Skip Navigation LinksHome > April 2011 - Volume 53 - Issue 4 > The Predictive Relationship of Physical Activity on the Inci...
Journal of Occupational & Environmental Medicine:
doi: 10.1097/JOM.0b013e31820d1633
Original Articles

The Predictive Relationship of Physical Activity on the Incidence of Low Back Pain in an Occupational Cohort

Thiese, Matthew S. PhD, MSPH; Hegmann, Kurt T. MD, MPH; Garg, Arun PhD, CPE; Porucznik, Christina PhD, MSPH; Behrens, Timothy PhD

Free Access
Article Outline
Collapse Box

Author Information

From the Department of Family and Preventive Medicine (Drs Thiese, Hegmann, and Porucznik), University of Utah, Salt Lake City, Utah; Department of Industrial and Manufacturing Engineering (Dr Garg), University of Wisconsin, Milwaukee, Wis; and Beth-El College of Nursing & Health Sciences (Dr Behrens), University of Colorado, Colorado Springs Colo.

Address correspondence to: Matthew S. Thiese, PhD, MSPH, Rocky Mountain Center for Occupational and Environmental Health, Department of Family and Preventive Medicine, University of Utah, 391 Chipeta Way Suite C, Salt Lake City, UT 84108 (matt.thiese@hsc.utah.edu).

Collapse Box

Abstract

Objective: Quantify the relationship between physical activity and development of incident low back pain (LBP).

Methods: This nested prospective cohort study utilized an objective measure of physical activity in 68 participants with 30 incident cases of LBP. Physical activity was divided into tertiles and quartiles. Univariate and multivariate relative risks and hazard ratios were calculated.

Results: Comparing highest to middle tertile of light activity demonstrated a statistically significant relative risk of 3.68 for developing incident LBP. Lowest and highest tertile of minutes of moderate/vigorous activity yielded statistically significant relative risks of 4.60 and 6.14, respectively. Multivariate analyses demonstrated similar associations.

Conclusions: Moderate amounts of physical activity were protective for the development of LBP in this cohort, after adjustment for risk factors. This nonlinear relationship suggests higher levels of activity do not confer increased LBP prevention.

Low back disorders are arguably the most common and expensive occupational health problems in developed countries.1 Low back pain (LBP) is the second most common reason for health care visits and are disproportionately expensive.2 It is estimated that more than 25% of all workers’ compensation claims in the United States are attributed to LBP.3 One study of Australian citizens found a 12-month LBP prevalence of 67.6% (95% CI, 65.5 to 69.7), and lifetime prevalence of LBP was 79.2% (95% CI, 77.3 to 81.0).4 The same study reported 6-month period prevalence of severely limiting disability due to LBP was 10.5% (95% CI, 9.2 to 11.9).4 A small proportion of individuals with LBP account for a large proportion of the costs.58 According to the World Health Organization (WHO), LBP should be one of the top occupational health problems within the WHO region of the Americas, due to the high prevalence and severity.9 Punnett et al10 conducted a review study evaluating the risk of LBP by occupational group within countries/regions as well as estimating the attributable proportion of LBP to occupational factors within WHO subregions and found significant risk of LBP in many regions and across a variety of job classifications within developed countries.

Low back pain is a health outcome that is quite distinctive, is difficult to diagnose, and has many proposed etiologies.11 It is both extraordinarily common, yet it is entirely subjective and without consensus on the cause of LBP in the vast majority of cases.58 Diagnostic imaging may attempt to find relationships between LBP and spinal structure abnormalities, but these relationships are unpredictable at best. A study by Carragee et al12 found that there was little association between magnetic resonance imaging changes and new LBP. Another review by de Vet et al8 concluded that of approximately 1200 epidemiologic studies on LBP, there was very little description of what defined a case of LBP.8 Furthermore, in the 81 studies where there were descriptions of the case definition, there was little uniformity between studies. This creates comparability problems between studies, as well as difficulties in replicating results.

There is some evidence that physical activity is an effective treatment for LBP.1319 Possible etiologic benefits of physical activity include increased blood flow, lower incidence of depression or other psychosocial issues, and both higher pain thresholds and pain tolerance. Although many studies included some aerobic physical activity as part of a battery of exercises, there are seven randomized controlled trials that appear to either solely or largely rely upon significant durations of aerobic physical activity for treatment of LBP.13,1621 Cited studies show benefits from aerobic physical activity including improvements in functional outcomes such as disability scores13,18 or measures of depression.13,1719 Aerobic exercise, particularly self-directed, is low cost, not invasive and has very low potential for adverse effects. Of two systematic reviews investigating risk factor modification for LBP one concluded that there was no quality evidence to support risk factor modification for the prevention of LBP,22 alternately the other concluded that “Only exercises provided sufficient evidence to conclude that they are an effective preventive intervention.”23(p.778) Physical activity may play a role in the prevention of LBP.

