the right shoulder (P < 0.05)
distribution of pain intensity was skewed with a tail towards higher (Fig. 1, right), whereas no significant difference between the two
values, the change in response to the intervention (i.e. delta values) exercise groups was found.
followed a normal distribution and was therefore used in the
analysis. Test for main effects for the pre- to post-training change in 3.3. Neck cases
pain intensity included Region (12 regions), Group (SRT, APE and
REF) and Status (Case or Control). When a significant main effect In neck pain cases compared with neck controls the additional
was found, e.g. for Region, subsequent ANOVAs with fewer factors number of pain regions (i.e. excluding the neck region) was
were used to locate differences. All values are reported as group 3.6  0.16 and 0.98  0.07 (P < 0.0001), respectively, at baseline.
mean  SE unless otherwise stated. Thus, the number of additional pain regions was significantly
higher for neck pain cases compared with individuals with no or
minor pain symptoms of the neck.
3. Results
In response to the intervention, the additional number of pain
regions in neck cases decreased in SRT (0.73  0.36, P < 0.05) and
3.1. Baseline
APE (0.91  0.31, P < 0.01), whereas no significant decrease from
baseline was observed in REF (0.40  0.32, n.s.).
At baseline, the most prevalent pain symptoms were in the neck
region, whereas the least prevalent symptoms were in the left
4. Discussion
hand, left elbow and hips (Table 1). Intensity of neck pain was
significantly related to pain in all other regions of the body
The main findings of this study are the beneficial effect of
(P < 0.0001), with correlation coefficients ranging from 0.17 to 0.21
exercise on musculoskeletal pain symptoms in several regions of
for the left elbow, left hand, hips, knees and feet, 0.26 to 0.27 for the
the upper body, as well as the decrease of additional number
right elbow and right hand, 0.36 for the low back, and 0.46 to 0.57
of pain regions in neck pain cases specifically. In contrast, the
for the left shoulder, right shoulder and upper back. The overall
overall preventative effect of exercise in individuals without pain at
prevalence of symptoms was generally higher in women compared
baseline was less convincing, due to minor development of pain in
with men (Table 1).
all three groups.
The baseline questionnaire survey confirmed the high preva-
lence of neck pain among office workers, and verified the positive
Table 1 relation between neck pain and pain in other regions of the body
Baseline prevalence of pain symptoms in different regions of the body in female (F) (Hagen et al., 2006; Juul-Kristensen et al., 2006). These cross-
and male (M) office workers, as well as for females and males together (F þ M, sectional baseline data illustrate the importance of reducing not
n ¼ 544).
only neck pain in office workers, but also pain in other regions of
% Cases (n ¼ 544) Pain intensity (0–9) the body.
F M FþM Cases Controls During the one-year intervention period, complaints decreased
Neck 53*** 29 44 4.66 (0.10) 0.95 (0.05) more in the neck, low back, right elbow and right hand in the two
Low Back 43** 30 39 4.48 (0.11) 0.91 (0.05) exercise groups compared with the reference group. These findings
R Shoulder 36*** 22 31 4.79 (0.13) 0.59 (0.04) were not significantly different between the two exercise groups.
Upper Back 33*** 17 27 4.29 (0.12) 0.57 (0.04) Whereas all-round physical exercise included overall body condi-
Knees 20 19 20 4.37 (0.14) 0.41 (0.03)
R Hand 22* 14 19 4.18 (0.15) 0.42 (0.03)
tioning, the specific resistance training program was designed to
L Shoulder 24*** 9 19 4.83 (0.16) 0.37 (0.03) target the neck/shoulder area specifically. However, a certain
Feet 18* 11 16 4.35 (0.16) 0.34 (0.03) conditioning of the elbow, hand, and low back region inevitable
R Elbow 16 13 15 4.56 (0.19) 0.30 (0.03) occurs in response to neck/shoulder training with dumbbells,
Hips 15*** 5 11 4.70 (0.20) 0.19 (0.02)
which may partially explain reduction of pain in these regions as
L Hand 10 6 9 4.51 (0.24) 0.20 (0.02)
L Elbow 10* 4 8 4.27 (0.25) 0.13 (0.02) well. This finding is of clinical relevance since these regions are
exposed to prolonged low-force muscle contractions throughout
*, **, *** Higher prevalence in F compared with M, P < 0.05, 0.01, 0.001, respectively.
On the right hand side of the table is mean (SE) pain intensity for cases and controls
the working day due to the monotony and repetitiveness of muscle
at baseline. ‘Cases’ and ‘Controls’ were defined for each body region as those activity during office work (Kilbom, 1994; Malchaire et al., 2001;
reporting pain intensities of 3 and 0–2, respectively, on a scale of 0–9. Sluiter et al., 2001).
 L.L. Andersen et al. / Manual Therapy 15 (2010) 100–104 103

