Impact of Workplace Exposure and Stress on Neck Pain and Disabilities in Women—A Longitudinal Follow-up After a Rehabilitation Intervention

Impact of Workplace Exposure and Stress on Neck Pain and Disabilities in Women—A Longitudinal... Abstract Introduction The aim was to evaluate if pain, disability, and work productivity are influenced by physical and psychosocial work exposures as well as by stress, up to 1 year after a randomized controlled trial treatment intervention, and to determine whether any such association differed between treatment and control groups. Methods Ninety-seven working women suffering non-specific neck pain (n = 67 treatment group, n = 30 control group) were followed from end of treatment intervention and at 9- and 15-month follow-ups, respectively. Physical and psychosocial exposures, as well as perceived stress, were assessed after the treatment intervention. Pain, neck disability, and work productivity were assessed at baseline, after intervention 3 months later and at 9- and 15-month follow-ups. Longitudinal assessment was conducted using the exposure level at 3 months as predictor of pain, disability, and work productivity at 3, 9, and 15 months, respectively. Mixed models were used to estimate longitudinal associations, accounting for within-individual correlation of repeated outcome measures by incorporation of a random intercept. Age and duration of neck pain were adjusted for in all models. To evaluate group differences, interactions between exposures and treatment groups were estimated. Results High perceived stress was associated with more neck pain, more neck disability, and decreased work productivity in both cross-sectional and longitudinal analyses. High ‘control of decision’ was associated with less neck pain, less neck disability, and higher work productivity in cross-sectional analyses but only to less disability and higher productivity in longitudinal analyses. Shoulder/arm load was the only physical exposure variable that was significantly associated with work productivity in the univariate analyses. Only small differences were observed between treatment and control groups. Conclusion High perceived stress and low ‘control of decision’ were associated with more neck pain, increased neck disability, and decreased work productivity. Treatment interventions for individuals with neck pain should take into account psychosocial workplace exposures and stress to improve intermediate and long-term results. non-specific neck pain, physiotherapy, shoulder pain, work productivity Introduction Neck pain is a growing global health problem that is ranked fourth in the world in terms of disability (Hoy et al., 2014). It is a common condition with studies reporting a 1-year prevalence from 27% to almost 50% (Cote et al., 2009). More than 50% of persons who experienced an episode of neck pain reported neck pain 1 year later, thus indicating a long-lasting problem (Carroll et al., 2008a). Among workers, 11–14% have limitations in their activities due to neck pain (Cote et al., 2009). In addition to the individual suffering, reduced productivity, increased use of healthcare, and great financial burden for the society may occur (Borghouts et al., 1999; Childs et al., 2008). Most studies show that women report more neck pain than men both in the general population (Hogg-Johnson et al., 2008; Hoy et al., 2014) and in the working population (Larsson et al., 2007; Wahlstedt et al., 2010; Paksaichol et al., 2012; Hoy et al., 2014). Neck pain is assumed to be of multifactorial origin with interacting individual, physical, and psychosocial risk factors. Repetitive work (Malchaire et al., 2001; Palmer and Smedley, 2007; Cote et al., 2009), upper arm elevation (Nordander et al., 2016), working with the neck flexed more than 20º during prolonged time periods (Palmer and Smedley, 2007), and precision-demanding work (Cote et al., 2009) have all been recognized as work-related physical risk factors for neck pain. Further, work-related psychosocial factors such as high quantitative demands, low decision latitude, and poor social climate are associated with increased risk of neck/shoulder pain (Hauke et al., 2011; Lang et al., 2012; Christensen and Knardahl, 2014). Older age and previous neck pain are examples of individual factors of importance in the development of neck pain (Carroll et al., 2008a; Cote et al., 2009). High perceived stress has also been observed to increase the risk for neck and upper limb symptoms (Bongers et al., 2006), but more knowledge is warranted (Christensen and Knardahl, 2014). In a recent longitudinal study, Fanavoll et al. (2016) observed associations between high work stress (i.e. ‘Does your work involve a lot of stress and hassles?’) and risk of chronic neck/shoulder pain in the general working population. The evidence for effects of treatment for neck pain is still modest, but neck exercises, strength training, and shoulder-blade muscle training may be beneficial (Gross et al., 2015). Several studies have shown that implementing short daily exercises in the workplace can be beneficial, e.g. group-based exercises 10 min per day (Jakobsen et al., 2015) or neck and shoulder exercises 2–20 min per day (Sjögren et al., 2005; Andersen et al., 2011; Lidegaard et al., 2013). However, the treatment and intervention results are predominantly short-term effects (Gross et al., 2015), and to achieve a more constant decrease of pain and disability, there is need for a better understanding on how treatment, working conditions, and other factors in combination affect the outcomes. In a recent randomized controlled trial (RCT) on neck pain rehabilitation, the two groups receiving active treatment programs improved significantly more regarding neck function and pain compared with the treatment-as-usual (TAU) control group, 1 week after intervention (Svedmark et al., 2016). However, the same study failed to show treatment effects in long-term follow-ups, which is in line with the conclusion that effective treatments of neck pain that lasts are rare (Cheng et al., 2015; Gross et al., 2015). Better knowledge on how work-related factors and perceived stress impact the long-term outcomes after rehabilitation may contribute to effective ways of preventing the high recurrence of neck pain. The primary aim of this study was to evaluate if pain, disability, and work productivity were influenced by physical and psychosocial workplace exposures as well as by stress measured directly after 11 weeks of rehabilitation. A control group without specific intervention was also included. Follow-ups were done at 9 and 15 months after start of intervention. A secondary aim was to determine whether there were any significant group differences between treatment and control groups regarding associations of exposure (physical and psychosocial workplace exposure and stress) and outcomes (pain, disability, and work productivity). Methods Study design and participants This longitudinal study is based on an RCT study that included three groups: tailored treatment (TT), non-tailored treatment (NTT), and a control group (TAU; Svedmark et al., 2016). Participants were recruited for the RCT by advertisements in local papers and local networks at Umea municipality that included the hospital and university. In that study, TT and NTT received 11 weeks of neck and shoulder rehabilitation that was led by a physiotherapist. The rehabilitation contained evidence-based treatments for non-specific neck pain including cervical range of motion exercises and manual cervical mobilization, strength, and endurance training for neck and shoulder muscles, head and neck motor control exercises, posture exercises, and electromyography biofeedback program for the trapezius muscle. The maximum number of treatment sessions was 27 during the 11-week intervention period. The TAU group did not receive any treatment from the study but were free to seek care themselves. After the intervention, this was the case for the TT and NTT groups as well. In the RCT (Svedmark et al., 2016), no significant differences were found between the TT and NTT groups regarding pain and disability at any of the follow-ups. For work productivity, small but significant differences were revealed between the treatment groups at 9 and 15 months. However, the significance of these small differences was reduced because of different baseline values and a possible regression-towards-mean effect (Svedmark et al., 2016). In the present longitudinal study, TT and NTT are pooled to one group and denoted as the treatment group. Thus, this present longitudinal study included a treatment group (TT and NTT) and a control group (TAU). Altogether, there were 97 working women aged 20–65 years with non-specific neck pain. Inclusion criteria were pain in the neck-shoulder region for a minimum of 6 weeks, disabilities due to neck symptoms—neck disability index (NDI) score ≥10% (more than no disability) and ≤68% (less than complete disability; Vernon and Mior, 1991) and self-reported work productivity loss (quality or quantity) due to neck symptoms (Martimo et al., 2009). Exclusion criteria were as follows: trauma-related neck pain, cervical radiculopathy or vestibular dysfunction; comorbid medical conditions such as cancer, type 1 diabetes, heart disease, rheumatic disease (including fibromyalgia), anxiety or depression, concurrent low back pain, temporomandibular disorders, surgery or spinal fracture, and severely restricted shoulder range of motion; catastrophizing thoughts; low treatment expectation; or substantial changes in physical or psychosocial work environment from the start of intervention to the 3-, 9-, or 15-month follow-ups (questionnaires). The definition for substantial changes in physical or psychosocial work environment was (i) small work environment changes in combination with major deterioration or markedly improved workload or (ii) large work environment changes in combination with major/somewhat deterioration or markedly/improved workload. Participant exclusion was based on criteria that were self-reported, first in a telephone interview with a study-administrator and second during an inclusion questionnaire screening, but the specific diagnosis should have been made by a medical doctor. A physiotherapy assessment complemented this at baseline measurement if there was uncertainty about exclusion for fibromyalgia, cervical radiculopathy, or vestibular dysfunction. Inclusions and exclusions are presented in detail in Fig. 1. The study was approved by the regional ethical review board in Uppsala, Sweden (No 2011/081). Participants were informed about the study and were provided written information that was signed indicating their consent prior to participation. Figure. 1. View largeDownload slide Flow diagram of the study. WP, work productivity; QEC, quick exposure check; QPS Nordic, the Nordic questionnaires for psychological and social factors at work. Asterisk indicates that data for two participants who scored NDI ≤ 10% and eight participants who reported no work productivity loss (quality or quantity) were excluded in analyses regarding these variables. Figure. 1. View largeDownload slide Flow diagram of the study. WP, work productivity; QEC, quick exposure check; QPS Nordic, the Nordic questionnaires for psychological and social factors at work. Asterisk indicates that data for two participants who scored NDI ≤ 10% and eight participants who reported no work productivity loss (quality or quantity) were excluded in analyses regarding these variables. Outcomes Outcomes (neck pain, neck disability, and work productivity) were assessed before the intervention period (baseline), after intervention (3 months), and at 9 (intermediate) and 15 months (long-term) after the start of intervention (Fig. 1). Participants rated average neck pain intensity during the last week with the 11-point Numeric Rating Scale (NRS) (Dworkin et al., 2005), where 0 = ‘no pain’ and 10 = ‘worst pain imaginable’. Furthermore, neck disability was assessed with the NDI (0–100; Vernon and Mior, 1991). NDI consists of 10 items addressing functional activities in daily life such as personal care, lifting, car driving, reading, work, and symptoms like pain intensity, headache, and concentration. Higher scores mean more disability. Finally, self-estimated impact of neck symptoms on work productivity was assessed with two questions: (i) ‘assess the impact of your neck symptoms last month and mark with a scale from 0 (practically nothing) to 10 (regular quantity)—how much work you were able to perform when compared with your normal workday’, (ii) ‘assess the impact of your neck symptoms last month and mark with a scale from 0 (very poor quality) to 10 (regular quality)—the quality of your work when compared with your normal workday (Martimo et al., 2009)’. Physical work exposure Three months after the start of intervention (comparable time in controls), an experienced ergonomist visited participants at their workplace and assessed the physical exposure with the ergonomic risk assessment tool Quick Exposure Check (QEC; David et al., 2008). The most common work task, as defined by the participant, was observed and assessed. According to the QEC, the ergonomist registered participant’s head posture and frequency of bent or rotated head, hand position in the horizontal plane, and frequency of shoulder/arm movements during work. The participant estimated the maximum weight of tools they handled, how many hours per day they performed the observed work task, and whether the task was visually demanding. The observed exposures were calculated and transformed to the total scores for neck load (range 4–18) and the total scores for shoulder/arm load (range 10–56). The QEC has shown good validity and moderate repeatability in similar contexts (David et al., 2008; Takala, 2010). Psychosocial work exposure At the workplace visit, participants filled in ‘The Nordic questionnaires for psychological and social factors at work’ (QPS Nordic; Wännström et al., 2009). Seven scales (36 questions) were included from the original questionnaire—‘quantitative demands’, ‘decision demands’, ‘learning demands’, ‘control of decision’, ‘control of work pacing’, ‘support from superior’, and ‘support from co-workers’. The response scale ranged from 1–5 (1 = ‘very seldom or never’, 5 = ‘very often or always’), and an index score was calculated for each scale by summing the answers from the individual questions and then dividing by the number of questions. The QPS Nordic is a psychometrically tested questionnaire and is considered to be of good quality (Wännström et al., 2009). Job stress and perceived stress To measure whether participants experienced their job as stressful, a specific question from the QEC assessment was included, i.e. ‘In general, how do you find your job?’ The response scale ranged from 1–4 (1 = ‘not at all stressful’, 4 = ‘very stressful’). Participants answered this question at the workplace visit after the intervention period. To measure the perceived stress level, a single-item question was answered (Elo et al., 2003). The question was phrased ‘Stress means a situation in which a person feels tense, restless, nervous or anxious or is unable to sleep at night because his/her mind is troubled all the time. Have you felt this kind of stress during the last month?’ The response alternatives ranged from 1 = ‘not at all’ to 5 = ‘very much’. Data analysis Levels and variability of outcomes, confounders, and potential risk factors were presented for normally distributed (or approximately normal) variables as means and standard deviations, for log-normal distributed variables as geometric means and geometric standard deviations, and otherwise as quartile levels. Confounding variables between outcomes and potential risk factors were determined a priori. Age and pain duration (at baseline) were considered potential confounders based on previous reviews (Walton et al., 2013; Bruls et al., 2015). First, the cross-sectional associations between outcomes and potential risk factor exposures were evaluated at the time of acquisition of risk factor and outcome data (3 months after baseline). Univariate associations between outcomes, confounders, and potential risk factor exposures, one at a time, were estimated using linear regression for the outcome neck disability and using log-linear regression for the outcomes neck pain and work productivity. Based on these univariate analyses, candidate risk factors with P-values <0.25 were included in a multiple linear regression model. Thereafter, a backwards stepwise approach was used to identify the most important risk factors among those identified in the univariate analyses. In each step, estimates and standard errors were recorded to gain understanding of specific exposure adjustments. The limit of the P-value in the stepwise procedure was 0.25. Secondly, longitudinal assessments were conducted using the risk factor exposure as predictor of outcomes at 3, 9, and 15 months after baseline. Mixed models were used to estimate longitudinal associations, accounting for within-individual correlation of repeated outcome measures by incorporation of a random intercept. The same method for variable selection as for the cross-sectional analyses was used to construct multiple longitudinal models. In these longitudinal models, interaction terms between time and risk factors were excluded before their corresponding main effects. The correlations between potential risk factors were calculated to assess potential multicollinearity. All statistical analyses were performed using the statistical software package SPSS 22.0 for Windows (SPSS Inc., Chicago, IL, USA). The significant level was set at P < 0.05. Results The participants had various occupations and employers with 80% working for the municipality in academia or in the healthcare sector. The most common work task was computer work (67%), e.g. administration, and the second most common was direct patient-related work (18%), e.g. nurse, assistant nurse, physiotherapist, or in dental care. The mean age was 47.4 years in the treatment group and 49.1 years in the TAU group, and the median duration of neck pain before entering the study was 60 months in both groups (Table 1). The scores of the QEC and QPS Nordic scales and the levels of stress are presented for the treatment group and the TAU group in Table 1. Table 1. Characteristics of participants and exposure levels. N represents the numbers of participants presented as mean values and standard deviations. Characteristics and exposure Treatment group (n = 67) TAU (n = 30) Age (years) 47.4 (11.7) 49.1 (9.7) BMI (kg/m2)* 24 (22–26.5)a 24 (22–28) Pain duration* (month) 60 (24–144) 60 (24–120) Neck load* (score range 4–18) 16 (14–16) 16 (13.5–16) Shoulder/arm load* (score range 10–56) 26 (22–32) 26 (26–34) Quantitative demands (score range 1–5) 3.0 (0.6) 2.9 (0.6) Decision demands (score range 1–5) 3.3 (0.6) 3.3 (0.7) Learning demands (score range 1–5) 2.5 (0.5)b 2.4 (0.7) Control of decision (score range 1–5) 2.9 (0.7) 2.9 (0.8) Control of work pacing (score range 1–5) 3.3 (1.1) 3.2 (1.2) Support from superior (score range 1–5) 3.7 (0.8)b 3.7 (0.8) Support from co-workers (score range 1–5) 4.0 (0.7)b 4.2 (0.8) Job stress (score range 1–4) 2.5 (0.8) 2.6 (0.9) Perceived stress (score range 1–5) 2.4 (1.0)b 2.6 (1.1)c Characteristics and exposure Treatment group (n = 67) TAU (n = 30) Age (years) 47.4 (11.7) 49.1 (9.7) BMI (kg/m2)* 24 (22–26.5)a 24 (22–28) Pain duration* (month) 60 (24–144) 60 (24–120) Neck load* (score range 4–18) 16 (14–16) 16 (13.5–16) Shoulder/arm load* (score range 10–56) 26 (22–32) 26 (26–34) Quantitative demands (score range 1–5) 3.0 (0.6) 2.9 (0.6) Decision demands (score range 1–5) 3.3 (0.6) 3.3 (0.7) Learning demands (score range 1–5) 2.5 (0.5)b 2.4 (0.7) Control of decision (score range 1–5) 2.9 (0.7) 2.9 (0.8) Control of work pacing (score range 1–5) 3.3 (1.1) 3.2 (1.2) Support from superior (score range 1–5) 3.7 (0.8)b 3.7 (0.8) Support from co-workers (score range 1–5) 4.0 (0.7)b 4.2 (0.8) Job stress (score range 1–4) 2.5 (0.8) 2.6 (0.9) Perceived stress (score range 1–5) 2.4 (1.0)b 2.6 (1.1)c BMI body mass index. an = 65. bn = 66. cn = 29. *For variables that had a skewed distribution, the median and the first and third quartiles are presented. View Large Table 1. Characteristics of participants and exposure levels. N represents the numbers of participants presented as mean values and standard deviations. Characteristics and exposure Treatment group (n = 67) TAU (n = 30) Age (years) 47.4 (11.7) 49.1 (9.7) BMI (kg/m2)* 24 (22–26.5)a 24 (22–28) Pain duration* (month) 60 (24–144) 60 (24–120) Neck load* (score range 4–18) 16 (14–16) 16 (13.5–16) Shoulder/arm load* (score range 10–56) 26 (22–32) 26 (26–34) Quantitative demands (score range 1–5) 3.0 (0.6) 2.9 (0.6) Decision demands (score range 1–5) 3.3 (0.6) 3.3 (0.7) Learning demands (score range 1–5) 2.5 (0.5)b 2.4 (0.7) Control of decision (score range 1–5) 2.9 (0.7) 2.9 (0.8) Control of work pacing (score range 1–5) 3.3 (1.1) 3.2 (1.2) Support from superior (score range 1–5) 3.7 (0.8)b 3.7 (0.8) Support from co-workers (score range 1–5) 4.0 (0.7)b 4.2 (0.8) Job stress (score range 1–4) 2.5 (0.8) 2.6 (0.9) Perceived stress (score range 1–5) 2.4 (1.0)b 2.6 (1.1)c Characteristics and exposure Treatment group (n = 67) TAU (n = 30) Age (years) 47.4 (11.7) 49.1 (9.7) BMI (kg/m2)* 24 (22–26.5)a 24 (22–28) Pain duration* (month) 60 (24–144) 60 (24–120) Neck load* (score range 4–18) 16 (14–16) 16 (13.5–16) Shoulder/arm load* (score range 10–56) 26 (22–32) 26 (26–34) Quantitative demands (score range 1–5) 3.0 (0.6) 2.9 (0.6) Decision demands (score range 1–5) 3.3 (0.6) 3.3 (0.7) Learning demands (score range 1–5) 2.5 (0.5)b 2.4 (0.7) Control of decision (score range 1–5) 2.9 (0.7) 2.9 (0.8) Control of work pacing (score range 1–5) 3.3 (1.1) 3.2 (1.2) Support from superior (score range 1–5) 3.7 (0.8)b 3.7 (0.8) Support from co-workers (score range 1–5) 4.0 (0.7)b 4.2 (0.8) Job stress (score range 1–4) 2.5 (0.8) 2.6 (0.9) Perceived stress (score range 1–5) 2.4 (1.0)b 2.6 (1.1)c BMI body mass index. an = 65. bn = 66. cn = 29. *For variables that had a skewed distribution, the median and the first and third quartiles are presented. View Large The levels of outcome variables for the treatment and the control groups at baseline, 3-, 9-, and 15-month follow-ups, respectively, are presented in Table 2. Both neck pain and neck disability decreased more in the treatment group compared with the control group from baseline to the 3-month follow-up. Table 2. Levels of outcome variables at different time points. Neck disability presented as means values with standard deviations within parentheses, neck pain and work productivity (quantity and quality) presented as geometric means and geometric standard deviations within parentheses. Baseline 3-month follow-up 9-month follow-up 15-month follow-up Neck pain (0–10) Treatment group 5.06 (1.49) 2.97 (1.77) 3.38 (1.81) 3.12 (1.95) Control group 5.25 (1.43) 4.44 (1.67) 3.99 (1.69) 4.11 (1.66) Neck disability (0–100) Treatment group 22.18 (8.16) 13.17 (9.38) 14.99 (9.83) 13.51 (9.83) Control group 23.8 (8.41) 21.48 (11.42) 19.45 (12.94) 17.27 (9.71) Work productivity (quantity; 0–10) Treatment group 8.65 (1.25) 8.87 (1.27) 8.87 (1.21) 8.91 (1.20) Control group 8.55 (1.29) 8.76 (1.28) 8.81 (1.22) 8.87 (1.20) Work productivity (quality; 0–10) Treatment group 8.45 (1.23) 8.78 (1.29) 8.78 (1.26) 8.85 (1.24) Control group 8.34 (1.25) 8.66 (1.30) 8.72 (1.30) 8.81 (1.29) Baseline 3-month follow-up 9-month follow-up 15-month follow-up Neck pain (0–10) Treatment group 5.06 (1.49) 2.97 (1.77) 3.38 (1.81) 3.12 (1.95) Control group 5.25 (1.43) 4.44 (1.67) 3.99 (1.69) 4.11 (1.66) Neck disability (0–100) Treatment group 22.18 (8.16) 13.17 (9.38) 14.99 (9.83) 13.51 (9.83) Control group 23.8 (8.41) 21.48 (11.42) 19.45 (12.94) 17.27 (9.71) Work productivity (quantity; 0–10) Treatment group 8.65 (1.25) 8.87 (1.27) 8.87 (1.21) 8.91 (1.20) Control group 8.55 (1.29) 8.76 (1.28) 8.81 (1.22) 8.87 (1.20) Work productivity (quality; 0–10) Treatment group 8.45 (1.23) 8.78 (1.29) 8.78 (1.26) 8.85 (1.24) Control group 8.34 (1.25) 8.66 (1.30) 8.72 (1.30) 8.81 (1.29) View Large Table 2. Levels of outcome variables at different time points. Neck disability presented as means values with standard deviations within parentheses, neck pain and work productivity (quantity and quality) presented as geometric means and geometric standard deviations within parentheses. Baseline 3-month follow-up 9-month follow-up 15-month follow-up Neck pain (0–10) Treatment group 5.06 (1.49) 2.97 (1.77) 3.38 (1.81) 3.12 (1.95) Control group 5.25 (1.43) 4.44 (1.67) 3.99 (1.69) 4.11 (1.66) Neck disability (0–100) Treatment group 22.18 (8.16) 13.17 (9.38) 14.99 (9.83) 13.51 (9.83) Control group 23.8 (8.41) 21.48 (11.42) 19.45 (12.94) 17.27 (9.71) Work productivity (quantity; 0–10) Treatment group 8.65 (1.25) 8.87 (1.27) 8.87 (1.21) 8.91 (1.20) Control group 8.55 (1.29) 8.76 (1.28) 8.81 (1.22) 8.87 (1.20) Work productivity (quality; 0–10) Treatment group 8.45 (1.23) 8.78 (1.29) 8.78 (1.26) 8.85 (1.24) Control group 8.34 (1.25) 8.66 (1.30) 8.72 (1.30) 8.81 (1.29) Baseline 3-month follow-up 9-month follow-up 15-month follow-up Neck pain (0–10) Treatment group 5.06 (1.49) 2.97 (1.77) 3.38 (1.81) 3.12 (1.95) Control group 5.25 (1.43) 4.44 (1.67) 3.99 (1.69) 4.11 (1.66) Neck disability (0–100) Treatment group 22.18 (8.16) 13.17 (9.38) 14.99 (9.83) 13.51 (9.83) Control group 23.8 (8.41) 21.48 (11.42) 19.45 (12.94) 17.27 (9.71) Work productivity (quantity; 0–10) Treatment group 8.65 (1.25) 8.87 (1.27) 8.87 (1.21) 8.91 (1.20) Control group 8.55 (1.29) 8.76 (1.28) 8.81 (1.22) 8.87 (1.20) Work productivity (quality; 0–10) Treatment group 8.45 (1.23) 8.78 (1.29) 8.78 (1.26) 8.85 (1.24) Control group 8.34 (1.25) 8.66 (1.30) 8.72 (1.30) 8.81 (1.29) View Large Cross-sectional associations, univariate analyses Associations between exposures (physical, psychosocial, and stress) and outcomes (neck pain, neck disability, and work productivity) at 3 months for the treatment and the control groups are presented in Table 3. High ‘control of decision’ and a high ‘control of work pacing’ were associated with lower neck disability. Also, high ‘control of decision’ was associated with lower neck pain and higher work productivity (quantity and quality). Both high perceived stress level last month and high job stress were associated with more neck pain, higher neck disability, and lower work productivity (quantity and quality). Finally, a high shoulder/arm load was related to lower work productivity (quantity). Table 3. Univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (neck pain, neck disability, work productivity). Cross-sectional analysis at 3 months after the start of intervention. Neck pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate (95% CI) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure Neck load 3% (−2; 8%) 0.31 (−0.56; 1.18) −1% (−3; 1%) −0.4% (−3; 2%) Shoulder/arm load 0.3% (−1.3; 1.9%) −0.08 (−0.38; 0.22) −1% (−2; −0.2%)* 0.6% (−1.3; 0.1) Quantitative demands 6% (−12; 27%) 1.29 (−2.09; 4.67) −1.4% (−9; 7%) −4% (−13; 4%) Decision demands 9% (−9; 31%) 1.36 (−1.97; 4.68) −2% (−10; 6%) −1% (−9; 7%) Learning demands 11% (−30; 4%) −1.20 (−4.86; 2.45) 7% (−3; 16%) 5% (−5; 15%) Control of decision −17% (−29; −2%)* −4.69 (−7.57; −1.8)** 9% (2; 16%)** 9% (1.4; 16%)* Control of work pacing −8% (−17; 3%) −2.21 (−4.09; −0.32)* 4% (−1; 8%) 4% (−1; 9%) Support from superior −2% (−16; 3%) −1.12 (−3.9; 1.66) −1% (−8; 6%) 2% (−5; 9%) Support from co-workers −12% (−26; 5%) −0.82 (−4.02; 2.37) 0.4% (−7; 8%) 0% (−8.2; 8.2 %) Job stress 19% (3; 37%)** 4.14 (1.58; 6.7)*** −7% (−13; −3%)* −9% (−15; −3%)** Perceived stress 17% (4; 31%)** 3.59 (1.54; 5.62)*** −6% (−11; −0.4)* −9% (−14; −4%)*** Pain duration 0% (−0.00; 0.002) 0.013 (−0.009; 0.035) 0% (0.00; 0.001%) 0% (0.00; 0.001%) Age 2% (−0.7; 12%) 0.87 (−0.94; 2.7) 0.6% (−4; 5%) −2% (−5; 4%) Neck pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate (95% CI) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure Neck load 3% (−2; 8%) 0.31 (−0.56; 1.18) −1% (−3; 1%) −0.4% (−3; 2%) Shoulder/arm load 0.3% (−1.3; 1.9%) −0.08 (−0.38; 0.22) −1% (−2; −0.2%)* 0.6% (−1.3; 0.1) Quantitative demands 6% (−12; 27%) 1.29 (−2.09; 4.67) −1.4% (−9; 7%) −4% (−13; 4%) Decision demands 9% (−9; 31%) 1.36 (−1.97; 4.68) −2% (−10; 6%) −1% (−9; 7%) Learning demands 11% (−30; 4%) −1.20 (−4.86; 2.45) 7% (−3; 16%) 5% (−5; 15%) Control of decision −17% (−29; −2%)* −4.69 (−7.57; −1.8)** 9% (2; 16%)** 9% (1.4; 16%)* Control of work pacing −8% (−17; 3%) −2.21 (−4.09; −0.32)* 4% (−1; 8%) 4% (−1; 9%) Support from superior −2% (−16; 3%) −1.12 (−3.9; 1.66) −1% (−8; 6%) 2% (−5; 9%) Support from co-workers −12% (−26; 5%) −0.82 (−4.02; 2.37) 0.4% (−7; 8%) 0% (−8.2; 8.2 %) Job stress 19% (3; 37%)** 4.14 (1.58; 6.7)*** −7% (−13; −3%)* −9% (−15; −3%)** Perceived stress 17% (4; 31%)** 3.59 (1.54; 5.62)*** −6% (−11; −0.4)* −9% (−14; −4%)*** Pain duration 0% (−0.00; 0.002) 0.013 (−0.009; 0.035) 0% (0.00; 0.001%) 0% (0.00; 0.001%) Age 2% (−0.7; 12%) 0.87 (−0.94; 2.7) 0.6% (−4; 5%) −2% (−5; 4%) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Table 3. Univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (neck pain, neck disability, work productivity). Cross-sectional analysis at 3 months after the start of intervention. Neck pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate (95% CI) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure Neck load 3% (−2; 8%) 0.31 (−0.56; 1.18) −1% (−3; 1%) −0.4% (−3; 2%) Shoulder/arm load 0.3% (−1.3; 1.9%) −0.08 (−0.38; 0.22) −1% (−2; −0.2%)* 0.6% (−1.3; 0.1) Quantitative demands 6% (−12; 27%) 1.29 (−2.09; 4.67) −1.4% (−9; 7%) −4% (−13; 4%) Decision demands 9% (−9; 31%) 1.36 (−1.97; 4.68) −2% (−10; 6%) −1% (−9; 7%) Learning demands 11% (−30; 4%) −1.20 (−4.86; 2.45) 7% (−3; 16%) 5% (−5; 15%) Control of decision −17% (−29; −2%)* −4.69 (−7.57; −1.8)** 9% (2; 16%)** 9% (1.4; 16%)* Control of work pacing −8% (−17; 3%) −2.21 (−4.09; −0.32)* 4% (−1; 8%) 4% (−1; 9%) Support from superior −2% (−16; 3%) −1.12 (−3.9; 1.66) −1% (−8; 6%) 2% (−5; 9%) Support from co-workers −12% (−26; 5%) −0.82 (−4.02; 2.37) 0.4% (−7; 8%) 0% (−8.2; 8.2 %) Job stress 19% (3; 37%)** 4.14 (1.58; 6.7)*** −7% (−13; −3%)* −9% (−15; −3%)** Perceived stress 17% (4; 31%)** 3.59 (1.54; 5.62)*** −6% (−11; −0.4)* −9% (−14; −4%)*** Pain duration 0% (−0.00; 0.002) 0.013 (−0.009; 0.035) 0% (0.00; 0.001%) 0% (0.00; 0.001%) Age 2% (−0.7; 12%) 0.87 (−0.94; 2.7) 0.6% (−4; 5%) −2% (−5; 4%) Neck pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate (95% CI) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure Neck load 3% (−2; 8%) 0.31 (−0.56; 1.18) −1% (−3; 1%) −0.4% (−3; 2%) Shoulder/arm load 0.3% (−1.3; 1.9%) −0.08 (−0.38; 0.22) −1% (−2; −0.2%)* 0.6% (−1.3; 0.1) Quantitative demands 6% (−12; 27%) 1.29 (−2.09; 4.67) −1.4% (−9; 7%) −4% (−13; 4%) Decision demands 9% (−9; 31%) 1.36 (−1.97; 4.68) −2% (−10; 6%) −1% (−9; 7%) Learning demands 11% (−30; 4%) −1.20 (−4.86; 2.45) 7% (−3; 16%) 5% (−5; 15%) Control of decision −17% (−29; −2%)* −4.69 (−7.57; −1.8)** 9% (2; 16%)** 9% (1.4; 16%)* Control of work pacing −8% (−17; 3%) −2.21 (−4.09; −0.32)* 4% (−1; 8%) 4% (−1; 9%) Support from superior −2% (−16; 3%) −1.12 (−3.9; 1.66) −1% (−8; 6%) 2% (−5; 9%) Support from co-workers −12% (−26; 5%) −0.82 (−4.02; 2.37) 0.4% (−7; 8%) 0% (−8.2; 8.2 %) Job stress 19% (3; 37%)** 4.14 (1.58; 6.7)*** −7% (−13; −3%)* −9% (−15; −3%)** Perceived stress 17% (4; 31%)** 3.59 (1.54; 5.62)*** −6% (−11; −0.4)* −9% (−14; −4%)*** Pain duration 0% (−0.00; 0.002) 0.013 (−0.009; 0.035) 0% (0.00; 0.001%) 0% (0.00; 0.001%) Age 2% (−0.7; 12%) 0.87 (−0.94; 2.7) 0.6% (−4; 5%) −2% (−5; 4%) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Cross-sectional associations, multiple analyses All cross-sectional multiple regression analyses models were adjusted for the confounders age and pain duration. ‘Control of decision’ and perceived stress were the only statistically significant risk factors for neck pain, neck disability, and work productivity (quality), i.e. neck pain (‘control of decision’ −19%, 95% CI; −36 to −4%, perceived stress 17%, 95% CI; 6 to 29%), neck disability (‘control of decision’, estimate −4.53, 95% CI; −7.32 to −1.76, perceived stress, estimate 3.53, 95% CI; 1.27 to 5.81), and work productivity (‘control of decision’ 11%, 95% CI; 4 to 18%, perceived stress −10%, 95% CI; −15 to −5%). ‘Control of decision’ (9%, 95% CI; 2 to 16%) was the only statistically significant risk factor for work productivity (quantity). Longitudinal associations, univariate analyses Associations between exposures (physical, psychosocial, and stress) and outcomes (neck pain, neck disability, and work productivity) at 3, 9, and 15 months are presented in Tables 4 and 5. Higher perceived stress was associated with higher neck pain and higher neck disability and lower work productivity (quality) at all follow-ups. Higher job stress was associated with higher neck disability at all follow-ups and with higher neck pain and lower work productivity (quality) at 3 and 9 months, and work productivity (quantity) at 3 months. For the psychosocial work exposure factors, higher ‘control of decision’ was associated with lower neck disability at all follow-ups and higher work productivity (quality and quantity) at 3 months. Higher ‘quantitative demands’ was associated with higher neck pain at 9 months. For the physical work exposure factors, only a higher shoulder/arm load was associated with lower work productivity (quantity) at 3 months. Table 4. Longitudinal univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (neck pain, neck disability; linear mixed model analyses). Neck Pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate(95%CI) 3 months 9 months 15 months 3 months 9 months 15 months Neck load 3%(−16; 8%) 4% (−1; 9%) 2% (−3; 6%) 0.5 (−0.3; 1.3) 0.5 (−0.3; 1.4) 0.3 (−0.5; 1.1) Shoulder/arm load 0.3% (−1; 2%) −1% (−2; 1%) −1% (−2; 1%) −0.09 (−0.4; 0.2) 0.02 (−0.3; 0.3) −0.01 (−0.3; 0.3) Quantitative demands 10% (−8; 32%) 22% (2; 46%)* 6% (−12; 27%) 1.9 (−1.2; 5.1) 3.1 (−0.1; 6.3) 1.8 (−1.3; 5) Decision demands 14% (−5; 36%) 3% (−14; 23%) 2% (−15; 22%) 2.3 (−0.8; 5.5) 0.4 (−2.7; 3.6) 1.4 (−1.8; 4.6) Learning demands −9% (−26; 13%) 1% (−8; 25%) −3% (−22; 21%) 0.14 (−3.5; 3.8) 2.1 (−1.5; 5.9) 3.2 (−0.5; 7.1) Control of decision −14% (−27; 1%) −7% (−21; 9%) −13% (−26; 2%) −4.2 (−7.0; −1.5)** −3.9 (−6.6; −1.1)** −2.9 (−5.7; −0.1)* Control of work pacing −6% (−16; 4%) −2% (−8; 14%) −5% (−14; 6%) −1.9 (−3.6; −0.04)* −1.6 (−3.4; 0.1) −1.8 (−3.7; −0.2)* Support from superior −3% (−6; 12%) −9% (−21; 6%) −0.6% (−14; 16%) −0.8 (−3.4; 1.7) −1.6 (−4.1; 0.9) 0.3 (−2.2; 2.9) Support from co-workers −12% (−26; 3%) −10% (−23; 6%) −14% (−30; 1%) −0.6 (−3.5; 2.3) 0.41 (−2.4; 3.3) 0.5 (−2.4; 3.3) Job stress 18% (3; 36%)* 16% (1; 35%)* 9% (−6; 25%) 4.0 (1.6; 6.5)*** 4.07 (1.6; 6.5)*** 3.1 (0.6; 5.6)** Perceived stress 15% (2; 28%)* 15% (3; 29%)* 