Customize your JAMA Network experience by selecting one or more topics from the list below.
Ylinen J, Takala E, Nykänen M, et al. Active Neck Muscle Training in the Treatment of Chronic Neck Pain in WomenA Randomized Controlled Trial. JAMA. 2003;289(19):2509–2516. doi:10.1001/jama.289.19.2509
Context Active physical training is commonly recommended for patients with chronic
neck pain; however, its efficacy has not been demonstrated in randomized studies.
Objective To evaluate the efficacy of intensive isometric neck strength training
and lighter endurance training of neck muscles on pain and disability in women
with chronic, nonspecific neck pain.
Design Examiner-blinded randomized controlled trial conducted between February
2000 and March 2002.
Setting Participants were recruited from occupational health care systems in
southern and eastern Finland.
Patients A total of 180 female office workers between the ages of 25 and 53 years
with chronic, nonspecific neck pain.
Interventions Patients were randomly assigned to either 2 training groups or to a
control group, with 60 patients in each group. The endurance training group
performed dynamic neck exercises, which included lifting the head up from
the supine and prone positions. The strength training group performed high-intensity
isometric neck strengthening and stabilization exercises with an elastic band.
Both training groups performed dynamic exercises for the shoulders and upper
extremities with dumbbells. All groups were advised to do aerobic and stretching
exercises regularly 3 times a week.
Main Outcome Measures Neck pain and disability were assessed by a visual analog scale, the
neck and shoulder pain and disability index, and the Vernon neck disability
index. Intermediate outcome measures included mood assessed by a short depression
inventory and by maximal isometric neck strength and range of motion measures.
Results At the 12-month follow-up visit, both neck pain and disability had decreased
in both training groups compared with the control group (P<.001). Maximal isometric neck strength had improved flexion by
110%, rotation by 76%, and extension by 69% in the strength training group.
The respective improvements in the endurance training group were 28%, 29%,
and 16% and in the control group were 10%, 10%, and 7%. Range of motion had
also improved statistically significantly in both training groups compared
with the control group in rotation, but only the strength training group had
statistically significant improvements in lateral flexion and in flexion and
Conclusions Both strength and endurance training for 12 months were effective methods
for decreasing pain and disability in women with chronic, nonspecific neck
pain. Stretching and fitness training are commonly advised for patients with
chronic neck pain, but stretching and aerobic exercising alone proved to be
a much less effective form of training than strength training.
Neck disorders remain a common problem in modern, industrialized countries.
Neck pain has been the most common chief complaint among working-aged women
visiting their physicians.1,2 In
a Canadian study,3 54% of the general population
had experienced neck pain during the past 6 months, and approximately 5% were
highly disabled by neck pain. The prevalence of chronic neck pain has been
reported to be 7% in women and 5% in men in Finland.4 Patients
with chronic neck pain used health care services twice as much as the population
on average.2 Sick leave, therapy, and specialist
care form the major part of the costs incurred by neck pain, whereas investigations
at the primary health care level play a minor role.5
The origin and exact pathophysiologic mechanisms of chronic neck pain
often remain obscure because trauma or severe degenerative conditions at working
age are found only in a few cases. The origin of neck pain is thought to be
multifactorial. Excessive physical strain may cause microtrauma in connective
tissues, and psychosocial stress may lead to increased muscular tension.6 Degenerative changes in cervical vertebrae and disks
are common and increase with advanced age in asymptomatic people. Thus, examination
using radiographs or magnetic resonance imaging does not elucidate the origin
of pain in most cases.7-10
Evidence for many of the standard treatment approaches to neck pain
is lacking.11 Conservative management of neck
disorders includes both passive and active therapies, neither of which have
been shown to be effective.12 However, these
treatments are widely prescribed by physicians.13 The
aim of our study was to investigate the efficacy of intensive isometric neck
strength training and lighter endurance training of neck muscles in rehabilitation
of women with chronic, nonspecific neck pain.