There are some disorders that are considered red flags for potentially serious low back pain conditions, including fractures, neoplasias, infections, cauda equine syndrome, or progressive neurologic deficit.11,2427

The goal of this study was to quantify the relationship between measured physical activity and the subsequent development of incident cases of LBP of any intensity nested within an established prospective cohort study. Our a priori hypothesis is that physical activity will be preventive for LBP. There are multiple possible etiological processes that would contribute to the protective effect. These include improved blood flow to painful regions of the low back combined with the mechanical motion of ambulation to improve diffusion to avascular structures including spinal disks to maintain healthy spinal structures.11 Additionally, individuals who participate in regular physical activity may have an increased pain tolerance and are therefore less likely to report mild LBP.28,29

Back to Top | Article Outline

METHODS

This prospective cohort study was approved by the University of Utah institutional review board (IRB 11889). The data for this study was collected between May 2006 and December 2007 and analyzed in 2008.

Back to Top | Article Outline

SUBJECTS

This study is a nested prospective cohort study. Participants were selected from an ongoing cohort study evaluating occupational and nonoccupational risk factors for job-related LBP. Selection criteria for participants in the original cohort included a stable workforce with no anticipated changes in jobs in the current prospective cohort study; no anticipated changes in jobs and/or workstations over the following 3 years; worker and management enthusiasm and cooperation for the study; consistent types of work; availability of jobs in low, medium, and high job physical factor exposure groups; and both male and female sex representation. A priori standards for accelerometer data are that participants must wear the accelerometer for 12 or more hours each day for at least 5 days, including one weekend day. All participants enrolled in the original cohort study at these three plants were eligible to participate in the nested physical activity study. Participants in the original cohort study were invited to participate in the nested cohort study if they had little or no exercise habits that would not be well captured by the accelerometer (eg, nonambulatory activities such as lifting weights, swimming). All participants who agreed to wear the accelerometer were given verbal and written instructions on its use and importance immediately before commencing the 7-day study period. Of these 119 were invited to participate, 14 declined to wear the accelerometer, and 27 additional participants wore the accelerometer but did not meet the a priori standards for accelerometer data. Of the 78 participants who met the a priori standards for wearing the accelerometer, 10 participants had prevalent LBP at the time of wearing the accelerometer and their pain did not resolve for more than 90 continuous days for the duration of the nested cohort study. Thus, the population for this study is 68.

Back to Top | Article Outline

ACCELEROMETERS

Participants (n = 68) wore the Actigraph accelerometers model GT1M (Actigraph LLC, Pensacola, Florida) monitor on their right hip during waking hours for a minimum continuous 7-day study period duration. Participants were instructed not to wear the accelerometer during periods of bathing and sleeping, however, the accelerometer was to be worn at all other times. Participants will be allowed to remove the accelerometer if it caused discomfort or is at risk of damage to the accelerometer.

The Actigraph is a small (2.0 × 1.6 × 0.6 in), light (0.09 lb), unobtrusive uniaxial accelerometer. Epoch length was 10 seconds, which were summed to 1-minute epoch lengths, as 1 minute epochs are standard for accelerometer field studies in adult populations.30 Minute-by-minute data were then summarized into daily averages for average (counts per minute per day) and total activity counts (counts per day) for times when the monitor was worn, and for activity durations (minutes per day) in the activity levels outlined earlier. Accelerometers have been demonstrated to be a reliable measure of ambulatory physical activity.3035

Accelerometer data were classified in two separate ways. The first was to calculate mean counts per minute, which were then divided into tertiles and quartiles. Second, time spent in light, moderate, and vigorous physical activity levels was calculated on a per minute basis. These were then divided into tertiles. To evaluate time spent in various activity levels, this study utilized modified cut-points of Matthews36 to reflect the optimal cut-points for moderate intensity activity but requiring less ambulation, as is common in the workplace setting,34,36,37 implemented at lower counts per minute, and Freedson cut points33 for moderate to vigorous activity. Caloric expenditure estimation was based on metabolic equivalent (MET) calculations. Time recorded at or below 250 counts per minute was coded inactive, counts from 251 to 760 counts per minute were interpreted as light activity (1 to 3 METs). Time recorded in the 761 to 5724 counts per minute range was interpreted to represent moderate (3 to 6 METs) activities. Time spent in activities 5725 counts per minute or more was interpreted as vigorous (>6 METs) activity. Minute-by-minute data were summarized into daily mean values for total activity counts (counts per day) for times when the monitor was worn and for activity durations (minutes per day) in the activity levels outlined earlier.