16% (3; 31%)* 3.4 (1.5; 5.3)*** 3.6 (1.6; 5.5)*** 3.1 (1.1; 5.6)** Pain duration 3% (−7; 13%) 12% (1; 23%)* 5% (−5; 16%) 1.04 (−0.7; 2.8) 1.4 (−0.3; 3.2) 0.9 (−0.8; 2.7) Age 0.4% (−1; 2%) 0.6% (−0.5; 2%) 0.3% (−1; 1%) 0.10 (−0.09; 0.3) 0.10 (−0.09; 0.3) 0.04 (−0.15; 0.2) Neck Pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate(95%CI) 3 months 9 months 15 months 3 months 9 months 15 months Neck load 3%(−16; 8%) 4% (−1; 9%) 2% (−3; 6%) 0.5 (−0.3; 1.3) 0.5 (−0.3; 1.4) 0.3 (−0.5; 1.1) Shoulder/arm load 0.3% (−1; 2%) −1% (−2; 1%) −1% (−2; 1%) −0.09 (−0.4; 0.2) 0.02 (−0.3; 0.3) −0.01 (−0.3; 0.3) Quantitative demands 10% (−8; 32%) 22% (2; 46%)* 6% (−12; 27%) 1.9 (−1.2; 5.1) 3.1 (−0.1; 6.3) 1.8 (−1.3; 5) Decision demands 14% (−5; 36%) 3% (−14; 23%) 2% (−15; 22%) 2.3 (−0.8; 5.5) 0.4 (−2.7; 3.6) 1.4 (−1.8; 4.6) Learning demands −9% (−26; 13%) 1% (−8; 25%) −3% (−22; 21%) 0.14 (−3.5; 3.8) 2.1 (−1.5; 5.9) 3.2 (−0.5; 7.1) Control of decision −14% (−27; 1%) −7% (−21; 9%) −13% (−26; 2%) −4.2 (−7.0; −1.5)** −3.9 (−6.6; −1.1)** −2.9 (−5.7; −0.1)* Control of work pacing −6% (−16; 4%) −2% (−8; 14%) −5% (−14; 6%) −1.9 (−3.6; −0.04)* −1.6 (−3.4; 0.1) −1.8 (−3.7; −0.2)* Support from superior −3% (−6; 12%) −9% (−21; 6%) −0.6% (−14; 16%) −0.8 (−3.4; 1.7) −1.6 (−4.1; 0.9) 0.3 (−2.2; 2.9) Support from co-workers −12% (−26; 3%) −10% (−23; 6%) −14% (−30; 1%) −0.6 (−3.5; 2.3) 0.41 (−2.4; 3.3) 0.5 (−2.4; 3.3) Job stress 18% (3; 36%)* 16% (1; 35%)* 9% (−6; 25%) 4.0 (1.6; 6.5)*** 4.07 (1.6; 6.5)*** 3.1 (0.6; 5.6)** Perceived stress 15% (2; 28%)* 15% (3; 29%)* 16% (3; 31%)* 3.4 (1.5; 5.3)*** 3.6 (1.6; 5.5)*** 3.1 (1.1; 5.6)** Pain duration 3% (−7; 13%) 12% (1; 23%)* 5% (−5; 16%) 1.04 (−0.7; 2.8) 1.4 (−0.3; 3.2) 0.9 (−0.8; 2.7) Age 0.4% (−1; 2%) 0.6% (−0.5; 2%) 0.3% (−1; 1%) 0.10 (−0.09; 0.3) 0.10 (−0.09; 0.3) 0.04 (−0.15; 0.2) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Table 4. Longitudinal univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (neck pain, neck disability; linear mixed model analyses). Neck Pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate(95%CI) 3 months 9 months 15 months 3 months 9 months 15 months Neck load 3%(−16; 8%) 4% (−1; 9%) 2% (−3; 6%) 0.5 (−0.3; 1.3) 0.5 (−0.3; 1.4) 0.3 (−0.5; 1.1) Shoulder/arm load 0.3% (−1; 2%) −1% (−2; 1%) −1% (−2; 1%) −0.09 (−0.4; 0.2) 0.02 (−0.3; 0.3) −0.01 (−0.3; 0.3) Quantitative demands 10% (−8; 32%) 22% (2; 46%)* 6% (−12; 27%) 1.9 (−1.2; 5.1) 3.1 (−0.1; 6.3) 1.8 (−1.3; 5) Decision demands 14% (−5; 36%) 3% (−14; 23%) 2% (−15; 22%) 2.3 (−0.8; 5.5) 0.4 (−2.7; 3.6) 1.4 (−1.8; 4.6) Learning demands −9% (−26; 13%) 1% (−8; 25%) −3% (−22; 21%) 0.14 (−3.5; 3.8) 2.1 (−1.5; 5.9) 3.2 (−0.5; 7.1) Control of decision −14% (−27; 1%) −7% (−21; 9%) −13% (−26; 2%) −4.2 (−7.0; −1.5)** −3.9 (−6.6; −1.1)** −2.9 (−5.7; −0.1)* Control of work pacing −6% (−16; 4%) −2% (−8; 14%) −5% (−14; 6%) −1.9 (−3.6; −0.04)* −1.6 (−3.4; 0.1) −1.8 (−3.7; −0.2)* Support from superior −3% (−6; 12%) −9% (−21; 6%) −0.6% (−14; 16%) −0.8 (−3.4; 1.7) −1.6 (−4.1; 0.9) 0.3 (−2.2; 2.9) Support from co-workers −12% (−26; 3%) −10% (−23; 6%) −14% (−30; 1%) −0.6 (−3.5; 2.3) 0.41 (−2.4; 3.3) 0.5 (−2.4; 3.3) Job stress 18% (3; 36%)* 16% (1; 35%)* 9% (−6; 25%) 4.0 (1.6; 6.5)*** 4.07 (1.6; 6.5)*** 3.1 (0.6; 5.6)** Perceived stress 15% (2; 28%)* 15% (3; 29%)* 16% (3; 31%)* 3.4 (1.5; 5.3)*** 3.6 (1.6; 5.5)*** 3.1 (1.1; 5.6)** Pain duration 3% (−7; 13%) 12% (1; 23%)* 5% (−5; 16%) 1.04 (−0.7; 2.8) 1.4 (−0.3; 3.2) 0.9 (−0.8; 2.7) Age 0.4% (−1; 2%) 0.6% (−0.5; 2%) 0.3% (−1; 1%) 0.10 (−0.09; 0.3) 0.10 (−0.09; 0.3) 0.04 (−0.15; 0.2) Neck Pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate(95%CI) 3 months 9 months 15 months 3 months 9 months 15 months Neck load 3%(−16; 8%) 4% (−1; 9%) 2% (−3; 6%) 0.5 (−0.3; 1.3) 0.5 (−0.3; 1.4) 0.3 (−0.5; 1.1) Shoulder/arm load 0.3% (−1; 2%) −1% (−2; 1%) −1% (−2; 1%) −0.09 (−0.4; 0.2) 0.02 (−0.3; 0.3) −0.01 (−0.3; 0.3) Quantitative demands 10% (−8; 32%) 22% (2; 46%)* 6% (−12; 27%) 1.9 (−1.2; 5.1) 3.1 (−0.1; 6.3) 1.8 (−1.3; 5) Decision demands 14% (−5; 36%) 3% (−14; 23%) 2% (−15; 22%) 2.3 (−0.8; 5.5) 0.4 (−2.7; 3.6) 1.4 (−1.8; 4.6) Learning demands −9% (−26; 13%) 1% (−8; 25%) −3% (−22; 21%) 0.14 (−3.5; 3.8) 2.1 (−1.5; 5.9) 3.2 (−0.5; 7.1) Control of decision −14% (−27; 1%) −7% (−21; 9%) −13% (−26; 2%) −4.2 (−7.0; −1.5)** −3.9 (−6.6; −1.1)** −2.9 (−5.7; −0.1)* Control of work pacing −6% (−16; 4%) −2% (−8; 14%) −5% (−14; 6%) −1.9 (−3.6; −0.04)* −1.6 (−3.4; 0.1) −1.8 (−3.7; −0.2)* Support from superior −3% (−6; 12%) −9% (−21; 6%) −0.6% (−14; 16%) −0.8 (−3.4; 1.7) −1.6 (−4.1; 0.9) 0.3 (−2.2; 2.9) Support from co-workers −12% (−26; 3%) −10% (−23; 6%) −14% (−30; 1%) −0.6 (−3.5; 2.3) 0.41 (−2.4; 3.3) 0.5 (−2.4; 3.3) Job stress 18% (3; 36%)* 16% (1; 35%)* 9% (−6; 25%) 4.0 (1.6; 6.5)*** 4.07 (1.6; 6.5)*** 3.1 (0.6; 5.6)** Perceived stress 15% (2; 28%)* 15% (3; 29%)* 16% (3; 31%)* 3.4 (1.5; 5.3)*** 3.6 (1.6; 5.5)*** 3.1 (1.1; 5.6)** Pain duration 3% (−7; 13%) 12% (1; 23%)* 5% (−5; 16%) 1.04 (−0.7; 2.8) 1.4 (−0.3; 3.2) 0.9 (−0.8; 2.7) Age 0.4% (−1; 2%) 0.6% (−0.5; 2%) 0.3% (−1; 1%) 0.10 (−0.09; 0.3) 0.10 (−0.09; 0.3) 0.04 (−0.15; 0.2) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Table 5. Longitudinal univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (work productivity; linear mixed model analyses). Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure 3 months 9 months 15 months 3 months 9 months 15 months Neck load −1.2% (−3; 0.5%) −1% (−3; 0.8%) −0.5% (−2; −1%) −0.4% (−2; –2) −0.9% (−3; 1%) −0.9% (−3; 1%) Shoulder/arm load −0.8% (−1.4; −0.2%)** −0.2% (−0.8; 0.3%) 1% (−0.4; 0.7%) −0.6% (−1; 0.1) −0.2% (−1; 0.5%) −0.1% (−1; −0.6%) Quantitative demands −2% (−9; 5%) −6% (−12; 2%) −4% (−11; 2%) −5% (−13; 2.8%) −7% (−15; 0.8) −5% (−13; 2.7) Decision demands −3% (−10%; 4%) −3% (−9; 4%) −3% (−10; 4%) −2% (−10; 6 %) 1.5% (−6; 9%) −2% (−10; 6%) Learning demands 6% (−2; 14%) −3% (−11; 5%) −2% (−10; 7%) 3% (−6; 12%) −0.8% (−10; 8 %) −3% (−13; 6%) Control of decision 8% (3; 14%)** 1% (−5; 7%) 2% (−5; 7%) 8% (1.7; 16%)* 2% (−5; 9%) 6% (−1.5; 12%) Control of work pacing 3% (−0.5; 7%) −0.2% (−5; 4%) 0.1% (−4; 4%) 4% (−9; 8%) 0.5% (−4; 5%) 3% (−1; 7%) Support from superior −0.4% (−6; 5%) 3% (−2; 9%) 5% (−0.7; 11%) 3% (−4; 10%) 2% (−4; 9%) 3% (−4; 10%) Support from co-workers 1% (−5; 8%) −1% (−8; 6%) 2% (−5; 9%) 1% (−6; 9%) −0.3% (−8; 7%) −3% (−11; 5%) Job stress −7% (−12; −1%)* −1% (−7; 4%) −3% (−8; 3%) −9% (−15; −3%)** −6% (−13; −0.3%)* −5% (−12; 0.9%) Perceived stress −5% (−9; −0.6)* −3% (−7; 1%) −4% (−8; 1%) −8% (−13; −3%)** −5% (−10; −0.1%)* −5% (−10; −3%)* Pain duration 0.5% (−3; 4%) −0.6% (−4; 3%) −0.7% (−4; 3%) −0.1% (−4; 4%) 2% (−2; 6%) 1% (−4; 5%) Age 0.2% (−0.2; 0.6%) 0.1% (−0.3; 5%) 0.3% (−0.8; 0.1) 0.3% (−0.2; 1%) 0.1% (−0.4; 0.6%) −0.1% (−0.6; 0.4%) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure 3 months 9 months 15 months 3 months 9 months 15 months Neck load −1.2% (−3; 0.5%) −1% (−3; 0.8%) −0.5% (−2; −1%) −0.4% (−2; –2) −0.9% (−3; 1%) −0.9% (−3; 1%) Shoulder/arm load −0.8% (−1.4; −0.2%)** −0.2% (−0.8; 0.3%) 1% (−0.4; 0.7%) −0.6% (−1; 0.1) −0.2% (−1; 0.5%) −0.1% (−1; −0.6%) Quantitative demands −2% (−9; 5%) −6% (−12; 2%) −4% (−11; 2%) −5% (−13; 2.8%) −7% (−15; 0.8) −5% (−13; 2.7) Decision demands −3% (−10%; 4%) −3% (−9; 4%) −3% (−10; 4%) −2% (−10; 6 %) 1.5% (−6; 9%) −2% (−10; 6%) Learning demands 6% (−2; 14%) −3% (−11; 5%) −2% (−10; 7%) 3% (−6; 12%) −0.8% (−10; 8 %) −3% (−13; 6%) Control of decision 8% (3; 14%)** 1% (−5; 7%) 2% (−5; 7%) 8% (1.7; 16%)* 2% (−5; 9%) 6% (−1.5; 12%) Control of work pacing 3% (−0.5; 7%) −0.2% (−5; 4%) 0.1% (−4; 4%) 4% (−9; 8%) 0.5% (−4; 5%) 3% (−1; 7%) Support from superior −0.4% (−6; 5%) 3% (−2; 9%) 5% (−0.7; 11%) 3% (−4; 10%) 2% (−4; 9%) 3% (−4; 10%) Support from co-workers 1% (−5; 8%) −1% (−8; 6%) 2% (−5; 9%) 1% (−6; 9%) −0.3% (−8; 7%) −3% (−11; 5%) Job stress −7% (−12; −1%)* −1% (−7; 4%) −3% (−8; 3%) −9% (−15; −3%)** −6% (−13; −0.3%)* −5% (−12; 0.9%) Perceived stress −5% (−9; −0.6)* −3% (−7; 1%) −4% (−8; 1%) −8% (−13; −3%)** −5% (−10; −0.1%)* −5% (−10; −3%)* Pain duration 0.5% (−3; 4%) −0.6% (−4; 3%) −0.7% (−4; 3%) −0.1% (−4; 4%) 2% (−2; 6%) 1% (−4; 5%) Age 0.2% (−0.2; 0.6%) 0.1% (−0.3; 5%) 0.3% (−0.8; 0.1) 0.3% (−0.2; 1%) 0.1% (−0.4; 0.6%) −0.1% (−0.6; 0.4%) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Table 5. Longitudinal univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (work productivity; linear mixed model analyses). Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure 3 months 9 months 15 months 3 months 9 months 15 months Neck load −1.2% (−3; 0.5%) −1% (−3; 0.8%) −0.5% (−2; −1%) −0.4% (−2; –2) −0.9% (−3; 1%) −0.9% (−3; 1%) Shoulder/arm load −0.8% (−1.4; −0.2%)** −0.2% (−0.8; 0.3%) 1% (−0.4; 0.7%) −0.6% (−1; 0.1) −0.2% (−1; 0.5%) −0.1% (−1; −0.6%) Quantitative demands −2% (−9; 5%) −6% (−12; 2%) −4% (−11; 2%) −5% (−13; 2.8%) −7% (−15; 0.8) −5% (−13; 2.7) Decision demands −3% (−10%; 4%) −3% (−9; 4%) −3% (−10; 4%) −2% (−10; 6 %) 1.5% (−6; 9%) −2% (−10; 6%) Learning demands 6% (−2; 14%) −3% (−11; 5%) −2% (−10; 7%) 3% (−6; 12%) −0.8% (−10; 8 %) −3% (−13; 6%) Control of decision 8% (3; 14%)** 1% (−5; 7%) 2% (−5; 7%) 8% (1.7; 16%)* 2% (−5; 9%) 6% (−1.5; 12%) Control of work pacing 3% (−0.5; 7%) −0.2% (−5; 4%) 0.1% (−4; 4%) 4% (−9; 8%) 0.5% (−4; 5%) 3% (−1; 7%) Support from superior −0.4% (−6; 5%) 3% (−2; 9%) 5% (−0.7; 11%) 3% (−4; 10%) 2% (−4; 9%) 3% (−4; 10%) Support from co-workers 1% (−5; 8%) −1% (−8; 6%) 2% (−5; 9%) 1% (−6; 9%) −0.3% (−8; 7%) −3% (−11; 5%) Job stress −7% (−12; −1%)* −1% (−7; 4%) −3% (−8; 3%) −9% (−15; −3%)** −6% (−13; −0.3%)* −5% (−12; 0.9%) Perceived stress −5% (−9; −0.6)* −3% (−7; 1%) −4% (−8; 1%) −8% (−13; −3%)** −5% (−10; −0.1%)* −5% (−10; −3%)* Pain duration 0.5% (−3; 4%) −0.6% (−4; 3%) −0.7% (−4; 3%) −0.1% (−4; 4%) 2% (−2; 6%) 1% (−4; 5%) Age 0.2% (−0.2; 0.6%) 0.1% (−0.3; 5%) 0.3% (−0.8; 0.1) 0.3% (−0.2; 1%) 0.1% (−0.4; 0.6%) −0.1% (−0.6; 0.4%) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure 3 months 9 months 15 months 3 months 9 months 15 months Neck load −1.2% (−3; 0.5%) −1% (−3; 0.8%) −0.5% (−2; −1%) −0.4% (−2; –2) −0.9% (−3; 1%) −0.9% (−3; 1%) Shoulder/arm load −0.8% (−1.4; −0.2%)** −0.2% (−0.8; 0.3%) 1% (−0.4; 0.7%) −0.6% (−1; 0.1) −0.2% (−1; 0.5%) −0.1% (−1; −0.6%) Quantitative demands −2% (−9; 5%) −6% (−12; 2%) −4% (−11; 2%) −5% (−13; 2.8%) −7% (−15; 0.8) −5% (−13; 2.7) Decision demands −3% (−10%; 4%) −3% (−9; 4%) −3% (−10; 4%) −2% (−10; 6 %) 1.5% (−6; 9%) −2% (−10; 6%) Learning demands 6% (−2; 14%) −3% (−11; 5%) −2% (−10; 7%) 3% (−6; 12%) −0.8% (−10; 8 %) −3% (−13; 6%) Control of decision 8% (3; 14%)** 1% (−5; 7%) 2% (−5; 7%) 8% (1.7; 16%)* 2% (−5; 9%) 6% (−1.5; 12%) Control of work pacing 3% (−0.5; 7%) −0.2% (−5; 4%) 0.1% (−4; 4%) 4% (−9; 8%) 0.5% (−4; 5%) 3% (−1; 7%) Support from superior −0.4% (−6; 5%) 3% (−2; 9%) 5% (−0.7; 11%) 3% (−4; 10%) 2% (−4; 9%) 3% (−4; 10%) Support from co-workers 1% (−5; 8%) −1% (−8; 6%) 2% (−5; 9%) 1% (−6; 9%) −0.3% (−8; 7%) −3% (−11; 5%) Job stress −7% (−12; −1%)* −1% (−7; 4%) −3% (−8; 3%) −9% (−15; −3%)** −6% (−13; −0.3%)* −5% (−12; 0.9%) Perceived stress −5% (−9; −0.6)* −3% (−7; 1%) −4% (−8; 1%) −8% (−13; −3%)** −5% (−10; −0.1%)* −5% (−10; −3%)* Pain duration 0.5% (−3; 4%) −0.6% (−4; 3%) −0.7% (−4; 3%) −0.1% (−4; 4%) 2% (−2; 6%) 1% (−4; 5%) Age 0.2% (−0.2; 0.6%) 0.1% (−0.3; 5%) 0.3% (−0.8; 0.1) 0.3% (−0.2; 1%) 0.1% (−0.4; 0.6%) −0.1% (−0.6; 0.4%) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Longitudinal associations, multiple analyses In the longitudinal multiple regression analysis, all models were adjusted for the confounders age and pain duration. Statistically significant risk factors for neck pain in the multiple model were ‘control of decision’ (3 months −20%, 95% CI; −36 to −3%) and perceived stress (3 months 18%, 95% CI; 6 to 30%, 9 months 16%, 95% CI; 5 to 28%, 15 months 18%, 95% CI; 6 to 30%). ‘Support from supervisor’ was close to significant (main effect 13%, 95% CI; −0.5 to 26%), and the association was positive, i.e. a high ‘support from supervisor’ was associated to more neck pain. Statistically significant risk factors for neck disability were ‘control of decision’ (3-month estimate −4.57, 95% CI; −7.29 to −1.86, 9-month estimate −4.14, 95% CI; −6.85 to −1.42, and 15-month estimate −3.09, 95% CI; −5.84 to −0.33) and perceived stress (3-month estimate 3.76, 95% CI; 1.6 to 5.92, 9-month estimate 3.78, 95% CI; 1.61 to 5.95, and 15-month estimate 3.11, 95% CI; 0.93 to 5.28). The only statistically significant risk factor for work productivity (quantity) in the multiple models was perceived stress at 3 months (−5%, 95% CI; −10 to −1%). Statistically significant risk factors for work productivity (quality) were ‘control of decision’ (main effect 7%, 95% CI; 2 to 12%) and perceived stress (3 months −9%, 95% CI; −14 to −4%, 9 months −6% 95% CI; −11 to −0.6%, and 15 months −6%, 95% CI; −11 to −1%). Consistency of results between treatment and control groups One physical work exposure factor, shoulder/arm load at 3 months, showed a small but significant difference between groups (P = 0.04); shoulder/arm load was significantly associated with work productivity (quantity) in the treatment group (−1.7%, 95% CI; −3.2 to −0.29%) but not in the control group (−0.6%, 95% CI; −1.7 to 3.1%). Also, one psychosocial work exposure factor, ‘quantitative demands’ at the 9-month follow-up differed significantly between groups (P = 0.04); ‘quantitative demands’ was statistically significant associated with neck disability in the control group (estimate 4.97, 95% CI; 1.3 to 8.65) but not in the treatment group (estimate −2.55, 95% CI; −8.99 to 3.88). There was no difference between groups regarding stress. Discussion Our longitudinal multiple regression analyses showed that perceived stress influenced neck pain, neck disability, and work productivity, and ‘control of decision’ influenced neck disability and work productivity (quality). These effects were evident at the intermediate (9 month) and the long-term (15 month) follow-ups. The only physical exposure related to outcomes was high shoulder/arm load that was associated with reduced work productivity (quantity) in the univariate analyses. These results imply that interventions focusing on the individual should consider workplace exposure/s and perceived stress since these exposures will influence the result both in the intermediate and long-term perspective. Our results indicate that handling the psychosocial factors and stress may be more important than the physical work exposures in order to optimize long-term treatment results in neck pain rehabilitation. Physical workplace exposure Our finding that only one physical exposure, i.e. shoulder/arm load, seemed important for work productivity (quantity) is surprising; the change in one-unit increase in shoulder/arm load was associated with only −1% of decreased work productivity. This rather weak influence differs from other studies that show a stronger relationship between high shoulder/arm load and neck and shoulder disorders (Larsson et al., 2007; Nordander et al., 2016). Similarly, other studies have shown that high neck load, such as with awkward posture (i.e. prolonged neck flexion), is associated with neck and shoulder pain (Sterud et al., 2014). This was, however, not investigated in our study. Nevertheless, the observation instrument that we used, QEC, addressed high neck load values (median 16 of maximum 18), but no associations to pain, disability, or work productivity were found. One explanation for this lack of associations could be the relatively low variation in neck load exposure since a lack of variance in the exposure variable will affect the possibility to observe associations. The low variance in exposure could partly be explained by the high proportion of computer workers (67%) who probably have more static work postures. Prior to our study, we piloted the QEC in staff in a hospital setting, and the tool revealed variations in exposure both between occupations and between workers (Ericsson et al., 2012). However, our pilot study included more occupations than in the sample of the present study. Despite the fact that QEC has shown good validity and moderate repeatability (David et al., 2008; Takala, 2010), the proposed categories in the instrument seem arbitrarily chosen, and longitudinal studies addressing exposure–response associations with musculoskeletal disorders are lacking. Psychosocial workplace exposure The QPS Nordic questionnaire analyses (cross-sectional) revealed that the ‘control of decision’ was statistically significant for all three outcomes—neck pain, neck disability, and work productivity. Also, in the longitudinal analysis, ‘control of decision’ was significantly associated with neck disability and work productivity but not with neck pain. In contrast, in two longitudinal studies (2 and 4 years in duration) using QPS Nordic, Christensen and Knardahl (2010, 2014) found that high-middle ‘control of decision’ had associations to neck pain. The comparison is complex due to different choices in the analyses. Christensen et al. (Christensen and Knardahl, 2010; Christensen and Knardahl, 2014) used category data such as high-middle ‘control of decision’ and ‘troubled by’ pain (1–4), and we used the index score of ‘control of decision’ (1–5) and pain intensity scale (NRS 0–10). Furthermore, the different follow-up times may limit the comparison with the results found in Christensen and Knardahl studies. In addition to ‘control of decision’, we found that high ‘quantitative demands’ was associated with more neck pain and more neck