A randomized controlled trial was conducted between February 2000 and
March 2002. The ethics committee of the Punkaharju Rehabilitation Center,
Punkaharju, Finland, approved the study design. In southern and eastern Finland,
occupational health care services that were known to provide health care for
women who work in offices were informed about the study and the inclusion
and exclusion criteria. On the basis of clinical examination, physicians working
in the occupational health care services referred suitable patients, who then
completed an application form in their local offices of the Social Insurance
Institution, which provides rehabilitation in Finland. The ordinary criteria
for state-financed rehabilitation were matched against the information given
on the form, and the accepted applications and referrals were sent to the
Punkaharju Rehabilitation Center. A questionnaire on their current health
and symptoms was mailed to the prospective participants to confirm their status
regarding the inclusion and exclusion criteria and to enable further selection
to be made (Figure 1).
All the participants gave written consent before entering the study.
The participants were randomized into 2 training groups and into a control
group, with 60 patients in each group. The block randomization into 3 groups
of 10 persons was performed blind before inviting the participants to the
rehabilitation center. As each group of 30 referrals meeting the inclusion
criteria were obtained, they were ranked by the neck and shoulder pain and
disability index14 and divided into 10 blocks
of 3 groups. From each block, 1 patient was randomized to 1 of the 2 training
groups or to the control group.15 This procedure
ensured that patients with equal severity of neck symptoms were present in
A total of 180 female office workers, recruited from various workplaces
through their respective occupational health care systems, were selected for
the study (Figure 1). The following
inclusion criteria were used: female sex, aged 25 to 53 years, office worker,
permanently employed, motivated to continue working, motivated for rehabilitation,
and constant or frequently occurring neck pain for more than 6 months. Exclusion
criteria were severe disorders of the cervical spine, such as disk prolapse,
spinal stenosis, postoperative conditions in the neck and shoulder areas,
history of severe trauma, instability, spasmodic torticollis, frequent migraine,
peripheral nerve entrapment, fibromyalgia, shoulder diseases (tendonitis,
bursitis, capsulitis), inflammatory rheumatic diseases, severe psychiatric
illness and other diseases that prevent physical loading, and pregnancy. These
states were assessed mainly by medical history and clinical examination before
entering the study.
Baseline variables included age, weight, height, duration of symptoms,
and smoking status. Outcome measurements were taken at the baseline and after
the 12-month intervention period in all 3 groups. Subjectively perceived neck
pain was assessed by a visual analog scale,16 and
the disability was assessed by the modified neck and shoulder pain and disability
index14 and Vernon neck disability index.17 On each scale, the theoretical range was 0 to 100.
At the 12-month follow-up visit, patients were asked to describe how the training
affected their neck pain on a 6-point scale (1 indicating much more pain and
6 indicating complete relief from pain).
Intermediate outcome measures were mood assessed by a short depression
inventory (theoretical range, 0-21)18 and by
maximal isometric neck strength19 and range
of motion (ROM).20 General physical function
was assessed with the grip-strength test21 and
a submaximal bicycle ergometer test by an exercise physiologist.22 All
the strength and ROM tests were performed blind by the same experienced physical
therapist before the intervention and after 12 months for all groups. The
training groups were also tested at 2 and 6 months. The chief researcher (J.Y.),
biostatistician (H.K.), and tester were not affiliated with the rehabilitation
Participants were asked about their use of analgesics before the study
and at the 12-month follow-up visit. They were also asked about visits made
to a physician and therapies performed to ameliorate neck pain during the
Training Groups. Each training group of 10
patients, alternating between strength and endurance training, started the
12-day institutional rehabilitation program at 1-month intervals during the
year. Both training regimens consisted of 5 sessions per week, each lasting
approximately 45 minutes. Thus, patients in each group participated in 9 practice
sessions to enable them to learn the program properly. Every other session
was performed at only half intensity to avoid excessive loading.
The endurance training group exercised neck flexor muscles by lifting
the head up from the supine position in 3 series of 20 repetitions. The strength
training group used an elastic rubber band (Theraband, Hygiene Corp, Akron,
Ohio) to train the neck flexor muscles in each session, performed in a sitting
position a single series of 15 repetitions directly forward, obliquely toward
right and left, and directly backward.23 The
aim was to maintain the level of resistance at 80% of the participant's maximum
isometric strength recorded at the baseline and at follow-up visits. The load
was checked with a handheld isometric strength testing device (Force-Five,
Wagner Instruments, Greenwich, Conn) during the training sessions at the baseline
and follow-up visits.