At the end of the 7-day period when the accelerometer was collected, a 7-day physical activity recall38 was administered to further assess both leisure and occupational activity components. This metric was designed to further capture details regarding the participants’ physical activity levels while they were wearing the accelerometer. Data from the questionnaires were used in conjunction with the activity diaries to validate the accelerometer data and ensure that the week of data captured by the accelerometer is representative of a typical week of activity for that participant.

Back to Top | Article Outline

COVARIATE DATA COLLECTION

Other data that may influence LBP, including potential confounders such as age, sex, tobacco use, job physical factors, personal psychosocial factors, and comorbid disease conditions such as history of diabetes mellitus, high cholesterol, and high blood pressure was collected in the baseline questionnaire or measured on the job and included in analyses. This questionnaire was individually administered using a laptop computer under the supervision of a health care professional or research assistant. These data were collected approximately 1 year before wearing the accelerometer and may have changed between the time of data collection and the beginning of the observation time in this study. Data influencing physical activity levels, including meteorological and seasonal data were recorded and analyzed to determine the impact, if any, on physical activity levels. Meteorological impacts were hopefully minimized through collection of these data at one time of the year (early summer).

Back to Top | Article Outline

HEALTH OUTCOMES

For primary outcomes, LBP recurrence was excluded and only the first incident case of LBP occurring after wearing the accelerometer was analyzed. Low back pain incidence requires a symptom free interval without treatment of at least 90 days prior to eligibility for development of a new incident case.

Back to Top | Article Outline

BASELINE ASSESSMENTS FOR DETERMINING PREVALENT CASES

For all participants in this study, the baseline-structured interview was administered by health care professionals (eg, occupational medicine residents). This interview emphasizes information related to the number, clinical course, and severity of historical episodes of LBP. It does not contain job-related information.

Following the structured interview, the trained health professionals performed a standardized physical examination on all participants. At baseline, all participants also received a second physical examination by an experienced, board certified occupational medicine physicians to verify the initial physical examination findings.

Back to Top | Article Outline

ASSESSMENTS DURING THE FOLLOW-UP PERIOD

During the observation period a research team consisting of a research assistant and a health care professional visited the participating plant monthly for follow-ups including focused standardized physical examination. Thus, every month, workers with new or altered low back symptoms received physical examinations. Preliminary information about the onset of new symptoms changes in previously reported symptoms, new injuries, other relevant changes in health status, and changes in contact information were ascertained monthly via a monthly symptom questionnaire, which was administered on a laptop computer by the research team. Monthly health outcomes assessments were blinded to physical activity exposure, job physical exposure, and baseline case status.

Back to Top | Article Outline

CASE DEFINITIONS

The locations of reported LBP were recorded using an anatomic pain diagram (Fig. 1). The perceived intensity of pain in the lumbar and lower extremity regions were recorded using 10-point visual analog pain scales and the results of particular physical examination maneuvers are recorded. In this study, low back pain refers to pain located primarily in the lumbar region of the low back (regions L, M, N, O, and/or P of Fig. 1). The durations of symptoms for incident episodes of LBP are ascertained from the monthly questionnaires and interviews and expressed as units of time (days). These data were collected by health care professionals during monthly questionnaire and interview follow-ups. Incidence rates were assessed via monthly data collection and were pinpointed to a day of incidence. Incident lob back pain is defined in this study as new pain of any intensity from any cause (eg, from work, outside of work, or unknown) in the lumbosacral region (areas L, M, N, O, or P) after being pain free for at least 90 days that occurred after having worn the accelerometer. A 90-day pain-free wash out period utilized for those individuals who had low back pain at the time of enrollment. This 90-day pain-free period was decided upon to try and reduce the misclassification of a recurrence of ongoing LBP.

Figure 1
Figure 1
Image Tools
Back to Top | Article Outline

STATISTICAL ANALYSES

All analyses were done with SAS 9.1.3 (SAS Institute, Cary, North Carolina). Univariate analyses of the predictive relationship, both relative risk or hazard ratio between the predictor variables (measured physical activity levels), confounders (age and psychosocial variables), and effect modifiers (prior LBP) and the incident cases of low back morbidity were calculated. In this initial univariate analysis, the job exposure variables (eg, revised National Institute for Occupational Safety and Health lifting equation, Lifting Index, estimated L5-S1 compressive force, maximum acceptable weights and forces, and low back moments) were analyzed both as continuous data and categorized into multiple categories (eg, prior published cut points and/or tertiles). Statistical significance was determined a priori with an α level of .05. Factors were considered trending toward statistical significance if there was approaching this level of statistical significance with a P < 0.10.

