Journal Logo

CLINICAL SCIENCES: Clinically Relevant

Dose-Response Relationship of Specific Training to Reduce Chronic Neck Pain and Disability


Author Information
Medicine & Science in Sports & Exercise: December 2006 - Volume 38 - Issue 12 - p 2068-2074
doi: 10.1249/01.mss.0000229105.16274.4b
  • Free


Neck and low-back disorders are the most common musculoskeletal problems causing visits to physicians (4). Chronic neck pain, a reduced range of motion of the cervical spine, and weakened neck-muscle force are more common among women than men, and they are related to impairments, functional limitations, and disability (11,14,30). The origin of chronic neck pain is normally multifactorial, including, for instance, physical strain and psychosocial stress (25). Its treatment has varied from passive rest to active treatments (31).

Earlier studies have suggested that active treatment may effectively reduce neck disorders (21). However, the results of active treatment vary in different studies depending on the training mode used, intensity, and duration. Several studies have shown that intensive strength training has resulted in a reduction of neck pain (5,16,24,31). However, the results of randomized studies with lower training intensity have been inconsistent (6,7,15,26,29,31).

In conservative treatments such as physiotherapy, the ultimate goal of therapeutic exercises is to achieve symptomless capacity to meet the physical load of work and other activities in everyday life. This tailored training should be based on knowledge of the effects of specific exercises, functional capacity status, the potential rate of recovery, complications, precautions, and contraindications (19). Whenever tailoring therapeutic exercises, one must be aware of possible interactions between the intervention and other activities outside the program. For the prescription of exercise, one should also be sensitive to the effects of the different types, quantities, and intensities of exercise, as well as the effects of temporal changes (12).

It is useful to get an insight into the dose-response relationship of different modes of physical activity. However, we still lack knowledge on the dose-response relationship for active treatments among neck patients. Therefore, the purpose of this study was to examine the dose-response relationship of specific strength- and endurance-training regimes for the cervical muscles that have been shown to be effective among women with chronic neck pain and disability (31).



This study was a secondary analysis of a randomized, controlled, examiner-blinded 12-month intervention trial with two training groups and a control group (31). In the original study, the participants were recruited from occupational health care services that were situated in the southern and eastern parts of Finland and known to provide health care for female employees. On the basis of a clinical examination, the 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. The accepted referrals were sent to the Punkaharju Rehabilitation Center. A questionnaire was sent to prospective patients to confirm their current health status regarding the inclusion and exclusion criteria. The following inclusion criteria were used: aged 25-55 yr, working in an office, permanently employed, motivated to continue working, motivated for rehabilitation, and constant or frequently occurring neck pain for more than 6 months. The exclusion criteria included 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, inflammatory rheumatic disease, severe psychiatric illness, other diseases that might prevent physical loading, and pregnancy.

The detailed subject flowchart has been described in an earlier report (31). A total of 180 female office workers were selected to participate in the study. The participants were randomly assigned into two training groups and a control group with 60 patients in each group (Table 1). Before the intervention, the training and control groups were divided into smaller classes of 10 patients each. The training groups participated in an institutional rehabilitation program, which they started in 1-month intervals alternating between strength and endurance training. The classes of the control group were measured between the classes of both training groups. The ethics committee of the Punkaharju Rehabilitation Center, Punkaharju, Finland, approved the study. All participants gave their written informed consent before entering the study.

Content of the training programs.


Measurements were obtained at baseline and after a 12-month intervention period. Maximal oxygen uptake was measured with a submaximal bicycle ergometer test by an exercise physiologist. Physical activity, defined as any bodily movement produced by skeletal muscle contractions that substantially increase energy expenditure (12,18), was estimated from data obtained by a 1-month (28 d, 4 wk) all-time recall questionnaire. Information on the specific training programs and other activities were also collected via a training diary throughout the 12-month intervention. Both the questionnaires and the diaries were administered by the same physical therapist during the institutional control visits.