After specific neck training, both groups performed dynamic exercises
for the shoulders and upper extremities by doing dumbell shrugs, presses,
curls, bent-over rows, flyes, and pullovers. The endurance training group
performed 3 sets of 20 repetitions for each exercise with a pair of dumbells
each weighing 2 kg. The strength training group exercised with an individually
adjusted single dumbbell. The group performed only 1 set for each exercise
with the highest load possible to perform 15 repetitions.
Members in both training groups thereafter performed exercises in the
same way for the trunk and leg muscles against their individual body weights
by doing a single series of squats, sit-ups, and back extension exercises.
Each training session concluded with stretching exercises for the neck, shoulder,
and upper limb muscles for 20 minutes. Training instruction was given by the
same specially trained physical therapist who had several years of experience.
The training groups were also advised to perform aerobic exercise 3 times
a week for a half hour. If they lacked the time to perform all of the exercises,
they were encouraged to perform muscle exercises, at a minimum, including
specific neck muscle exercises.
Because many of our patients say they are not willing to visit health
clubs on a regular basis primarily because of costs, time spent traveling,
and time away from family, exercises for both training groups were planned
so that they could be performed at home. The participants received written
information about the exercises to be practiced at home and were taught to
keep a weekly exercise diary throughout the training year. They were encouraged
to exercise regularly 3 times a week at home. Exercise intensity and technique
were checked at follow-up visits at 2 and 6 months and at the end of the year.
Both training groups also underwent a common multimodal rehabilitation
program, including aspects commonly associated with traditional treatment:
relaxation training, aerobic training, behavioral support to reduce fear of
pain and improve exercise motivation, and lectures and practical exercises
in ergonomics. During the rehabilitation course, each patient received 4 sessions
of physical therapy, which consisted mainly of massage and mobilization to
alleviate neck pain and to enable those with severe neck pain to perform active
Control Group. Each control group of 10 participants
was asked to come in for baseline measurements of strength and ROM at 2 monthly
intervals between training groups during the same year, according to the recruiting
and block randomizing procedure. They spent 3 days at the rehabilitation center
and performed recreational activities in addition to the tests. The participants
were advised to perform aerobic exercise 3 times a week for a half hour. They
received written information about the same stretching exercises that were
performed by the training groups, which they were to practice at home for
approximately 20 minutes regularly 3 times a week. Each group of participants
was trained once in the proper way to perform these exercises. The participants
were not encouraged to perform any exercises to improve muscle strength, and
they received no treatments for this. After the 1-year follow-up measurements,
they were given the opportunity to participate in the same rehabilitation
training course as the active intervention participants, including training
and follow-up, which was also financed by the Social Insurance Institution.
The initial sample size for each treatment group was estimated on the
assumption that the endurance training group and control group would show
a 15% improvement in pain measured with the visual analog scale and that the
active strength training group would have to produce a 50% improvement to
be better than either the endurance training group or the control group. To
test the null hypothesis of equality of treatment at α = .05 with 95%
power and assuming a uniform dropout rate of 5%, it was calculated that 60
patients in each group would be sufficient.
Clinical outcome variables were analyzed by intention to treat. The
results are expressed by means and SDs, medians, and interquartile ranges.
Statistical comparison among the groups was made using the t test, Mann-Whitney U test, analysis of variance
with Tukey honestly significant difference test, and Kruskal-Wallis test.
Hommel adjustments were used to correct significance levels for multiple and
post hoc testing. Hodges-Lehmann estimate of median difference with 95% confidence
intervals was used to show changes in clinical outcome variables. The normality
of variables was evaluated by the Shapiro-Wilk statistics. The α level
was set at .05 for all tests. Analyses were performed using Stata Statistical
Software version 8.0 (Stata Corp, College Station, Tex).
