The annual incidence and course of neck pain in the general population: a population-based cohort study.
ABSTRACT Although neck pain is a common source of disability, little is known about its incidence and course. We conducted a population-based cohort study of 1100 randomly selected Saskatchewan adults to determine the annual incidence of neck pain and describe its course. Subjects were initially surveyed by mail in September 1995 and followed-up 6 and 12 months later. The age and gender standardized annual incidence of neck pain is 14.6% (95% confidence interval: 11.3, 17.9). Each year, 0.6% (95% confidence interval: 0.0-1.1) of the population develops disabling neck pain. The annual rate of resolution of neck pain is 36.6% (95% confidence interval: 32.7, 40.5) and another 32.7% (95% confidence interval: 25.5, 39.9) report improvement. Among subjects with prevalent neck pain at baseline, 37.3% (95% confidence interval: 33.4, 41.2) report persistent problems and 9.9% (95% confidence interval: 7.4, 12.5) experience an aggravation during follow-up. Finally, 22.8% (95% confidence interval: 16.4, 29.3) of those with prevalent neck pain at baseline report a recurrent episode. Women are more likely than men to develop neck pain (incidence rate ratio=1.67, 95% confidence interval 1.08-2.60); more likely to suffer from persistent neck problems (incidence rate ratio=1.19, 95% confidence interval 1.03-1.38) and less likely to experience resolution (incidence rate ratio=0.75, 95% confidence interval 0.63-0.88). Neck pain is a disabling condition with a course marked by periods of remission and exacerbation. Contrary to prior belief, most individuals with neck pain do not experience complete resolution of their symptoms and disability.
- SourceAvailable from: PubMed Central[show abstract] [hide abstract]
ABSTRACT: Since oxidative stress plays a pathogenetic role in chronic neck pain (CNP), we investigated whether a combination of α-lipoic acid (ALA) and superoxide dismutase (SOD) might improve pain control and the efficacy of physiotherapy ("multimodal therapy") in patients with CNP. This study was conducted in the Rehabilitation Unit of the Department of Surgical and Oncological Sciences at the University Policlinic in Palermo, Italy. This was a prospective, randomized, open study in outpatients. Patients randomly received either physiotherapy alone (group 2; n = 45) or a combination of ALA 600 mg and SOD 140 IU daily in addition to physiotherapy (group 1; n = 51), for 60 days. Pain was assessed by a visual analogue scale (VAS) and a modified Neck Pain Questionnaire (mNPQ). Treatment compliance and safety were also evaluated. Both groups experienced a significant reduction in the VAS and mNPQ scores after 1 month; however, while no further improvement was observed in group 2 at 60 days, group 1 showed a further VAS reduction (p < 0.001). In addition, in the mNPQ at 60 days, more patients in group 1 than in group 2 reported that their neck pain was improved (p < 0.01), and they showed greater compliance with prescribed physiotherapy (p = 0.048). No drug reaction was observed. Use of ALA/SOD in combination with physiotherapy may be a useful approach to CNP, being antioxidants that act on nerve inflammation and disease progression. These preliminary observations suggest that some interesting goals (better pain control and physical wellbeing) can be achieved by multimodal therapy in CNP patients.Drugs in R&D. 01/2014;
- [show abstract] [hide abstract]
ABSTRACT: Our purpose was to compare the effectiveness of three manual therapy techniques: high velocity, low amplitude (HVLA), mobilization (Mob) and sustained natural apophyseal glide (SNAG) in patients with chronic neck pain (CNP). The randomized controlled trial included patients with mechanically reproducible CNP, who were randomized to the treatment group. Outcome measures were the Visual Analogue scale (VAS), Neck Disability Index (NDI), Global Rating of Change (GROC) and Cervical Range of Motion (CROM). Two-way repeated measures analysis of variance compared outcomes at baseline, at the end of treatment and 1, 2 and 3 months after treatment. A total of 51 subjects completed the trial. No signiﬁcant differences were found between HVLA, Mob and SNAG at the end of treatment and during the follow-up in any of the analysed outcomes. There were no differences in satisfaction for all techniques. The results lead to the conclusion that there is no long-term difference between the application of HVLA, Mob and SNAG in pain, disability and cervical range of motion for patients with CNP.Manual therapy 01/2014; Available online. · 2.32 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: INTRODUCTION:Cervical functional capacity outcome measures that are simple and reliable are urgently needed in order permit accurate assessment/reassessment during treatments and rehabilitation. Induced neck muscle fatigue has been shown to alter functional capacities such as balance and kinaesthetic sense in the standing posture. The Rod and Frame Test has also shown promise as a method of assessing the effects of chronic neck pain and injury, but currently only in the sitting position. The objectives of this project were therefore 1) to validate the computerised rod and frame test in the standing posture, and 2) to measure the effects that different cervical muscle fatigue protocol would have on the assessment of the subjective visual vertical and horizontal. METHOD:The validation of the standing computerised rod and frame test in the standing posture was obtained by comparing results (n = 74) between the sitting and standing positions with the Spearman's correlation coefficient. In addition, agreement between the two methods was analysed with the Bland-Altman method.Participants (n = 56) resisted with their neck muscles approximately 35% maximum isometric voluntary contraction force for 15 minutes on a purpose built apparatus in eight different directions. Wilcoxon signed rank tests analysed changes in horizontal and vertical rod and frame test between the neutral and all different directions of contraction. The changes of recorded unsigned vertical and horizontal errors for the combined frame condition in all situations of isometric contraction were analysed with two respective one-way repeated measures analysis of variance (ANOVA). DISCUSSION:The Spearman's rho and Bland-Altman plots show that the Rod and Frame Test works equally well in sitting and standing positions.After muscle contraction, there were significant increases in error in all participants for both horizontal and vertical rod and frame tests, except after flexion. These errors were predominantly present after fatigue of muscles in the coronal plane of contraction. Proprioception alone cannot explain the difference in the rod and frame results between different muscle groups. It is suggested that an evolutionary advantage of developing improved subjective verticality awareness in the same direction as the main visual field could explain these findings.SpringerPlus 02/2014; 3(1):78.
