Introduction
Methods
Search strategy
Selection criteria
Methodological quality assessment
Study objective | |
1. Positive if the study had a specific and clearly stated objective description | PC/CS |
Study population | |
2. Positive if the main features of the study population were described (sampling frame and distribution of the population according to age and sex) | PC/CS |
3. Positive if the participation rate is >70% (data presented) | PC/CS |
4. Positive if the response at main moment of follow up is >70% (data presented) | PC |
Exposure assessment | |
5. Positive if data are collected and presented about physical activity at work time | PC/CS |
6. Method for measuring physical activity: direct measurement and observation (+), interview or questionnaire only (−) | PC/CS |
7. Positive if more than one dimension of physical activity is assessed: duration, frequency or amplitude | PC/CS |
8. Positive if data are collected and presented about physical activity at leisure time | PC/CS |
9. Positive if data are collected and presented about a history of neck or back disorders | PC/CS |
10. Positive if the exposure assessment is blinded to disease status | CS |
Outcome assessment | |
11. Positive if data were collected for at least 1 year | PC |
12. Positive if data were collected at least every 3 months or obtained from a continuous registration system | PC |
13. Method for assessing neck or back pain: physical examination blinded to exposure status (+), self reported: specific questions relating to neck and back disability or use of manikin (+), single question (−) | PC/CS |
Analysis and data presentation | |
14. Positive if the appropriate statistical model is used (univariate or multivariate model) | PC/CS |
15. Positive if measures of association are presented (OR/RR), including 95% CIs and numbers in the analysis (totals) | PC/CS |
16. Positive if the analysis is adjusted for confounding or effect modification is studied | PC/CS |
17. Positive if the number of cases in the multivariate analysis is at least 10 times the number of independent variables in the analysis (final model) | PC/CS |
Data extraction and analysis
Strength of evidence
-
Strong evidence consistent findings in at least 50% of high-quality cohort studies.
-
Moderate evidence consistent findings in one high-quality cohort study and at least 50% of two or more high-quality cross-sectional studies or at least 50% of high-quality cross-sectional studies.
-
Limited evidence consistent findings in one high-quality cohort study or in at least 50% of two or more high-quality cross-sectional studies.
-
Conflicting evidence inconsistent findings among multiple studies.
-
No evidence when one low-quality cohort or cross-sectional study or no study provided findings for or against an association.
Results
Search strategy
Methodological quality assessment
Quality item/Study | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | Total score (%) |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Cohort study | ||||||||||||||||||
Wedderkopp et al. [47] | + | + | ? | + | + | + | + | + | + | + | − | + | + | + | + | + | 14/16 (88) | |
van den Heuvel et al. [41] | + | − | + | ? | − | − | − | + | − | + | + | + | + | + | + | + | 10/16 (62) | |
Mikkelsson et al. [33] | + | + | − | − | − | − | + | + | − | + | − | − | + | + | + | + | 9/16 (56) | |
Picavet and Schuit [37] | + | − | − | − | + | − | − | + | − | + | − | ? | + | + | + | + | 8/16 (50) | |
Hartvigsen and Christensen [17] | + | ? | ? | + | ? | − | + | ? | − | + | − | + | + | + | − | ? | 7/16 (44) | |
Cross-sectional study | ||||||||||||||||||
Auvinen et al. [4] | + | + | − | − | − | + | + | + | + | + | + | + | + | + | 11/14 (78) | |||
Auvinen et al. [5] | + | + | − | − | − | + | + | + | + | + | + | + | + | + | 11/14 (78) | |||
Heneweer et al. [20] | + | ? | − | + | − | + | + | + | + | + | + | + | + | + | 11/14 (78) | |||
Østerås et al. [36] | + | + | + | − | − | − | + | + | + | + | + | + | + | + | 11/14 (78) | |||
Björck-van Dijken et al. [6] | + | + | − | + | ? | + | + | + | ? | ? | + | + | + | + | 10/14 (71) | |||
Sjolie [39] | + | + | + | − | − | + | + | − | + | + | + | + | + | ? | 10/14 (71) | |||
Diepenmaat et al. [14] | + | ? | + | ? | − | + | ? | + | + | + | + | + | ? | + | 9/14 (64) | |||
Brown et al. [9] | + | − | + | ? | − | + | + | − | + | − | + | + | + | ? | 8/14 (57) | |||
Kujala et al. [29] | + | + | ? | − | − | + | + | + | + | + | + | − | ? | ? | 8/14 (57) | |||
Wedderkopp et al. [46] | + | + | + | − | − | + | + | + | + | − | + | − | ? | ? | 8/14 (57) | |||
Jacob et al. [24] | + | − | − | + | − | − | + | + | + | + | + | − | − | ? | 7/14 (50) | |||
Andersen et al. [3] | + | + | − | + | − | − | ? | + | + | ? | + | − | ? | ? | 6/14 (43) |
Assessment of physical activity
Study | Study design (follow-up period) | Study population and participation rate | Type and measure of physical activity | Measure of pain (recall period) | Results (level of physical activity, otherwise stated) | Conclusion |