The dropout rate during the year was 1.7%. One patient in the endurance
training group was diagnosed as having polymyalgia rheumatica after randomization
and was thus excluded from the study. In addition, there was 1 withdrawal
from the endurance training group and 1 from the control group (Figure 1). These 2 patients were included in the intention-to-treat
The distributions of age, anthropometric measures, short depression
inventory, grip strength, and oxygen uptake were similar in the 3 groups at
the baseline (Table 1). The pain
and disability indices in the 3 groups were also at the same level (Table 2). All these outcome measures were
significantly lower in the 2 training groups compared with the control group
at the 12-month follow-up visit. There was no statistically discernible difference
between the endurance and strength training groups. Considerable or complete
relief from pain was obtained by 73% of participants in the strength training
group, 59% in the endurance training group, and 21% in the control group.
Only 3% in all groups felt that their pain had become worse due to the training.
Neck strength showed a significant increase in the directions tested
in all groups at the 12-month follow-up visit compared with baseline (Figure 2). In the strength training group,
maximal isometric neck strength increased in flexion by 110%, in rotation
by 76%, and in extension by 69%. The results in the endurance training group
improved by 28%, 29%, and 16% and in the control group by 10%, 10%, and 7%,
respectively. The strength increases were significantly different among all
3 groups (P<.001). Transient neck pain was sometimes
reported throughout the few days succeeding the isometric maximal neck strength
tests, but all patients were able to continue with the training without interruption.
No major complications occurred.
All groups achieved a statistically significant increase in ROM in the
sagittal plane (flexion-extension) and frontal plane (lateral flexion) at
the 12-month follow-up visit compared with baseline (Table 3). However, only the training groups achieved a statistically
significant increase in rotation. The increase in ROM in the strength training
group was statistically significant in all directions when compared with the
control group. Between the endurance and control groups, a statistically significant
discernible difference emerged only in the measurements of rotation. No statistically
significant changes in mean grip strength were observed in any groups at the
12-month follow-up visit compared with baseline. There were also no significant
changes in maximal oxygen uptake.
According to their exercise diaries, the patients continued the training
throughout the year; the mean (SD) training frequency was 2.0 (0.8) times
a week in the endurance training group and 1.7 (0.6) times a week in the strength
training group. Stretching exercises were performed 2.0 (0.8) times a week
by both groups.
The number of patients taking analgesics at the 12-month follow-up visit
had decreased in all groups although this decline was more evident in the
training groups than in the control group (Table 4). Visits to a physician and use of therapies due to neck
pain also decreased during the follow-up year compared with the previous year
in all groups. The change was more marked in the training groups. The therapies
most commonly used were massage and stretching (65%). Hot and ice packs, electrotherapy,
acupuncture, traction, and zone therapy each accounted for between 5% and
Our study showed that participation in 1-year endurance and strength
training programs led to a considerable reduction in average neck pain and
disability compared with the control group. Neck function, including neck
strength and ROM, was improved significantly in both training groups compared
with the control group.
Previous randomized studies24-31 have
not shown active training to be effective or the results have been short-lived
and have disappeared within a few months. Similarly, multidisciplinary rehabilitation
has not been shown to be better at reducing pain and disability than the more
commonly used methods of care, despite the fact that the former requires substantial
staffing and financial resources.32 This raises
the question of why previous intervention studies have failed to produce any
significant long-term change. There are several possible reasons for this.
In many studies the inclusion and, especially, exclusion criteria have
not been well defined, which may be critical in studies with smaller numbers
of participants; several other conditions may act as confounding factors.
Epidemiologic studies have shown that women experience chronic neck pain more
often than men, and thus study populations have commonly consisted of women.
It should be recognized that a considerable number of working-aged women experience
fibromyalgia, the prevalence of which was reported to be 10.5% in Norway.33 The prevalence may be even higher among women with
chronic neck pain. Fibromyalgia has not been an exclusion criteria in previous
studies although it is a major confounding factor in treating painful conditions
in women. In addition, patients with neck and shoulder pain are often combined
into a single group. However, ailments in the shoulder joint are distinctly
different and should be differentiated from neck pain. There are also several
other conditions that cause neck pain, which should be treated by means other
than special neck exercises. In our study, we tried to exclude these diseases
as much as possible.