For precise estimation, the information on physical activity collected from the diary and the questionnaire was divided into activity during work, commuting, leisure time, specific training, and miscellaneous time such as gardening or home work (Fig. 1). All activities were converted into metabolic equivalents (METs) by specific software (MetPro 2.03.7, Sci Reha, Jyväskylä, Finland). One MET represents the approximate rate of oxygen consumption of a seated individual at rest (3.5 mL·kg−1·min−1) (2,12). Physical activity at work was assessed with a seven-point scale, which was accompanied by illustrations and descriptions of various types of work. The scale varied from 1.5 (light work) to 10 METs (extremely heavy work). Activity in commuting was categorized as motor vehicle, bicycle, or walking. An additional question was asked about the intensity of the exercise, which was assessed as METs on the basis of breathing and sweating.

tegories of physical activity.

The assessment of intensity was based on 1700 activities described in the MetPro-database (17), which includes the values of previous publications such as Ainsworth et al. (1,2). The test-retest intraclass correlation coefficients obtained for the questionnaire ranged between 0.62 and 0.94, and the earlier version has also been proven to be acceptable (kappa coefficient from 0.46 to 0.78) (17). Subjectively perceived neck pain (0-100 mm) was assessed on a visual analog scale (VAS) (8,23). Subjectively perceived disability (0-100 mm) was assessed similarly on a modified neck- and shoulder-pain and disability index (28). The questionnaire includes 13 questions about pain and subjective disability in different daily activities. The results are expressed as a mean of the scores of the answers.


The training groups participated in a 12-d consecutive institutional rehabilitation period to ensure that the training program was learned properly. Participants were instructed by the same experienced physical therapist and they were instructed to perform the exercises at home three times a week. Written information about the exercises was given to all participants. The content of the training programs is shown in Table 1.

The endurance-training group exercised neck flexor muscles by lifting the head up from the supine position in three series of 20 repetitions. The strength-training group used an elastic rubber band (Theraband, Hygiene Corp, Akron, OH) to train the neck flexor muscles in each session. Single series of 15 repetitions directly forward, obliquely towards right and left, and directly backward were performed in a sitting position. The aim was to maintain the level of resistance at 80% of the participant's maximum isometric strength. The load was checked with a handheld isometric strength testing device (Force-Five, Wagner Instruments, Greenwich, CT) during the training sessions at the baseline and after 2-, 6-, and 12-month follow-up visits for controlling the progress of the training.

After specific neck training, both groups performed dynamic exercises for the shoulders and upper extremities by doing dumbbell shrugs, presses, curls, bent-over rows, flyes, and pullovers. The endurance-training group performed three sets of 20 repetitions for each exercise with a pair of dumbbells, each of which weighed 2 kg. The strength-training group exercised with an individually adjusted single dumbbell with a weight ranging from 4 to 13 kg at a gradually increasing load. The group performed only one set for each exercise with the highest load possible to perform 15 repetitions. Members of 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 finished with stretching exercises for the neck, shoulder, and upper-limb muscles for 20 min. The training groups were also advised to perform aerobic exercise three times a week for a half an hour. If they lacked the time to perform all of the exercises, they were encouraged to perform a minimum amount of muscle exercises specifically for the neck.

Both training groups also underwent a multimodal rehabilitation program, including relaxation training, aerobic training, behavioral support to reduce fear of pain, and improve exercise motivation, as well as lectures and practical exercises in ergonomics. During the rehabilitation course, each patient received four 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 physical exercises.

Each control class of 10 participants was asked to come in for baseline strength measurements. According to the recruiting and the block randomizing procedure, these measurements took place in 2-month intervals between the training classes during the same year. The control participants spent 3 d at the rehabilitation center and performed recreational activities in addition to the tests. They were advised to perform aerobic exercises three times a week for a half hour each time. They received written information about the same stretching exercises that were performed by the training groups, and they were supposed to practice the exercises at home for approximately 20 min, three times a week, on a regular basis. Each control participant was trained once to perform these exercises properly. The participants were not introduced to any exercises to improve muscle strength, and they received no advice for this. After the 1-yr follow-up measurements, they were offered the option to participate in the same rehabilitation training course as the active intervention participants, including training and follow-up. This was also financed by the Social Insurance Institution.

Statistical Analysis.