disability. ‘High ‘quantitative demands’ and ‘low ‘control of decision’, also commonly known as ‘job demands’ and ‘decision latitude’ (Karasek et al., 1998), are concluded to be psychosocial workplace factors with associations to neck pain (Bongers et al., 2006) even though the associations are not very strong or very specific We have not found any other study with a similar design, i.e. longitudinal follow-ups after a neck pain intervention that address the impact of workplace exposure and stress on the outcome of the rehabilitation. Perceived stress and job stress In a cross-sectional study, Jay et al. (Jay et al., 2015) investigated the associations between combined perceived stress (The Perceived Stress scale) and neck/shoulder pain with work ability among working women. They found that high stress and more pain were independently associated with lower work ability. This is in line with our findings that high perceived stress was associated with lower self-rated work ability. In addition to the previous study (Jay et al., 2015), we found that perceived stress and work productivity were associated in intermediate and long-term follow-ups as well. Recently, a longitudinal study (Lindegård et al., 2014) investigated associations between perceived stress (Elo et al., 2003), musculoskeletal pain (neck and low back), and work productivity in healthcare workers. The authors found that frequent musculoskeletal pain in combination with perceived stress was associated with lower work ability; however, perceived stress alone was not clearly associated with work ability. In our study, we used two stress questions—one regarding job stress (QEC) and one regarding perceived stress last month (Elo et al., 2003). These questions had a rather high correlation (0.53) although they measure different aspects. The job stress question is related to workplace situations, and the perceived stress question is related to individuals’ perceived stress as an effect of a stressful job or other stressful life events such as a divorce, a sick child, or economic difficulties. The univariate analysis revealed that both measures of stress were associated with neck pain, disability, and work productivity after intervention, and for the intermediate and the long-term follow-ups. However, the multiple analyses revealed that only perceived stress was associated with neck pain, disability, and work productivity in the intermediate and the long-term follow-ups. Therefore, based on our results, it seems that perceived stress in full is a more important factor than job stress. Differences in treatment group and controls Only small differences between treatment and control group regarding associations of exposure and outcomes were observed. This indicates that the RCT intervention did not have an impact on self-estimated psychosocial work exposure, job stress, and perceived stress. That was not the aim with the previous RCT, but the results from this longitudinal study highlight the importance in accounting for work and stress factors in future trials for neck pain. Time points Time points for evaluations were set at 3, 9, and 15 months following the start of intervention and corresponded to frequently used alternatives representing short, intermediate, and long-term follow-ups after finalized treatment. These time points were thus considered as adequate in the RCT, which was the base for our study material (Svedmark et al., 2016). However, for the purpose of the present study, it could have been desirable with an even longer follow-up time than 15 months, but the risk of further attrition with time due to changes for the employees and the work environment made us refrain from an extended time for evaluation. Strength and limitations of the study In general, individuals with non-specific neck pain are regarded as a heterogeneous group with respect to underlying mechanisms and presentation of pain and disability (Cote et al., 2008; Carroll et al., 2008b). The purpose of the exclusion and inclusion criteria in the RCT study, on which the present study was based, was to make the study group more homogenous in order to fit the objectives of the RCT (Bjorklund et al., 2012; Svedmark et al., 2016). Thus, participants with depression or anxiety symptoms, catastrophizing thoughts, and low treatment expectation were excluded. This should be considered with regard to generalization of findings from the present study. The outcomes together with psychosocial work exposure and stress levels were self-reported questionnaires or questions. Even though the employed instruments have been proven valid, self-reporting always constitutes a challenge due to potential misunderstanding of the questions or careless mistakes because of lack of time or other circumstances. The focus of the present study was on exposures at the workplace while non-work exposures were not specifically addressed. Exposures from outside of work could have been potential confounders. Since the present study used data from a RCT, the sample size was determined with power calculations for primary outcomes of the RCT (Bjorklund et al., 2012; Svedmark et al., 2016) and not primarily for the present study. Despite this, there were consistent findings with regard to the impact of exposure on neck pain rehabilitation outcomes. We believe that the present study approach helped to identify factors crucial for sustainable rehabilitation results. Conclusion High perceived stress and low ‘control of decision’ were associated with more neck pain and increased neck disability but also to decreased work productivity in women after a rehabilitation intervention. Our findings therefore strongly suggest that neck rehabilitation interventions (e.g. manual therapy, exercises) focusing on the individual should take workplace exposures and perceived stress into account since these exposures could influence the treatment results both in the intermediate and the long-term perspectives. Funding The project was funded by the Swedish Council for Working Life and Social Research (registration number 2009–1403), AFA Insurance (registration number 090288). The funders have no other role in the study other than to provide funding. Conflict of Interest The authors declare that they have no competing interests. Acknowledgements Majken Rahm is acknowledged for valuable contribution to administrative routine and logistics, Pernilla Jansson for data collection, and Göran Sandström for database support. References Andersen LL , Saervoll CA , Mortensen OS , et al. ( 2011 ) Effectiveness of small daily amounts of progressive resistance training for frequent neck/shoulder pain: randomised controlled trial . Pain ; 152 : 440 – 6 . Google Scholar CrossRef Search ADS PubMed Björklund M , Djupsjöbacka M , Svedmark A , et al. ( 2012 ) Effects of tailored neck-shoulder pain treatment based on a decision model guided by clinical assessments and standardized functional tests. A study protocol of a randomized controlled trial . BMC Musculoskelet Disord ; 13 : 75 . Google Scholar CrossRef Search ADS PubMed Bongers PM , Ijmker S , van den Heuvel S , et al. ( 2006 ) Epidemiology of work related neck and upper limb problems: psychosocial and personal risk factors (Part I) and effective interventions from a bio behavioural perspective (Part II) . J Occup Rehabil ; 16 : 279 – 302 . Google Scholar CrossRef Search ADS PubMed Borghouts JA , Koes BW , Vondeling H , et al. ( 1999 ) Cost-of-illness of neck pain in The Netherlands in 1996 . Pain ; 80 : 629 – 36 . Google Scholar CrossRef Search ADS PubMed Bruls VE , Bastiaenen CH , de Bie RA . ( 2015 ) Prognostic factors of complaints of arm, neck, and/or shoulder: a systematic review of prospective cohort studies . Pain ; 156 : 765 – 88 . Google Scholar CrossRef Search ADS PubMed Carroll LJ , et al. ( 2008a ) Course and prognostic factors for neck pain in workers: results of the Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders . Spine (Phila Pa 1976) ; 33 ( Suppl. 4 ): S93 – 100 . Google Scholar CrossRef Search ADS Carroll LJ , et al. ( 2008b ) Course and prognostic factors for neck pain in the general population: results of the Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders . Spine (Phila Pa 1976) ; 33 ( Suppl. 4 ): S75 – 82 . Google Scholar CrossRef Search ADS Cheng CH , Su HT , Yen LW , et al. ( 2015 ) Long-term effects of therapeutic exercise on nonspecific chronic neck pain: a literature review . J Phys Ther Sci ; 27 : 1271 – 6 . Google Scholar CrossRef Search ADS PubMed Childs JD , Cleland JA , Elliott JM , et al. ; American Physical Therapy Association . ( 2008 ) Neck pain: clinical practice guidelines linked to the International Classification of Functioning, Disability, and Health from the Orthopedic Section of the American Physical Therapy Association . J Orthop Sports Phys Ther ; 38 : A1 – 34 . Google Scholar CrossRef Search ADS PubMed Christensen JO , Knardahl S . ( 2010 ) Work and neck pain: a prospective study of psychological, social, and mechanical risk factors . Pain ; 151 : 162 – 73 . Google Scholar CrossRef Search ADS PubMed Christensen JO , Knardahl S . ( 2014 ) Time-course of occupational psychological and social factors as predictors of new-onset and persistent neck pain: a three-wave prospective study over 4 years . Pain ; 155 : 1262 – 71 . Google Scholar CrossRef Search ADS PubMed Cote P , et al. ( 2008 ) The burden and determinants of neck pain in workers: results of the Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders . Spine (Phila Pa 1976) ; 33 ( Suppl. 4 ): S60 – 74 . Google Scholar CrossRef Search ADS PubMed Côté P , van der Velde G , Cassidy JD , et al. ( 2009 ) The burden and determinants of neck pain in workers: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders . J Manipulative Physiol Ther ; 32 ( Suppl. 2 ): S70 – 86 . Google Scholar CrossRef Search ADS PubMed David G , Woods V , Li G , et al. ( 2008 ) The development of the Quick Exposure Check (QEC) for assessing exposure to risk factors for work-related musculoskeletal disorders . Appl Ergon ; 39 : 57 – 69 . Google Scholar CrossRef Search ADS PubMed Dworkin RH , Turk DC , Farrar JT , et al. ; IMMPACT . ( 2005 ) Core outcome measures for chronic pain clinical trials: IMMPACT recommendations . Pain ; 113 : 9 – 19 . Google Scholar CrossRef Search ADS PubMed Elo AL , Leppänen A , Jahkola A . ( 2003 ) Validity of a single-item measure of stress symptoms . Scand J Work Environ Health ; 29 : 444 – 51 . Google Scholar CrossRef Search ADS PubMed Ericsson P , Bjorklund M , Wahlstrom J ( 2012 ) Exposure assessment in different occupational groups at a hospital using Quick Exposure Check (QEC)—a pilot study . Work ; 41 ( Suppl. 1 ): 5718 – 20 . Google Scholar PubMed Fanavoll R , Nilsen TI , Holtermann A , et al. ( 2016 ) Psychosocial work stress, leisure time physical exercise and the risk of chronic pain in the neck/shoulders: longitudinal data from the Norwegian HUNT Study . Int J Occup Med Environ Health ; 29 : 585 – 95 . Google Scholar CrossRef Search ADS PubMed Gross A , Kay TM , Paquin JP , et al. ; Cervical Overview Group . ( 2015 ) Exercises for mechanical neck disorders . Cochrane Database Syst Rev ; 1 : CD004250 . Google Scholar PubMed Hauke A , Flintrop J , Brun E , Rugulies R ( 2011 ) The impact of work-related psychosocial stressors on the onset of musculoskeletal disorders in specific body regions: a review and meta-analysis of 54 longitudinal studies . Work & Stress ; 25 : 243 – 56 . Google Scholar CrossRef Search ADS Hogg-Johnson S , van der Velde G , Carroll LJ , et al. ; Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders . ( 2008 ) The burden and determinants of neck pain in the general population: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders . Spine (Phila Pa 1976) ; 33 ( Suppl. 4 ): S39 – 51 . Google Scholar CrossRef Search ADS PubMed Hoy D , et al. ( 2014 ) The global burden of neck pain: estimates from the global burden of disease 2010 study . Ann Rheum Dis ; 73 : 1309 – 15 . Google Scholar CrossRef Search ADS PubMed Jakobsen MD , Sundstrup E , Brandt M , et al. ( 2015 ) Effect of workplace- versus home-based physical exercise on musculoskeletal pain among healthcare workers: a cluster randomized controlled trial . Scand J Work Environ Health ; 41 : 153 – 63 . Google Scholar CrossRef Search ADS PubMed Jay K , Friborg MK , Sjøgaard G , et al. ( 2015 ) The consequence of combined pain and stress on work ability in female laboratory technicians: a cross-sectional study . Int J Environ Res Public Health ; 12 : 15834 – 42 . Google Scholar CrossRef Search ADS PubMed Karasek R , Brisson C , Kawakami N , et al. ( 1998 ) The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics . J Occup Health Psychol ; 3 : 322 – 55 . Google Scholar CrossRef Search ADS PubMed Lang J , Ochsmann E , Kraus T , et al. ( 2012 ) Psychosocial work stressors as antecedents of musculoskeletal problems: a systematic review and meta-analysis of stability-adjusted longitudinal studies . Soc Sci Med ; 75 : 1163 – 74 . Google Scholar CrossRef Search ADS PubMed Larsson B , Søgaard K , Rosendal L . ( 2007 ) Work related neck-shoulder pain: a review on magnitude, risk factors, biochemical characteristics, clinical picture and preventive interventions . Best Pract Res Clin Rheumatol ; 21 : 447 – 63 . Google Scholar CrossRef Search ADS PubMed Lidegaard M , Jensen RB , Andersen CH , et al. ( 2013 ) Effect of brief daily resistance training on occupational neck/shoulder muscle activity in office workers with chronic pain: randomized controlled trial . Biomed Res Int ; 2013 : 262386 . Google Scholar CrossRef Search ADS PubMed Lindegård A , Larsman P , Hadzibajramovic E , et al. ( 2014 ) The influence of perceived stress and musculoskeletal pain on work performance and work ability in Swedish health care workers . Int Arch Occup Environ Health ; 87 : 373 – 9 . Google Scholar CrossRef Search ADS PubMed Malchaire J , Cock N , Vergracht S . ( 2001 ) Review of the factors associated with musculoskeletal problems in epidemiological studies . Int Arch Occup Environ Health ; 74 : 79 – 90 . Google Scholar CrossRef Search ADS PubMed Martimo KP , Shiri R , Miranda H , et al. ( 2009 ) Self-reported productivity loss among workers with upper extremity disorders . Scand J Work Environ Health ; 35 : 301 – 8 . Google Scholar CrossRef Search ADS PubMed Nordander C , Hansson GÅ , Ohlsson K , et al. ( 2016 ) Exposure-response relationships for work-related neck and shoulder musculoskeletal disorders—analyses of pooled uniform data sets . Appl Ergon ; 55 : 70 – 84 . Google Scholar CrossRef Search ADS PubMed Paksaichol A , Janwantanakul P , Purepong N , et al. ( 2012 ) Office workers’ risk factors for the development of non-specific neck pain: a systematic review of prospective cohort studies . Occup Environ Med ; 69 : 610 – 8 . Google Scholar CrossRef Search ADS PubMed Palmer KT , Smedley J . ( 2007 ) Work relatedness of chronic neck pain with physical findings—a systematic review . Scand J Work Environ Health ; 33 : 165 – 91 . Google Scholar CrossRef Search ADS PubMed Sjögren T , Nissinen KJ , Järvenpää SK , et al. ( 2005 ) 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 – 28 . Google Scholar CrossRef Search ADS PubMed Sterud T , Johannessen HA , Tynes T . ( 2014 ) Work-related psychosocial and mechanical risk factors for neck/shoulder pain: a 3-year follow-up study of the general working population in Norway . Int Arch Occup Environ Health ; 87 : 471 – 81 . Google Scholar CrossRef Search ADS PubMed Svedmark Å , Djupsjöbacka M , Häger C , et al. ( 2016 ) Is tailored treatment superior to non-tailored treatment for pain and disability in women with non-specific neck pain? A randomized controlled trial . BMC Musculoskelet Disord ; 17 : 408 . Google Scholar CrossRef Search ADS PubMed Takala EP , Pehkonen I , Forsman M , et al. ( 2010 ) Systematic evaluation of observational methods assessing biomechanical exposures at work . Scand J Work Environ Health ; 36 : 3 – 24 . Google Scholar CrossRef Search ADS PubMed Wahlstedt K , Norbäck D , Wieslander G , et al. ( 2010 ) Psychosocial and ergonomic factors, and their relation to musculoskeletal complaints in the Swedish workforce . Int J Occup Saf Ergon ; 16 : 311 – 21 . Google Scholar CrossRef Search ADS PubMed Walton DM , Carroll LJ , Kasch H , et al. ; ICON . ( 2013 ) An overview of systematic reviews on prognostic factors in neck pain: results from the International Collaboration on Neck Pain (ICON) project . Open Orthop J ; 7 : 494 – 505 . Google Scholar CrossRef Search ADS PubMed Wännström I , Peterson U , Asberg M , et al. ( 2009 ) Psychometric properties of scales in the General Nordic Questionnaire for Psychological and Social Factors at Work (QPS): confirmatory factor analysis and prediction of certified long-term sickness absence . Scand J Psychol ; 50 : 231 – 44 . Google Scholar CrossRef Search ADS PubMed Vernon H , Mior S . ( 1991 ) The Neck Disability Index: a study of reliability and validity . J Manipulative Physiol Ther ; 14 : 409 – 15 . Google Scholar PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Occupational Hygiene Society. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene) Oxford University Press

Impact of Workplace Exposure and Stress on Neck Pain and Disabilities in Women—A Longitudinal Follow-up After a Rehabilitation Intervention

Loading next page...
 
/lp/ou_press/impact-of-workplace-exposure-and-stress-on-neck-pain-and-disabilities-vgOfmBQygV
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the British Occupational Hygiene Society.