REF
APE
SRT Cases Controls

R Hand
***
L Hand

R Elbow
*
L Elbow

&
R Shoulder

L Shoulder
REGION

Feet #

Knee

Hip

Low Back
*
Upper Back

Neck
**

-3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.0 0.5 1.0 1.5

Δ Pain Intensity (Scale 0-9)

Fig. 1. Changes in pain intensity for each separate body region in cases (left) and controls (right) in response to the one-year intervention. ‘Cases’ and ‘Controls’ were defined for each
body region as those reporting pain intensities at baseline of 3 and 0–2, respectively, on a scale of 0–9. (*, **, ***) Larger decrease from baseline in SRT and APE compared with REF,
P < 0.05, 0.001, 0.0001, respectively. (&) Larger increase from baseline in REF compared with APE, P < 0.05. (#) Larger decrease from baseline in APE compared with SRT and REF,
P < 0.001–0.05. SRT: Specific Resistance Training, APE: All-round Physical Exercise, REF: Reference intervention.

Another clinically relevant finding was the decrease of addi- Further, no significant differences between the three intervention
tional number of pain regions in neck cases in response to both groups were observed for the remaining eleven body regions.
types of exercise intervention. Several mechanisms can be specu- Altogether, the overall preventative effect of exercise compared
lated upon to cause this finding. Pain perception is known to be with the reference intervention on development of complaints in
altered in chronic pain conditions, leading to central pain sensiti- different regions of the body was very limited. This may be due
zation (Arendt-Nielsen and Graven-Nielsen, 2008). The present to the length of the intervention period during which develop-
findings show beneficial changes in overall pain perception in ment of complaints was minor in those without symptoms at
response to exercise, indicating decreased pain sensitization. baseline in all three groups. On a scale of 0–9, the overall
A significant difference between the two exercise groups was development of pain during the one-year period was less than
only observed for pain intensity in the feet, which may be explained 0.5 in most regions of the body. Likewise, previous preventative
by the fact that all-round physical exercise also included condi- efforts have shown minor effects (Linton and van Tulder, 2001;
tioning of the lower extremities. However, this could also be van Poppel et al., 2004). Thus, future studies are recommended
a statistical type I error. to employ long-term prospective follow-ups of several years to
The overall preventative effect of exercise on pain develop- investigate possible preventative effects of exercise on develop-
ment in office workers without pain at baseline was less ment of musculoskeletal complaints in office workers, or to
convincing. Whereas all-round physical exercise compared with investigate preventative effects in more heavily exposed occu-
the reference intervention had a preventative effect of develop- pational groups, e.g. blue-collar workers.
ment of pain symptoms in the right shoulder, no significant Based on the present results, both all-round physical exercise
differences between the two exercise groups were found. and specific strength training can reduce pain in several regions of
104 L.L. Andersen et al. / Manual Therapy 15 (2010) 100–104