Chronic neck pain symptoms are known to exhibit seasonal variation,
worsening in the autumn and decreasing in the spring.34 Our
study was performed so that both training groups started at regular intervals
throughout the year, and control groups were tested between these intervals
to avoid effects due to seasonal variation. In addition, the baseline and
final follow-up measurements were performed at the same time of the year to
exclude seasonal effect.
Type of exercise, frequency, and intensity of training are key factors
in the effectiveness of training. Intensive muscle training at a fitness center,
which excludes specific neck exercises, does not increase neck muscle strength.35 The conventional stretching and aerobic exercises
the control group were advised to perform had a poor effect on the functional
parameters of the neck, and they had only a weak effect on chronic neck pain
and disability. However, one fifth of the patients in the control group reported
feeling better, which may be the result of endurance training, stretching
exercises, periodic variation of symptoms, or spontaneous recovery, which
has been observed to occur in patients with chronic neck pain.36
The results of several previous randomized studies24-31 do
not fully elucidate the effect of the exercise used in inducing change in
neck function because the descriptions given are poor and functional measures
have not been included or the improvement in function has been minor. However,
it is well established that patients with chronic, nonspecific neck pain can
tolerate intensive neck strength training and that training performed 2 times
a week produces significant changes in functional measures in these participants.23,37 The goal of the training programs
in our study was to improve neck function by the use of a regular exercise
program, which could be performed at home. In our study, all the neck strength
and ROM measures improved considerably in the strength training group. A significant
improvement in neck strength also occurred in the endurance training group.
This is not surprising, since the initial strength values of several patients
were so low that lifting their heads when performing dynamic exercises in
flexion caused them a great deal of effort. Thus, they were in fact performing
strength training. Several women had difficulty lifting their heads from a
supine position while doing neck flexion exercises or sit-ups. Thus, the weight
of their heads initally provided enough resistance for neck strength training.
The small strength increase in the control group was probably due to biological
variation and learning effect due to repeated testing.19
These results cannot necessarily be applied to men; women attain approximately
50% to 80% of the neck strength that men attain.23,38-40 It
is not certain that the endurance training, in which intensity was lighter
and remained unchanged during the follow-up year, would produce a significant
change in men even though it was effective among women.
It has been well established by previous studies41 that
the results achieved by strength training soon disappear if training is performed
for a few months only. In our study, the patients were encouraged to adhere
to their training program, which was designed to continue throughout the year.
The importance of adherence was emphasized to the training groups initially
and at the intermediate follow-up visits in the rehabilitation center.
In our study, the exercise diaries were examined at the intervening
follow-up visits, and the supervision may be an important factor in maintaining
a high level of compliance. Exercise diaries have not commonly been used in
previous randomized studies. Only Taimela et al30 have
monitored the training frequency of the intensive group after supervised training
and compared this to the frequency of the home training group, which received
instructions at a few meetings only. Surprisingly, in this study exercising
was far more frequent in the home training group.
Pain often accompanies the start of intensive exercise. Difficulty coping
with this pain may cause a patient to stop exercising. The institutional rehabilitation
program was planned to enable patients to overcome fear, which might prevent
them from performing stressful exercises. The etiology of nonspecific neck
pain and the effects of exercising were discussed in detail with the rehabilitation
groups. The same physical therapist met with the group every day, giving instructions
and leading the group in exercises. The patients received physical therapy
during the institutional rehabilitation period to better enable them to cope
with the initial stage of intensive training. However, the main aim of the
rehabilitation unit staff was to enhance the idea of self-care and increase
adherence to training.
In conclusion, both isometric strength training and dynamic endurance
training effectively decreased pain and disability in women with chronic neck
pain during the 1-year follow-up period. An important practical observation
was that long-term benefits could be obtained by training as infrequently
as twice a week. Aerobic and stretching exercises proved to be much less effective
than controlled endurance and strength training of the neck muscles.