The annual incidence and course of neck pain in the general
population: a population-based cohort study
Pierre Co ˆte ´a,b,*, J. David Cassidyb,c, Linda J. Carrolld, Vicki Kristmana,b
aInstitute for Work and Health, 481 University Avenue, Suite 800, Toronto, Ont., Canada M5G 2E9
bDepartment of Public Health Sciences, University of Toronto, Toronto, Ont., Canada
cDivision of Outcomes & Population Health, Toronto Western Hospital Research Institute, University Health Network, Toronto, Ont., Canada
dDepartment of Public Health Sciences and the Alberta Centre for Injury Control and Research, University of Alberta, Edmonton, Alta., Canada
Received 12 April 2004; received in revised form 24 August 2004; accepted 1 September 2004
Although neck pain is a common source of disability, little is known about its incidence and course. We conducted a population-based
cohort study of 1100 randomly selected Saskatchewan adults to determine the annual incidence of neck pain and describe its course. Subjects
were initially surveyed by mail in September 1995 and followed-up 6 and 12 months later. The age and gender standardized annual incidence
of neck pain is 14.6% (95% confidence interval: 11.3, 17.9). Each year, 0.6% (95% confidence interval: 0.0–1.1) of the population develops
disabling neck pain. The annual rate of resolution of neck pain is 36.6% (95% confidence interval: 32.7, 40.5) and another 32.7% (95%
confidence interval: 25.5, 39.9) report improvement. Among subjects with prevalent neck pain at baseline, 37.3% (95% confidence interval:
33.4, 41.2) report persistent problems and 9.9% (95% confidence interval: 7.4, 12.5) experience an aggravation during follow-up. Finally,
22.8% (95% confidence interval: 16.4, 29.3) of those with prevalent neck pain at baseline report a recurrent episode. Women are more likely
than men to develop neck pain (incidence rate ratioZ1.67, 95% confidence interval 1.08–2.60); more likely to suffer from persistent neck
problems (incidence rateratioZ1.19, 95% confidence interval 1.03–1.38) and less likely to experience resolution (incidence rate ratioZ0.75,
95% confidence interval 0.63–0.88). Neck pain is a disabling condition with a course marked by periods of remission and exacerbation.
Contrary to prior belief, most individuals with neck pain do not experience complete resolution of their symptoms and disability.
q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
Keywords: Neck pain; Epidemiology; Incidence; Cohort study; Prognosis; Disability; Population-based
Neck pain is a common health problem that is
associated with significant disability in the general
population (Co ˆte ´ et al., 1998; Picavet and Schouten,
2003; van der Windt et al., 2002). During any 6-month
period, 54% of adults suffer from neck pain and 4.6%
experience important activity limitations because of neck
problems (Co ˆte ´ et al., 1998). While numerous cross-
sectional studies have described the prevalence of neck
pain, very few cohort studies have measured its incidence
and course. Consequently, very little is known about
the natural history of neck pain. To our knowledge, the
only population-based incidence cohort study of neck
pain suggests that 17.9% of adults develop an episode of
neck pain every year (Croft et al., 2001). While the
cumulative incidence of neck pain may be high, the
incidence of disabling neck pain remains unknown.