The assessment of LBP-related incidence rates involved careful adjustment for confounders and assessments for effect modifiers, for example, prior exposures and past injuries. Factors were selected for the multivariate model based on biological plausibility, prior evidence related to low back pain (eg, tobacco use) and statistical significance. Proper handling of confounders was performed by taking both analytic and practical appropriateness into account. The proportional hazards assumption was tested. Ties in time to event were calculated using exact calculations, not an estimate.

Many of the covariates included in the adjusted model were trending toward statistical significance, including BMI and age. There was no estimate for current tobacco use because only two participants reported current tobacco use. Other potential risk factors, including high cholesterol (total cholesterol >200 mg/dL), high blood pressure (systolic >140 mm Hg), and diabetes mellitus were not included in the final model because of small numbers of participants with the diagnosis, or no statistical relationship in either univariate or multivariate models.

Back to Top | Article Outline

RESULTS

Demographic data (Table 1) show the baseline characteristics for the 68 participants included in this study. The mean age of this population was 38.2 (SD = 11.3) years. Their mean body mass index (BMI) was 28.4 (SD = 5.7) kg/m2. There were 18 (26.5%) females. There were 55 (80.9%) never smokers, 11 former smokers (16.2%) and two (2.9%) current smokers. There were 21 (30.9%) participants who reported that they had a prior back injury diagnosed by a health care practitioner. When asked how often they felt down, blue, or depressed 20 (29.4%) participants responded never, 38 (55.9%) responded seldom, and 10 (14.7%) responded often, with no participants responding always. A large proportion of participants strongly agreed (n = 13, 19.1%) or agreed (n = 46, 67.7%) with the statement “I have enough time to get my job done,” with three participants (4.4%) neither agreeing nor disagreeing, and six (8.8%) disagreeing.

Table 1
Table 1
Image Tools

During follow-up of the 68 participants, which summed to 92.7 person-years of follow-up time, there were 30 first incident cases of LBP in the follow-up period. The mean duration of pain-free days before onset of LBP or censure was 241.7 (SD = 109.4) days.

The relationship between this case definition of LBP and both mean counts per minute as recorded by the accelerometer and minutes in each activity level as defined by the Matthews cut points were analyzed. Univariate analyses (Table 2) suggest a U-shaped relationship between these measures of physical activity and any cause LBP. Estimates for moderate or moderate and vigorous combined are virtually identical, because there were few minutes of vigorous activity as defined by these cut points in this population. The middle group was selected as the reference category to demonstrate an increased risk when compared to both the lowest and highest groups of different accelerometer measures of physical activity.

Table 2
Table 2
Image Tools

There were statistically significant relative risk (RR) estimates between both tertiles and quartiles of mean counts per minute. When comparing the highest tertile of moderate or vigorous activity with the middle tertile there was a statistically significant RR of 3.68 (95% confidence interval [CI], 1.06 to 12.77). The estimate for the lowest tertile as compared with the middle tertile did not achieve statistical significance (P = 0.10) and had a RR estimate of 2.83 (95% CI, 0.81 to 9.90).

Similar relationships were found when estimating hazard ratios using a Cox proportional hazard model. There was a statistically significantly increased risk of development of LBP among the most active participants compared to those with moderate activity as measured by accelerometer counts per minute with a hazard ratio (HR) of 2.85 (95% CI, 1.08 to 7.51). When comparing the lowest tertile of mean counts per minute with the middle tertile there was a trend toward statistical significance (HR = 2.11, 95% CI, 0.78 to 5.70). Similarly, when analyzing minutes spent in various activity levels, hazard ratio estimates were significant when comparing the middle tertile of moderate or vigorous physical activity with both highest tertile (HR = 4.39, 95% CI, 1.43 to 13.51) and lowest tertile (HR = 3.29, 95% CI, 1.07 to 10.09) as defined by the cut points. The hazard ratio estimate was statistically significant for the highest tertile as compared with the middle tertile (HR = 3.48, 95% CI, 1.24 to 9.78) and was trending toward significance (P = 0.059) with an HR estimate of 2.74 (95% CI, 0.96 to 7.77).

Multivariate analyses were performed, and risk estimates presented are adjusted for all other factors in the model (Table 3). After adjustment for age, sex, BMI, feeling depressed, peak low back compressive force of the workers typical job, tobacco use, and having seen a health care professional for past LBP, nearly all estimates relating physical activity and LBP remained statistically significant, often with increasing risk estimates with increasing levels of activity. Other multivariate models were analyzed, which included earlier-mentioned factors alone and in groups. Many other job physical factors were substituted for peak back compressive force of the typical job, including different measures of the National Institute for Occupational Safety and Health cumulative lifting index, lifting height, frequency, low back moments, and exertion ratings. All models demonstrated similar trends and statistical significance, so the models presented were selected to adjust for suspected or known risk factors.

