Neck Pain in Office Workers: Consequences and Supervised

Training

RESEARCH ARTICLE Open Access

Consequences and management of neckpain by female office workers: results of asurvey and clinical assessmentVenerina Johnston

Abstract

Background: Neck pain is common in office workers. However, the functional consequences of this pain to theindividual and how they are managed are not well known. The objective of this study is understand the impactof neck pain and the strategies female office workers use to manage their pain while remaining at work.

Methods: Female office workers with neck pain (n = 174) completed a survey about the impact of their neck pain,with 51 attending a university clinic for further assessment. Consequences of neck pain were evaluated withquestions on self-reported work absence, workers’ compensation claims, health care use, impact on work andleisure activity, and management strategies. Responses to survey questions were analysed using descriptiveanalyses.

Results: The results showed that during the preceding 12 months, 57.5 % of participants had consulted a healthprofessional due to neck pain; 42 % had reduced their leisure activities; 22.4 % had reduced their work activity and20.7 % had been absent from work. Only 5.2 % had ever submitted a workers’ compensation claim and 9 %indicated changing jobs due to neck pain. Of the 51 participants who attended for further assessment, 35.3 %indicated they ‘self-managed’ their neck pain with conventional medical strategies. Common strategies utilizedwere: prescription or over-the-counter medications (82.5 %), physiotherapy (64.7 %) and visiting their generalmedical practitioner (54.9 %).

Conclusions: Although the severity of neck pain experienced by female office workers in this study was low, theimpact on work and leisure was substantial. These workers tended to self-manage their pain by reducing workand/or leisure activity and utilizing passive coping strategies to remain at work. Physiotherapists are ideally suitedto provide self-management strategies to ensure workers remain healthy while working.

Keywords: Neck problem, Pain management, Public health

BackgroundNeck pain is a common problem in the working population[1, 2]. In particular, the prevalence of neck pain in officeworkers has been reported to be between 50 and 76 % inAustralia and 45–63 % internationally [3–8]. Despite this,little is known about the consequences of this problem tothe individual office worker and which strategies, if any, areutilized to ensure they remain at work [4]. Evidencesuggests that neck pain may lead to care seeking behaviour,

sickness absences and workers’ compensation claims[4, 9, 10]. In Australia, neck pain account for a small pro-portion of all serious workers’ compensation claims (2.2 %)[11], yet the contribution of persistent neck pain to thetotal burden of chronic pain in Australian society is 20 %[12]. Neck pain accounts for 20 % of the $34 billion eachyear spent on chronic pain in the Australian community[13]. Thus, gaining a greater understanding of how officeworkers manage their pain can enhance the developmentof validated and cost effective interventions and reduce theburden on the individual and the employer.Current strategies utilized by office workers to remain

at work with neck pain are unknown. Exploring theCorrespondence: v.johnston@uq.edu.auSchool of Health and Rehabilitation Sciences, The University of Queensland,Brisbane, Australia

© 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Johnston Archives of Physiotherapy (2016) 6:8 DOI 10.1186/s40945-016-0023-3

coping strategies office workers employ to manage theirneck pain may provide some insight into the importanceof this problem and help direct interventions in theworkplace. A recent study of patients’ experience andmanagement of neck pain in general practice found thatmany self-managed their pain with techniques likemassage and “over-the-counter medication” [14]. How-ever, this study was conducted in Germany, which has adifferent health care system than Australia, whereconservative interventions such as physiotherapy mustbe prescribed by a medical practitioner. Furthermore, itwas not specifically conducted in a working population.The aims of this study were to 1) describe the conse-quences of neck pain for female office workers and 2)explore their self-reported management and copingstrategies. It was hypothesised that female office workerswith neck pain generally, do not submit worker’scompensation claims but remain in the workforce bymanaging symptoms at the individual level by takingsick leave, visiting a health care professional and self-medication.

MethodsStudy designData from eligible female office workers was obtainedthrough a cross-sectional survey and clinical assessment.This research focused on neck pain in female officeworkers as females consistently demonstrate an increasedprevalence of neck disorders and are usually over-represented in the office worker population [5, 15].

ParticipantsOffice workers with neck pain over the age of 18 yearsand working at a computer more than 20 h per weekwere invited from 12 public and private institutions inthe banking, local government and health industry sec-tors. A total of 333 office workers volunteered (overallresponse rate of 30 %) and were screened for eligibilitybased on the severity of neck complaints and history ofneck trauma [3]. Those scoring greater than 8 % on theNeck Disability Index (NDI) and free of neck traumawere deemed as having neck complaints (N = 174,response rate 52.3 %). The score on the NDI was se-lected as the cut-off as this severity is indicative of mildto severe neck problems [16]. History of neck traumawas established by one question as previous musculo-skeletal trauma of the neck, shoulder or arm has beenshown to be a significant predictor of work absence [17].The characteristics of all employees were establishedthrough an employee profile provided by each organizationincluding age, gender, job titles, employment status, hoursworked per week, and type of work performed. This profiledid not differ between participants and non-participants.

ProcedureThe comprehensive online survey collected demographicdata and information about the consequences of theirneck problem on their work and home activities, work-place psychosocial demands and physical ergonomicdemands of their work [3]. The 174 eligible officeworkers were invited to attend a university clinic forfurther assessment of their neck pain of which 51 partic-ipants (29 %) agreed. To limit selection bias, no financialincentives or offers of treatment were provided. Figure 1demonstrates the flow of participants in the study. Thetime lapse between survey and assessment was on aver-age, 1 month. The assessment was undertaken duringthe worker’s lunch time or after hours and consisted ofan interview and physical examination performed by aqualified physiotherapist. The purpose of the interviewwas to further understand and describe the strategiesused to manage their pain and consisted of questionsabout the severity and history of pain and the type ofmanagement sought. The physical examination consistedof manual palpation of the neck, assessment of activephysiological movements of the neck and shoulders.

Ethical considerationsAll participants were informed that participation wasvoluntary and that no remuneration or incentives wouldbe provided by the investigators or their employer.Information on individual results was not released to theemployer. Ethics for this study was granted by TheUniversity of Queensland Medical Ethics Committee(#2004000225).

MeasuresSeverity of neck painThe NDI assessed the severity of disability due to neckpain [16] as this tool has been shown to have good test-retest reliability and internal consistency [18]. This indexincludes ten items that address functional activitiesincluding sleeping, reading, lifting, personal care, recre-ation, driving and work. There are six possible responsesfor each item which are scored from 0 (no disability) to 5(complete disability). The final score is obtained as a per-centage after adding the scores for each of the 10 items. Ahigher score indicates greater pain and disability.

Consequences of neck complaintsThe Nordic Musculoskeletal Questionnaire (NMQ) wasused to determine the duration of neck pain with thequestion “What is the total length of time that you havehad neck trouble (ache, pain or discomfort) during thelast 12 months?” [19] Five response options werepossible from No days, 1–7days, 8–30days, >30 days butnot every day, and every day. A body map was includedto assist participants to understand the area defined as

Johnston Archives of Physiotherapy (2016) 6:8 Page 2 of 8

the neck. This tool is widely used in occupationalresearch [20] and is a validated and reliable tool [21].Absence from work due to neck pain was assessed by

one question from the NMQ: “Have you been absentfrom work during the last 12 months because of troublein the neck?” This question showed high specificity andsensitivity when used to measure the occurrence of sick-ness absence due to back pain [22]. It has also been usedin the assessment of sickness absence due to neck andupper extremity pain [23].Health care use was determined by one question from

the NMQ, “Have you been seen by a doctor, physiotherap-ist, chiropractor or other health professional because oftrouble in the neck during the last 12 months?”. In theAustralian health care system, workers do not require amedical referral to consult with allied health practitionersand may self-refer without the knowledge of the work-place or the worker’s general practitioner.Consequences on the workers’ work activity, leisure

activities, and submission of workers’ compensationclaims were evaluated with one question each with adichotomous response option of No/Yes. These ques-tions were from the NMQ [19]:

� “Have you ever had to change jobs or duties becauseof neck trouble”

� “Has neck trouble caused you to reduce your activityat work during the last 12 months”

� “Has neck trouble caused you to reduce your leisureactivity during the last 12 months?”