Clinicians generally accept that the course of neck pain is
favourable (Binder, 2003). According to this view, most
affected individuals recover and few develop chronic neck
pain and disability. However, a recent cross-sectional study,
(Picavet and Schouten, 2003) found that only 6.3% of
individuals who suffered from neck pain in the previous
year reported that their pain was non-recurrent. This
suggests that neck pain may be similar to low back pain
and follow an episodic course marked by periods of
0304-3959/$20.00 q 2004 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
Pain 112 (2004) 267–273
* Corresponding author. Tel.: C1 416 927 2027.
E-mail address: email@example.com (P. Co ˆte ´) (www.iwh.on.ca).
remissions and exacerbations (Hestbaek et al., 2003a,b;
Picavet and Schouten, 2003).
Cross-sectional studies consistently report that the
prevalence of neck pain increases with age and that it is
higher in women (Bovim et al., 1994; Co ˆte ´ et al., 1998;
Ma ¨kela ¨ et al., 1991; Picavet and Schouten, 2003). These
findings were supported by Croft et al. (2001) who found
that the incidence of neck pain increased slightly with age
and peaked between the ages of 30 and 45 years. Similarly,
they found that the incidence of neck pain was slightly
higher in women than men. However, much less is known
about the age and gender-specific course of neck pain. The
higher prevalence of chronic neck pain in older individuals
and in women suggests that the prognosis of neck pain
varies with age and gender (Guez et al., 2002; Ma ¨kela ¨ et al.,
In this paper, we report the annual age and gender
standardized incidence of neck pain in a population-based
cohort study of Saskatchewan adults. In addition, we present
the age, gender and severity-specific incidence of neck pain
and describe the course of neck pain by classifying subjects
into one offive paths: resolution, improvement, persistence,
aggravation and recurrence. Finally, we describe the age and
gender differences in the annual incidence and course of
2.1. Source population and design
We conducted the Saskatchewan Health and Back Pain
Survey, a population-based cohort study of Saskatchewan
adults (Carroll et al., 2000; Cassidy et al., 1998; Co ˆte ´ et al.,
1998). The purpose of the Saskatchewan Health and Back
Pain Survey was to study the epidemiology of neck pain,
back pain and depression in the general population.
Saskatchewan is a Canadian province of approximately
one million inhabitants that provides universal health care
coverage. We formed a weighted, age-stratified random
sample of residents using the Saskatchewan Health
Insurance Registration File, which includes over 99% of
the Saskatchewan population. Eligible for the study were
Saskatchewan residents between the ages of 20 and 69 who
held a valid Health Services card on August 31, 1995
(Saskatchewan Health, 1995). Excluded were inmates of
correctional facilities, residents under the Office of the
Public Trustee, foreign students and workers holding
employment or immigration visas, and residents of special
care homes. Details on the survey methodology and sample
size calculations are reported elsewhere (Carroll et al.,
2000; Cassidy et al., 1998; Co ˆte ´ et al., 1998).
Participation in the survey was voluntary. Saskatchewan
Health randomly selected the subjects and mailed all
surveys to protect the confidentiality of the participants.
Potential subjects were contacted by mail in September
1995 and again 6 and 12 months later. Returning the
completed baseline questionnaire signified implied consent
for participation. The 6-month follow-up questionnaire was
sent to respondents of the baseline questionnaire, and the
12-month follow-up to respondents of the 6-month follow-
up. There were three mailing waves for each survey period.
The first wave included an introduction letter and the survey
questionnaire. Non-respondents received a reminder card
and later a duplicate questionnaire. The University of
Saskatchewan Advisory Committee on Ethics in Human
Experimentation approved this study.
2.2. Study sample
Of the eligible 593,464 individuals, 2184 were randomly
selected. One hundred and nineteen baseline questionnaires
were returned due to mailing errors, five because of ‘health
reasons’, four individuals had emigrated and one had died.
Of the 2055 remaining subjects, 1133 returned baseline
questionnaires. Two questionnaires were completed by
subjects outside of the pre-determined age range and 30
subjects did not complete the neck pain questionnaire.
Therefore, the study sample for this analysis includes 1100
subjects. A comparison of the target population and the
randomly selected sample revealed no important differences
in age, gender and geographic location of residence (Co ˆte ´
et al., 1998). A comparison of participants and non-
participants suggested that older individuals, women,
those married, individuals residing on Indian Reserves and
those with intense non-disabling neck pain were slightly
more likelyto participate (Carroll et al., 2000; Cassidy et al.,
1998; Co ˆte ´ et al., 1998).
2.3. Measurement of neck pain
We used the Chronic Pain Questionnaire to measure the
presence of neck pain in the past 6 months. We made the
Chronic Pain Questionnaire specific to neck pain, which was
located by a mannequin drawing as the area from the
occiput and the third thoracic vertebra (Cassidy et al., 1998;
Co ˆte ´ et al., 1998). The Chronic Pain Questionnaire is a valid
and reliable 7-item index that is used to classify pain into
five grades of severity (Von Korff et al., 1988, 1990, 1992).