Table 3
Table 3
Image Tools

When analyzing multivariate relative risk estimates for mean counts per minutes in tertiles, the highest tertile had a statistically significant relative risk estimate of 4.95 (95% CI, 1.12 to 21.92) and the lowest tertile had a nonsignificant relative risk of 2.49 (95% CI, 0.58 to 10.79, P = 0.221) as compared with the middle tertile. Multivariate relative risk estimates comparing the middle tertile of moderate or vigorous activity with the highest tertile were statistically significant after adjustment for factors noted earlier, with an adjusted RR of 7.46 (95% CI, 1.50 to 37.19). Comparing the lowest with the middle tertile yields a nonsignificant relative risk of 4.29 (95% CI, 0.92 to 19.99, P = 0.063). As compared to the lowest tertile of light activity, the middle and highest tertile of light activity was trending toward statistical significance with a RR of 5.21 (95% CI, 0.82 to 33.14, P = 0.08) and 5.08 (95% CI, 0.67 to 38.49, P = 0.12), respectively. When an adjusted model including the aforementioned factors and both light and moderate or vigorous activity was included, risk estimates for light activity were virtually unchanged, however, estimates for both the lowest tertile (RR 13.01, 95% CI, 1.79 to 94.58) and the highest tertile (RR 9.53, 95% CI, 1.56 to 58.02) increased markedly. The relative risk estimate for often feeling depressed also trended toward statistical significance at 5.35 (95% CI, 0.83 to 34.61, P = 0.08). Having seen a health care provider for prior LBP remained statistically significant 4.98 (95% CI, 1.30 to 19.11). Other covariates in the relative risk model were not statistically significant.

For estimates of hazard ratios, the multivariate model differed from the relative risk model in risk estimates. Multivariate hazard ratio estimates for mean counts per minute yielded statistically significantly increased risk for development of LBP of the highest tertile as compared to the middle tertile (HR = 4.15, 95% CI, 1.39 to 12.42). Multivariate hazard ratio estimates for the lowest tertile of mean counts per minute as compared to the middle tertile were not statistically significant (HR = 1.74, 95% CI, 0.59 to 5.15). Multivariate hazard ratio estimates of moderate or vigorous activity as defined by cut points were significant for the highest tertile (HR = 6.33, 95% CI, 1.82 to 21.99) as compared with the middle tertile. The hazard ratio estimate for lowest tertile of moderate or vigorous activity was not statistically significant, although it was trending toward significance (HR = 3.10, 95% CI, 0.94 to 10.24, P = 0.064). Multivariate hazard ratio estimates for second and third tertiles of minutes of light activity as compared to the lowest tertile were not statistically significant with HR estimates of 2.38 (95% CI, 0.78 to 7.30, P = 0.13) and 2.81 (95% CI, 0.73 to 10.84, P = 0.13), respectively. When both tertiles of light activity and tertiles of moderate or vigorous activity were introduced into the adjusted model, estimates for the both tertiles of light activity (HR = 2.41 and 2.91, P = 0.12 for both) and the highest tertile of moderate or vigorous activity (HR = 6.93, 95% CI, 1.72 to 27.81) remained similar, however, the estimate for the lowest tertile of moderate or vigorous activity increased and became statistically significant (HR = 5.36, 95% CI 1.38 to 20.95). Often feeling depressed was statistically significant in hazard ratio modeling (HR = 3.53, 95% CI, 1.06 to 11.73) and never feeling depressed was trending toward statistical significance (HR = 2.56, 95% CI, 0.97 to 6.77, P = 0.06). Having seen a health care provider for prior back injury diagnosed by a health care practitioner had a hazard ratio of 3.48 (95% CI, 1.58 to 7.65). Interactions among select, biologically plausible variables were tested, and none were statistically significant.

Back to Top | Article Outline

DISCUSSION

There was a strong protective relationship between intermediate levels of objectively measured physical activity and subsequent first incident case of LBP from any cause after a minimum of 90 days pain free. This association was present in univariate modeling of both relative risk and proportional hazards, as well as being present in models adjusted for other suspected risk factors. When dividing both mean counts per minute and minutes of moderate activity as defined by Matthews cut points, there were statistically significant relationships suggesting that the middle tertile of both measures of physical activity. Furthermore, this association did not demonstrate a linear dose–response relationship. There were strong associations between middle categories and both higher and lower groups in mean counts per minute, minutes of moderate or vigorous activity. All of these relationships demonstrate a “U” shaped curve where very low or very high measures of activity are associated with higher risk for the development of LBP. This “U” shaped relationship was not evident in the measure of light activity. These relationships persist, even after adjustment for a host of individual and job physical factors. These data do not differentiate between different types of physical activity. The type of activity may impact the relationship between time spent participating in physical activity and the development of LBP.