The analyses were carried out with the SPSS (Statistics Package for Social Sciences) 11.0 statistical program. The means and standard deviations (SD) have been used for the descriptive statistics. A baseline comparison was made between the groups using a one-way analysis of variance (ANOVA). Paired t-tests were used to evaluate the within-group change. A covariance analysis (ANCOVA) with the baseline measurements as covariates was used to evaluate the differences between the groups at follow-up. A P value < 0.05 was considered statistically significant.

For determining predictors of the changed neck pain (VAS), a correlation analysis was conducted first. Offered variables for correlation with change in neck pain (VAS) were the energy expenditure of the specific neck training; change in energy expenditure of work, commuting, leisure, and miscellaneous time activity; neck pain at baseline; the study group (the strength group vs endurance group); and age. Secondly, based on the suggested variables by the correlation analysis, a forced entry model was conducted to determine the contribution of different variables to the change in neck pain. Accordingly, neck pain at baseline, strength group, age, energy expenditure of work, change in the energy expenditure of leisure-time physical activity, and energy expenditure for a specific training program were entered into the forced model. Similarly, a forced entry model was conducted to determine the contribution of different variables to the change in the disability index. A P value < 0.05 was considered statistically significant.


At baseline, the participants had similar characteristics in general. Their characteristics, physical performance, and neck-pain score at baseline are given in Table 2. The training program was completed by 60, 58, and 59 persons in the strength, endurance, and control groups, respectively. One person was excluded after randomization because of diagnosed polymyalgia rheumatica, another withdrew because of pregnancy, and a third one had personal reasons for quitting. Seven participants (three in the strength-training group and four in the endurance-training group) did not fill out the physical activity questionnaire properly, and the collected information could not be analyzed.

Baseline characteristics of the participants.

The training adherence (at least once a week) was 86, 93, and 65% for the strength, endurance, and control groups, respectively. According to training diaries, the patients continued their training throughout the year; the mean (SD) for the weekly training sessions of the three groups were 1.7 (0.6), 2.0 (0.8), and 2.0 (0.8), respectively. The maximal oxygen uptake was about 30 mL·kg−1·min−1 and did not change significantly within or between the groups during the 12-month study period. Actual neck-pain scores (VAS; mean (SD)) were 57 (20), 57 (21), and 58 (20) at baseline and 18 (22), 23 (22), and 42 (23) at 12-month follow-up of the three groups, respectively. Similarly, actual disability scores were 35 (13), 38 (14), and 38 (15) at baseline and 12 (13), 16 (16), and 26 (16) at 12-month follow-up of the three groups.

The average metabolic rate and the energy expenditure of the specific training programs and leisure-time activity are shown in Table 3. There was no statistically discernible difference between groups for leisure-time activity, whereas total activity of the training groups was significantly greater because of their specific training.

Metabolic rate and the monthly energy expenditure of the specific training programs and leisure-time activity in the strength, endurance, and control groups at the 12-month follow-up.

The independent predictors of changes in chronic neck pain are shown in Table 4. The total variation predicted by the model was 50%. The energy expenditure of the specific training correlated negatively with pain scores, accounting for 10% of the total variation predicted by the model. In the specific training program, 1 MET·h accounted for an 0.8-mm (95% CI from 0.5 to 1.1 mm) decrease in experienced pain. The level of neck pain at baseline accounted for 36% of the total variation predicted by the model; that is, the participants with the highest pain scores at baseline benefited most from the training program. The intervention group (strength vs endurance) was not a predictor of change in neck pain; both training modes relieved the participants' neck pain significantly. Among the members of the strength and endurance groups whose training load was under 20 MET·h per month, the pain score was observed to both increase and decrease (Fig. 2). Among those who trained from 20 to 35 MET·h per month, the level of pain reduction varied from 0 to more than 80 mm (VAS). Furthermore, all those who had trained for more than 35 MET·h decreased their neck pain by more than 20 mm (VAS). The energy expenditure during work or aerobic leisure-time activities did not explain the decrease in neck pain.

Determinants of change in chronic neck pain.
Scatter diagram of change in neck pain (VAS) and in energy expenditure of the specific training program at follow-up. Fitted regression lines are shown.