The grades are computed by combining scores from three
questions on neck pain intensity (current, worst, and
average), one question regarding the number of disability
days in the past 6 months (days kept from usual activities
such as work, school or housework), and three questions on
how neck pain has interfered with activities (daily activities;
recreational, social or family activities; and ability to work,
including housework). Grade 0 refers to no neck pain; Grade
I represents pain of low intensity and few activity
limitations; Grade II is pain of high intensity, but few
activity limitations; Grade III is pain associated with high
levels of disability and moderate limitations in activities;
and Grade IV refers to pain with high levels of disability
P. Co ˆte ´ et al. / Pain 112 (2004) 267–273268
and several activity limitations. Because of the low number
of subjects with Grades III and IV neck pain, we combined
them to describe disabling pain.
We defined two cohorts from our study sample. The first
cohort included subjects who were free from neck pain at
baseline and the second one included subjects with
prevalent neck pain at baseline. The two cohorts were
formed to measure the incidence and describe the course of
neck pain, respectively.
2.4.1. Incidence cohort
The ‘incidence’ cohort included 513 subjects who were
free of neck pain at baseline. These subjects had Grade 0
neck pain indicating that they had not experienced any neck
pain or disability in the previous 6 months. This group
comprised the sample at risk of developing an episode of
neck pain during the follow-up. Subject in this cohort
developed the outcome if they reported that they had
suffered from Grade I–IV neck pain in the past 6 months at
either the 6- or 12-month follow-up. Subjects were followed
until they developed neck pain, until they were lost to
follow-up or to the end of the study.
2.4.2. Course cohort
The ‘course’ cohort included 587 subjects with prevalent
neck pain at baseline. These subjects experienced neck pain
that varied from Grades I–IV in the 6-month period before
the survey. Overall, 72.4% experienced Grade I neck pain,
18.6% had Grade II neck pain and 9.0% had Grade III–IV
The course cohort included subjects who were at risk of
developing one of four outcomes with regard to their pain
and disability: (1) resolution, (2), improvement, (3)
persistence and (4) aggravation (Table 1). Subjects had
resolution of their neck pain if their baseline condition
improved to Grade 0 at either the 6- or 12-month follow-up.
Subjects were considered improved if the severity of their
pain decreased at the 6- or 12-month follow-up. We
classified subjects as suffering from persistent neck pain if
their condition remained at the same grade of severity in at
least two consecutive questionnaires. Finally, subjects
whose baseline neck pain increased in severity during
follow-up were classified as suffering from an aggravation.
The sample at risk varied with the outcomes (Table 1).
While it was 587 for resolution and persistence, it was 534
and 162, respectively, for aggravation and improvement.
This is attributable to the floor and ceiling restraints
imposed by our classification of pain severity. By definition,
improvement is restricted to those who transit from Grades
III–IV to Grades II or I and from Grade II to Grade I neck
pain. Subjects who transit to Grade 0 are considered to have
resolution of their neck pain and were not included in the
improvement computation. Similarly, the computation of
the incidence of aggravation excludes subjects with Grade
III–IV baseline neck pain because they have reached the
most severe grade of neck pain. For the purpose of this
analysis, subjects were followed until they developed one of
the four outcomes of interest, until they were lost to follow-
up or to the end of the study.
2.4.3. Recurrence sub-cohort
Subjects from the course cohort were classified as having
recurrent neck pain if their baseline condition improved to
Grade 0 at 6 months but returned to Grade I–IV at the
12-month follow-up. The sample at risk of recurrent neck
pain included 162 subjects whose baseline neck pain had
resolved at the 6-month follow-up and completed the
2.5. Statistical analysis
2.5.1. Incidence and course of neck pain
We computed the 1-year cumulative, age-group-specific,
gender-specific and grade-specific incidence of new epi-
sodes of neck pain. Similar incidence measures were
computed for each of the course outcomes. We directly
standardized the cumulative incidence (based on the 1995
Saskatchewan covered population) to control for age and
gender differences between our sample and the Saskatch-
ewan population (Saskatchewan Health, 1995). Six subjects
in the incidence cohort who were free of neck pain at
baseline and at 1-year did not complete the Chronic Pain
Questionnaire at the 6-month follow-up. We assumed that
they had no neck pain at the 6-month follow-up period.