� “Have you ever been absent from work during thelast 12 months because of neck trouble?”

� “Have you ever submitted a worker’s compensationclaim because of neck trouble?”

The reliability of the NMQ to collect data on theprevalence and consequences of musculoskeletal painhas been established and shown to range from moderate

to almost perfect [24]. The duration of time that neckpain affected work activity was evaluated with one ques-tion “What is the total length of time that neck troublehas prevented you from doing your normal work (athome or away from home) during the last 12 months?”with four response options: No days, 1–7 days, 8–30days or greater than 30 days.In the interview at the university clinic, participants

were asked two-open ended questions: ‘What do youthink has caused your neck problems?’ and ‘How do youmanage your neck pain?’ The open ended questions wereto reduce potential bias in responses. Responses to thefirst question were used to establish the work-relatednessof neck pain and were grouped into similar concepts withany ambiguity checked by another researcher.The responses to the second question were used to

determine the coping strategies adopted. Coping mecha-nisms refer to the specific thoughts and behaviourspeople use to manage their pain or their emotionalreactions to their pain [25]. These self-managementstrategies were classified as either active or passivecoping strategies based on the categorization tables ofBrown and Nicassio [26]. Blyth et al. [27] subsequentlyused these groupings to examine the relationship be-tween self-management strategies, disability and healthcare utilization in a population-based study of individ-uals with chronic pain. These authors identified passivecoping strategies as any treatment where something wasdone to, or given to the patient. This was further dividedinto the two sub-categories of passive behavioural (e.g.massage, rest, heat) and conventional medical manage-ment strategies (e.g. medication, physiotherapy as theseare given to or one to the patient). Active copingstrategies were described as any instrumental activityinitiated by the individual to manage their pain, if notcharacterised by avoidance or escape. For example, whilerest may be initiated by the individual, it is considered apassive strategy as it is intended to escape from pain

Fig. 1 Flow chart of participants through the study. *NDI Neck Disability Index

Johnston Archives of Physiotherapy (2016) 6:8 Page 3 of 8

rather than to function despite the pain. These strategieswere further divided into active behavioural with a phys-ical component (e.g. exercise, postural modification) andcognitive (e.g. relaxation, mental distraction).

AnalysisResponses to the survey questions were analysed using de-scriptive statistical analyses to determine the percentage ofpositive and negative responses to each question aboutthe impact of neck problems. SPSS version 22 (IBMCorporation New York, USA) was used to manage thedata. Strategies to manage neck problems reported by the51 participants were categorised as active (behavioural orcognitive) strategies or passive (behavioural or conven-tional medical) strategies. The percentage of participantsnominating any of these four categories was calculated.

ResultsIn the last 12 months, 82 % of the office workers whovolunteered reported experiencing neck pain. The meanscore on the NDI of all 174 participants was 20.2 %(SD = 9.1) indicating mild neck pain and disability.The median age of participants was between 40 and44 years. The mean age of the subsample of 51 par-ticipants was 44.3 years (SD = 9.5) with a mean NDIscore of 22 % (SD = 8.9). This NDI score indicatesthat those who volunteered for further assessmentwere representative of the total sample population.

Survey resultsTable 1 displays the reported consequences of neck pain insymptomatic female office workers. The results revealed

that 20.7 % of participants who reported neck pain in thelast 12 months were absent from work due to this pain. Inaddition, 14.9 % of workers did not participate in activitiesof daily living or normal work duties for 8 or more days inthe last 12 months due to pain. Nearly a quarter of thosesurveyed indicated reducing their work activity due to neckpain but the impact on leisure activities was greater with42 % affected. Only a small proportion of subjects reportedchanging jobs as a consequence of their neck pain. Adviceand/or treatment from a health care professional wassought by 57.5 % of participants with only 5.2 % reportedmaking a workers’ compensation claim in relation to theirneck pain.

Subsample results (N = 51)Figure 2 displays the self-reported source of neck symp-toms. The majority of participants believed performingcomputer work including keyboard work, mouse work andreading at computer was the source of their pain. Manyparticipants indicated more than one source of pain.Table 2 displays the active and passive self-management

strategies utilised by the subgroup of female office workersto manage their neck pain. It can be seen that passivestrategies were nominated more often than active self-management strategies. Conventional medical strategieswere the most common strategies utilised. Use of pre-scription or over-the-counter medication (most com-monly paracetamol and ibuprofen), physiotherapy, andvisiting a medical practitioner were the most commonlyreported management techniques used. Cognitive strat-egies were not employed at all by this population and ex-ercise as an active behavioural strategy was only used by a

Table 1 Responses to survey questions on the consequences of neck pain (N = 174)

Survey question Percent (n)

Have you ever had to change jobs or duties because ofneck trouble?

No 90.8 % (158)

Yes 9.2 % (16)

Has neck trouble caused you to reduce your activity atwork during the last 12 months?

No 77.6 % (135)

Yes 22.4 % (39)

Has neck trouble caused you to reduce your leisureactivity during the last 12 months?

No 58.0 % (101)

Yes 42.0 % (73)

Have you ever been absent from work during the last12 months because of neck trouble?

No 42.5 % (74)

Yes 57.5 % (100)

What is the total length of time that neck trouble hasprevented you from doing your normal work (at homeor away from home) during the last 12 months?

0 days 54.6 % (95)

1-7days 30.5 % (53)

8-30days 8.6 % (15)

>30 days 6.3 % (11)

Have you ever been absent from work during the last12 months because of neck trouble?

No 79.3 % (138)

Yes 20.7 % (36)

Have you ever submitted a worker’s compensationclaim because of neck trouble?

No 94.8 % (165)

Yes 5.2 % (9)

Johnston Archives of Physiotherapy (2016) 6:8 Page 4 of 8

small percentage of female office workers with neck pain.It is apparent that most participants employed more thanone strategy to manage their symptoms.Table 3 shows the categories and combination of self-

management strategies used by female office workerswith neck pain, the most common being the use of con-ventional medical management. Only five female officeworkers indicated not using anything to manage theirpain and six used a combination of strategies.

DiscussionThis study identified that the severity of neck pain infemale office workers is mild but that it has negativeimpact on their work and leisure time activity. Approxi-mately half of the sample reported that their partici-pation in usual activities of daily living was reduceddue to their neck pain. The number of workerscompensations claims submitted was low with mostparticipants using passive coping strategies to managetheir pain and remain at work. The significant impactof neck pain on function has been highlighted byother researchers [4, 28, 29].Nearly half of the participants nominated computer

work or the ergonomic environment at work as thesource of their neck pain. While there is ample evidencelinking musculoskeletal symptoms of the neck withcomputer use or features of the ergonomic workstation[6, 30–32], causation cannot be inferred in this cross-sectional study. Prospective studies and systematic re-views offer mixed evidence for the relationship between

Fig. 2 Reported source of neck pain in 51 female office workers

Table 2 Self-management strategies used by female office workerswith neck pain (n= 51)

Strategy used Individuals reporting % (n)

Active Coping Strategies

Behavioural:

Exercise 5.88 % (3)

Stretches 1.96 % (1)

Posture Modification 1.96 % (1)

Cognitive:

Mental Distraction, relaxation 0 % (0)

Passive Coping Strategies

Behavioural:

Massage 11.76 % (6)

Aromatherapy 1.96 % (1)

Conventional Medical:

Medication 82.35 % (42)

Physiotherapy 64.71 % (33)

General Practitioner 54.90 % (28)

Chiropractor 19.61 % (10)

Medical Specialist 13.73 % (7)

Other health care practitioner 7.84 % (4)

Acupuncture 3.92 % (2)

Table 3 Combinations of self-management strategies used byfemale office workers to manage their neck pain (n = 51)

Management combinations Percent (n)

Conventional medical only 70.59 % (36)

No management 9.80 % (5)

Passive behavioural & conventional medical 7.84 % (4)

Active behavioural only 5.88 % (3)

Passive behavioural only 3.92 % (2)