ISSN
2398-7308
eISSN
2398-7316
D.O.I.
10.1093/annweh/wxy018
Publisher site
See Article on Publisher Site

Abstract

Abstract Introduction The aim was to evaluate if pain, disability, and work productivity are influenced by physical and psychosocial work exposures as well as by stress, up to 1 year after a randomized controlled trial treatment intervention, and to determine whether any such association differed between treatment and control groups. Methods Ninety-seven working women suffering non-specific neck pain (n = 67 treatment group, n = 30 control group) were followed from end of treatment intervention and at 9- and 15-month follow-ups, respectively. Physical and psychosocial exposures, as well as perceived stress, were assessed after the treatment intervention. Pain, neck disability, and work productivity were assessed at baseline, after intervention 3 months later and at 9- and 15-month follow-ups. Longitudinal assessment was conducted using the exposure level at 3 months as predictor of pain, disability, and work productivity at 3, 9, and 15 months, respectively. Mixed models were used to estimate longitudinal associations, accounting for within-individual correlation of repeated outcome measures by incorporation of a random intercept. Age and duration of neck pain were adjusted for in all models. To evaluate group differences, interactions between exposures and treatment groups were estimated. Results High perceived stress was associated with more neck pain, more neck disability, and decreased work productivity in both cross-sectional and longitudinal analyses. High ‘control of decision’ was associated with less neck pain, less neck disability, and higher work productivity in cross-sectional analyses but only to less disability and higher productivity in longitudinal analyses. Shoulder/arm load was the only physical exposure variable that was significantly associated with work productivity in the univariate analyses. Only small differences were observed between treatment and control groups. Conclusion High perceived stress and low ‘control of decision’ were associated with more neck pain, increased neck disability, and decreased work productivity. Treatment interventions for individuals with neck pain should take into account psychosocial workplace exposures and stress to improve intermediate and long-term results. non-specific neck pain, physiotherapy, shoulder pain, work productivity Introduction Neck pain is a growing global health problem that is ranked fourth in the world in terms of disability (Hoy et al., 2014). It is a common condition with studies reporting a 1-year prevalence from 27% to almost 50% (Cote et al., 2009). More than 50% of persons who experienced an episode of neck pain reported neck pain 1 year later, thus indicating a long-lasting problem (Carroll et al., 2008a). Among workers, 11–14% have limitations in their activities due to neck pain (Cote et al., 2009). In addition to the individual suffering, reduced productivity, increased use of healthcare, and great financial burden for the society may occur (Borghouts et al., 1999; Childs et al., 2008). Most studies show that women report more neck pain than men both in the general population (Hogg-Johnson et al., 2008; Hoy et al., 2014) and in the working population (Larsson et al., 2007; Wahlstedt et al., 2010; Paksaichol et al., 2012; Hoy et al., 2014). Neck pain is assumed to be of multifactorial origin with interacting individual, physical, and psychosocial risk factors. Repetitive work (Malchaire et al., 2001; Palmer and Smedley, 2007; Cote et al., 2009), upper arm elevation (Nordander et al., 2016), working with the neck flexed more than 20º during prolonged time periods (Palmer and Smedley, 2007), and precision-demanding work (Cote et al., 2009) have all been recognized as work-related physical risk factors for neck pain. Further, work-related psychosocial factors such as high quantitative demands, low decision latitude, and poor social climate are associated with increased risk of neck/shoulder pain (Hauke et al., 2011; Lang et al., 2012; Christensen and Knardahl, 2014). Older age and previous neck pain are examples of individual factors of importance in the development of neck pain (Carroll et al., 2008a; Cote et al., 2009). High perceived stress has also been observed to increase the risk for neck and upper limb symptoms (Bongers et al., 2006), but more knowledge is warranted (Christensen and Knardahl, 2014). In a recent longitudinal study, Fanavoll et al. (2016) observed associations between high work stress (i.e. ‘Does your work involve a lot of stress and hassles?’) and risk of chronic neck/shoulder pain in the general working population. The evidence for effects of treatment for neck pain is still modest, but neck exercises, strength training, and shoulder-blade muscle training may be beneficial (Gross et al., 2015). Several studies have shown that implementing short daily exercises in the workplace can be beneficial, e.g. group-based exercises 10 min per day (Jakobsen et al., 2015) or neck and shoulder exercises 2–20 min per day (Sjögren et al., 2005; Andersen et al., 2011; Lidegaard et al., 2013). However, the treatment and intervention results are predominantly short-term effects (Gross et al., 2015), and to achieve a more constant decrease of pain and disability, there is need for a better understanding on how treatment, working conditions, and other factors in combination affect the outcomes. In a recent randomized controlled trial (RCT) on neck pain rehabilitation, the two groups receiving active treatment programs improved significantly more regarding neck function and pain compared with the treatment-as-usual (TAU) control group, 1 week after intervention (Svedmark et al., 2016). However, the same study failed to show treatment effects in long-term follow-ups, which is in line with the conclusion that effective treatments of neck pain that lasts are rare (Cheng et al., 2015; Gross et al., 2015). Better knowledge on how work-related factors and perceived stress impact the long-term outcomes after rehabilitation may contribute to effective ways of preventing the high recurrence of neck pain. The primary aim of this study was to evaluate if pain, disability, and work productivity were influenced by physical and psychosocial workplace exposures as well as by stress measured directly after 11 weeks of rehabilitation. A control group without specific intervention was also included. Follow-ups were done at 9 and 15 months after start of intervention. A secondary aim was to determine whether there were any significant group differences between treatment and control groups regarding associations of exposure (physical and psychosocial workplace exposure and stress) and outcomes (pain, disability, and work productivity). Methods Study design and participants This longitudinal study is based on an RCT study that included three groups: tailored treatment (TT), non-tailored treatment (NTT), and a control group (TAU; Svedmark et al., 2016). Participants were recruited for the RCT by advertisements in local papers and local networks at Umea municipality that included the hospital and university. In that study, TT and NTT received 11 weeks of neck and shoulder rehabilitation that was led by a physiotherapist. The rehabilitation contained evidence-based treatments for non-specific neck pain including cervical range of motion exercises and manual cervical mobilization, strength, and endurance training for neck and shoulder muscles, head and neck motor control exercises, posture exercises, and electromyography biofeedback program for the trapezius muscle. The maximum number of treatment sessions was 27 during the 11-week intervention period. The TAU group did not receive any treatment from the study but were free to seek care themselves. After the intervention, this was the case for the TT and NTT groups as well. In the RCT (Svedmark et al., 2016), no significant differences were found between the TT and NTT groups regarding pain and disability at any of the follow-ups. For work productivity, small but significant differences were revealed between the treatment groups at 9 and 15 months. However, the significance of these small differences was reduced because of different baseline values and a possible regression-towards-mean effect (Svedmark et al., 2016). In the present longitudinal study, TT and NTT are pooled to one group and denoted as the treatment group. Thus, this present longitudinal study included a treatment group (TT and NTT) and a control group (TAU). Altogether, there were 97 working women aged 20–65 years with non-specific neck pain. Inclusion criteria were pain in the neck-shoulder region for a minimum of 6 weeks, disabilities due to neck symptoms—neck disability index (NDI) score ≥10% (more than no disability) and ≤68% (less than complete disability; Vernon and Mior, 1991) and self-reported work productivity loss (quality or quantity) due to neck symptoms (Martimo et al., 2009). Exclusion criteria were as follows: trauma-related neck pain, cervical radiculopathy or vestibular dysfunction; comorbid medical conditions such as cancer, type 1 diabetes, heart disease, rheumatic disease (including fibromyalgia), anxiety or depression, concurrent low back pain, temporomandibular disorders, surgery or spinal fracture, and severely restricted shoulder range of motion; catastrophizing thoughts; low treatment expectation; or substantial changes in physical or psychosocial work environment from the start of intervention to the 3-, 9-, or 15-month follow-ups (questionnaires). The definition for substantial changes in physical or psychosocial work environment was (i) small work environment changes in combination with major deterioration or markedly improved workload or (ii) large work environment changes in combination with major/somewhat deterioration or markedly/improved workload. Participant exclusion was based on criteria that were self-reported, first in a telephone interview with a study-administrator and second during an inclusion questionnaire screening, but the specific diagnosis should have been made by a medical doctor. A physiotherapy assessment complemented this at baseline measurement if there was uncertainty about exclusion for fibromyalgia, cervical radiculopathy, or vestibular dysfunction. Inclusions and exclusions are presented in detail in Fig. 1. The study was approved by the regional ethical review board in Uppsala, Sweden (No 2011/081). Participants were informed about the study and were provided written information that was signed indicating their consent prior to participation. Figure. 1. View largeDownload slide Flow diagram of the study. WP, work productivity; QEC, quick exposure check; QPS Nordic, the Nordic questionnaires for psychological and social factors at work. Asterisk indicates that data for two participants who scored NDI ≤ 10% and eight participants who reported no work productivity loss (quality or quantity) were excluded in analyses regarding these variables. Figure. 1. View largeDownload slide Flow diagram of the study. WP, work productivity; QEC, quick exposure check; QPS Nordic, the Nordic questionnaires for psychological and social factors at work. Asterisk indicates that data for two participants who scored NDI ≤ 10% and eight participants who reported no work productivity loss (quality or quantity) were excluded in analyses regarding these variables. Outcomes Outcomes (neck pain, neck disability, and work productivity) were assessed before the intervention period (baseline), after intervention (3 months), and at 9 (intermediate) and 15 months (long-term) after the start of intervention (Fig. 1). Participants rated average neck pain intensity during the last week with the 11-point Numeric Rating Scale (NRS) (Dworkin et al., 2005), where 0 = ‘no pain’ and 10 = ‘worst pain imaginable’. Furthermore, neck disability was assessed with the NDI (0–100; Vernon and Mior, 1991). NDI consists of 10 items addressing functional activities in daily life such as personal care, lifting, car driving, reading, work, and symptoms like pain intensity, headache, and concentration. Higher scores mean more disability. Finally, self-estimated impact of neck symptoms on work productivity was assessed with two questions: (i) ‘assess the impact of your neck symptoms last month and mark with a scale from 0 (practically nothing) to 10 (regular quantity)—how much work you were able to perform when compared with your normal workday’, (ii) ‘assess the impact of your neck symptoms last month and mark with a scale from 0 (very poor quality) to 10 (regular quality)—the quality of your work when compared with your normal workday (Martimo et al., 2009)’. Physical work exposure Three months after the start of intervention (comparable time in controls), an experienced ergonomist visited participants at their workplace and assessed the physical exposure with the ergonomic risk assessment tool Quick Exposure Check (QEC; David et al., 2008). The most common work task, as defined by the participant, was observed and assessed. According to the QEC, the ergonomist registered participant’s head posture and frequency of bent or rotated head, hand position in the horizontal plane, and frequency of shoulder/arm movements during work. The participant estimated the maximum weight of tools they handled, how many hours per day they performed the observed work task, and whether the task was visually demanding. The observed exposures were calculated and transformed to the total scores for neck load (range 4–18) and the total scores for shoulder/arm load (range 10–56). The QEC has shown good validity and moderate repeatability in similar contexts (David et al., 2008; Takala, 2010). Psychosocial work exposure At the workplace visit, participants filled in ‘The Nordic questionnaires for psychological and social factors at work’ (QPS Nordic; Wännström et al., 2009). Seven scales (36 questions) were included from the original questionnaire—‘quantitative demands’, ‘decision demands’, ‘learning demands’, ‘control of decision’, ‘control of work pacing’, ‘support from superior’, and ‘support from co-workers’. The response scale ranged from 1–5 (1 = ‘very seldom or never’, 5 = ‘very often or always’), and an index score was calculated for each scale by summing the answers from the individual questions and then dividing by the number of questions. The QPS Nordic is a psychometrically tested questionnaire and is considered to be of good quality (Wännström et al., 2009). Job stress and perceived stress To measure whether participants experienced their job as stressful, a specific question from the QEC assessment was included, i.e. ‘In general, how do you find your job?’ The response scale ranged from 1–4 (1 = ‘not at all stressful’, 4 = ‘very stressful’). Participants answered this question at the workplace visit after the intervention period. To measure the perceived stress level, a single-item question was answered (Elo et al., 2003). The question was phrased ‘Stress means a situation in which a person feels tense, restless, nervous or anxious or is unable to sleep at night because his/her mind is troubled all the time. Have you felt this kind of stress during the last month?’ The response alternatives ranged from 1 = ‘not at all’ to 5 = ‘very much’. Data analysis Levels and variability of outcomes, confounders, and potential risk factors were presented for normally distributed (or approximately normal) variables as means and standard deviations, for log-normal distributed variables as geometric means and geometric standard deviations, and otherwise as quartile levels. Confounding variables between outcomes and potential risk factors were determined a priori. Age and pain duration (at baseline) were considered potential confounders based on previous reviews (Walton et al., 2013; Bruls et al., 2015). First, the cross-sectional associations between outcomes and potential risk factor exposures were evaluated at the time of acquisition of risk factor and outcome data (3 months after baseline). Univariate associations between outcomes, confounders, and potential risk factor exposures, one at a time, were estimated using linear regression for the outcome neck disability and using log-linear regression for the outcomes neck pain and work productivity. Based on these univariate analyses, candidate risk factors with P-values <0.25 were included in a multiple linear regression model. Thereafter, a backwards stepwise approach was used to identify the most important risk factors among those identified in the univariate analyses. In each step, estimates and standard errors were recorded to gain understanding of specific exposure adjustments. The limit of the P-value in the stepwise procedure was 0.25. Secondly, longitudinal assessments were conducted using the risk factor exposure as predictor of outcomes at 3, 9, and 15 months after baseline. Mixed models were used to estimate longitudinal associations, accounting for within-individual correlation of repeated outcome measures by incorporation of a random intercept. The same method for variable selection as for the cross-sectional analyses was used to construct multiple longitudinal models. In these longitudinal models, interaction terms between time and risk factors were excluded before their corresponding main effects. The correlations between potential risk factors were calculated to assess potential multicollinearity. All statistical analyses were performed using the statistical software package SPSS 22.0 for Windows (SPSS Inc., Chicago, IL, USA). The significant level was set at P < 0.05. Results The participants had various occupations and employers with 80% working for the municipality in academia or in the healthcare sector. The most common work task was computer work (67%), e.g. administration, and the second most common was direct patient-related work (18%), e.g. nurse, assistant nurse, physiotherapist, or in dental care. The mean age was 47.4 years in the treatment group and 49.1 years in the TAU group, and the median duration of neck pain before entering the study was 60 months in both groups (Table 1). The scores of the QEC and QPS Nordic scales and the levels of stress are presented for the treatment group and the TAU group in Table 1. Table 1. Characteristics of participants and exposure levels. N represents the numbers of participants presented as mean values and standard deviations. Characteristics and exposure Treatment group (n = 67) TAU (n = 30) Age (years) 47.4 (11.7) 49.1 (9.7) BMI (kg/m2)* 24 (22–26.5)a 24 (22–28) Pain duration* (month) 60 (24–144) 60 (24–120) Neck load* (score range 4–18) 16 (14–16) 16 (13.5–16) Shoulder/arm load* (score range 10–56) 26 (22–32) 26 (26–34) Quantitative demands (score range 1–5) 3.0 (0.6) 2.9 (0.6) Decision demands (score range 1–5) 3.3 (0.6) 3.3 (0.7) Learning demands (score range 1–5) 2.5 (0.5)b 2.4 (0.7) Control of decision (score range 1–5) 2.9 (0.7) 2.9 (0.8) Control of work pacing (score range 1–5) 3.3 (1.1) 3.2 (1.2) Support from superior (score range 1–5) 3.7 (0.8)b 3.7 (0.8) Support from co-workers (score range 1–5) 4.0 (0.7)b 4.2 (0.8) Job stress (score range 1–4) 2.5 (0.8) 2.6 (0.9) Perceived stress (score range 1–5) 2.4 (1.0)b 2.6 (1.1)c Characteristics and exposure Treatment group (n = 67) TAU (n = 30) Age (years) 47.4 (11.7) 49.1 (9.7) BMI (kg/m2)* 24 (22–26.5)a 24 (22–28) Pain duration* (month) 60 (24–144) 60 (24–120) Neck load* (score range 4–18) 16 (14–16) 16 (13.5–16) Shoulder/arm load* (score range 10–56) 26 (22–32) 26 (26–34) Quantitative demands (score range 1–5) 3.0 (0.6) 2.9 (0.6) Decision demands (score range 1–5) 3.3 (0.6) 3.3 (0.7) Learning demands (score range 1–5) 2.5 (0.5)b 2.4 (0.7) Control of decision (score range 1–5) 2.9 (0.7) 2.9 (0.8) Control of work pacing (score range 1–5) 3.3 (1.1) 3.2 (1.2) Support from superior (score range 1–5) 3.7 (0.8)b 3.7 (0.8) Support from co-workers (score range 1–5) 4.0 (0.7)b 4.2 (0.8) Job stress (score range 1–4) 2.5 (0.8) 2.6 (0.9) Perceived stress (score range 1–5) 2.4 (1.0)b 2.6 (1.1)c BMI body mass index. an = 65. bn = 66. cn = 29. *For variables that had a skewed distribution, the median and the first and third quartiles are presented. View Large Table 1. Characteristics of participants and exposure levels. N represents the numbers of participants presented as mean values and standard deviations. Characteristics and exposure Treatment group (n = 67) TAU (n = 30) Age (years) 47.4 (11.7) 49.1 (9.7) BMI (kg/m2)* 24 (22–26.5)a 24 (22–28) Pain duration* (month) 60 (24–144) 60 (24–120) Neck load* (score range 4–18) 16 (14–16) 16 (13.5–16) Shoulder/arm load* (score range 10–56) 26 (22–32) 26 (26–34) Quantitative demands (score range 1–5) 3.0 (0.6) 2.9 (0.6) Decision demands (score range 1–5) 3.3 (0.6) 3.3 (0.7) Learning demands (score range 1–5) 2.5 (0.5)b 2.4 (0.7) Control of decision (score range 1–5) 2.9 (0.7) 2.9 (0.8) Control of work pacing (score range 1–5) 3.3 (1.1) 3.2 (1.2) Support from superior (score range 1–5) 3.7 (0.8)b 3.7 (0.8) Support from co-workers (score range 1–5) 4.0 (0.7)b 4.2 (0.8) Job stress (score range 1–4) 2.5 (0.8) 2.6 (0.9) Perceived stress (score range 1–5) 2.4 (1.0)b 2.6 (1.1)c Characteristics and exposure Treatment group (n = 67) TAU (n = 30) Age (years) 47.4 (11.7) 49.1 (9.7) BMI (kg/m2)* 24 (22–26.5)a 24 (22–28) Pain duration* (month) 60 (24–144) 60 (24–120) Neck load* (score range 4–18) 16 (14–16) 16 (13.5–16) Shoulder/arm load* (score range 10–56) 26 (22–32) 26 (26–34) Quantitative demands (score range 1–5) 3.0 (0.6) 2.9 (0.6) Decision demands (score range 1–5) 3.3 (0.6) 3.3 (0.7) Learning demands (score range 1–5) 2.5 (0.5)b 2.4 (0.7) Control of decision (score range 1–5) 2.9 (0.7) 2.9 (0.8) Control of work pacing (score range 1–5) 3.3 (1.1) 3.2 (1.2) Support from superior (score range 1–5) 3.7 (0.8)b 3.7 (0.8) Support from co-workers (score range 1–5) 4.0 (0.7)b 4.2 (0.8) Job stress (score range 1–4) 2.5 (0.8) 2.6 (0.9) Perceived stress (score range 1–5) 2.4 (1.0)b 2.6 (1.1)c BMI body mass index. an = 65. bn = 66. cn = 29. *For variables that had a skewed distribution, the median and the first and third quartiles are presented. View Large The levels of outcome variables for the treatment and the control groups at baseline, 3-, 9-, and 15-month follow-ups, respectively, are presented in Table 2. Both neck pain and neck disability decreased more in the treatment group compared with the control group from baseline to the 3-month follow-up. Table 2. Levels of outcome variables at different time points. Neck disability presented as means values with standard deviations within parentheses, neck pain and work productivity (quantity and quality) presented as geometric means and geometric standard deviations within parentheses. Baseline 3-month follow-up 9-month follow-up 15-month follow-up Neck pain (0–10) Treatment group 5.06 (1.49) 2.97 (1.77) 3.38 (1.81) 3.12 (1.95) Control group 5.25 (1.43) 4.44 (1.67) 3.99 (1.69) 4.11 (1.66) Neck disability (0–100) Treatment group 22.18 (8.16) 13.17 (9.38) 14.99 (9.83) 13.51 (9.83) Control group 23.8 (8.41) 21.48 (11.42) 19.45 (12.94) 17.27 (9.71) Work productivity (quantity; 0–10) Treatment group 8.65 (1.25) 8.87 (1.27) 8.87 (1.21) 8.91 (1.20) Control group 8.55 (1.29) 8.76 (1.28) 8.81 (1.22) 8.87 (1.20) Work productivity (quality; 0–10) Treatment group 8.45 (1.23) 8.78 (1.29) 8.78 (1.26) 8.85 (1.24) Control group 8.34 (1.25) 8.66 (1.30) 8.72 (1.30) 8.81 (1.29) Baseline 3-month follow-up 9-month follow-up 15-month follow-up Neck pain (0–10) Treatment group 5.06 (1.49) 2.97 (1.77) 3.38 (1.81) 3.12 (1.95) Control group 5.25 (1.43) 4.44 (1.67) 3.99 (1.69) 4.11 (1.66) Neck disability (0–100) Treatment group 22.18 (8.16) 13.17 (9.38) 14.99 (9.83) 13.51 (9.83) Control group 23.8 (8.41) 21.48 (11.42) 19.45 (12.94) 17.27 (9.71) Work productivity (quantity; 0–10) Treatment group 8.65 (1.25) 8.87 (1.27) 8.87 (1.21) 8.91 (1.20) Control group 8.55 (1.29) 8.76 (1.28) 8.81 (1.22) 8.87 (1.20) Work productivity (quality; 0–10) Treatment group 8.45 (1.23) 8.78 (1.29) 8.78 (1.26) 8.85 (1.24) Control group 8.34 (1.25) 8.66 (1.30) 8.72 (1.30) 8.81 (1.29) View Large Table 2. Levels of outcome variables at different time points. Neck disability presented as means values with standard deviations within parentheses, neck pain and work productivity (quantity and quality) presented as geometric means and geometric standard deviations within parentheses. Baseline 3-month follow-up 9-month follow-up 15-month follow-up Neck pain (0–10) Treatment group 5.06 (1.49) 2.97 (1.77) 3.38 (1.81) 3.12 (1.95) Control group 5.25 (1.43) 4.44 (1.67) 3.99 (1.69) 4.11 (1.66) Neck disability (0–100) Treatment group 22.18 (8.16) 13.17 (9.38) 14.99 (9.83) 13.51 (9.83) Control group 23.8 (8.41) 21.48 (11.42) 19.45 (12.94) 17.27 (9.71) Work productivity (quantity; 0–10) Treatment group 8.65 (1.25) 8.87 (1.27) 8.87 (1.21) 8.91 (1.20) Control group 8.55 (1.29) 8.76 (1.28) 8.81 (1.22) 8.87 (1.20) Work productivity (quality; 0–10) Treatment group 8.45 (1.23) 8.78 (1.29) 8.78 (1.26) 8.85 (1.24) Control group 8.34 (1.25) 8.66 (1.30) 8.72 (1.30) 8.81 (1.29) Baseline 3-month follow-up 9-month follow-up 15-month follow-up Neck pain (0–10) Treatment group 5.06 (1.49) 2.97 (1.77) 3.38 (1.81) 3.12 (1.95) Control group 5.25 (1.43) 4.44 (1.67) 3.99 (1.69) 4.11 (1.66) Neck disability (0–100) Treatment group 22.18 (8.16) 13.17 (9.38) 14.99 (9.83) 13.51 (9.83) Control group 23.8 (8.41) 21.48 (11.42) 19.45 (12.94) 17.27 (9.71) Work productivity (quantity; 0–10) Treatment group 8.65 (1.25) 8.87 (1.27) 8.87 (1.21) 8.91 (1.20) Control group 8.55 (1.29) 8.76 (1.28) 8.81 (1.22) 8.87 (1.20) Work productivity (quality; 0–10) Treatment group 8.45 (1.23) 8.78 (1.29) 8.78 (1.26) 8.85 (1.24) Control group 8.34 (1.25) 8.66 (1.30) 8.72 (1.30) 8.81 (1.29) View Large Cross-sectional associations, univariate analyses Associations between exposures (physical, psychosocial, and stress) and outcomes (neck pain, neck disability, and work productivity) at 3 months for the treatment and the control groups are presented in Table 3. High ‘control of decision’ and a high ‘control of work pacing’ were associated with lower neck disability. Also, high ‘control of decision’ was associated with lower neck pain and higher work productivity (quantity and quality). Both high perceived stress level last month and high job stress were associated with more neck pain, higher neck disability, and lower work productivity (quantity and quality). Finally, a high shoulder/arm load was related to lower work productivity (quantity). Table 3. Univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (neck pain, neck disability, work productivity). Cross-sectional analysis at 3 months after the start of intervention. Neck pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate (95% CI) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure Neck load 3% (−2; 8%) 0.31 (−0.56; 1.18) −1% (−3; 1%) −0.4% (−3; 2%) Shoulder/arm load 0.3% (−1.3; 1.9%) −0.08 (−0.38; 0.22) −1% (−2; −0.2%)* 0.6% (−1.3; 0.1) Quantitative demands 6% (−12; 27%) 1.29 (−2.09; 4.67) −1.4% (−9; 7%) −4% (−13; 4%) Decision demands 9% (−9; 31%) 1.36 (−1.97; 4.68) −2% (−10; 6%) −1% (−9; 7%) Learning demands 11% (−30; 4%) −1.20 (−4.86; 2.45) 7% (−3; 16%) 5% (−5; 15%) Control of decision −17% (−29; −2%)* −4.69 (−7.57; −1.8)** 9% (2; 16%)** 9% (1.4; 16%)* Control of work pacing −8% (−17; 3%) −2.21 (−4.09; −0.32)* 4% (−1; 8%) 4% (−1; 9%) Support from superior −2% (−16; 3%) −1.12 (−3.9; 1.66) −1% (−8; 6%) 2% (−5; 9%) Support from co-workers −12% (−26; 5%) −0.82 (−4.02; 2.37) 0.4% (−7; 8%) 0% (−8.2; 8.2 %) Job stress 19% (3; 37%)** 4.14 (1.58; 6.7)*** −7% (−13; −3%)* −9% (−15; −3%)** Perceived stress 17% (4; 31%)** 3.59 (1.54; 5.62)*** −6% (−11; −0.4)* −9% (−14; −4%)*** Pain duration 0% (−0.00; 0.002) 0.013 (−0.009; 0.035) 0% (0.00; 0.001%) 0% (0.00; 0.001%) Age 2% (−0.7; 12%) 0.87 (−0.94; 2.7) 0.6% (−4; 5%) −2% (−5; 4%) Neck pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate (95% CI) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure Neck load 3% (−2; 8%) 0.31 (−0.56; 1.18) −1% (−3; 1%) −0.4% (−3; 2%) Shoulder/arm load 0.3% (−1.3; 1.9%) −0.08 (−0.38; 0.22) −1% (−2; −0.2%)* 0.6% (−1.3; 0.1) Quantitative demands 6% (−12; 27%) 1.29 (−2.09; 4.67) −1.4% (−9; 7%) −4% (−13; 4%) Decision demands 9% (−9; 31%) 1.36 (−1.97; 4.68) −2% (−10; 6%) −1% (−9; 7%) Learning demands 11% (−30; 4%) −1.20 (−4.86; 2.45) 7% (−3; 16%) 5% (−5; 15%) Control of decision −17% (−29; −2%)* −4.69 (−7.57; −1.8)** 9% (2; 16%)** 9% (1.4; 16%)* Control of work pacing −8% (−17; 3%) −2.21 (−4.09; −0.32)* 4% (−1; 8%) 4% (−1; 9%) Support from superior −2% (−16; 3%) −1.12 (−3.9; 1.66) −1% (−8; 6%) 2% (−5; 9%) Support from co-workers −12% (−26; 5%) −0.82 (−4.02; 2.37) 0.4% (−7; 8%) 0% (−8.2; 8.2 %) Job stress 19% (3; 37%)** 4.14 (1.58; 6.7)*** −7% (−13; −3%)* −9% (−15; −3%)** Perceived stress 17% (4; 31%)** 3.59 (1.54; 5.62)*** −6% (−11; −0.4)* −9% (−14; −4%)*** Pain duration 0% (−0.00; 0.002) 0.013 (−0.009; 0.035) 0% (0.00; 0.001%) 0% (0.00; 0.001%) Age 2% (−0.7; 12%) 0.87 (−0.94; 2.7) 0.6% (−4; 5%) −2% (−5; 4%) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Table 3. Univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (neck pain, neck disability, work productivity). Cross-sectional analysis at 3 months after the start of intervention. Neck pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate (95% CI) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure Neck load 3% (−2; 8%) 0.31 (−0.56; 1.18) −1% (−3; 1%) −0.4% (−3; 2%) Shoulder/arm load 0.3% (−1.3; 1.9%) −0.08 (−0.38; 0.22) −1% (−2; −0.2%)* 0.6% (−1.3; 0.1) Quantitative demands 6% (−12; 27%) 1.29 (−2.09; 4.67) −1.4% (−9; 7%) −4% (−13; 4%) Decision demands 9% (−9; 31%) 1.36 (−1.97; 4.68) −2% (−10; 6%) −1% (−9; 7%) Learning demands 11% (−30; 4%) −1.20 (−4.86; 2.45) 7% (−3; 16%) 5% (−5; 15%) Control of decision −17% (−29; −2%)* −4.69 (−7.57; −1.8)** 9% (2; 16%)** 9% (1.4; 16%)* Control of work pacing −8% (−17; 3%) −2.21 (−4.09; −0.32)* 4% (−1; 8%) 4% (−1; 9%) Support from superior −2% (−16; 3%) −1.12 (−3.9; 1.66) −1% (−8; 6%) 2% (−5; 9%) Support from co-workers −12% (−26; 5%) −0.82 (−4.02; 2.37) 0.4% (−7; 8%) 0% (−8.2; 8.2 %) Job stress 19% (3; 37%)** 4.14 (1.58; 6.7)*** −7% (−13; −3%)* −9% (−15; −3%)** Perceived stress 17% (4; 31%)** 3.59 (1.54; 5.62)*** −6% (−11; −0.4)* −9% (−14; −4%)*** Pain duration 0% (−0.00; 0.002) 0.013 (−0.009; 0.035) 0% (0.00; 0.001%) 0% (0.00; 0.001%) Age 2% (−0.7; 12%) 0.87 (−0.94; 2.7) 0.6% (−4; 5%) −2% (−5; 4%) Neck pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate (95% CI) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure Neck load 3% (−2; 8%) 0.31 (−0.56; 1.18) −1% (−3; 1%) −0.4% (−3; 2%) Shoulder/arm load 0.3% (−1.3; 1.9%) −0.08 (−0.38; 0.22) −1% (−2; −0.2%)* 0.6% (−1.3; 0.1) Quantitative demands 6% (−12; 27%) 1.29 (−2.09; 4.67) −1.4% (−9; 7%) −4% (−13; 4%) Decision demands 9% (−9; 31%) 1.36 (−1.97; 4.68) −2% (−10; 6%) −1% (−9; 7%) Learning demands 11% (−30; 4%) −1.20 (−4.86; 2.45) 7% (−3; 16%) 5% (−5; 15%) Control of decision −17% (−29; −2%)* −4.69 (−7.57; −1.8)** 9% (2; 16%)** 9% (1.4; 16%)* Control of work pacing −8% (−17; 3%) −2.21 (−4.09; −0.32)* 4% (−1; 8%) 4% (−1; 9%) Support from superior −2% (−16; 3%) −1.12 (−3.9; 1.66) −1% (−8; 6%) 2% (−5; 9%) Support from co-workers −12% (−26; 5%) −0.82 (−4.02; 2.37) 0.4% (−7; 8%) 0% (−8.2; 8.2 %) Job stress 19% (3; 37%)** 4.14 (1.58; 6.7)*** −7% (−13; −3%)* −9% (−15; −3%)** Perceived stress 17% (4; 31%)** 3.59 (1.54; 5.62)*** −6% (−11; −0.4)* −9% (−14; −4%)*** Pain duration 0% (−0.00; 0.002) 0.013 (−0.009; 0.035) 0% (0.00; 0.001%) 0% (0.00; 0.001%) Age 2% (−0.7; 12%) 0.87 (−0.94; 2.7) 0.6% (−4; 5%) −2% (−5; 4%) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Cross-sectional associations, multiple analyses All cross-sectional multiple regression analyses models were adjusted for the confounders age and pain duration. ‘Control of decision’ and perceived stress were the only statistically significant risk factors for neck pain, neck disability, and work productivity (quality), i.e. neck pain (‘control of decision’ −19%, 95% CI; −36 to −4%, perceived stress 17%, 95% CI; 6 to 29%), neck disability (‘control of decision’, estimate −4.53, 95% CI; −7.32 to −1.76, perceived stress, estimate 3.53, 95% CI; 1.27 to 5.81), and work productivity (‘control of decision’ 11%, 95% CI; 4 to 18%, perceived stress −10%, 95% CI; −15 to −5%). ‘Control of decision’ (9%, 95% CI; 2 to 16%) was the only statistically significant risk factor for work productivity (quantity). Longitudinal associations, univariate analyses Associations between exposures (physical, psychosocial, and stress) and outcomes (neck pain, neck disability, and work productivity) at 3, 9, and 15 months are presented in Tables 4 and 5. Higher perceived stress was associated with higher neck pain and higher neck disability and lower work productivity (quality) at all follow-ups. Higher job stress was associated with higher neck disability at all follow-ups and with higher neck pain and lower work productivity (quality) at 3 and 9 months, and work productivity (quantity) at 3 months. For the psychosocial work exposure factors, higher ‘control of decision’ was associated with lower neck disability at all follow-ups and higher work productivity (quality and quantity) at 3 months. Higher ‘quantitative demands’ was associated with higher neck pain at 9 months. For the physical work exposure factors, only a higher shoulder/arm load was associated with lower work productivity (quantity) at 3 months. Table 4. Longitudinal univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (neck pain, neck disability; linear mixed model analyses). Neck Pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate(95%CI) 3 months 9 months 15 months 3 months 9 months 15 months Neck load 3%(−16; 8%) 4% (−1; 9%) 2% (−3; 6%) 0.5 (−0.3; 1.3) 0.5 (−0.3; 1.4) 0.3 (−0.5; 1.1) Shoulder/arm load 0.3% (−1; 2%) −1% (−2; 1%) −1% (−2; 1%) −0.09 (−0.4; 0.2) 0.02 (−0.3; 0.3) −0.01 (−0.3; 0.3) Quantitative demands 10% (−8; 32%) 22% (2; 46%)* 6% (−12; 27%) 1.9 (−1.2; 5.1) 3.1 (−0.1; 6.3) 1.8 (−1.3; 5) Decision demands 14% (−5; 36%) 3% (−14; 23%) 2% (−15; 22%) 2.3 (−0.8; 5.5) 0.4 (−2.7; 3.6) 1.4 (−1.8; 4.6) Learning demands −9% (−26; 13%) 1% (−8; 25%) −3% (−22; 21%) 0.14 (−3.5; 3.8) 2.1 (−1.5; 5.9) 3.2 (−0.5; 7.1) Control of decision −14% (−27; 1%) −7% (−21; 9%) −13% (−26; 2%) −4.2 (−7.0; −1.5)** −3.9 (−6.6; −1.1)** −2.9 (−5.7; −0.1)* Control of work pacing −6% (−16; 4%) −2% (−8; 14%) −5% (−14; 6%) −1.9 (−3.6; −0.04)* −1.6 (−3.4; 0.1) −1.8 (−3.7; −0.2)* Support from superior −3% (−6; 12%) −9% (−21; 6%) −0.6% (−14; 16%) −0.8 (−3.4; 1.7) −1.6 (−4.1; 0.9) 0.3 (−2.2; 2.9) Support from co-workers −12% (−26; 3%) −10% (−23; 6%) −14% (−30; 1%) −0.6 (−3.5; 2.3) 0.41 (−2.4; 3.3) 0.5 (−2.4; 3.3) Job stress 18% (3; 36%)* 16% (1; 35%)* 9% (−6; 25%) 4.0 (1.6; 6.5)*** 4.07 (1.6; 6.5)*** 3.1 (0.6; 5.6)** Perceived stress 15% (2; 28%)* 15% (3; 29%)* 16% (3; 31%)* 3.4 (1.5; 5.3)*** 3.6 (1.6; 5.5)*** 3.1 (1.1; 5.6)** Pain duration 3% (−7; 13%) 12% (1; 23%)* 5% (−5; 16%) 1.04 (−0.7; 2.8) 1.4 (−0.3; 3.2) 0.9 (−0.8; 2.7) Age 0.4% (−1; 2%) 0.6% (−0.5; 2%) 0.3% (−1; 1%) 0.10 (−0.09; 0.3) 0.10 (−0.09; 0.3) 0.04 (−0.15; 0.2) Neck Pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate(95%CI) 3 months 9 months 15 months 3 months 9 months 15 months Neck load 3%(−16; 8%) 4% (−1; 9%) 2% (−3; 6%) 0.5 (−0.3; 1.3) 0.5 (−0.3; 1.4) 0.3 (−0.5; 1.1) Shoulder/arm load 0.3% (−1; 2%) −1% (−2; 1%) −1% (−2; 1%) −0.09 (−0.4; 0.2) 0.02 (−0.3; 0.3) −0.01 (−0.3; 0.3) Quantitative demands 10% (−8; 32%) 22% (2; 46%)* 6% (−12; 27%) 1.9 (−1.2; 5.1) 3.1 (−0.1; 6.3) 1.8 (−1.3; 5) Decision demands 14% (−5; 36%) 3% (−14; 23%) 2% (−15; 22%) 2.3 (−0.8; 5.5) 0.4 (−2.7; 3.6) 1.4 (−1.8; 4.6) Learning demands −9% (−26; 13%) 1% (−8; 25%) −3% (−22; 21%) 0.14 (−3.5; 3.8) 2.1 (−1.5; 5.9) 3.2 (−0.5; 7.1) Control of decision −14% (−27; 1%) −7% (−21; 9%) −13% (−26; 2%) −4.2 (−7.0; −1.5)** −3.9 (−6.6; −1.1)** −2.9 (−5.7; −0.1)* Control of work pacing −6% (−16; 4%) −2% (−8; 14%) −5% (−14; 6%) −1.9 (−3.6; −0.04)* −1.6 (−3.4; 0.1) −1.8 (−3.7; −0.2)* Support from superior −3% (−6; 12%) −9% (−21; 6%) −0.6% (−14; 16%) −0.8 (−3.4; 1.7) −1.6 (−4.1; 0.9) 0.3 (−2.2; 2.9) Support from co-workers −12% (−26; 3%) −10% (−23; 6%) −14% (−30; 1%) −0.6 (−3.5; 2.3) 0.41 (−2.4; 3.3) 0.5 (−2.4; 3.3) Job stress 18% (3; 36%)* 16% (1; 35%)* 9% (−6; 25%) 4.0 (1.6; 6.5)*** 4.07 (1.6; 6.5)*** 3.1 (0.6; 5.6)** Perceived stress 15% (2; 28%)* 15% (3; 29%)* 16% (3; 31%)* 3.4 (1.5; 5.3)*** 3.6 (1.6; 5.5)*** 3.1 (1.1; 5.6)** Pain duration 3% (−7; 13%) 12% (1; 23%)* 5% (−5; 16%) 1.04 (−0.7; 2.8) 1.4 (−0.3; 3.2) 0.9 (−0.8; 2.7) Age 0.4% (−1; 2%) 0.6% (−0.5; 2%) 0.3% (−1; 1%) 0.10 (−0.09; 0.3) 0.10 (−0.09; 0.3) 0.04 (−0.15; 0.2) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Table 4. Longitudinal univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (neck pain, neck disability; linear mixed model analyses). Neck Pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate(95%CI) 3 months 9 months 15 months 3 months 9 months 15 months Neck load 3%(−16; 8%) 4% (−1; 9%) 2% (−3; 6%) 0.5 (−0.3; 1.3) 0.5 (−0.3; 1.4) 0.3 (−0.5; 1.1) Shoulder/arm load 0.3% (−1; 2%) −1% (−2; 1%) −1% (−2; 1%) −0.09 (−0.4; 0.2) 0.02 (−0.3; 0.3) −0.01 (−0.3; 0.3) Quantitative demands 10% (−8; 32%) 22% (2; 46%)* 6% (−12; 27%) 1.9 (−1.2; 5.1) 3.1 (−0.1; 6.3) 1.8 (−1.3; 5) Decision demands 14% (−5; 36%) 3% (−14; 23%) 2% (−15; 22%) 2.3 (−0.8; 5.5) 0.4 (−2.7; 3.6) 1.4 (−1.8; 4.6) Learning demands −9% (−26; 13%) 1% (−8; 25%) −3% (−22; 21%) 0.14 (−3.5; 3.8) 2.1 (−1.5; 5.9) 3.2 (−0.5; 7.1) Control of decision −14% (−27; 1%) −7% (−21; 9%) −13% (−26; 2%) −4.2 (−7.0; −1.5)** −3.9 (−6.6; −1.1)** −2.9 (−5.7; −0.1)* Control of work pacing −6% (−16; 4%) −2% (−8; 14%) −5% (−14; 6%) −1.9 (−3.6; −0.04)* −1.6 (−3.4; 0.1) −1.8 (−3.7; −0.2)* Support from superior −3% (−6; 12%) −9% (−21; 6%) −0.6% (−14; 16%) −0.8 (−3.4; 1.7) −1.6 (−4.1; 0.9) 0.3 (−2.2; 2.9) Support from co-workers −12% (−26; 3%) −10% (−23; 6%) −14% (−30; 1%) −0.6 (−3.5; 2.3) 0.41 (−2.4; 3.3) 0.5 (−2.4; 3.3) Job stress 18% (3; 36%)* 16% (1; 35%)* 9% (−6; 25%) 4.0 (1.6; 6.5)*** 4.07 (1.6; 6.5)*** 3.1 (0.6; 5.6)** Perceived stress 15% (2; 28%)* 15% (3; 29%)* 16% (3; 31%)* 3.4 (1.5; 5.3)*** 3.6 (1.6; 5.5)*** 3.1 (1.1; 5.6)** Pain duration 3% (−7; 13%) 12% (1; 23%)* 5% (−5; 16%) 1.04 (−0.7; 2.8) 1.4 (−0.3; 3.2) 0.9 (−0.8; 2.7) Age 0.4% (−1; 2%) 0.6% (−0.5; 2%) 0.3% (−1; 1%) 0.10 (−0.09; 0.3) 0.10 (−0.09; 0.3) 0.04 (−0.15; 0.2) Neck Pain (NRS)a change (95% CI) for one-unit increase in exposure Neck disability (NDI) slope estimate(95%CI) 3 months 9 months 15 months 3 months 9 months 15 months Neck load 3%(−16; 8%) 4% (−1; 9%) 2% (−3; 6%) 0.5 (−0.3; 1.3) 0.5 (−0.3; 1.4) 0.3 (−0.5; 1.1) Shoulder/arm load 0.3% (−1; 2%) −1% (−2; 1%) −1% (−2; 1%) −0.09 (−0.4; 0.2) 0.02 (−0.3; 0.3) −0.01 (−0.3; 0.3) Quantitative demands 10% (−8; 32%) 22% (2; 46%)* 6% (−12; 27%) 1.9 (−1.2; 5.1) 3.1 (−0.1; 6.3) 1.8 (−1.3; 5) Decision demands 14% (−5; 36%) 3% (−14; 23%) 2% (−15; 22%) 2.3 (−0.8; 5.5) 0.4 (−2.7; 3.6) 1.4 (−1.8; 4.6) Learning demands −9% (−26; 13%) 1% (−8; 25%) −3% (−22; 21%) 0.14 (−3.5; 3.8) 2.1 (−1.5; 5.9) 3.2 (−0.5; 7.1) Control of decision −14% (−27; 1%) −7% (−21; 9%) −13% (−26; 2%) −4.2 (−7.0; −1.5)** −3.9 (−6.6; −1.1)** −2.9 (−5.7; −0.1)* Control of work pacing −6% (−16; 4%) −2% (−8; 14%) −5% (−14; 6%) −1.9 (−3.6; −0.04)* −1.6 (−3.4; 0.1) −1.8 (−3.7; −0.2)* Support from superior −3% (−6; 12%) −9% (−21; 6%) −0.6% (−14; 16%) −0.8 (−3.4; 1.7) −1.6 (−4.1; 0.9) 0.3 (−2.2; 2.9) Support from co-workers −12% (−26; 3%) −10% (−23; 6%) −14% (−30; 1%) −0.6 (−3.5; 2.3) 0.41 (−2.4; 3.3) 0.5 (−2.4; 3.3) Job stress 18% (3; 36%)* 16% (1; 35%)* 9% (−6; 25%) 4.0 (1.6; 6.5)*** 4.07 (1.6; 6.5)*** 3.1 (0.6; 5.6)** Perceived stress 15% (2; 28%)* 15% (3; 29%)* 16% (3; 31%)* 3.4 (1.5; 5.3)*** 3.6 (1.6; 5.5)*** 3.1 (1.1; 5.6)** Pain duration 3% (−7; 13%) 12% (1; 23%)* 5% (−5; 16%) 1.04 (−0.7; 2.8) 1.4 (−0.3; 3.2) 0.9 (−0.8; 2.7) Age 0.4% (−1; 2%) 0.6% (−0.5; 2%) 0.3% (−1; 1%) 0.10 (−0.09; 0.3) 0.10 (−0.09; 0.3) 0.04 (−0.15; 0.2) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Table 5. Longitudinal univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (work productivity; linear mixed model analyses). Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure 3 months 9 months 15 months 3 months 9 months 15 months Neck load −1.2% (−3; 0.5%) −1% (−3; 0.8%) −0.5% (−2; −1%) −0.4% (−2; –2) −0.9% (−3; 1%) −0.9% (−3; 1%) Shoulder/arm load −0.8% (−1.4; −0.2%)** −0.2% (−0.8; 0.3%) 1% (−0.4; 0.7%) −0.6% (−1; 0.1) −0.2% (−1; 0.5%) −0.1% (−1; −0.6%) Quantitative demands −2% (−9; 5%) −6% (−12; 2%) −4% (−11; 2%) −5% (−13; 2.8%) −7% (−15; 0.8) −5% (−13; 2.7) Decision demands −3% (−10%; 4%) −3% (−9; 4%) −3% (−10; 4%) −2% (−10; 6 %) 1.5% (−6; 9%) −2% (−10; 6%) Learning demands 6% (−2; 14%) −3% (−11; 5%) −2% (−10; 7%) 3% (−6; 12%) −0.8% (−10; 8 %) −3% (−13; 6%) Control of decision 8% (3; 14%)** 1% (−5; 7%) 2% (−5; 7%) 8% (1.7; 16%)* 2% (−5; 9%) 6% (−1.5; 12%) Control of work pacing 3% (−0.5; 7%) −0.2% (−5; 4%) 0.1% (−4; 4%) 4% (−9; 8%) 0.5% (−4; 5%) 3% (−1; 7%) Support from superior −0.4% (−6; 5%) 3% (−2; 9%) 5% (−0.7; 11%) 3% (−4; 10%) 2% (−4; 9%) 3% (−4; 10%) Support from co-workers 1% (−5; 8%) −1% (−8; 6%) 2% (−5; 9%) 1% (−6; 9%) −0.3% (−8; 7%) −3% (−11; 5%) Job stress −7% (−12; −1%)* −1% (−7; 4%) −3% (−8; 3%) −9% (−15; −3%)** −6% (−13; −0.3%)* −5% (−12; 0.9%) Perceived stress −5% (−9; −0.6)* −3% (−7; 1%) −4% (−8; 1%) −8% (−13; −3%)** −5% (−10; −0.1%)* −5% (−10; −3%)* Pain duration 0.5% (−3; 4%) −0.6% (−4; 3%) −0.7% (−4; 3%) −0.1% (−4; 4%) 2% (−2; 6%) 1% (−4; 5%) Age 0.2% (−0.2; 0.6%) 0.1% (−0.3; 5%) 0.3% (−0.8; 0.1) 0.3% (−0.2; 1%) 0.1% (−0.4; 0.6%) −0.1% (−0.6; 0.4%) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure 3 months 9 months 15 months 3 months 9 months 15 months Neck load −1.2% (−3; 0.5%) −1% (−3; 0.8%) −0.5% (−2; −1%) −0.4% (−2; –2) −0.9% (−3; 1%) −0.9% (−3; 1%) Shoulder/arm load −0.8% (−1.4; −0.2%)** −0.2% (−0.8; 0.3%) 1% (−0.4; 0.7%) −0.6% (−1; 0.1) −0.2% (−1; 0.5%) −0.1% (−1; −0.6%) Quantitative demands −2% (−9; 5%) −6% (−12; 2%) −4% (−11; 2%) −5% (−13; 2.8%) −7% (−15; 0.8) −5% (−13; 2.7) Decision demands −3% (−10%; 4%) −3% (−9; 4%) −3% (−10; 4%) −2% (−10; 6 %) 1.5% (−6; 9%) −2% (−10; 6%) Learning demands 6% (−2; 14%) −3% (−11; 5%) −2% (−10; 7%) 3% (−6; 12%) −0.8% (−10; 8 %) −3% (−13; 6%) Control of decision 8% (3; 14%)** 1% (−5; 7%) 2% (−5; 7%) 8% (1.7; 16%)* 2% (−5; 9%) 6% (−1.5; 12%) Control of work pacing 3% (−0.5; 7%) −0.2% (−5; 4%) 0.1% (−4; 4%) 4% (−9; 8%) 0.5% (−4; 5%) 3% (−1; 7%) Support from superior −0.4% (−6; 5%) 3% (−2; 9%) 5% (−0.7; 11%) 3% (−4; 10%) 2% (−4; 9%) 3% (−4; 10%) Support from co-workers 1% (−5; 8%) −1% (−8; 6%) 2% (−5; 9%) 1% (−6; 9%) −0.3% (−8; 7%) −3% (−11; 5%) Job stress −7% (−12; −1%)* −1% (−7; 4%) −3% (−8; 3%) −9% (−15; −3%)** −6% (−13; −0.3%)* −5% (−12; 0.9%) Perceived stress −5% (−9; −0.6)* −3% (−7; 1%) −4% (−8; 1%) −8% (−13; −3%)** −5% (−10; −0.1%)* −5% (−10; −3%)* Pain duration 0.5% (−3; 4%) −0.6% (−4; 3%) −0.7% (−4; 3%) −0.1% (−4; 4%) 2% (−2; 6%) 1% (−4; 5%) Age 0.2% (−0.2; 0.6%) 0.1% (−0.3; 5%) 0.3% (−0.8; 0.1) 0.3% (−0.2; 1%) 0.1% (−0.4; 0.6%) −0.1% (−0.6; 0.4%) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Table 5. Longitudinal univariate associations between exposures (physical, psychosocial, stress), confounders (pain duration, age), and outcomes (work productivity; linear mixed model analyses). Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure 3 months 9 months 15 months 3 months 9 months 15 months Neck load −1.2% (−3; 0.5%) −1% (−3; 0.8%) −0.5% (−2; −1%) −0.4% (−2; –2) −0.9% (−3; 1%) −0.9% (−3; 1%) Shoulder/arm load −0.8% (−1.4; −0.2%)** −0.2% (−0.8; 0.3%) 1% (−0.4; 0.7%) −0.6% (−1; 0.1) −0.2% (−1; 0.5%) −0.1% (−1; −0.6%) Quantitative demands −2% (−9; 5%) −6% (−12; 2%) −4% (−11; 2%) −5% (−13; 2.8%) −7% (−15; 0.8) −5% (−13; 2.7) Decision demands −3% (−10%; 4%) −3% (−9; 4%) −3% (−10; 4%) −2% (−10; 6 %) 1.5% (−6; 9%) −2% (−10; 6%) Learning demands 6% (−2; 14%) −3% (−11; 5%) −2% (−10; 7%) 3% (−6; 12%) −0.8% (−10; 8 %) −3% (−13; 6%) Control of decision 8% (3; 14%)** 1% (−5; 7%) 2% (−5; 7%) 8% (1.7; 16%)* 2% (−5; 9%) 6% (−1.5; 12%) Control of work pacing 3% (−0.5; 7%) −0.2% (−5; 4%) 0.1% (−4; 4%) 4% (−9; 8%) 0.5% (−4; 5%) 3% (−1; 7%) Support from superior −0.4% (−6; 5%) 3% (−2; 9%) 5% (−0.7; 11%) 3% (−4; 10%) 2% (−4; 9%) 3% (−4; 10%) Support from co-workers 1% (−5; 8%) −1% (−8; 6%) 2% (−5; 9%) 1% (−6; 9%) −0.3% (−8; 7%) −3% (−11; 5%) Job stress −7% (−12; −1%)* −1% (−7; 4%) −3% (−8; 3%) −9% (−15; −3%)** −6% (−13; −0.3%)* −5% (−12; 0.9%) Perceived stress −5% (−9; −0.6)* −3% (−7; 1%) −4% (−8; 1%) −8% (−13; −3%)** −5% (−10; −0.1%)* −5% (−10; −3%)* Pain duration 0.5% (−3; 4%) −0.6% (−4; 3%) −0.7% (−4; 3%) −0.1% (−4; 4%) 2% (−2; 6%) 1% (−4; 5%) Age 0.2% (−0.2; 0.6%) 0.1% (−0.3; 5%) 0.3% (−0.8; 0.1) 0.3% (−0.2; 1%) 0.1% (−0.4; 0.6%) −0.1% (−0.6; 0.4%) Work productivity (quantity)a change (95% CI) for one-unit increase in exposure Work productivity (quality)a change (95% CI) for one-unit increase in exposure 3 months 9 months 15 months 3 months 9 months 15 months Neck load −1.2% (−3; 0.5%) −1% (−3; 0.8%) −0.5% (−2; −1%) −0.4% (−2; –2) −0.9% (−3; 1%) −0.9% (−3; 1%) Shoulder/arm load −0.8% (−1.4; −0.2%)** −0.2% (−0.8; 0.3%) 1% (−0.4; 0.7%) −0.6% (−1; 0.1) −0.2% (−1; 0.5%) −0.1% (−1; −0.6%) Quantitative demands −2% (−9; 5%) −6% (−12; 2%) −4% (−11; 2%) −5% (−13; 2.8%) −7% (−15; 0.8) −5% (−13; 2.7) Decision demands −3% (−10%; 4%) −3% (−9; 4%) −3% (−10; 4%) −2% (−10; 6 %) 1.5% (−6; 9%) −2% (−10; 6%) Learning demands 6% (−2; 14%) −3% (−11; 5%) −2% (−10; 7%) 3% (−6; 12%) −0.8% (−10; 8 %) −3% (−13; 6%) Control of decision 8% (3; 14%)** 1% (−5; 7%) 2% (−5; 7%) 8% (1.7; 16%)* 2% (−5; 9%) 6% (−1.5; 12%) Control of work pacing 3% (−0.5; 7%) −0.2% (−5; 4%) 0.1% (−4; 4%) 4% (−9; 8%) 0.5% (−4; 5%) 3% (−1; 7%) Support from superior −0.4% (−6; 5%) 3% (−2; 9%) 5% (−0.7; 11%) 3% (−4; 10%) 2% (−4; 9%) 3% (−4; 10%) Support from co-workers 1% (−5; 8%) −1% (−8; 6%) 2% (−5; 9%) 1% (−6; 9%) −0.3% (−8; 7%) −3% (−11; 5%) Job stress −7% (−12; −1%)* −1% (−7; 4%) −3% (−8; 3%) −9% (−15; −3%)** −6% (−13; −0.3%)* −5% (−12; 0.9%) Perceived stress −5% (−9; −0.6)* −3% (−7; 1%) −4% (−8; 1%) −8% (−13; −3%)** −5% (−10; −0.1%)* −5% (−10; −3%)* Pain duration 0.5% (−3; 4%) −0.6% (−4; 3%) −0.7% (−4; 3%) −0.1% (−4; 4%) 2% (−2; 6%) 1% (−4; 5%) Age 0.2% (−0.2; 0.6%) 0.1% (−0.3; 5%) 0.3% (−0.8; 0.1) 0.3% (−0.2; 1%) 0.1% (−0.4; 0.6%) −0.1% (−0.6; 0.4%) CI confidence interval. aLogarithmically transformed. *P < 0.05; **P < 0.01; ***P < 0.001. View Large Longitudinal associations, multiple analyses In the longitudinal multiple regression analysis, all models were adjusted for the confounders age and pain duration. Statistically significant risk factors for neck pain in the multiple model were ‘control of decision’ (3 months −20%, 95% CI; −36 to −3%) and perceived stress (3 months 18%, 95% CI; 6 to 30%, 9 months 16%, 95% CI; 5 to 28%, 15 months 18%, 95% CI; 6 to 30%). ‘Support from supervisor’ was close to significant (main effect 13%, 95% CI; −0.5 to 26%), and the association was positive, i.e. a high ‘support from supervisor’ was associated to more neck pain. Statistically significant risk factors for neck disability were ‘control of decision’ (3-month estimate −4.57, 95% CI; −7.29 to −1.86, 9-month estimate −4.14, 95% CI; −6.85 to −1.42, and 15-month estimate −3.09, 95% CI; −5.84 to −0.33) and perceived stress (3-month estimate 3.76, 95% CI; 1.6 to 5.92, 9-month estimate 3.78, 95% CI; 1.61 to 5.95, and 15-month estimate 3.11, 95% CI; 0.93 to 5.28). The only statistically significant risk factor for work productivity (quantity) in the multiple models was perceived stress at 3 months (−5%, 95% CI; −10 to −1%). Statistically significant risk factors for work productivity (quality) were ‘control of decision’ (main effect 7%, 95% CI; 2 to 12%) and perceived stress (3 months −9%, 95% CI; −14 to −4%, 9 months −6% 95% CI; −11 to −0.6%, and 15 months −6%, 95% CI; −11 to −1%). Consistency of results between treatment and control groups One physical work exposure factor, shoulder/arm load at 3 months, showed a small but significant difference between groups (P = 0.04); shoulder/arm load was significantly associated with work productivity (quantity) in the treatment group (−1.7%, 95% CI; −3.2 to −0.29%) but not in the control group (−0.6%, 95% CI; −1.7 to 3.1%). Also, one psychosocial work exposure factor, ‘quantitative demands’ at the 9-month follow-up differed significantly between groups (P = 0.04); ‘quantitative demands’ was statistically significant associated with neck disability in the control group (estimate 4.97, 95% CI; 1.3 to 8.65) but not in the treatment group (estimate −2.55, 95% CI; −8.99 to 3.88). There was no difference between groups regarding stress. Discussion Our longitudinal multiple regression analyses showed that perceived stress influenced neck pain, neck disability, and work productivity, and ‘control of decision’ influenced neck disability and work productivity (quality). These effects were evident at the intermediate (9 month) and the long-term (15 month) follow-ups. The only physical exposure related to outcomes was high shoulder/arm load that was associated with reduced work productivity (quantity) in the univariate analyses. These results imply that interventions focusing on the individual should consider workplace exposure/s and perceived stress since these exposures will influence the result both in the intermediate and long-term perspective. Our results indicate that handling the psychosocial factors and stress may be more important than the physical work exposures in order to optimize long-term treatment results in neck pain rehabilitation. Physical workplace exposure Our finding that only one physical exposure, i.e. shoulder/arm load, seemed important for work productivity (quantity) is surprising; the change in one-unit increase in shoulder/arm load was associated with only −1% of decreased work productivity. This rather weak influence differs from other studies that show a stronger relationship between high shoulder/arm load and neck and shoulder disorders (Larsson et al., 2007; Nordander et al., 2016). Similarly, other studies have shown that high neck load, such as with awkward posture (i.e. prolonged neck flexion), is associated with neck and shoulder pain (Sterud et al., 2014). This was, however, not investigated in our study. Nevertheless, the observation instrument that we used, QEC, addressed high neck load values (median 16 of maximum 18), but no associations to pain, disability, or work productivity were found. One explanation for this lack of associations could be the relatively low variation in neck load exposure since a lack of variance in the exposure variable will affect the possibility to observe associations. The low variance in exposure could partly be explained by the high proportion of computer workers (67%) who probably have more static work postures. Prior to our study, we piloted the QEC in staff in a hospital setting, and the tool revealed variations in exposure both between occupations and between workers (Ericsson et al., 2012). However, our pilot study included more occupations than in the sample of the present study. Despite the fact that QEC has shown good validity and moderate repeatability (David et al., 2008; Takala, 2010), the proposed categories in the instrument seem arbitrarily chosen, and longitudinal studies addressing exposure–response associations with musculoskeletal disorders are lacking. Psychosocial workplace exposure The QPS Nordic questionnaire analyses (cross-sectional) revealed that the ‘control of decision’ was statistically significant for all three outcomes—neck pain, neck disability, and work productivity. Also, in the longitudinal analysis, ‘control of decision’ was significantly associated with neck disability and work productivity but not with neck pain. In contrast, in two longitudinal studies (2 and 4 years in duration) using QPS Nordic, Christensen and Knardahl (2010, 2014) found that high-middle ‘control of decision’ had associations to neck pain. The comparison is complex due to different choices in the analyses. Christensen et al. (Christensen and Knardahl, 2010; Christensen and Knardahl, 2014) used category data such as high-middle ‘control of decision’ and ‘troubled by’ pain (1–4), and we used the index score of ‘control of decision’ (1–5) and pain intensity scale (NRS 0–10). Furthermore, the