Similarly, seven subjects in the course cohort did not
complete the Chronic Pain Questionnaire at the 6-month
follow-up. We assumed that their condition had remained
constant at the 6-monthfollow-up.Finally, we computed the
incidence rate for each outcome and arbitrarily allocated 3
months of follow-up time to individuals who had dropped
out of the study by the 6-month mail out and 9-months of
follow-up time to those who were lost to follow-up at the
Definition of outcomes for the incidence and course cohorts
Outcome Sample at
Baseline statusFollow-up status
Grade I to
Improvement 162Grade II or Grade I
Grade II to Grade IV
P. Co ˆte ´ et al. / Pain 112 (2004) 267–273 269
2.5.2. Age, gender and neck pain
We computed age-specific and gender-specific incidence
rates by dividing the number of men and women with a new
episode of neck pain by the sum of their individual time at
risk. We arbitrarily allocated 3 months of follow-up time to
individuals who had dropped out of the study by the
6-month mail-out and 9-month of follow-up time to those
who were lost to follow-up at the 1-year mail out. We
computed the crude incidence rate ratio (IRR) and 95%
confidence interval (CI) for subjects older than 46 years
(mean age of sample at baseline) of age relative to younger
subjects and for women relative to men.
All analyses were conducted using SAS (SAS Institute,
1990) except the direct standardizations, which were done
using STATA (StataCorp, 2000).
3.1. Sample characteristics
The mean age of the incidence cohort was 45.8 years (SD
13.4) and 47.2% were women. Half of the incidence cohort
worked full time and 13.8% were retired. According to the
Chronic Pain Questionnaire, most subjects in the incidence
cohort (61.3%) had experienced low back pain in the
previous 6 months (Von Korff et al., 1988, 1990, 1992).
Fourteen percent reported on the CES-D questionnaire that
they had suffered from significant depressive symptomatol-
ogy in the previous week (Boyd et al., 1982; Devins et al.,
1988; Orme et al., 1986; Radloff, 1977; Schulbert et al.,
The mean age of the course cohort was 43.4 years (SD
13.0) and 58.6% were women. Fifty percent of the course
cohort worked full time and 9.7% were retired. Almost 80%
experienced low back pain in the previous 6 months and
27.7% suffered from significant depressive symptomatology
in the previous week.
The follow-up rate was 73.7% at 6 months and 61.4% at
12 months. Subjects younger than 41 years of age,
unemployed, without post-secondary education and those
with worse general health on the SF-36 questionnaire
(Brazier et al., 1992; Garratt et al., 1993; Kravitz et al.,
1992; McHorney et al., 1992; Ware et al., 1993) were more
likely to drop out of the incidence cohort. However, subjects
who reported diabetes on the Comorbidity Questionnaire
(Carroll et al., 2004; Co ˆte ´ et al., 2000; Jaroszynski et al.,
1996) were less likely to drop out. In the course cohort,
subjects who were younger than 47.5 years of age and
reported cardiovascular disease were more likely to drop out
of the study, but those who had sedentary work were less
likely to be lost to follow-up.
3.3. Annual incidence of a new episode of neck pain
The age- and gender-standardized annual cumulative
incidence was 14.6% (95 percent confidence interval: 11.3,
Cumulative, gender specific and age- and gender standardized incidence (percent, 95 percent confidence interval) of an episode of neck pain in a Saskatchewan
population, 1995–1996 (n at riskZ513)
Any neck painGrade I Grade II Grade III–IV
Cumulative incidence (%)
Age-, gender-standardized* incidence (%)
Gender-specific incidence (%)
*Standard populationZSaskatchewan health insurance registration population, 1995.
Age-specific incidence rates, incidence rate ratios (IRR) and 95% confidence interval (CI)*
Incidence rate per person–yr (95% CI) Crude IRR (95% CI)
Any neck pain
%46 years: 0.226 (0.166, 0.287)
O46 years: 0.135 (0.087, 0.183)
%46 years: 0.644 (0.578, 0.710)
O 46 years: 0.542 (0.465, 0.620)
%46 years: 0.542 (0.408, 0.676)
O 46 years: 0.515 (0.362, 0.669)
%46 years: 0.684 (0.614, 0.754)
O46 years: 0.888 (0.830, 0.945)
%46 years: 0.183 (0.137, 0.233)
O46 years: 0.193 (0.132, 0.254)
%46 years: 0.211 (0.129, 0.293)
O46 years: 0.290 (0.174, 0.407)
0.60 (0.38, 0.93)
Resolution of pain and disability0.84 (0.70, 1.00)
Improvement0.95 (0.64, 1.41)
Persistence of symptoms1.30 (1.15, 1.47)
Aggravation of condition1.05 (0.69, 1.61)
Recurrence of neck pain1.37 (0.79, 2.38)
*Age%46 is used as the reference group.
P. Co ˆte ´ et al. / Pain 112 (2004) 267–273 270
17.9) (Table 2). Most new episodes were mild (Grade I)
(12.8 percent; 95 percent confidence interval: 9.6, 15.9).
Less than one percent of the population developed disabling
neck pain (Grade III–IV) (0.6%, 95 percent confidence
interval: 0, 1.1). The incidence of Grade I neck pain was
higher in women, but men were more likely to experience
Grade II–IV neck disorders (Table 2).