the body. Thus, the preferred type of exercise may be chosen when Juul-Kristensen B, Kadefors R, Hansen K, Bystrom P, Sandsjo L, Sjogaard G. Clinical
signs and physical function in neck and upper extremities among elderly female
the goal is to reduce overall musculoskeletal pain.
computer users: the NEW study. Eur J Appl Physiol 2006;96:136–45.
In conclusion, both specific resistance training and all-round Kilbom Å. Repetitive work of the upper extremity: part I – guidelines for the
physical exercise for office workers caused better effects than practitioner. Int J Ind Erg 1994;14:51–7.
a reference intervention on musculoskeletal pain symptoms in Kraemer WJ, Adams K, Cafarelli E, Dudley GA, Dooly C, Feigenbaum MS, et al,
American College of Sports Medicine Position Stand. Progression models in
several regions of the upper body, as well as on the additional resistance training for healthy adults. Med Sci Sports Exerc 2002;34:
number of pain regions in individuals with neck pain specifically. In 364–80.
contrast, the overall preventative effect of exercise in individuals Kuorinka I, Jonsson B, Kilbom Å, Vinterberg H, Biering-Sørensen F, Andersson G,
et al. Standardised Nordic questionnaires for the analysis of musculoskeletal
without pain at baseline was less convincing, due to minor devel- symptoms. Appl Ergon 1987;18:233–7.
opment of pain in all three groups. Linton SJ, van Tulder MW. Preventive interventions for back and neck pain prob-
lems: what is the evidence? Spine 2001;26:778–87.
Malchaire JB, Roquelaure Y, Cock N, Piette A, Vergracht S, Chiron H. Musculoskeletal
References complaints, functional capacity, personality and psychosocial factors. Int Arch
Occup Environ Health 2001;74:549–57.
Andersen JH, Haahr JP, Frost P. Risk factors for more severe regional musculoskeletal National Research Council, Institute of Medicine. Musculoskeletal disorders and the
symptoms: a two-year prospective study of a general working population. workplase – low back and upper extremities. Washington D.C.: National
Arthritis Rheum 2007;56:1355–64. Research Council, Institute of Medicine; 2001. ISBN-10: 0-309-07284-0.
Andersen LL, Jorgensen MB, Blangsted AK, Pedersen MT, Hansen EA, Sjogaard G. A Pedersen MT, Blangsted AK, Andersen LL, Jorgensen MB, Hansen EA, Sjogaard G.
randomized controlled intervention trial to relieve and prevent neck/shoulder The effect of worksite physical activity intervention on physical capacity, health,
pain. Med Sci Sports Exerc 2008a;40:983–90. and productivity: a 1-year randomized controlled trial. J Occup Environ Med
Andersen LL, Kjaer M, Sogaard K, Hansen L, Kryger AI, Sjogaard G. Effect of two 2009;51:759–70.
contrasting types of physical exercise on chronic neck muscle pain. Arthritis Punnett L, Wegman DH. Work-related musculoskeletal disorders: the epidemio-
Rheum 2008b;59:84–91. logic evidence and the debate. J Electromyogr Kinesiol 2004;14:13–23.
Arendt-Nielsen L, Graven-Nielsen T. Muscle pain: sensory implications and inter- Sjøgren P, Ekholm O, Peuckmann V, Grønbaek M. Epidemiology of chronic pain in
action with motor control. Clin J Pain 2008;24:291–8. Denmark: an update. Eur J Pain 2009;13:287–92.
Blangsted AK, Sogaard K, Hansen EA, Hannerz H, Sjogaard G. One-year randomized Sluiter JK, Rest KM, Frings-Dresen MH. Criteria document for evaluating the work-
controlled trial with different physical-activity programs to reduce musculo- relatedness of upper-extremity musculoskeletal disorders. Scand J Work
skeletal symptoms in the neck and shoulders among office workers. Scand J Environ Health 2001;(27 Suppl. 1):1–102.
Work Environ Health 2008;34:55–65. van Poppel MN, Hooftman WE, Koes BW. An update of a systematic review of
Brauer C, Thomsen JF, Loft IP, Mikkelsen S. Can we rely on retrospective pain controlled clinical trials on the primary prevention of back pain at the work-
assessments? Am J Epidemiol 2003;157:552–7. place. Occup Med (Lond) 2004;54:345–52.
Ferrari R, Russell AS. Regional musculoskeletal conditions: neck pain. Best Pract Res Waling K, Sundelin G, Ahlgren C, Jarvholm B. Perceived pain before and after three
Clin Rheumatol 2003;17:57–70. exercise programs – a controlled clinical trial of women with work-related
Hagen EM, Svensen E, Eriksen HR, Ihlebaek CM, Ursin H. Comorbid subjective trapezius myalgia. Pain 2000;85:201–7.
health complaints in low back pain. Spine 2006;31:1491–5. Ylinen J. Physical exercises and functional rehabilitation for the management of
Hayden JA, van Tulder MW, Malmivaara A, Koes BW. Exercise therapy for treatment chronic neck pain. Eur Medicophys 2007;43:119–32.
of non-specific low back pain. Cochrane Database Syst Rev 2005. CD000335. Ylinen J, Ruuska J. Clinical use of neck isometric strength measurement in reha-
Henderson M, Glozier N, Holland EK. Long term sickness absence. BMJ bilitation. Arch Phys Med Rehabil 1994;75:465–9.
2005;330:802–3. Ylinen J, Takala EP, Nykanen M, Hakkinen A, Malkia E, Pohjolainen T, et al. Active
Ihlebaek C, Brage S, Eriksen HR. Health complaints and sickness absence in Norway. neck muscle training in the treatment of chronic neck pain in women:
Occup Med (Lond) 2007;57:43–9. a randomized controlled trial. JAMA 2003;289:2509–16.