The protective relationship between physical activity and LBP in this study is consistent with some prior reports, primarily of subjective measures of physical activity.13,39,40 As the relationship between physical activity and LBP in this population is not linear, prior reports finding no statistical relationship between physical activity and LBP may not have obtained sufficient power or data accuracy to accurately measure these differences. These data were also analyzed in quartiles and quintiles, and yielded similar U-shaped trends among moderate physical activity.

Many studies investigate the relationship between physical activity and the treatment of existing LBP, but there is scant research that assesses the potential preventive relationship that physical activity may play in the etiology of LBP. Of the relatively few studies that have been published, nearly all are cross sectional in nature. In the only prospective cohort study evaluating the preventive relationship between physical activity and LBP, authors41 found no statistical association between maximum oxygen uptake (Vo2max), a marker of fitness, and prevention of LBP. A possible explanation for this may be the nonlinear dose response demonstrated by these data. Peak Vo2max is a measure of fitness, which is related to total time spent in higher levels of physical activity. As these data indicate that there may not be a dose–response relationship, peak Vo2max may not be the most accurate measure of physical activity to elucidate the potential relationship between physical activity and LBP. For comparison between our study and those utilizing peak Vo2max, the peak Vo2max has been reported to be underestimated utilizing accelerometer data alone for high speed running as uniaxial accelerometer data plateau at high speeds, although correlations remain relatively high.39 As there were virtually no participants reporting participation in sprinting activities, this underestimation is assumed to be minimal. A cross-sectional study42 reported that there was a significant reduction in lifetime, 1-year period prevalence, and point prevalence in the adult population for those that were physically active for 3 or more hours per week. Another study13 reported that in a select population, participation in physical activity has a statistically significant protective effect for the development of LBP after adjustment for suspected risk factors. Although there are some published studies that document both no effect and a protective effect, there is little basis for recommending lifestyle changes to prevent low back pain in a working population.

The observed protective effect for physical activity is consistent with prior reports that physical activity may be effective for treatment of LBP. A review article43 found that a wide variety of outcome measures were used. Exercise had a positive effect in all 16 trials43 with 12 of 16 programs incorporated strengthening exercise, of which 10 maintained their positive results at follow-up. Another review article44 concluded that trials provide strong evidence that exercise significantly reduces sick days during the first follow-up year.

There is a statistical relationship between the psychosocial measure of depression and incidence of LBP. A recent study reported “psychosocial occupational discomfort” was directly associated with both stenosis in the lumbar back (P = 0.041) and number of stenotic levels (P = 0.019).45 Another article found significant differences between cases and age matched controls regarding work perception (occupational mental stress, intensity of concentration, job satisfaction, and resignation; P < 0.027) and psychosocial factors (anxiety, depression, self-control, marital status; P < 0.0001).46 Furthermore, the relationship between psychosocial factors and LBP may be confounded by participation in physical activity. There have been few epidemiologic studies in this area and important questions remain to be addressed in well-designed intervention studies. Additionally, other factors such as social support or physical activity as a marker for healthy behavior may confound this relationship.

There may be interactive effects between physical activity and many of these comorbid diseases, such as diabetes, cardiovascular disease, hyperlipidemia, and may play a role in reduction of psychosocial factors including anxiety and depression.4754 This study has multiple implications for prevention of LBP and other diseases, as well as increased impetus for recommendations and policy to encourage physical activity on a population basis. The prospective study design has the benefit of being able to suggest temporality in the relationship between physical activity measures and LBP. With the large prevalence of LBP in the general population, it is possible that results from this study may help initiate behavior change.

Strengths of this study include the use of a well-defined, reproducible case definition of LBP with incidence pinpointed to the day because of the monthly follow-up protocol. The main exposure measure is an objective measure of physical activity, with a priori cut points for activity levels and inclusion criteria to increase generalizability of data collected. The prospective cohort study design has a stronger ability to demonstrate temporality. Lastly, this study has the ability to control for multiple factors, including personal, psychosocial, and job physical factors, many of which were objectively measured.

Weaknesses include the fact that the cut points utilized may not precisely reflect metabolic expenditure; light activity may actually be inactivity which is consistent with a Cochrane review of bed rest for LBP55 concluded that “bed rest compared with advice to stay active at best has no effect, and at worst may have slightly harmful effects on LBP”44 suggesting that physical inactivity, yet another measure of activity, may actually be detrimental in the recovery process for LBP. There were very few minutes of vigorous activity as defined by the cut points used in this study. This study evaluating the relationship between first incidence of LBP after at least 90 days pain free during the follow-up period which is different than true primary prevention of first ever incidence of LBP. Therefore, this study cannot differentiate between prevention of first ever LBP and subsequent prevention of LBP. To compensate for this weakness the implementation of a 90-day washout period was incorporated into the case definition.