3.4. Course of neck pain
Of the 587 subjects with prevalent neck pain at baseline,
36.6% (95 percent confidence interval: 32.7, 40.5) reported
resolution of their pain and disability and 32.7% (95 percent
confidence interval: 25.5, 39.9) reported that their condition
improved during follow-up. Persistent neck pain was
experienced by 37.3% (95 percent confidence interval:
33.4, 41.2) of subjects and 9.9% (95 percent confidence
interval: 7.4, 12.5) reported an aggravation of their
condition. Finally, 22.8% (95 percent confidence interval:
16.4, 29.3) had recurrent neck pain.
3.5. Age, gender and neck pain
The incidence of a new episode of neck pain was lower in
older subjects than in younger ones (IRR 0.60; 95% CI 0.38,
0.93) (Table 3). Older subjects were slightly less likely than
younger subjects to report complete resolution of their neck
pain (IRR 0.84; 95% CI 0.70, 1.00) and more likely to
experience persistent symptoms (IRR 1.30; 95% CI 1.15,
1.47). The rates of improvement, aggravation, or recurrence
of neck pain did not vary with age.
Women were more likely (IRR 1.67; 95% 1.08, 2.60) to
experience an episode of neck pain then men (Table 4).
Women were more likely than men to report persistent neck
pain (IRR 1.19; 95% CI 1.03–1.38) and less likely to
experience complete resolution of their pain and disability
(IRR 0.75; 95% CI 0.63–0.88). The rates of improvement,
aggravation or recurrence of neck pain were similar for men
We conducted a population-based, cohort study of
randomly selected adults to determine the incidence and
course of neck pain. Our study is the first one to quantify the
burden of disability associated with neck pain in the general
population. Each year in Saskatchewan, 600/100,000 adults
experience a new episode of disabling neck pain and only
about one-third of those with neck pain experience a
complete resolution of their condition. These findings
contradict the commonly held view that neck pain is
a benign disorder with a highly favourable prognosis.
Rather, neck pain is a chronic episodic condition charac-
terized by episodes of persistent, recurrent or fluctuating
pain and disability.
Our study was designed to follow prospectively two
cohorts of subjects, one that is at risk of developing neck
pain and the other composed of subjects with prevalent neck
pain. This design is optimal to study episodic disorders
because subjects with and without the condition can be
followed simultaneously and therefore document incidence
and course in the same population at the same time.
Studying chronic episodic disorders requires that we adapt
the traditional definition of incidence to reflect its course.
Specifically, we defined incidence as the development of
neck pain in a cohort of subjects who were free from neck
pain in the previous 6 months rather than the development
of neck pain in a cohort of subjects who had never
experienced neck pain.
Although we used a different definition of neck pain, our
incidence estimates agree with those reported by Croft et al.
(2001). In South Manchester, United Kingdom, the
cumulative annual incidence of neck pain lasting for more
than 1 day was 17.9% among individuals who were free
from neck pain in the month prior to the survey. We
measured neck pain with the Chronic Pain Questionnaire
and found that age and gender standardized incidence of any
neck pain was 14.6% among individuals who had been free
from neck pain for the past 6 months. Both studies found a
higher incidence of neck pain in women. These results
Gender-specific incidence rates, incidence rate ratios (IRR) and 95% confidence interval (CI)
Incidence rate per person–yr (95% CI) Crude IRR (95% CI)
Any neck pain Men: 0.136 (0.088–0.183)
Women: 0.227 (0.166–0.289)
Men: 0.673 (0.598, 0.749)
Women: 0.503 (0.438, 0.568)
Men: 0.500 (0.321, 0.679)
Women: 0.571 (0.452, 0.690)
Men: 0.684 (0.599, 0.770)
Women: 0.815 (0.757, 0.873)
Men: 0.119 (0.070, 0.168)
Women: 0.145 (0.100, 0.191)
Men: 0.218 (0.123, 0.312)
Women: 0.263 (0.167, 0.360)
1.67 (1.08, 2.60)
Resolution of pain and disability0.75 (0.63, 0.88)
Improvement1.14 (0.76, 1.73)
Persistence of symptoms1.19 (1.03, 1.38)
Aggravation of condition1.22 (0.73, 2.05)
Recurrence of neck pain 1.21 (0.68, 2.13)
Males are used as the reference group.