The independent predictors of changes in modified neck and shoulder pain and disability index are shown in Table 5. The energy expenditure of the specific training correlated negatively with the disability index, accounting for 11% of the total variation predicted by the model. In the specific training program, 1 MET·h accounted for a 0.5-mm (95% CI from 0.3 to 0.7 mm) decrease in disability index. Again, the intervention group (strength vs endurance) was not a predictor of the change in disability; both training modes relieved the participants' disabilities significantly.

Determinants of change in modified neck and shoulder pain and disability index.


In this study, the high compliance and adherence of the training enabled detailed analysis of the amount of training in the treatment of chronic neck pain and disability. We showed a clear dose-response relationship for two neck-training programs. One MET-hour of training per week accounted for an 0.8-mm decrease of neck pain on the visual analog scale (VAS) and a 0.5-mm decrease on the disability index. The training dose of 20 MET·h per month (5 MET·h·wk−1) represented a 16-mm decline in the VAS. Furthermore, the study showed a clear decrease in VAS in all those subjects who trained more than 8.75 MET·h (525 MET·min) per week, or 35 MET·h of training per month.

Training compliance and adherence are important issues. The training program was ineffective when performed twice a week at a total load of less than 5 MET·h·wk−1. However, the same training was effective if it was performed three times per week, for 8.75 MET·h altogether. In other words, a greater dosage of the specific training had a greater effect on neck-pain symptoms in the studied patients. This may be attributable to improved local metabolism and muscle strength. All of the patients who complied with training of more than that of 35 MET·h per month belonged to the strength group. It may be more probable to complete 40 min of specific higher-load strength exercises than 60 min of endurance training with 2-kg dumbbells only. Despite the moderate metabolic rate of both training groups, the strength group's training at 80% muscular capacity was more intense compared with the endurance group's dumbbell training. Accordingly, quite high muscular strain was continually achieved with the rubber-band method, because the 80% load was controlled during the follow-up visits. These regular follow-up visits may also have increased the motivation of some patients in both training groups to continue training throughout the year.

The study had several strengths, such as the randomized controlled design and the low dropout rate. Also, the validity of the VAS in measuring the change in pain has been shown to be good (23). The careful set of inclusion and exclusion criteria, as well as gender, age, work, and seasonal changes were considered possible confounding factors. The inquired information was collected for a 4-wk period according to recommendations (3). MET values have been shown to have an acceptable level of reliability, and to minimize misinterpretations, we compared every exercise value with those of other similar activities (1,13,22). In addition to recalled questionnaires, we used prospective training diaries for a more detailed analysis of the specific training. According to the training diaries, the specific training stayed constant during the 12-month period. There may have been some limitations in the study, however. In the assessment of exercise intensity, light functioning is typically overestimated and heavy functioning is underestimated (12,20). Nevertheless, these possible inaccuracies counteract each other.

Both the strength- and the endurance-training protocols seemed to be effective in decreasing neck pain and disability if enough training was performed. Accordingly, it was possible to achieve an average pain relief of 40 mm (60 to 20) in VAS with both of the training modes, which has high clinical relevance (10). With such an amount in pain reduction, it is likely that patients would report successful treatment. Also, the patients' disability declined considerably in both of the training groups: more than 50% from their baseline values (9). However, their level of disability was only mild at baseline. The positive long-term effect of endurance-type training may be attributable to improved metabolism and increased muscle strength (32). The study provided good evidence that both training methods can be effective for rehabilitating women who work in offices and who have average, moderate levels of chronic neck pain and mild disability. However, as an additional limitation, patients with more acute or severe neck pain and disability may have demonstrated different results (27). Also, the study's findings are not directly generalizable to males, individuals in other occupations or without work, or individuals who have chronic neck pain associated with any of the disorders listed in the exclusion criteria. Thus, further studies are needed.

In this study, the level of chronic neck pain and disability at baseline accounted for 36 and 25% of the total variation predicted by the model. Accordingly, the greater the chronic pain and disability, the more they are likely to benefit from the training. However, one should be aware that patients with different levels of pain and disability, and other types of exercise programs, may demonstrate different dose responses (5-7,15,26,29). In addition to the present study, several other studies have found clinically relevant pain reduction (~50%) as a consequence of exercise therapy, at least in the short term (5,15,26). In these previous studies investigating exercise dose responses among neck-pain patients, duration of interventions, compliance, training mode, and intensity have all varied greatly. Thus, it is difficult to compare findings between the studies.