Active behavioural, passive behavioural& conventional medical

1.96 % (1)

Active behavioural & passive behavioural 0.00 % (0)

Active behavioural & conventional medical 0.00 % (0)

Cognitive only 0.00 % (0)

Johnston Archives of Physiotherapy (2016) 6:8 Page 5 of 8

the ergonomic environment and incident neck pain.Côté et al. [33] found that poor computer workstationdesign and work posture were two of several factorsassociated with the development of an episode of neckpain. In contrast, another systematic review found strongevidence that high keyboard usage time and poorperception of computer placement have no predictivevalue for the onset of neck pain [34].It is interesting to note that 22.4 % of the participants

identified that their work activity had decreased. This issimilar to two previous studies which also identified self-perceived reductions in productivity at work due to neckproblems of 26 % in the Netherlands [35] and 13 % inSweden [36]. Our study is the first in Australia tosuggest the possibility of reduced productivity due toneck problems in office workers. This is an importantfinding as it suggests that “presenteeism” may be aconcern in the workplace. Presenteeism has been de-fined as the decrease in performance associated with aworker remaining at work whilst impaired by a healthcondition [37]. Although challenging to measure, it hasbeen estimated that presenteeism can result in up to fourtimes greater loss in productivity than absenteeism[37, 38]. Thus, the cost of presenteeism for the workplacemay be greater than direct health care costs [39, 40]. Ourstudy suggests that this may be occurring in office workerswith neck pain and requires further investigation.The sub-sample of office workers used a range of self-

management strategies, mostly passive in nature. This isconsistent with findings in other neck pain related stud-ies [41]. Passive coping strategies were reported moreoften than active strategies in our sub-sample of femaleoffice workers and usually consisted of over the counteror prescription medications, physiotherapy and/orconsultation with a general practitioner.It is surprising that exercise was reported by so few

participants as a strategy to manage neck pain. Twosystematic reviews have found strong evidence to sup-port the positive effect of muscle strengthening andendurance exercises for controlling neck pain in officeworkers [42, 43]. When combined with manual therapy,these interventions, which consist of both passive andactive strategies, produce greater improvements in pain,function, quality of life and patient satisfaction com-pared to manipulation or mobilisation alone for chronicneck pain [44]. Health practitioners are encouraged torecommend such interventions for office workers withneck pain.There is evidence that the use of active self-management

strategies can substantially reduce the likelihood of devel-oping disabling pain [27, 45, 46]. Conversely, the use ofpassive coping strategies has been associated with an in-creased risk of developing disabling trouble or pain relateddisability [35, 45, 47]. Carroll et al. [41] identified that high

use of passive coping strategies to manage neck or lowerback pain can lead to the inability to work or carry outusual activities of daily living. This is reflected in our study,where passive coping strategies were mainly used, andcould explain why neck pain had a significant impact onleisure activities, activities of daily living and work activity.Hence, it would be useful for clinicians to include educa-tion on active self-management strategies for neck pain.Recently, active self-management strategies have beenshown to be more effective than passive physiotherapytechniques by increasing self-efficacy [46].In our sample, only a small proportion submitted a

workers’ compensation claim for their neck pain support-ing our primary hypothesis. There are several reasons forthe lack of claims submitted for neck problems in theworking population. Firstly, neck pain is common in thegeneral population [48] thus it would be difficult to estab-lish the relative contribution of work. Secondly, the neckpain may be effectively managed with conservative treat-ments and of insufficient severity and duration to warrantthe trouble of submitting a claim.

Study limitationsThe conclusions drawn from this study need to be con-sidered in light of several limitations. The findings ofthis study cannot be generalised beyond female officeworkers. This sample may not be a true representationof office workers as many potential participants mayhave already left the workplace or sought alternateemployment due to neck pain. As only 29 % of the studysample agreed to attend for further assessment of theirneck pain, it is possible that selection bias contributed tothe interview findings of the 51 office workers with onlythose with significant pain attending. Thus the results ofthis research cannot be generalized to all female officeworkers with neck pain. However, the level of neck painand disability of this subsample did not differ from thesample population suggesting their responses were rep-resentative of the larger sample of office workers. Whilea strength of this study was the use of validated tools tomeasure health outcomes, they were based on self-reports rather than objective measures. This has beenlinked with both over and under estimation of preva-lence of health outcomes [49, 50]. It is possible thatmeasurement error was introduced due to commonmethod bias or information bias (e.g. item characteristicsor context), however the magnitude is thought to besmall [51]. Attempts were made to limit method bias.For example, during the interview, the researcher usedopen-ended questions to determine participant’s percep-tion of the source of their neck pain or the strategiesused to manage it. Self-report measures of work absenceand workers’ compensation data were not validated withemployee records due to lack of access to these

Johnston Archives of Physiotherapy (2016) 6:8 Page 6 of 8

databases. However, the use of questionnaires to estab-lish rates of sickness absences is considered a valuablesource of information correlating relatively well withcompany records [52]. Despite the limitations identified,the findings of this research provide insight into the self-reported impact on neck pain in a working populationof female office workers.

ConclusionThis study has important implications for the workplaceand health professionals. Although the severity of neckpain in this sample of female office workers is low, theimpact on work is of concern. This study suggests thelevel of presenteeism in the female office workers maybe significant, with many self-managing by reducingwork and/or leisure activity and utilizing passive copingstrategies. Future research should investigate the benefitsof active coping strategies for office workers with neckproblems to the individual and employer. Physiothera-pists are well placed to assist office workers self-managetheir symptoms to ensure they remain healthy at work.

AbbreviationsNDI, neck disability index; NMQ, nordic musculoskeletal questionnaire

AcknowledgementsI would like to thank Nadia Brandon Black, Emma Ellison, Jane Marshall,Jemma Morland, Matthew O’Rourke and Paula Wilson for assisting with thedata analysis.

FundingNo funding was received to support this study. The Manuscript submitteddoes not contain information about medical device(s)/drug(s).

Availability of data and materialsThe dataset supporting the conclusions of this article will be made availableon request to the author.

Author’s contributionVJ conceived and designed the study, was responsible for data collectionand analysis, and manuscript writing.

Competing interestsThe author declare that she has no competing interests.

Consent for publicationNot applicable.

Ethics approval and consent to participateThis study was approved by the Medical Research Ethics Committee of TheUniversity of Queensland, Australia. Informed consent was received from allparticipants prior to commencement (#2004000225).

Received: 21 January 2016 Accepted: 12 July 2016

References1. Carroll LJ, Hogg-Johnson S, Cote P, et al. Course and prognostic factors for

neck pain in workers. Spine. 2008;33:93–100.2. Nolet PS, Cote P, Cassidy JD, et al. The association between a lifetime

history of a work-related neck injury and future neck pain: a populationbased cohort study. J Manipulative Physiol Ther. 2011;34:348–55.

3. Johnston V, Jimmieson NL, Souvlis T, et al. Interaction of psychosocial riskfactors explain increased neck problems among female office workers. Pain.2007;129:311–20.

4. Côté P, Kristman V, Vidmar M, et al. The prevalence and incidence of workabsenteeism involving neck pain. A cohort of Ontario lost-time claimants.Spine. 2008;33:S192–S8.

5. Korhonen T, Ketola R, Toivonen R, et al. Work related and individualpredictors for incident neck pain among office employees working withvideo display units. Occup Environ Med. 2003;60:475–82.

6. Hush JM, Michaleff Z, Maher CG, et al. Individual, physical and psychologicalrisk factors for neck pain in Australian office workers: a 1-year longitudinalstudy. Eur Spine J. 2009;18:1532–40.

7. Griffiths KL, Mackey MG, Adamson BJ. Behavioral and psychophysiologicalresponses to job demands and association with musculoskeletal symptomsin computer work. J Occup Rehab. 2011;21:482–92.

8. Eltayeb S, Staal JB, Kennes J, et al. Prevalence of complaints of arm, neckand shoulder among computer office workers and psychometric evaluationof a risk factor questionnaire. BMC Musculoskelet Disord. 2007;8:68.