---|---|---|---|---|---|---|
Wedder-kopp et al. [47] | Prospective cohorts (3 years) | School children ?% | Overall physical activity at least 10 h per day assessed by using MTI-accelerometer | Interviewing (the past month) | Back pain at baseline | No significant association between physical activity and neck pain |
High physical activity (HPA) | ||||||
Bivariate analysis 3.3 (0.1–72.8) | ||||||
No back pain at baseline | ||||||
High physical activity (HPA) | ||||||
Multivariate analysis 1.0 (0.2–5.1) | ||||||
Bivariate analysis 1.0 (0.1–8.5) | ||||||
Auvinen et al. [4] | Cross-sectional | School children 64% | Leisure time (outside school hours) assessed by using self-reported questionnaire | Self-reported questionnaire (the past 6 months) | Very active (>6 h of brisk physical activity per week) | No significant association between physical activity and neck pain |
0.87–1.54 (0.69–2.86) | ||||||
Active (4–6 h of brisk physical activity per week) | ||||||
0.87–1.62 (0.69–2.66) | ||||||
Moderately active (2–3 h of brisk physical activity per week) 1.00 | ||||||
Lightly active (1 h of brisk physical activity or <0.5 h of brisk physical activity together with >2 h of light or commuting physical activity per week) | ||||||
1.00–1.12 (0.64–1.68) | ||||||
Inactive (<0.5 h of brisk physical activity and <2 h of light or commuting physical activity per week) | ||||||
0.89–1.15 (0.58–2.28) | ||||||
Diepenmaat et al. [14] | Cross-sectional | School children 92% | ? (type of physical activity) assessed by using self-reported questionnaire | Self-reported questionnaire (the past month) | 0–0.5 h per day 1.00 | No significant association between physical activity and neck pain |
0.51–1.0 h per day 0.8 (0.6–1.1) | ||||||
≥1.01 h per day 0.8 (0.6–1.0) | ||||||
Mikkelsson et al. [33] | Prospective cohorts (25 years) | School children 67% | Leisure time (outside school hours) assessed by using self-reported questionnaire | Self-reported questionnaire (lifetime) | Physical activity at follow up | No significant association between physical activity and neck pain |
<1 time a week 1.00 | ||||||
1–4 times a week 1.06–1.43 (0.60–2.29) | ||||||
5–7 times a week 1.01–1.20 (0.53–2.71) | ||||||
Østerås et al. [36] | Cross-sectional | School children 85% | Leisure time (outside school hours) assessed by using self-reported questionnaire | Self-reported questionnaire (the past 4 weeks) | Low level PA (active <1 time/week) 1.00 | No significant association between physical activity and neck/upper back pain |
Medium level PA (active 1–3 times/week) | ||||||
1.04–1.42 (0.28–7.17) | ||||||
High level PA (active >3 times/week) | ||||||
0.95–3.72 (0.34–20.19) | ||||||
Van den Heuvel et al. [41] | Prospective cohorts (3 years) | Working population 87% | Leisure time assessed by using self-reported questionnaire | Self-reported questionnaire (the past year) | No walking/cycling 1.00 | No significant association between physical activity and neck pain |
Walking/cycling 10–150 min per week | ||||||
1.13 (0.95–1.35) | ||||||
Walking/cycling at least 150 min per week | ||||||
0.90 (0.66–1.21) | ||||||
Kujala et al. [29] | Cross-sectional | School children ?% | Leisure time (outside school hours) assessed by using self-reported questionnaire | Self-reported questionnaire (the preceding 12 months) | Prevalence of neck pain | No significant association between physical activity and neck pain |
Low MET 18.4% | ||||||
Middle MET 24.4% | ||||||
High MET 19.4% |
Study | Study design (follow-up period) | Study population and participation rate | Type and measure of physical activity | Measure of pain (recall period) | Results (level of physical activity, otherwise stated) | Conclusion |
---|---|---|---|---|---|---|
Heneweer et al. [20] | Cross-sectional | General population 47% | Routine daily activities and leisure time physical activity assessed by using Short Questionnaire to Assess Health enhancing physical activity (SQUASH) | Self-reported questionnaire (the previous 12 months) | Low physical activity level (not fulfilling the recommended activity level of at least 0.5 h of moderate activity per day for at least 5 days a week) | Both extremely low and high levels of physical activity were associated with an increased risk of chronic low back pain |
1.31 (1.08–1.58) | ||||||
Moderate physical activity level 1.00 | ||||||
High physical activity level (physical activity level with the highest quartile of the amount of physical activity and the performance of high intensive sport activities) | ||||||
1.22 (1.00–1.49) | ||||||
Wedderkopp et al. [47] | prospective cohorts (3 years) | School children ?% | Overall physical activity at least 10 h per day assessed by using MTI-accelerometer | Interviewing (the past month) | Back pain at baseline | The tertile with the lowest HPA had an increased odds ratio of having low back pain |
High physical activity (HPA) | ||||||