Back to Top | Article Outline

CONCLUSIONS

Multiple measures of physical activity, both at work and outside of work, have statistically significant relationships with the development of LBP. The most basic objective measure, mean activity counts per minute, demonstrated statistically significantly increased risk for very low or very high measures as compared to the middle group. Moderate physical activity, as defined by applying cut points on activity levels on a per minute basis, was protective for the development of subsequent LBP of any intensity in this cohort. These relationships were present, often strengthened, after adjustment for other known and suspected risk factors. This relationship was nonlinear suggesting that both higher and lower levels of participation increased risk for the development of incident cases of LBP as compared with middle groups. Conversely, increasing amounts of light physical activity or inactivity may be an increased risk for the development of LBP of any intensity. As there may be a protective relationship between both intensity of physical activity and duration of physical activity and LBP then individuals have an opportunity to reduce their likelihood of developing LBP, as well as other morbid conditions, through physical activity. Additional research into the types of activity and more refined divisions of categories of objective measures of physical activity and their relationship with the development of LBP is needed.

Back to Top | Article Outline

REFERENCES

1. Walker BF. The prevalence of low back pain: a systematic review of the literature from 1966 to 1998. J Spinal Disord. 2000;13:205–217.

2. Hart LG, Deyo RA, Cherkin DC. Physician office visits for low back pain. Frequency clinical evaluation and treatment patterns from a U.S. national survey. Spine. 1995;20:11–19.

3. Guo HR, Tanaka S, Cameron LL, et al.. Back pain among workers in the United States: national estimates and workers at high risk. Am J Ind Med. 1995;28:591–602.

4. Walker BF, Muller R, Grant WD. Low back pain in Australian adults: prevalence and associated disability. J Manipulative Physiol Ther. 2004;27:238–244.

5. Neubauer E, Junge A, Pirron P, Seemann H, Schiltenwolf M. HKF-R 10—screening for predicting chronicity in acute low back pain (LBP): a prospective clinical trial. Eur J Pain. 2006;10:559–566.

6. Bouter LM, van Tulder MW, Koes BW. Methodologic issues in low back pain research in primary care. Spine. 1998;23:2014–2020.

7. Cedraschi C, Robert J, Goerg D, Perrin E, Fischer W, Vischer TL. Is chronic non-specific low back pain chronic? Definitions of a problem and problems of a definition. Br J Gen Pract. 1999;49:358–362.

8. de Vet HC, Heymans MW, Dunn KM, et al.. Episodes of low back pain: a proposal for uniform definitions to be used in research. Spine. 2002;27:2409–2416.

9. Choi BC, Eijkemans GJ, Tennassee LM. Prioritization of occupational sentinel health events for workplace health and hazard surveillance: the Pan American Health Organization experience. J Occup Environ Med. 2001;43:147–157.

10. Punnett L, Pruss-Utun A, Nelson DI, et al.. Estimating the global burden of low back pain attributable to combined occupational exposures. Am J Ind Med. 2005;48:459–469.

11. American College of Occupational and Environmental Medicine (ACOEM) Hegmann KT, ed. In: Occupational Medicine Practice Guidelines: Evaluation and Management of Common Health Problems and Functional Recovery of Workers, Third Edition. 3rd ed. Beverly Farms, MA: OEM Press; 2010.

12. Carragee E, Alamin T, Cheng I, Franklin T, van den Haak E, Hurwitz E. Are first-time episodes of serious LBP associated with new MRI findings? Spine J. 2006;6:624–635.

13. Hurwitz EL, Morgenstern H, Chiao C. Effects of recreational physical activity and back exercises on low back pain and psychological distress: findings from the UCLA Low Back Pain Study. Am J Public Health. 2005;95:1817–1824.

14. Hurwitz EL, Morgenstern H, Harber P, et al.. A randomized trial of medical care with and without physical therapy and chiropractic care with and without physical modalities for patients with low back pain: 6-month follow-up outcomes from the UCLA low back pain study. Spine. 2002;27:2193–2204.
15. Mannion AF, Taimela S, Muntener M, Dvorak J. Active therapy for chronic low back pain part 1. Effects on back muscle activation, fatigability, and strength. Spine. 2001;26:897–908.

16. Mannion AF, Muntener M, Taimela S, Dvorak J. Comparison of three active therapies for chronic low back pain: results of a randomized clinical trial with one-year follow-up. Rheumatology (Oxford). 2001;40:772–778.

17. Chatzitheodorou D, Kabitsis C, Malliou P, Mougios V. A pilot study of the effects of high-intensity aerobic exercise versus passive interventions on pain, disability, psychological strain, and serum cortisol concentrations in people with chronic low back pain. Phys Ther. 2007;87:304–312.