9. Chiu TT, Lam TH, Hedley AJ. A randomized controlled trial on the efficacy ofexercise for patients with chronic neck pain. Spine. 2005;30:E1–7.

10. Luime JJ, Koes B, Miedema HS, et al. High incidence and recurrence ofshoulder and neck pain in nursing home employees was demonstratedduring a 2-year follow-up. J Clin Epidem. 2005;58:407–13.

11. Safe Work Australia. Key Workers’ Compensation Information, Australia. 2010,Commonwealth of Australia: Available from: http://www.safeworkaustralia.gov.au/. Accessed 19 Apr 2015.

12. Access Economics. The high price of pain: the economic impact ofpersistent pain in Australia. Sydney: University of Sydney Pain ManagementResearch Institute; 2007. p. 101.

13. Blyth F, March LM, Brnabic AJM, et al. Chronic pain in Australia: a prevalencestudy. Pain. 2001;89:127–34.

14. Scherer M, Schaefer H, Blozik E, et al. The experience and management ofneck pain in general practice: the patients’ perspective. Euro Spine J.2010;19:963–71.

15. Madeleine P, Vangsgaard S, Hviid Andersen J, et al. Computer work andself-reported variables on anthropometrics, computer usage, work ability,productivity, pain, and physical activity. BMC Musculoskelet Disord.2013;14:226.

16. Vernon H, Mior S. The neck disability index: a study of reliability and validity.J Manipulative Physiol Ther. 1991;14:409–15.

17. Karels CH, Bierma-Zeinstra SM, Verhagen AM, et al. Sickness absence inpatients with arm, neck and shoulder complaints presenting in physicaltherapy practice: 6 months follow-up. Man Ther. 2010;15:476–81.

18. Vernon H. The neck disability index: patient assessment and outcomemonitoring in whiplash. J Musculoskelet Pain. 1996;4:95–104.

19. Kuorinka I, Jonsson B, Kilbom A, et al. Standarised Nordic questionnaires forthe analysis of musculoskeletal symptoms. Appl Ergon. 1987;18:233–7.

20. Ariens G, Bongers PM, Hoogendoorn WE. High physical and psychosocialload at work and sickness absence due to neck pain. Scand J Work EnvironHealth. 2002;28:222–31.

21. Bergqvist U, Wolgast E, Nilsson B, et al. Musculoskeletal disorders amongvisual display terminal workers: individual, ergonomic, and workorganizational factors. Ergonomics. 1995;38:763–76.

22. Blatter BM, Bongers PM. Duration of computer use and mouse use inrelation to musculoskeletal disorders of neck or upper limb. Intern J IndustErgon. 2002;30:295–306.

23. Chiu TTW, Ku WY, Sum WK, et al. A study of the prevalence of and riskfactors for neck pain among university academic staff in Hong Kong.J Occup Rehab. 2002;12:77–91.

24. Dawson AP, Steele EJ, Hodges PW. Development and test–retest reliabilityof an extended version of the Nordic Musculoskeletal Questionnaire (NMQ-E):a screening instrument for musculoskeletal pain. J Pain. 2009;10:517–26.

25. Turner JA, Clancy S. Strategies for coping with chronic low back pain:relationship to pain and disability. Pain. 1986;24:355–64.

26. Brown GK, Nicassio PM. Development of a questionnaire for the assessment ofactive and passive coping strategies in chronic pain patients. Pain. 1987;31:53–64.

27. Blyth FM, March LM, Nicholas MK, et al. Self-management of chronic pain: apopulation-based study. Pain. 2005;113:285–92.

28. Long M, Bogossian F, Johnston V. Functional consequences of work-relatedspinal musculoskeletal symptoms in a cohort of Australian midwives.Women Birth. 2013;26:e50–e8.

29. Hush JM, Lin CC, Michaleff ZA, et al. Prognosis of acute idiopathic neck painis poor: a systematic review and meta-analysis. Arch Phys Med Rehabil.2011;92:824–9.

Johnston Archives of Physiotherapy (2016) 6:8 Page 7 of 8

Clinical StudyEffect of Training Supervision on Effectiveness of StrengthTraining for Reducing Neck/Shoulder Pain and Headache inOffice Workers: Cluster Randomized Controlled Trial

Bibi Gram,1,2 Christoffer Andersen,3 Mette K. Zebis,1

Thomas Bredahl,1 Mogens T. Pedersen,4 Ole S. Mortensen,3,5 Rigmor H. Jensen,6

Lars L. Andersen,3 and Gisela Sjøgaard1

1 Institute of Sports Science and Clinical Biomechanics, University of Southern Denmark, 5230 Odense, Denmark2 Institute of Regional Health Research, University of Southern Denmark, 5230 Odense, Denmark3National Research Centre for the Working Environment, 2100 Copenhagen, Denmark4Department of Nutrition, Exercise and Sports, University of Copenhagen, 2200 Copenhagen, Denmark5 Department of Occupational Medicine Køge Hospital, 4600 Køge, Denmark6Danish Headache Center, Department of Neurology, Glosrtup Hospital, University of Copenhagen, 2600 Copenhagen, Denmark

Correspondence should be addressed to Bibi Gram; bgram@health.sdu.dk

Received 21 November 2013; Accepted 5 January 2014; Published 19 February 2014

Academic Editor: David G. Behm

Copyright © 2014 Bibi Gram et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Objective. To investigate the effect of workplace neck/shoulder strength training with and without regular supervision onneck/shoulder pain and headache among office workers.Method. A 20-week cluster randomized controlled trial among 351 officeworkers was randomized into three groups: two training groups with the same total amount of planned exercises three times perweek (1) with supervision (3WS) throughout the intervention period, (2) with minimal supervision (3MS) only initially, and (3) areference group (REF). Main outcome is self-reported pain intensity in neck and shoulder (scale 0–9) and headache (scale 0–10).Results. Intention-to-treat analyses showed a significant decrease in neck pain intensity the last 7 days in 3MS compared with REF:−0.5 ± 0.2 (𝑃 < 0.02) and a tendency for 3WS versus REF: −0.4 ± 0.2 (𝑃 < 0.07). Intensity of headache the last month decreasedin both training groups: 3WS versus REF: −1.1 ± 0.2 (𝑃 < 0.001) and 3MS versus REF: −1.1 ± 0.2 (𝑃 < 0.001). Additionally, days ofheadache decreased 1.0 ± 0.5 in 3WS and 1.3 ± 0.5 in 3MS versus REF.There were no differences between the two training groups forany of the variables. Conclusion. Neck/shoulder training at the workplace reduced neck pain and headache among office workersindependently of the extent of supervision. This finding has important practical implications for future workplace interventions.

1. Introduction

Work-related symptoms in neck and shoulder are commonamong occupational computer users and other sedentaryoccupations [1, 2] although the evidence of causality isinconclusive [3, 4]. Along with pain in the neck and shoul-ders, office work is associated with frequent headache andcooccurrence of headache is estimated to be fourfold inworkers with musculoskeletal symptoms [5]. Neck pain andheadaches are closely related too, although the reportedheadaches only are rarely diagnosed further into tension-typeheadache or migraine [6].

Additionally, studies have shown a 31% decrease inquality of life among workers with neck/shoulder symptoms[7] and self-reported health is inversely correlated withneck/shoulder pain [8] and headache [9, 10]. Thus, thereis a need for initiatives to reduce the pain problem amongoffice workers who are exposed to repetitive low intensitymusculoskeletal load in the neck and shoulder region.

In the past decade, exercise interventions at theworkplacehave becomemore common and studies have shown positiveeffect of physical exercise at work in managing muscu-loskeletal pain [11–13]. Especially neck pain seems to respondpositively to specific strength training, while evidence of

Hindawi Publishing CorporationBioMed Research InternationalVolume 2014, Article ID 693013, 9 pageshttp://dx.doi.org/10.1155/2014/693013

2 BioMed Research International

strength training impact on shoulder pain is sparse [11–17]. Furthermore, exercise interventions at the workplacehave shown significant reduction headache intensity [18] andheadache frequency [19].