Our intervention was also intended to increase leisure-time recreational aerobic activities of the patients. In the present study, however, considerable changes in patients leisure-time activity levels were not observed. This may be attributable to the relatively high demands of the intervention itself.


The energy expenditure of the described specific exercise protocols were associated with decreased chronic neck pain and disability among women who worked in offices. The effective dose of the specific training program to decrease chronic neck pain was 8.75 MET·h (525 MET·min) per week. The amount of training required was feasible and safe to perform among female office workers.

The authors thank all of the study participants for their great effort; in particular, the work of the staff of the Punkaharju rehabilitation center is greatly appreciated. We would also like to thank Carlos Saavedra, MSc, Laval University, Canada and Désirée Wilks, MSc, Manchester Metropolitan University, UK for their constructive criticism, Tomi Strengell, MSc, PT, for the MetPro-analyses, Matti Pasanen, MSc, for his statistical consultation, Georgianna Oja for her linguistic revision of the manuscript, and Seppo Niemi for his assistance in preparing the figures and tables. Arja Häkkinen, DSc, and the staff of the department of Physical and Rehabilitation Medicine in the Jyväskylä Central Hospital were a great help in organizing analyses. This study was financed by the Social Insurance Institution, Helsinki, Finland.

Professor Mälkiä has a decision-making position in SciReha Company (MetPro 2.03.7, SciReha, Jyväskylä, Finland).