Bivariate analysis 1.1 (0.1–9.8) | ||||||
No back pain at baseline | ||||||
High physical activity (HPA) | ||||||
Multivariate analysis 4.6 (1.9–11.2) | ||||||
Bivariate analysis 4.9 (1.7–14.0) | ||||||
Auvinen et al. [5] | Cross-sectional | School children 64% | Leisure time (outside school hours) assessed by using self-reported questionnaire | Self-reported questionnaire (the past 6 months) | Very active (>6 h of brisk physical activity per week) | Being physically very active was associated with increased prevalence of low back pain |
1.16–3.93 (0.91–6.65) | ||||||
Active (4–6 h of brisk physical activity per week) | ||||||
0.94–1.61 (0.53–2.65) | ||||||
Moderately active (2–3 h of brisk physical activity per week) 1.00 | ||||||
Lightly active (1 h of brisk physical activity or <0.5 h of brisk physical activity together with >2 h of light or commuting physical activity per week) | ||||||
0.75–1.51 (0.41–2.36) | ||||||
Inactive (<0.5 h of brisk physical activity and <2 h of light or commuting physical activity per week) | ||||||
0.65–1.14 (0.29–1.93) | ||||||
Björck-van Dijken et al. [6] | Cross-sectional | General population 69% | Both work and leisure time assessed by using self-reported questionnaire | Self-reported questionnaire (lifetime) | Sitting work 1.00 | High levels of physical activity at work and low physical activity at leisure time was associated with increased prevalence of low back pain |
Light physical work 1.13 (0.95–1.35) | ||||||
Moderate heavy work 1.37 (1.14–1.65) | ||||||
Heavy work 1.46 (1.09–1.94) | ||||||
Low physical activity during leisure time in the past year | ||||||
1.16 (1.02–1.33) | ||||||
Diepenmaat et al. [14] | Cross-sectional | School children 92% | ? (type of Physical activity) assessed by using self-reported questionnaire | Self-reported questionnaire (the past month) | 0–0.5 h per day 1.00 | No significant association between physical activity and low back pain |
0.51–1.0 h per day 1.2 (0.8–1.7) | ||||||
≥1.01 h per day 1.0 (0.8–1.3) | ||||||
Mikkelsson et al. [33] | Prospective cohorts (25 years) | School children 67% | Leisure time (outside school hours) assessed by using self-reported questionnaire | Self-reported questionnaire (lifetime) | School age physical activity | Men who were physically active in adolescence were at a lower risk of recurrent low back pain |
Men | ||||||
Inactive (<2 times per week for at least 30 min per session of physical activity outside school hours) 1.00 | ||||||
Active (≥2 times per week for at least 30 min per session of physical activity outside school hours) 0.62 (0.39–0.98) | ||||||
Women | ||||||
Inactive 1.00 | ||||||
Active 0.80 (0.48–1.32) | ||||||
Physical activity at follow up | ||||||
<1 time a week 1.00 | ||||||
1–4 times a week 0.65–1.31 (0.37–2.14) | ||||||
5–7 times a week 0.54–0.88 (0.25–1.87) | ||||||
Sjolie [39] | Cross-sectional | School children 84% | Leisure time assessed by using self-reported questionnaire | Self-reported questionnaire (the preceding year) | Physical activity (hour), quartiles 0.6 (0.4–0.8) | Physical activity was significantly associated with decreased risk of low back pain, in particular with regular walking or bicycling |
Walking or bicycling 0.4 (0.2–0.8) | ||||||
Wedderkopp et al. [46] | Cross-sectional | School children 79% | Leisure time assessed by using self-reported questionnaire and the CSA accelerometer | Interviewing (the preceding month) | No significant positive or negative associations were noted between self-reported physical inactivity and low back pain (p = 0.41) | No significant association between physical activity and low back pain |
There were no associations between low back pain and the objectively measured level of physical activity (data not showed) | ||||||
Brown et al. [9] | Cross-sectional | General population (women) 99% | Leisure time assessed by using self-reported questionnaire | Self-reported questionnaire (lifetime) | PA score <5 (none or very low; equivalent to no PA or moderate PA once per week) 1.00 | Physical activity was significantly associated with decreased risk of low back pain |
PA score 5 to <15 (low to moderate; moderate PA 2–4 times or vigorous PA 1–2 times per week, or equivalent combination) | ||||||
0.83–0.91 (0.74–1.02) | ||||||
PA score 15 to <25 (moderate to high; moderate PA 5–8 times or vigorous PA 3–5 times per week, or equivalent combination) 0.76–0.85 (0.68–0.95) | ||||||
PA score ≥25 (high; moderate PA 8–13 times or vigorous PA 5–8 times per week, or equivalent combination) | ||||||
0.67–0.84 (0.58–0.95) | ||||||
Kujala et al. [29] | Cross-sectional | School children ?% | Leisure time (outside school hours) assessed by using self-reported questionnaire | Self-reported questionnaire (the preceding 12 months) | Prevalence of low back pain; | High leisure physical activity was associated with increased prevalence of low back pain |
Low MET 9.7% | ||||||
Middle MET 11.7% | ||||||
High MET 14.6% | ||||||
p = 0.022 |