Exercise programs with supervision are most likely to bebeneficial in reducing pain amongpatientswith lowback pain[20]. Likewise weekly supervision in maintenance traininghad significantly better effect than unsupervised training [21],and the effect was closely related to adherence to the program.In a systematic review Coury et al. [12] concluded thatthere was indication for strong evidence of ineffectivenessfor unsupervised training. Exercise programs supervised byinstructors enable the participants regularly to tailor theprogram to the instructions which may have a physiologicaland motivational value that the unsupervised participants donot benefit from. On the other hand some participants doprefer—after being introduced to the program—to exercisewhen it fits into their daily routines.

When conducting workplace interventions one may pre-sume that the use of supervision is important to maximizetraining effects and compliance. Studies that make use ofsupervision report positive impact on neck/shoulder pain[11, 12, 14–17]; however, the specific effect of the supervisionon pain relief is not well established [22]. A study conductingexercise intervention at the workplace showed that only witha single introductory session of supervision a significantreduction of neck/shoulder pain intensity in office workerswas attained [23]. However, the study was uncontrolled anddid not compare supervised exercises with unsupervisedexercises. Supervised training can be expensive and notalways an available resource at workplaces. Therefore it ispertinent to reveal the minimum amount of supervisionneeded for safe and effective exercise training for painreduction when implementing exercise at the workplaces.

This study is part of a larger intervention program:Workplace adjusted intelligent physical exercise training forreducingmusculoskeletal pain in shoulder/neck (VIMS) [24]and investigates the effect of instructor supervised versusminimally supervised exercise training on neck/shoulderpain. The concept of “intelligent physical exercise training”is to balance the individual physiological capacity relativelyto occupational exposure, tailor the exercise to individualcapacities and disorders, allow for flexibility and personalpreferences of the participant, and to be as cost effective forthe company as possible.

The aim of this paper is to investigate the relevance oftraining supervision for safe and effective training, in order tominimize expenses for workplace physical exercise training.

The hypotheses are as follows.(1) Both training interventions have positive effects on

neck/shoulder pain and headache compared withreference.

(2) Regular supervision of the exercise trainingwill resultin a larger effect compared with initial instructiononly.

(3) Regular supervision of the exercisewill have a positiveinfluence on compliance compared with minimalsupervision.

2. Methods

2.1. Study Design. The study was a cluster randomizedcontrolled trial and the intervention period was 20 weeks.The participants were office workers of a national publicadministrative authority recruited from 12 geographicallydifferent units that were located in major cities throughoutDenmark.

Randomization was performed on a cluster level tominimize contamination between the participants; for detailssee Andersen et al. [24]. In short, the clusters were naturallyoccurring groups of employees working together on a dailybasis, being located at the same floor, same office or the like.To ensure the comparability of the training groups and thereference group, the geographical sites were categorized into13 strata [24]. Adjustments weremade in respect to the clusterallocation due to 26 participants being relocated to otherwork sites between the time of randomization and the start-up of the different interventions (approx. 3 weeks) in orderto have these participants follow the intervention for thecluster of their new colleagues. No subsequent reallocationswere performed. The participants were randomized into fivegroups: one reference group (REF) without exercise train-ing and four training groups performing specific strengthtraining. The present study addressed only two of thesetraining groups: one was scheduled for training 3 × 20minutes per week with supervising half of the sessionsthroughout the training period (3WS) and the other groupwas likewise scheduled for training 3 times per week butonly received minimal supervision (3MS), which was giveninitially in terms of instructions for 2 sessions to learn theexercises correctly. The total number of planned training was60 sessions (3 × 20 weeks) of which 3WS had instructorssupervising the training 10 hours (30 sessions × 20min),while 3MS had instructor supervision for 40–60min (due toabsence by some participants instructors would usually haveto come for 2-3 training sessions for this group). A previouspaper addresses the other training groups [14].

Written informed consent was obtained from all partici-pants before they entered the study.

The study protocol was approved by the local ethics com-mittee (H-C-2008-103) and registered in ClinicalTrials.gov(no. NCT01027390).

2.2. Participants. About half of the participants wererecruited from the Capital Region and the other half fromother parts of Denmark.Thereby the population is nationallyrepresentative and strengths external validity. The eligibilitycriterion was employees performing office work for at leasthalf of their working hours. The exclusion criteria were(i) hypertension (systolic BP > 160, diastolic BP > 100) orcardiovascular diseases, (ii) symptomatic herniated disc orsevere disorders of the cervical spine, (iii) postoperativeconditions in the neck or shoulder region, (iv) history ofsevere trauma, (v) pregnancy, (vi) or serious disease.

2.3. Intervention. The two training groups had the same totalamount of exercises and repetitions planned three timesper week. The training groups performed specific strength

BioMed Research International 3

training with 4 different dumbbell exercises for the neck andshoulder muscles and one for the wrist as described in detailpreviously [24]: front raise, lateral raise, reverse flies, andshrugs.

The participants performed warm-up exercises in thebeginning of each training session (10 repetition of eachexercise with 50% of 1 repetition maximum (RM)). At thebeginning and halfway through the intervention period, theparticipants were tested for optimizing the training intensityand the loads were progressively increased according to theprinciple of periodization and progressive overload [25]. Theintensity of the program increased gradually from 20 RM atthe beginning of the intervention period to 8 RM furtheralong in the process.

2.4. Outcome Measures. Structured e-mail based question-naires were applied before and after the intervention. Theprimary outcome was musculoskeletal pain symptoms inneck/shoulder and secondary outcome was headache char-acteristics [24].

The standardized nordic questionnaire [26] was appliedat baseline before the randomization and repeated after theintervention. The questions were “How many days have youhad trouble in body part during the last three months?” (0days; 1–7 days; 8–30 days; >30 days; everyday) for symptomduration, and “On average, how intensewas your pain in bodypart during the last three months on a scale ranging from 0to 9?” where 0 is no pain and 9 is worst imaginable pain forsymptom intensity.The same questionwas also asked for painduring the previous seven days.

Secondary outcome variables were headache characteris-tics (frequency and pain intensity).

Question about duration of headache was “How manydays have you had a headache during the previous month?”The following response options were 0, 1–3, 4–7, 8–14, and>14 days. For subsequent analyses 1–3 days were recorded to 2days, 4–7 days to 5.5 days, 8–14 to 11 days, and >14 to 20 days.Intensity of headache was also inquired about “On average,how bad were your headaches when you experienced themduring the previous month?” where 0 is no pain and 10 isworst imaginable pain [19].

Compliancewas based on follow-up questionnaire replieson training frequency (completers).

The response categories were (1) “regular exercise train-ing 40–60min/week,” (2) “regular exercise training 20–40min/week,” (3) “not regular but at least 80min/month,”and (4) “not regular but at least 40–60min/month.” Regulartraining was collapsed into “regular exercise training 20–60min per week.”

2.5. Statistical Analyses. The statistical analyses were basedon an intention-to-treat approach (ITT) via Stata SE12(StataCorp LP, College Station, Texas). Missing values inpostmeasurements were substituted with the last observationcarried forward [27, 28]. Differences between groups inneck/shoulder pain and headache frequency and intensitywere tested using analyses of covariance (ANCOVA) with thelevel at baseline and sex as a covariate.

In addition to the ITT analyses we performed analysesusing ANCOVA only on completers, defined as those whohad answered the questionnaire before and after the interven-tion and the rest were defined as noncompleters. Relationshipbetween neck pain and headache was estimated using Spear-man rank correlation.We also defined a subgroup, neck-paincases, as those who at baseline reported pain intensity in theneck during the last 3 months of 3 or more (scale 0–9) [29].This sub group analysis was performed on ITT data as well ascompleters only.

Results were considered statistically significant if the 2-tailed 𝑃 value was ≤ 0.05.

3. Results

3.1. Baseline. Flow of participants through the trial is pre-sented in Figure 1.Thepresent study included 351 participantscluster randomized in 3 groups: 3WS (𝑛 = 126), 3MS (𝑛 =124), and REF (𝑛 = 101). At baseline there were no significantdifferences between the groups (Table 1).