1. Ainsworth, B. E., W. L. Haskell, and M. C. Whitt. Compendium of physical activities: an update of activity codes and MET intensities. Med. Sci. Sports Exerc. 32:S498-S516, 2000.
2. American College of Sports Medicine. ACSM's Guidelines for Exercise Testing and Prescription, Philadelphia, PA: Lippincott Williams & Wilkins, 2000.
3. Baranowski, T., M. Smith, and W. O. Thompson. Intraindividual variability and reliability in a 7-day exercise record. Med. Sci. Sports Exerc. 31:1619-1622, 1999.
4. Bot, S. D., J. M. van der Waal, C. B. Terwee, et al. Incidence and prevalence of complaints of the neck and upper extremity in general practice. Ann. Rheum. Dis. 64:118-123, 2005.
5. Bronfort, G., R. Evans, and B. Nelson. A randomized clinical trial of exercise and spinal manipulation for patients with chronic neck pain. Spine 26:788-799, 2001.
6. Chiu, T. T., C. W. Hui-Chan, and G. Chein. A randomized clinical trial of TENS and exercise for patients with chronic neck pain. Clin. Rehabil. 19:850-860, 2005.
7. Chiu, T. T., T. H. Lam, and A. J. Hedley. A randomized controlled trial on the efficacy of exercise for patients with chronic neck pain. Spine 30:E1-E7, 2005.
8. Dixon, J. S., and H. A. Bird. Reproducibility along a 10-cm vertical visual analogue scale. Ann. Rheum. Dis. 40:87-89, 1981.
9. Farrar, J. T., J. P. Young, JR., L. LaMoreaux, J. L. Werth, and R. M. Poole. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain 94:149-158, 2001.
10. Forouzanfar, T., W. E. J. Weber, M. Kemler, and M. van Kleef. What is a meaningful pain reduction in patients with complex regional pain syndrome Type 1? Clin. J. Pain 19:281-285, 2003.
11. Herman, K. M., and C. S. Reese. Relationships among selected measures of impairment, functional limitation, and disability in patients with cervical spine disorders. Phys. Ther. 3:903-912, 2001.
12. Howley, E. T. Type of activity: resistance, aerobic and leisure versus occupational physical activity. Med. Sci. Sports. Exerc. 33:S364-S369, 2001.
13. International Organization for Standardization. Ergonomics of the Thermal Environment-Determination of Metabolic Rate. ISO/FDIS 8996. Geneva: ISO, 2004.
14. Jordan, A., J. Mehlsen, and K. Ostergaard. A comparison of physical characteristics between patients seeking treatment for neck pain and age-matched healthy people. J. Manipulative Physiol. Ther. 20:468-475, 1997.
15. Jull, G., P. Trott, H. Potter, et al. A randomized controlled trial of exercise and manipulative therapy for cervicogenic headache. Spine 27:1835-1843, 2002.
16. Levoska, S., and S. Keinänen-Kiukaanniemi. Active or passive physiotherapy for occupational cervicobrachial disorders? A comparison of two treatment methods with a 1-year follow-up. Arch. Phys. Med. Rehabil. 74:425-430, 1993.
17. Mälkiä, E. MET based questionnaire for the study of physical activity. In: Assessment of Function and Movement: Selected Papers of the Third Nordic Symposium on Physiotherapy, E. Mälkiä and S. Sihvonen (Eds.). Jyväskylä: Paino Porras, pp. 98-103, 1996.
18. Mälkiä, E., O. Impivaara, and M. Heliövaara. The physical activity of healthy and chronically ill adults in Finland at work, at leisure, and during commuting. Scand. J. Med. Sci. Sports 4:82-87, 1994.
19. Mälkiä, E., and P. Kannus. Low back pain - to exercise or not to exercise? Scand. J. Med. Sci. Sports 6:63-72, 1996.
20. Pernold, G., E. Wigaeus-Tornqvist, and C. Wiktorin. Validity of occupational energy expenditure assessed by interview. AIHA J. 63:29-33, 2002.
21. Philadelphia Panel. Philadelphia panel evidence - based clinical practice guidelines on selected rehabilitation interventions for neck pain. Phys. Ther. 81:1701-1717, 2001.
22. Phillips, W. T., and J. A. Ziuraitis. Energy cost of the ACSM single-set resistance training protocol. J. Strength Cond. Res. 17:350-355, 2003.
23. Price, D. D., P. A. McGrath, and A. Rafii. The validation of visual analogue scales as ratio scale measures for chronic and experimental pain. Pain 17:45-55, 1983.
24. RandlØv, A., M. Østergaard, and C. Manniche. Intensive dynamic training for females with chronic neck/shoulder pain. A randomized, controlled trial. Clin. Rehabil. 12:200-210, 1998.
25. SjØgaard, G., U. Lundberg, and R. Kadefors. The role of muscle activity and mental load in the development of pain and degenerative processes at the muscle cell level during computer work. Eur. J. Appl. Physiol. 83:99-105, 2000.
26. Sjögren, T., K. J. Nissinen, S. K. Järvenpää, M. T. Ojanen, H.Vanharanta, and E. A. Mälkiä. Effects of a workplace physical exercise intervention on the intensity of headache and neck and shoulder symptoms and upper extremity muscular strength of office workers: a cluster randomized controlled cross-over trial. Pain 116:119-128, 2005.
27. Taylor, N. F., K. J. Dodd, and D. L. Damiano. Progressive resistance exercise in physical therapy: a summary of systematic reviews. Phys. Ther. 85:1208-1223, 2005.
28. Viikari-Juntura, E., E.-P. Takala, and H. Alaranta. Neck and shoulder pain and disability. Scand. J. Rehabil. Med. 20:167-173, 1988.
29. Viljanen, M., A. Malmivaara, and J. Uitti. Effectiveness of dynamic muscle training, relaxation training, or ordinary activity for chronic neck pain: randomized controlled trial. BMJ 327:475-479, 2003.
30. Webb, R., T. Brammah, and M. Lunt. Prevalence and predictors of intense, chronic, and disabling neck and back pain in the UK general population. Spine 28:1195-1202, 2003.
31. Ylinen, J., E-P. Takala, and M. Nykänen. Active neck muscle training in the treatment of chronic neck pain in women. JAMA 289:2509-2516, 2003.
32. Ylinen, J., A. H. Häkkinen, E. P. Takala, et al. Effects of neck muscle training in women with chronic neck pain: one-year follow-up study. J. Strength Cond. Res. 20:6-13, 2006.


©2006The American College of Sports Medicine