P. Co ˆte ´ et al. / Pain 112 (2004) 267–273 271
suggest that, as for low back pain, neck pain is a condition
that affects individuals at a similar rate across industrialized
Our results highlight the importance of measuring neck
pain with a standard instrument that discriminates between
low-intensity, high-intensity and disabling pain. Using a
valid and reliable questionnaire such as the Chronic Pain
Questionnaire presents several advantages. First, it com-
bines the constructs of pain and activity limitations into one
outcome that can be used to measure the prevalence and
incidence of neck pain. Second, it allows for the accurate
measurement and reporting of the burden of disability
associated with neck pain. This is particularly important
because most individuals with neck pain experience low
intensity, non-disabling conditions. The use of a single
generic question about neck pain can therefore lead to an
overestimation of the magnitude of the problem. Third,
classifying subjects into mutually exclusive severity grades
provides a method to follow the progression of neck pain.
This information is essential to adequately describe the
course of neck pain in the general population. Finally, using
a standard questionnaire to measure neck pain and disability
would allow valid comparisons of prevalence and incidence
estimates across populations.
The current evidence suggests that neck and low back
pain run similar courses. Recent studies of the general
population indicate that the course of low back pain is also
characterized by periods of remissions and exacerbations
(Hestbaek et al., 2003a,b). These observations and the report
of strong associations between neck and low back pain give
support to the hypothesis that they represent components of
a larger chronic pain syndrome (Cassidy et al., 2000; Co ˆte ´ et
al., 2000; Croft et al., 2001, 2003).
The main threat to the validity of our results is loss to
follow-up. To understand whether attrition biased our
results, we determined which factors predicted loss to
follow-up. If attrition were random, then the measured
baseline characteristics would be equally distributed
between those who remained in the study and those who
dropped out, and loss to follow-up would not introduce bias
into the study. However, bias is present when characteristics
are both related to the outcome and loss to follow-up.
Our analysis suggests that attrition was not random. In the
incidence cohort, subjects who dropped out were more
likely to be young, unemployed, less educated and less
healthy. We have previously reported that these subjects are
more likely to report prevalent neck pain (Co ˆte ´ et al., 2000).
Although we controlled for age and gender through direct
standardization, it is possible that our study underestimated
the incidence of new episodes of neck pain. Fewer
factors were related to attrition in the ‘course’ cohort and
no systematic pattern of attrition was observed. Again the
impact of age was controlled through standardization,
and it is unclear whether sitting or standing at work, or
the presence of heart disease is related to the course of
The current knowledge on the epidemiology of neck pain
is limited. Our study provides clinicians and policy makers
with population-based estimates of the incidence and course
of neck pain. Moreover, our results emphasize that neck
pain is related to significant activity limitations for an
important proportion of the population. This information is
essential for health-care providers to accurately assess
prognosis and determine treatment expectations. We
demonstrated that neck pain follows an episodic course
that can lead to significant disability. Although a significant
proportion improve, most individuals with neck pain do not
experience complete resolution of their pain and disability.
In this respect, neck pain represents an important chronic
This work was supported by a research grant from the
Chiropractor’s Association of Saskatchewan. Dr Co ˆte ´ is
supported by the Canadian Institute of Health Research
through a New Investigator Award and by the Institute for
Work & Health by the Workplace Safety and Insurance
Board of Ontario. Dr Cassidy is supported by an endowed
research chair from the University Health Network. Dr
Carroll is supported by a Health Scholar Award from the
Alberta Heritage Foundation for Medical Research. Vicki
Kristman is supported by a Doctoral Training Award from
the Canadian Institutes of Health Research in partnership
with the Physical Medicine Research Foundation’s Wood-
bridge Grants and Awards Program and by the Institute for
Work & Health by the Workplace Safety and Insurance
Board of Ontario.
Binder A. Neck pain. Clin Evid 2003;9:1277–91.
Bovim G, Schrader H, Sand T. Neck pain in the general population. Spine
Boyd JH, Weissman MM, Thompson WD, Myers JK. Screening for
depression in a community sample. Arch Gen Psychiatry 1982;39:
Brazier JE, HarperR, Jones NMB,O’CathainA,ThomasKJ, UsherwoodT,
Westlake L. Validating the SF-36 Health Survey questionnaire: new
outcome measure for primary care. Br Med J 1992;305:160–4.
Carroll LJ, Cassidy JD, Co ˆte ´ P. The Saskatchewan Health and Back Pain
Survey: the prevalence and factors associated with depressive
symptomatology in Saskatchewan adults. Can J Public Health 2000;
Carroll L, Cassidy JD, Co ˆte ´ P. Depression as a risk factor for onset of
troublesome neck and low back pain. Pain 2004;107:134–9.
Cassidy JD, Carroll LJ, Co ˆte ´ P. The Saskatchewan Health and Back Pain
Survey. The prevalence of low back pain and related disability in
Saskatchewan adults. Spine 1998;23:1860–7.
Cassidy JD, Carroll L, Co ˆte ´ P, Lemstra M, Berglund A, Nygren A˚.