Baseline data on demographics and pain variables for theentire study group is presented in Table 1 and for completersand neck-pain cases in Table 2. Analyses on completers(𝑛 = 220) versus noncompleters (𝑛 = 131) showed nosignificant differences between these groups at baseline onpain variables. However, noncompleters were significantlyyounger than completers; mean age 44 ± 1.4 versus 47 ± 0.7(𝑃 < 0.05).

Four participants reported nonpermanent injuries duringthe intervention period: back pain (𝑛 = 2), shoulder/wristpain (𝑛 = 1), and pain in the knee (𝑛 = 1).

Mean values on neck pain were∼3 on a scale 0–9 (Table 1)and neck-pain cases accounted for 56% (pain intensity 3 ormore the last 3 months). Of note is further that relativelymany participants reported pain intensity corresponding to4 or more: 41% (the last 3 months) and 32% (the last 7 days).

Regarding headache, approximately 15% of the partici-pants reported having headache above 7 days the previousmonth with an average intensity at 7.0± 1.9. Average numberof days in which participants used medication because ofheadache was for WS: 2.3 ± 1.0, MS: 2.2 ± 1.0, and REF:2.4 ± 0.9.

Among neck-pain cases (𝑛 = 197), 90% also reportedheadache; mean headache frequency at baseline is 6.5 ± 5.9days of previous month.

A statistical significant relationship between intensity ofneck pain the last three months and intensity of headacheduring the previous month was identified (Spearman corre-lation, 𝑟: 0.39 (𝑃 < 0.001)).

3.2. Intervention

3.2.1. Primary OutcomeNeck and Shoulder Pain (ITT analyses). Intention-to-treatanalyses showed a significant decrease in neck pain intensitythe last 7 days in 3MS compared with REF: −0.5 ± 0.2 (𝑃 <0.02) and a tendency for 3WS versus REF: −0.4 ± 0.2 (𝑃 <0.07) (Table 3). Analyses on neck pain the last 3 months andshoulder pain did not show any significant changes (Table 3).

4 BioMed Research International

Allocated to exercise 9x

Allocated to exercise 1x

Ana

lysis

Inte

rven

tion

Follo

w-u

pA

lloca

tion

Analysed

Lost to follow-up:

Declined participation or did not meet

Screening questionnaires sent

Allocated to exercise 3x weekly with minimal

Allocated to exercise 3x weekly with supervision

Allocated to reference

Lost to follow-up: Lost to follow-up:

Analysed Analysed

Enro

llmen

t

(n = 2114)

Replied (n = 990)

Did not reply (n = 1124)

eligibility criteria (n = 417)

Cluster randomized (n = 573)

weekly (n = 116) weekly (n = 106)

Present study (n = 351)

3WS (n = 126) supervision 3MS (n = 124)

group REF (n = 101)

Completers: (n = 75) Completers: (n = 64) Completers: (n = 81)

(n = 51)Noncompleters (n = 60) (n = 20)

ITT analyses: (n = 126) ITT analyses: (n = 124) ITT analyses: (n = 101)

Non completers Non completers

Figure 1: Flowchart of the trial.

ITT-analyses for the group defined as neck-pain cases(𝑛 = 197) showed a significant decrease in neck pain the last 7days, 3WS versus REF: −0.7 ± 0.4 (𝑃 < 0.05) but not for 3MSversus REF. Similarly, analyses on neck pain over the last 3months showed significant decrease in neck pain only in 3WSversus REF: −0.7 ± 0.4 (𝑃 = 0.05). There were no significantchanges in shoulder painwithin the group of neck-pain cases.

Neck and Shoulder Pain (completers). In both training groupsthere were significant decreases in the intensity of neck pain(last 3 months): 3WS versus REF: −1.0 ± 0.3 (𝑃 < 0.001)and 3MS versus REF: −0.9 ± 0.3 (𝑃 < 0.001) (Table 4).Additionally, both groups showed a significant decrease in theintensity of neck pain (7 days): 3WS versus REF: −1.0 ± 0.3(𝑃 < 0.001) and 3MS versus REF: −1.1 ± 0.3 (𝑃 < 0.001).The same applied to the intensity of shoulder pain the last 3

months in both training groups: 3WS versus REF: −0.7 ± 0.3(𝑃 < 0.01) and 3MS versus REF: −0.6 ± 0.3 (𝑃 < 0.05). Therewere no significant changes in the intensity of shoulder painthe last 7 days (Table 4).

At baseline, completers defined as neck-pain cases (𝑛 =124) had intensity of neck pain (last 3months) correspondingto 5.1±1.6 (3WS), 5.2± 2.0 (3MS), and 4.1±2.1 (REF). Furthercensored analyses on this sub-group showed significantchanges in the intensity of neck pain the last 3 months forboth training groups compared to REF: 3WS: −1.9 ± 0.4 (𝑃 <0.001) and 3MS: −1.1 ± 0.5 (𝑃 < 0.03) as well as in intensityof neck pain the past 7 days: 3WS: −1.7 ± 0.5 (𝑃 < 0.001) and3MS: −1.4 ± 0.5 (𝑃 < 0.004) (Table 5). Furthermore, amongcompleters defined as neck-pain cases there was a significantdifference between 3WS and 3MS in intensity of neck pain(last 3 months) with better improvement in 3WS; 0.8 ± 0.4

BioMed Research International 5

Table 1: Baseline demographic and clinical characteristics of trial groups.

Characteristics Exercise group 3WS(𝑛 = 126)

Exercise group 3MS(𝑛 = 124)

Reference group(𝑛 = 101)

Min–Max(𝑛 = 351) 𝑃

Sex, (m/f) 39/87 52/72 42/59 NSAge, (y) 46 ± 10 45 ± 11 46 ± 10 22–66 NSBMI 24.7 ± 4.3 25.6 ± 3.8 26.0 ± 4.5 14–45 NSPain (on scale 0–9)

Neck pain (3 months) 3.1 ± 2.4 3.2 ± 2.4 3.2 ± 2.3 0–9 NSNeck pain (7 days) 2.6 ± 2.5 2.4 ± 2.4 2.5 ± 2.5 0–9 NSRight shoulder pain (3 months) 2.3 ± 2.4 2.0 ± 2.4 2.0 ± 2.4 0–8 NSRight shoulder pain (7 days) 1.8 ± 2.3 1.6 ± 2.2 1.6 ± 2.3 0–8 NSLeft shoulder pain (3 months) 1.8 ± 2.3 1.6 ± 2.3 1.5 ± 1.9 0–9 NSLeft shoulder pain (7 days) 1.4 ± 2.0 1.3 ± 2.0 1.3 ± 1.9 0–8 NS

Headache (pain scale 0–10) 3.4 ± 2.5 3.6 ± 2.8 3.6 ± 3.0 1–10 (𝑛 = 211) NSHeadache (days of last month) 3.8 ± 4.3 4.1 ± 4.7 4.2 ± 4.9 0–>14 days NSValues are mean (SD) and numbers. 𝑃 values for the 1-way analysis of variances.

Table 2: Baseline neck pain, right shoulder pain, and headache in completers and neck-pain cases, respectively.