different follow-up times may limit the comparison with the results found in Christensen and Knardahl studies. In addition to ‘control of decision’, we found that high ‘quantitative demands’ was associated with more neck pain and more neck disability. ‘High ‘quantitative demands’ and ‘low ‘control of decision’, also commonly known as ‘job demands’ and ‘decision latitude’ (Karasek et al., 1998), are concluded to be psychosocial workplace factors with associations to neck pain (Bongers et al., 2006) even though the associations are not very strong or very specific We have not found any other study with a similar design, i.e. longitudinal follow-ups after a neck pain intervention that address the impact of workplace exposure and stress on the outcome of the rehabilitation. Perceived stress and job stress In a cross-sectional study, Jay et al. (Jay et al., 2015) investigated the associations between combined perceived stress (The Perceived Stress scale) and neck/shoulder pain with work ability among working women. They found that high stress and more pain were independently associated with lower work ability. This is in line with our findings that high perceived stress was associated with lower self-rated work ability. In addition to the previous study (Jay et al., 2015), we found that perceived stress and work productivity were associated in intermediate and long-term follow-ups as well. Recently, a longitudinal study (Lindegård et al., 2014) investigated associations between perceived stress (Elo et al., 2003), musculoskeletal pain (neck and low back), and work productivity in healthcare workers. The authors found that frequent musculoskeletal pain in combination with perceived stress was associated with lower work ability; however, perceived stress alone was not clearly associated with work ability. In our study, we used two stress questions—one regarding job stress (QEC) and one regarding perceived stress last month (Elo et al., 2003). These questions had a rather high correlation (0.53) although they measure different aspects. The job stress question is related to workplace situations, and the perceived stress question is related to individuals’ perceived stress as an effect of a stressful job or other stressful life events such as a divorce, a sick child, or economic difficulties. The univariate analysis revealed that both measures of stress were associated with neck pain, disability, and work productivity after intervention, and for the intermediate and the long-term follow-ups. However, the multiple analyses revealed that only perceived stress was associated with neck pain, disability, and work productivity in the intermediate and the long-term follow-ups. Therefore, based on our results, it seems that perceived stress in full is a more important factor than job stress. Differences in treatment group and controls Only small differences between treatment and control group regarding associations of exposure and outcomes were observed. This indicates that the RCT intervention did not have an impact on self-estimated psychosocial work exposure, job stress, and perceived stress. That was not the aim with the previous RCT, but the results from this longitudinal study highlight the importance in accounting for work and stress factors in future trials for neck pain. Time points Time points for evaluations were set at 3, 9, and 15 months following the start of intervention and corresponded to frequently used alternatives representing short, intermediate, and long-term follow-ups after finalized treatment. These time points were thus considered as adequate in the RCT, which was the base for our study material (Svedmark et al., 2016). However, for the purpose of the present study, it could have been desirable with an even longer follow-up time than 15 months, but the risk of further attrition with time due to changes for the employees and the work environment made us refrain from an extended time for evaluation. Strength and limitations of the study In general, individuals with non-specific neck pain are regarded as a heterogeneous group with respect to underlying mechanisms and presentation of pain and disability (Cote et al., 2008; Carroll et al., 2008b). The purpose of the exclusion and inclusion criteria in the RCT study, on which the present study was based, was to make the study group more homogenous in order to fit the objectives of the RCT (Bjorklund et al., 2012; Svedmark et al., 2016). Thus, participants with depression or anxiety symptoms, catastrophizing thoughts, and low treatment expectation were excluded. This should be considered with regard to generalization of findings from the present study. The outcomes together with psychosocial work exposure and stress levels were self-reported questionnaires or questions. Even though the employed instruments have been proven valid, self-reporting always constitutes a challenge due to potential misunderstanding of the questions or careless mistakes because of lack of time or other circumstances. The focus of the present study was on exposures at the workplace while non-work exposures were not specifically addressed. Exposures from outside of work could have been potential confounders. Since the present study used data from a RCT, the sample size was determined with power calculations for primary outcomes of the RCT (Bjorklund et al., 2012; Svedmark et al., 2016) and not primarily for the present study. Despite this, there were consistent findings with regard to the impact of exposure on neck pain rehabilitation outcomes. We believe that the present study approach helped to identify factors crucial for sustainable rehabilitation results. Conclusion High perceived stress and low ‘control of decision’ were associated with more neck pain and increased neck disability but also to decreased work productivity in women after a rehabilitation intervention. Our findings therefore strongly suggest that neck rehabilitation interventions (e.g. manual therapy, exercises) focusing on the individual should take workplace exposures and perceived stress into account since these exposures could influence the treatment results both in the intermediate and the long-term perspectives. Funding The project was funded by the Swedish Council for Working Life and Social Research (registration number 2009–1403), AFA Insurance (registration number 090288). The funders have no other role in the study other than to provide funding. Conflict of Interest The authors declare that they have no competing interests. Acknowledgements Majken Rahm is acknowledged for valuable contribution to administrative routine and logistics, Pernilla Jansson for data collection, and Göran Sandström for database support. References Andersen LL , Saervoll CA , Mortensen OS , et al. ( 2011 ) Effectiveness of small daily amounts of progressive resistance training for frequent neck/shoulder pain: randomised controlled trial . Pain ; 152 : 440 – 6 . Google Scholar CrossRef Search ADS PubMed Björklund M , Djupsjöbacka M , Svedmark A , et al. ( 2012 ) Effects of tailored neck-shoulder pain treatment based on a decision model guided by clinical assessments and standardized functional tests. A study protocol of a randomized controlled trial . BMC Musculoskelet Disord ; 13 : 75 . Google Scholar CrossRef Search ADS PubMed Bongers PM , Ijmker S , van den Heuvel S , et al. ( 2006 ) Epidemiology of work related neck and upper limb problems: psychosocial and personal risk factors (Part I) and effective interventions from a bio behavioural perspective (Part II) . J Occup Rehabil ; 16 : 279 – 302 . Google Scholar CrossRef Search ADS PubMed Borghouts JA , Koes BW , Vondeling H , et al. ( 1999 ) Cost-of-illness of neck pain in The Netherlands in 1996 . Pain ; 80 : 629 – 36 . Google Scholar CrossRef Search ADS PubMed Bruls VE , Bastiaenen CH , de Bie RA . ( 2015 ) Prognostic factors of complaints of arm, neck, and/or shoulder: a systematic review of prospective cohort studies . Pain ; 156 : 765 – 88 . Google Scholar CrossRef Search ADS PubMed Carroll LJ , et al. ( 2008a ) Course and prognostic factors for neck pain in workers: results of the Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders . Spine (Phila Pa 1976) ; 33 ( Suppl. 4 ): S93 – 100 . Google Scholar CrossRef Search ADS Carroll LJ , et al. ( 2008b ) Course and prognostic factors for neck pain in the general population: results of the Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders . Spine (Phila Pa 1976) ; 33 ( Suppl. 4 ): S75 – 82 . Google Scholar CrossRef Search ADS Cheng CH , Su HT , Yen LW , et al. ( 2015 ) Long-term effects of therapeutic exercise on nonspecific chronic neck pain: a literature review . J Phys Ther Sci ; 27 : 1271 – 6 . Google Scholar CrossRef Search ADS PubMed Childs JD , Cleland JA , Elliott JM , et al. ; American Physical Therapy Association . ( 2008 ) Neck pain: clinical practice guidelines linked to the International Classification of Functioning, Disability, and Health from the Orthopedic Section of the American Physical Therapy Association . J Orthop Sports Phys Ther ; 38 : A1 – 34 . Google Scholar CrossRef Search ADS PubMed Christensen JO , Knardahl S . ( 2010 ) Work and neck pain: a prospective study of psychological, social, and mechanical risk factors . Pain ; 151 : 162 – 73 . Google Scholar CrossRef Search ADS PubMed Christensen JO , Knardahl S . ( 2014 ) Time-course of occupational psychological and social factors as predictors of new-onset and persistent neck pain: a three-wave prospective study over 4 years . Pain ; 155 : 1262 – 71 . Google Scholar CrossRef Search ADS PubMed Cote P , et al. ( 2008 ) The burden and determinants of neck pain in workers: results of the Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders . Spine (Phila Pa 1976) ; 33 ( Suppl. 4 ): S60 – 74 . Google Scholar CrossRef Search ADS PubMed Côté P , van der Velde G , Cassidy JD , et al. ( 2009 ) The burden and determinants of neck pain in workers: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders . J Manipulative Physiol Ther ; 32 ( Suppl. 2 ): S70 – 86 . Google Scholar CrossRef Search ADS PubMed David G , Woods V , Li G , et al. ( 2008 ) The development of the Quick Exposure Check (QEC) for assessing exposure to risk factors for work-related musculoskeletal disorders . Appl Ergon ; 39 : 57 – 69 . Google Scholar CrossRef Search ADS PubMed Dworkin RH , Turk DC , Farrar JT , et al. ; IMMPACT . ( 2005 ) Core outcome measures for chronic pain clinical trials: IMMPACT recommendations . Pain ; 113 : 9 – 19 . Google Scholar CrossRef Search ADS PubMed Elo AL , Leppänen A , Jahkola A . ( 2003 ) Validity of a single-item measure of stress symptoms . Scand J Work Environ Health ; 29 : 444 – 51 . Google Scholar CrossRef Search ADS PubMed Ericsson P , Bjorklund M , Wahlstrom J ( 2012 ) Exposure assessment in different occupational groups at a hospital using Quick Exposure Check (QEC)—a pilot study . Work ; 41 ( Suppl. 1 ): 5718 – 20 . Google Scholar PubMed Fanavoll R , Nilsen TI , Holtermann A , et al. ( 2016 ) Psychosocial work stress, leisure time physical exercise and the risk of chronic pain in the neck/shoulders: longitudinal data from the Norwegian HUNT Study . Int J Occup Med Environ Health ; 29 : 585 – 95 . Google Scholar CrossRef Search ADS PubMed Gross A , Kay TM , Paquin JP , et al. ; Cervical Overview Group . ( 2015 ) Exercises for mechanical neck disorders . Cochrane Database Syst Rev ; 1 : CD004250 . Google Scholar PubMed Hauke A , Flintrop J , Brun E , Rugulies R ( 2011 ) The impact of work-related psychosocial stressors on the onset of musculoskeletal disorders in specific body regions: a review and meta-analysis of 54 longitudinal studies . Work & Stress ; 25 : 243 – 56 . Google Scholar CrossRef Search ADS Hogg-Johnson S , van der Velde G , Carroll LJ , et al. ; Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders . ( 2008 ) The burden and determinants of neck pain in the general population: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders . Spine (Phila Pa 1976) ; 33 ( Suppl. 4 ): S39 – 51 . Google Scholar CrossRef Search ADS PubMed Hoy D , et al. ( 2014 ) The global burden of neck pain: estimates from the global burden of disease 2010 study . Ann Rheum Dis ; 73 : 1309 – 15 . Google Scholar CrossRef Search ADS PubMed Jakobsen MD , Sundstrup E , Brandt M , et al. ( 2015 ) Effect of workplace- versus home-based physical exercise on musculoskeletal pain among healthcare workers: a cluster randomized controlled trial . Scand J Work Environ Health ; 41 : 153 – 63 . Google Scholar CrossRef Search ADS PubMed Jay K , Friborg MK , Sjøgaard G , et al. ( 2015 ) The consequence of combined pain and stress on work ability in female laboratory technicians: a cross-sectional study . Int J Environ Res Public Health ; 12 : 15834 – 42 . Google Scholar CrossRef Search ADS PubMed Karasek R , Brisson C , Kawakami N , et al. ( 1998 ) The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics . J Occup Health Psychol ; 3 : 322 – 55 . Google Scholar CrossRef Search ADS PubMed Lang J , Ochsmann E , Kraus T , et al. ( 2012 ) Psychosocial work stressors as antecedents of musculoskeletal problems: a systematic review and meta-analysis of stability-adjusted longitudinal studies . Soc Sci Med ; 75 : 1163 – 74 . Google Scholar CrossRef Search ADS PubMed Larsson B , Søgaard K , Rosendal L . ( 2007 ) Work related neck-shoulder pain: a review on magnitude, risk factors, biochemical characteristics, clinical picture and preventive interventions . Best Pract Res Clin Rheumatol ; 21 : 447 – 63 . Google Scholar CrossRef Search ADS PubMed Lidegaard M , Jensen RB , Andersen CH , et al. ( 2013 ) Effect of brief daily resistance training on occupational neck/shoulder muscle activity in office workers with chronic pain: randomized controlled trial . Biomed Res Int ; 2013 : 262386 . Google Scholar CrossRef Search ADS PubMed Lindegård A , Larsman P , Hadzibajramovic E , et al. ( 2014 ) The influence of perceived stress and musculoskeletal pain on work performance and work ability in Swedish health care workers . Int Arch Occup Environ Health ; 87 : 373 – 9 . Google Scholar CrossRef Search ADS PubMed Malchaire J , Cock N , Vergracht S . ( 2001 ) Review of the factors associated with musculoskeletal problems in epidemiological studies . Int Arch Occup Environ Health ; 74 : 79 – 90 . Google Scholar CrossRef Search ADS PubMed Martimo KP , Shiri R , Miranda H , et al. ( 2009 ) Self-reported productivity loss among workers with upper extremity disorders . Scand J Work Environ Health ; 35 : 301 – 8 . Google Scholar CrossRef Search ADS PubMed Nordander C , Hansson GÅ , Ohlsson K , et al. ( 2016 ) Exposure-response relationships for work-related neck and shoulder musculoskeletal disorders—analyses of pooled uniform data sets . Appl Ergon ; 55 : 70 – 84 . Google Scholar CrossRef Search ADS PubMed Paksaichol A , Janwantanakul P , Purepong N , et al. ( 2012 ) Office workers’ risk factors for the development of non-specific neck pain: a systematic review of prospective cohort studies . Occup Environ Med ; 69 : 610 – 8 . Google Scholar CrossRef Search ADS PubMed Palmer KT , Smedley J . ( 2007 ) Work relatedness of chronic neck pain with physical findings—a systematic review . Scand J Work Environ Health ; 33 : 165 – 91 . Google Scholar CrossRef Search ADS PubMed Sjögren T , Nissinen KJ , Järvenpää SK , et al. ( 2005 ) 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 – 28 . Google Scholar CrossRef Search ADS PubMed Sterud T , Johannessen HA , Tynes T . ( 2014 ) Work-related psychosocial and mechanical risk factors for neck/shoulder pain: a 3-year follow-up study of the general working population in Norway . Int Arch Occup Environ Health ; 87 : 471 – 81 . Google Scholar CrossRef Search ADS PubMed Svedmark Å , Djupsjöbacka M , Häger C , et al. ( 2016 ) Is tailored treatment superior to non-tailored treatment for pain and disability in women with non-specific neck pain? A randomized controlled trial . BMC Musculoskelet Disord ; 17 : 408 . Google Scholar CrossRef Search ADS PubMed Takala EP , Pehkonen I , Forsman M , et al. ( 2010 ) Systematic evaluation of observational methods assessing biomechanical exposures at work . Scand J Work Environ Health ; 36 : 3 – 24 . Google Scholar CrossRef Search ADS PubMed Wahlstedt K , Norbäck D , Wieslander G , et al. ( 2010 ) Psychosocial and ergonomic factors, and their relation to musculoskeletal complaints in the Swedish workforce . Int J Occup Saf Ergon ; 16 : 311 – 21 . Google Scholar CrossRef Search ADS PubMed Walton DM , Carroll LJ , Kasch H , et al. ; ICON . ( 2013 ) An overview of systematic reviews on prognostic factors in neck pain: results from the International Collaboration on Neck Pain (ICON) project . Open Orthop J ; 7 : 494 – 505 . Google Scholar CrossRef Search ADS PubMed Wännström I , Peterson U , Asberg M , et al. ( 2009 ) Psychometric properties of scales in the General Nordic Questionnaire for Psychological and Social Factors at Work (QPS): confirmatory factor analysis and prediction of certified long-term sickness absence . Scand J Psychol ; 50 : 231 – 44 . Google Scholar CrossRef Search ADS PubMed Vernon H , Mior S . ( 1991 ) The Neck Disability Index: a study of reliability and validity . J Manipulative Physiol Ther ; 14 : 409 – 15 . Google Scholar PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Occupational Hygiene Society. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene)Oxford University Press

Published: Mar 17, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off