Effects of eliminating compensation for pain and suffering on the
incidence and prognosis of whiplash claims. New Engl J Med
P. Co ˆte ´ et al. / Pain 112 (2004) 267–273 272
Co ˆte ´ P, Cassidy JD, Carroll L. The Saskatchewan Health and Back Pain
Survey: the prevalence of neck pain and related disability in
Saskatchewan adults. Spine 1998;23:1689–98.
Co ˆte ´ P, Cassidy JD, Carroll L. The factors associated with neck pain and its
related disability in the Saskatchewan population. Spine 2000;20:
Croft PR, Lewis M, Papageorgiou AC, Thomas E, Jayson MIV,
Macfarlane GJ, Silman AJ. Risk factors for neck pain: a longitudinal
study in the general population. Pain 2001;93:317–25.
Croft P, Lewis M, Hannaford P. Is all chronic pain the same? A 25-year
follow-up study Pain 2003;105:309–17.
Devins GM, Orme CM, Costello CG, Binik YM, Frizzell B, Stam HJ,
Pulin WM. Measuring depressive symptoms in illness populations:
psychometric properties of the Center for Epidemiologic Studies
Depression (CES-D) Scale. Psychol Health 1988;2:139–56.
Garratt AM, Ruta DA, Abdalla MI, Buckingham JK, Russell IT. The SF-36
Health Survey Questionnaire: an outcomes measure suitable for routine
use within the NHS? Br Med J 1993;306:1440–4.
Guez M, Hildingsson C, Nilsson M, Toolanen G. The prevalence of neck
pain. A population-based study from Northern Sweden. Acta Orthop
Hestbaek L, Leboeuf-Yde C, Manniche C. Low back pain what is the long-
term course? A review of studies of general patients populations Eur
Spine J 2003a;12:149–65.
Hestbaek L, Leboeuf-Yde C, Engberg M, Lauritzen T, Bruun NH,
Manniche C. The course of low back pian in a general population.
Results from a 5-year prospective study. J Manipulative Physiol Ther
Jaroszynski G, Cassidy JD, Carroll LJ, Co ˆte ´ P, Yong-Hing K. Development
and validation of a comorbidity scale. Canadian Orthopaedic Research
Association, Quebec, June, 1996.
Kravitz RL, Greenfield S, Rogers WH, Manning WG, Zubkoff M,
Nelson E, Tarlov AR, Ware JE. Differences in the mix of patients
among medical specialities and symptoms of care: results from the
Medical Outcomes Study. J Am Med Assoc 1992;267:1617–23.
Ma ¨kela ¨ M, Helio ¨vaara M, Sievers K, Impivaara O, Knekt P, Aromaa A.
Prevalence, determinants, and consequences of chronic neck pain in
Finland. Am J Epidemiol 1991;134:1356–67.
McHorney CA, Ware JE, Rogers W, Raczek A, Lu JFR. The validity and
relatie precision of MOS Short- and Long-Form Health Status Scales
and Dartmouth COOP Charts. Med Care 1992;20:253–65.
Orme JG, Reis J, Herz EJ. Factorial and discriminant validity of the Center
for Epidemiological Studies Depression (CES-D) Scale. J Clin Psychol
Picavet HSJ, Schouten JSAG. Musculoskeletal pain in the Netherlands:
prevalence, consequencesand risk groups, the DMC3-Study. Pain 2003;
Radloff LS. The CES-D Scale. A self-reported depression scale for research
in the general population. Appl Psychol Meas 1977;1:385–401.
SAS Institute. SAS/STAT User’s Guide, Version 6, Volume 1 and 2. Cary,
NC: SAS Institute; 1990.
Saskatchewan Health, Health Insurance Registration. Covered Population,
Saskatchewan Health; 1995.
Schulbert HC, Saul M, McClellant M, Ganguli M, Christy W, Frank R.
Assessing depression in primary medical and psychiatric practices.
Arch Gen Psychiatry 1985;42:1164–70.
StataCorp, STATA Statistical Software: Release 7.0, Stata Corporation,
College Station, TX, 2000.
van der Windt D, CroftP, Pennix B. Neck and upperlimb pain: more painis
associated with psychological distress and consultation rate in primary
care. J Rheumatol 2002;29:564–9.
Von Korff M, Dworkin SF, Le Resche L, Kruger A. An epidemiologic
comparison of pain complaints. Pain 1988;32:173–83.
Von Korff M, Dworkin SF, LeResche L. Graded chronic pain status: an
epidemiologic evaluation. Pain 1990;40:279–91.
Von Korff M, Ormel J, Keefe FJ, Dworkin SF. Grading the severity of
chronic pain. Pain 1992;50:133–49.
Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey: Manual
and interpretation guide. Boston: The Health Institute, New England
Medical Centre; 1993.
P. Co ˆte ´ et al. / Pain 112 (2004) 267–273 273