CharacteristicsExercise group 3WS Exercise group 3MS Reference group

𝑃Completers(𝑛 = 75)

Neck-pain cases(𝑛 = 69)

Completers(𝑛 = 64)

Neck-pain cases(𝑛 = 70)

Completers(𝑛 = 81)

Neck-pain cases(𝑛 = 58)

Completers (neck-pain cases)Neck pain (3 months) 3.0 ± 2.5 5.0 ± 1.6 3.5 ± 2.5 4.9 ± 1.7 3.4 ± 2.3 4.8 ± 1.6 NSNeck pain (7 days) 2.4 ± 2.5 4.2 ± 2.1 2.5 ± 2.4 3.8 ± 2.3 2.7 ± 2.5 3.9 ± 2.2 NSRight shoulder pain (3months) 1.8 ± 2.2 3.3 ± 2.4 2.1 ± 2.4 2.7 ± 2.6 1.9 ± 2.4 3.2 ± 2.5 NSRight shoulder pain (7 days) 1.3 ± 1.9 2.8 ± 2.5 1.5 ± 2.2 2.2 ± 2.4 1.6 ± 2.3 2.7 ± 2.6 NSLeft shoulder pain (3months) 1.4 ± 2.0 2.8 ± 2.5 1.8 ± 2.4 2.2 ± 2.3 1.6 ± 2.0 2.2 ± 2.2 NSLeft shoulder pain (7 days) 1.0 ± 1.7 2.2 ± 2.3 1.4 ± 2.1 2.0 ± 2.3 1.5 ± 2.1 2.0 ± 2.2 NS

Headache (pain scale 0–10) 4.9 ± 1.8 5.5 ± 2.0 5.5 ± 2.8 6.1 ± 2.4 5.7 ± 2.5 6.4 ± 2.5 NSHeadache (days last month) 2.2 ± 1.0 2.6 ± 1.1 2.5 ± 2.8 2.8 ± 1.2 4.4 ± 1.1 2.8 ± 1.2 NSValues are mean (SD) and numbers. 𝑃 values for the 1-way analysis of variances. Completers had answered the questionnaire before and after the intervention.Neck-pain cases reported pain intensity in neck last 3 months of 3 or more at baseline (scale ranging from 0 to 9).

(𝑃 = 0.05). Concerning shoulder pain within this sub-groupthere was a significant decrease in intensity of shoulder painthe last 3months in the 3WS group compared to the referencegroup: −1.2 ± 0.4 (𝑃 < 0.003) (right shoulder) and −0.8 ± 0.4(𝑃 < 0.03) (left shoulder) but not in the 3MS group (Table 5).There were no significant changes in intensity of shoulderpain the last 7 days, neither in 3WS nor in 3MS.

3.2.2. Secondary Outcome

Headache (ITT analyses). Results of secondary outcome vari-ables are shown in Table 3. There was a significant reductionin days with headache in both training groups comparedto the reference group. Furthermore, there was a significantdecrease in pain intensity in both training groups comparedto the reference group. In the group of neck-pain cases therewas a significant decrease in headache intensity in 3WS versusREF: −0.9 ± 0.4 (𝑃 < 0.02) and 3MS versus REF: −0.9 ± 0.3(𝑃 < 0.01).

After the intervention, there were no changes in the useof medication because of headache.

Headache (completers). Days with headache last monthdecreased significantly in both groups compared to thereference group: 3WS versus REF: −0.6 ± 0 (𝑃 < 0.001) and3MS versus REF:−0.6± 0.1 (𝑃 < 0.001), and the pain intensitydecreased significantly in both training groups compared tothe reference group: 3WS versus REF: −1.6 ± 0.4 (𝑃 < 0.00) 1and 3MS versus REF: −1.5 ± 0.3 (𝑃 < 0.001) (Table 4).

Compliance. Among completers 60% in the 3WS group and47% in the 3MS group reported that they were exercisingon a regular basis 20–60min a week in the interventionperiod. There was no significant difference in compliancebetween the groups (𝑃 < 0.14) with an overall value of 54%participating at a regular basis.

Regarding primary and secondary outcomes, the analyseson the entire group showed no significant difference betweenthe two training groups.

4. Discussion

The major findings of this study were significant reductionsof similar magnitude in neck/shoulder pain and in headache

6 BioMed Research International

Table 3: Summary results for each study group after 20 weeks of intervention (ITT data).

Characteristics3WS

Post-pre (SD)(𝑛 = 126)

3MSPost-pre (SD)(𝑛 = 124)

Ref. groupPost-pre (SD)(𝑛 = 101)

Difference3WS versus REF(95% CI) (SE)

𝑃Difference

3MS versus REF(95% CI) (SE)

𝑃

Pain (a scale ranging from 0 to 9)

Neck pain (3 months) −0.9 ± 2.1 −0.9 ± 1.5 −0.6 ± 2.0 −0.4 ± 0.2(−0.8 to 0.1) 0.11 −0.3 ± 0.2

(−0.7 to 0.1) 0.15

Neck pain (7 days) −0.7 ± 2.1 −0.6 ± 1.5 −0.2 ± 2.0 −0.4 ± 0.2(−0.9 to 0.03) 0.07 −0.5 ± 0.2

(−0.9 to −0.1) 0.02∗

Right shoulder pain (3months) −0.5 ± 1.9 −0.5 ± 2.0 −0.2 ± 1.9 −0.1 ± 0.2(−0.6 to 0.3) 0.50 −0.3 ± 0.2

(−0.8 to 0.1) 0.13

Right shoulder pain (7 days) −0.3 ± 1.8 −0.4 ± 1.8 −0.2 ± 1.9 0.0 ± 0.2(−0.4 to 0.4) 0.97 −0.2 ± 0.2

(−0.6 to 0.3) 0.43

Left shoulder pain (3months) −0.4 ± 1.5 −0.5 ± 1.5 −0.3 ± 1.8 0.0 ± 0.2(−0.4 to 0.4) 0.99 −0.2 ± 0.2

(−0.5 to 0.2) 0.46

Left shoulder pain (7 days) −0.8 ± 1.6 −0.3 ± 1.4 −0.4 ± 1.8 0.3 ± 0.2(−0.1 to 0.7) 0.19 0.1 ± 0.2

(−0.3 to 0.4) 0.77

Headache (pain scale, 0–10) −0.4 ± 1.8 −0.4 ± 1.4 0.7 ± 2.2 −1.1 ± 0.2(−1.6 to −0.6) 0.00∗ −1.1 ± 0.2

(−1.5 to −0.6) 0.00∗

Headache (days of last month) −0.4 ± 3.7 −0.7 ± 2.6 0.6 ± 4.4 −1.1 ± 0.5(−2.1 to −0.1) 0.03∗ −1.3 ± 0.5

(−2.2 to −0.5) 0.00∗

Changes in post-pre values are absolute and not adjusted. Differences are estimated as the difference between means, with 95% confidence intervals, based onthe 1-factor analyses of covariance with the level at baseline and sex as a covariate. ∗Significant change.

Table 4: Summary results for each study group after 20 weeks of intervention (completers).

Characteristics 3WS: post-pre(SD) (𝑛)

3MS: post-pre(SD) (𝑛)

Ref.: post-pre(SD) (𝑛)

Difference 3WSversus REF

(95% CI) (SE)𝑃

Difference 3MSversus REF

(95% CI) (SE)𝑃

Neck pain (3 months) −1.5 ± 2.5 (75) −1.8 ± 1.7 (64) −0.7 ± 2.2 (81) −1.0 ± 0.3(−1.5 to −0.4) 0.00∗ −0.9 ± 0.3

(−1.5 to −0.4) 0.00∗

Neck pain (7 days) −1.1 ± 2.6 (75) −1.2 ± 1.9 (64) −0.2 ± 2.2 (81) −1.0 ± 0.3(−1.6 to −0.4) 0.00∗ −1.1 ± 0.3

(−1.6 to −0.5) 0.00∗

Right shoulder pain (3months) −0.8 ± 2.4 (75) −1.0 ± 2.7 (64) −0.2 ± 2.1 (81) −0.7 ± 0.3(−1.2 to −0.2) 0.01∗ −0.6 ± 0.3

(−1.2 to 0.0) 0.04∗

Right shoulder pain (7 days) −0.5 ± 2.3 (75) −0.7 ± 2.5 (64) −0.2 ± 2.1 (81) −0.5 ± 0.3(−1.0 to 0.0) 0.07 −0.5 ± 0.3

(−1.1 to 0.1) 0.09

Left shoulder pain (3months) −0.7 ± 1.9 (75) −0.9 ± 2.0 (64) −0.4 ± 2.0 (81) −0.0 ± 0.2(−0.5 to 0.4) 0.89 0.4 ± 0.2

(−0.1 to 0.9) 0.10

Left shoulder pain (7 days) −0.3 ± 2.1 (75) −0.6 ± 1.9 (64) −0.5 ± 2.0 (81) −0.2 ± 0.3(−0.7 to 0.4) 0.59 0.0 ± 0.3

(−0.5 to 0.6) 0.90

Headache (pain scale 0–10) −2.0 ± 2.2 (58) −2.1 ± 1.6 (55) −0.7 ± 2.2 (64) −1.6 ± 0.4(−2.4 to −0.8) 0.00∗ −1.5 ± 0.3

(−2.2 to −0.9) 0.00∗

Headache (days of last month) −0.6 ± 4.9 (75) −1.4 ± 3.6 (64) −0.7 ± 4.9 (81) −1.9 ± 0.7(−3.3 to −0.5) 0.01∗ −2.2 ± 0.6

(−3.4 to −1.0) 0.00∗

Changes in post-pre values are absolute and not adjusted. Values are means with standard deviation presented for each group (completers). Differences areestimated as the difference between means (SE), with 95% confidence intervals, based on the 1-factor analyses of covariance with the level at baseline and sexas a covariate. ∗Significant change.

for 3WS and 3MS compared to REF after a 20-week trainingperiod. Furthermore, the training program was consideredto be safe since only 4 out of 351 participants (1%) reportedtransitory adverse events of short duration.

Our study hypothesis was that training with supervisionwould be more effective on neck/shoulder pain as well asheadache reduction and that the supervision would causebetter compliance than training without supervision. The

study could not confirm this hypothesis since the sizes ofthese effects of pain reduction among the intervention groupswere of the same order ofmagnitude. By and large, the relativedifference between baseline and postmeasurements was 10–20% for neck pain and ∼30% for headache intensity for bothtraining groups in the ITT analyses when compared to REF.Similarly, the relative difference was ∼30–40% for neck painand ∼30% for headache intensity among completers. In the

BioMed Research International 7

Table 5: Summary results for each neck-pain cases group after 20 weeks of intervention (completers only).

Characteristics 3WS: post-pre(SD) (𝑛 = 39)

3MS: post-pre(SD) (𝑛 = 37)

Ref.: post-pre(SD) (𝑛 = 48)

Difference 3WS versusREF (95% CI) (SE) 𝑃

Difference 3MSversus REF (95% CI)

(SE)𝑃

Neck pain (3 months) −3.1 ± 2.2 −2.4 ± 1.9 −1.2 ± 2.6 −1.9 ± 0.4 0.000∗ −1.1 ± 0.5 0.022∗

Neck pain (7 days) −2.4 ± 2.7 −1.8 ± 2.1 −0.6 ± 2.5 −1.7 ± 0.5 0.000∗ −1.4 ± 0.5 0.003∗

Right shoulder pain (3months) −1.7 ± 2.5 −1.3 ± 3.2 −0.7 ± 2.4 −1.1 ± 0.4 0.002∗ −0.8 ± 0.5 0.096Right shoulder pain (7 days) −1.2 ± 2.8 −0.9 ± 2.9 −0.7 ± 2.5 −0.8 ± 0.4 0.062 −0.7 ± 0.5 0.170Left shoulder pain (3months) −1.4 ± 2.2 −1.4 ± 2.2 −0.6 ± 2.2 −0.8 ± 0.4 0.024∗ −0.6 ± 0.4 0.156Left shoulder pain (7 days) −0.8 ± 2.6 −1.0 ± 2.3 −0.9 ± 2.4 −0.2 ± 0.4 0.634 −0.2 ± 0.4 0.727Headache (pain scale 0–10) −0.5 ± 2.8 −0.9 ± 1.5 0.5 ± 2.5 −1.4 ± 0.5 0.008∗ −1.3 ± 0.4 0.001∗

Headache (days of last month) −0.9 ± 4.7 −1.4 ± 3.1 0.9 ± 4.6 −2.3 ± 0.9 0.010∗ −2.1 ± 0.8 0.006∗

Changes in post-pre values are absolute and not adjusted. Values are means with standard deviation presented for each group (neck-pain and completers).Differences are estimated as the difference between means (SE), with 95% confidence intervals, based on the 1-factor analyses of covariance with the level atbaseline and sex as a covariate. ∗Significant change.

group of pain cases the relative difference was 16–8% forheadache intensity.

The results showed relatively high intensity of neck painsince approx. 40% of the participants reported neck paincorresponding to 3 or above 3 on a scale 1–10. Furthermore,relatively high frequency of days with headache was reportedat baseline.

Reduction of neck pain associated with work site inter-vention program in this study is consistent with resultsfrom previous studies [11–15] and confirms also the positiveeffect of exercise training to reduce headache [17, 18, 30].In this study, the headache could not be further classifiedor quantified into tension-type headache or migraine, as itrequired a detailed diagnostic interview and a neurologicalexamination plus a prospective diagnostic diary which isquite time consuming and complicated to be applied in largescale working place studies like the present.

It is highly relevant to investigate the importance oftraining supervisionwhen conducting exercise training at theworkplace because recruitment of training instructors is elab-orative and expensive andmay result in restriction of trainingtimes. Training supervision as used in present study requiresan annual basis salary of approx. 78 hours for instructors.Thismay hinder implementation of exercise training programsduring working hours. Based on the scientific literature theeffect of supervision shows conflicting findings. Zavanela etal. [31] showeddecreasedneck/shoulder pain andheadache ofbus drivers after 24weeks of supervised training intervention.On the other hand, Mongini et al. [32] reported decreasein neck/shoulder pain and headache in a large randomizedcontrolled trial using an unsupervised program. However,none of these two studies were designed to measure the effectof supervision compared to no supervision. Interestingly, ITTanalyses in the present study for the group defined as neck-pain cases showed significant decrease in neck pain the last7 days and the last 3 months in the group with trainingsupervision (3WS) but not for the minimally supervisedgroup (3MS).Thus, we cannot exclude that the pain conditionfor participants may influence the need for supervision suchthat patients or those in pain to a larger extend benefit fromproper instruction than pain-free participants. To evaluate

the effect of the supervision, compliance is crucial. Fortypercent of the 3WS group, 48% of 3MS group, and 20% of theREF did not answer the questionnaire after the intervention.This is limiting factor for the study and might induce type 2error.

Only 60% of completers in the 3WS group and 47%in the 3MS group reported that they were exercising ona regular basis in the intervention period, which may bea limitation when evaluating the effect of the supervision.The reason for these levels of participation may be thatsupervision is motivating but also causing time constrains,while training without supervision gives more flexibility butmay lack motivating actions.

Since noncompleters did not return the questionnaire, weare not able to conclude upon reasons behind not respondingon the second questionnaire or upon questions regardingsupervision. Noncompleters were in this study defined asthose who did not reply the second questionnaire, but asa term, noncompleter is not unambiguous. Although thefinal questionnaire is not completed, participants could, inprinciple, have been training a large part of the interventionperiod. Ongoing evaluation of the included population anda detailed interview could possibly elucidate their trainingprocess and outcome.

Dropout and poor compliance are always a challengein intervention studies [11, 15, 33] and balanced strategiesto maintain long-term motivation in studies with exercisesinterventions are pertinent [17, 18].

5. Conclusion

One hour of physical exercise training per week for 20weeks at the workplace was highly effective to reduce neckpain and headache and the effect was overall independentof the level of supervision. A well-performed introductionand supervision of the exercises only in the beginningof the relatively simple training program was sufficient toachieve pain-relieving effects. Greater flexibility in planningand conducting exercise training at the workplace due tono constraint with supervised training schedules may have

“This course was developed and edited from the open access article: Consequences and management of neck

pain by female office workers: results of a survey and clinical assessment - Johnston Archives of Physiotherapy

(2016) 6:8. (DOI 10.1186/s40945-016-0023-3), used under the Creative Commons Attribution License.”

“This course was developed and edited from the open access article: Effect of Training Supervision on Effectiveness of

Strength Training for Reducing Neck/Shoulder Pain and Headache in Office Workers: Cluster Randomized Controlled

Trial - Hindawi Publishing Corporation, BioMed Research International: Volume 2014, Article ID 693013, 9 pages,

(http://dx.doi.org/10.1155/2014/693013), used under the Creative Commons Attribution License.”