18. Machado LA, Azevedo DC, Capanema MB, Neto TN, Cerceau DM. Client-centered therapy vs exercise therapy for chronic low back pain: a pilot randomized controlled trial in Brazil. Pain Med. 2007;8:251–258.

19. Sculco AD, Paup DC, Fernhall B, Sculco MJ. Effects of aerobic exercise on low back pain patients in treatment. Spine J. 2001;1:95–101.

20. Turner JA, Clancy S, McQuade KJ, Cardenas DD. Effectiveness of behavioral therapy for chronic low back pain: a component analysis. J Consult Clin Psychol. 1990;58:573–579.

21. Tritilanunt T, Wajanavisit W. The efficacy of an aerobic exercise and health education program for treatment of chronic low back pain. J Med Assoc Thai. 2001;84(Suppl 2):S528–S533.

22. Lahad A, Malter AD, Berg AO, Deyo RA. The effectiveness of four interventions for the prevention of low back pain. JAMA. 1994;272:1286–1291.

23. Linton SJ, van Tulder MW. Preventive interventions for back and neck pain problems: what is the evidence? Spine. 2001;26:778–787.

24. Airaksinen O, Brox JI, Cedraschi C et al. Chapter 4. European guidelines for the management of chronic nonspecific low back pain. Eur Spine J. 2006;15(Suppl 2):S192–S300.

25. Cohen SP, Argoff CE, Carragee EJ. Management of low back pain. BMJ. 2008;337:a2718.

26. Krismer M, van Tulder M. Strategies for prevention and management of musculoskeletal conditions. Low back pain (non-specific). Best Pract Res Clin Rheumatol. 2007;21:77–91.

27. Waddell G, Burton AK. Occupational health guidelines for the management of low back pain at work: evidence review. Occup Med (Lond). 2001;51:124–135.

28. Farrell PA. Exercise and endorphins—male responses. Med Sci Sports Exerc. 1985;17:89–93.

29. Goldfarb AH, Jamurtas AZ. Beta-endorphin response to exercise. An update. Sports Med. 1997;24:8–16.

30. Trost SG, McIver KL, Pate RR. Conducting accelerometer-based activity assessments in field-based research. Med Sci Sports Exerc. 2005;37(Suppl 11):S531–S543.

31. Bassett DR Jr. Validity and reliability issues in objective monitoring of physical activity. Res Q Exerc Sport. 2000;71(Suppl 2):S30–S36.

32. Crouter SE, Clowers KG, Bassett DR Jr. A novel method for using accelerometer data to predict energy expenditure. J Appl Physiol. 2006;100:1324–1331.

33. Freedson PS, Melanson E, Sirard J. Calibration of the computer science and applications, Inc accelerometer. Med Sci Sports Exerc. 1998;30:777–781.

34. Hendelman D, Miller K, Baggett C, Debold E, Freedson P. Validity of accelerometry for the assessment of moderate intensity physical activity in the field. Med Sci Sports Exerc. 2000;32(Suppl 9):S442–S449.

35. Matthews CE, Ainsworth BE, Thompson RW, Bassett DR Jr. Sources of variance in daily physical activity levels as measured by an accelerometer. Med Sci Sports Exerc. 2002;34:1376–1381.

36. Matthews CE. Calibration of accelerometer output for adults. Med Sci Sports Exerc. 2005;37(Suppl 11):S512–S522.

37. Strath SJ, Bassett DR Jr, Swartz AM. Comparison of MTI accelerometer cut-points for predicting time spent in physical activity. Int J Sports Med. 2003;24:298–303.

38. Sallis JF, Haskell WL, Wood PD, et al.. Physical activity assessment methodology in the Five-City Project. Am J Epidemiol. 1985;121:91–106.

39. Fudge BW, Wilson J, Easton C, et al.. Estimation of oxygen uptake during fast running using accelerometry and heart rate. Med Sci Sports Exerc. 2007;39:192–198.

40. Pillastrini P, Mugnai R, Bertozzi L, et al.. Effectiveness of an at-work exercise program in the prevention and management of neck and low back complaints in nursery school teachers. Ind Health. 2009;47:349–354.

41. Battie MC, Bigos SJ, Fisher LD, et al.. A prospective study of the role of cardiovascular risk factors and fitness in industrial back pain complaints. Spine. 1989;14:141–147.

42. Harreby M, Hesselsoe G, Kjer J, Neergaard K. Low back pain and physical exercise in leisure time in 38-year-old men and women: a 25-year prospective cohort study of 640 school children. Eur Spine J. 1997;6:181–186.

43. Liddle SD, Baxter GD, Gracey JH. Exercise and chronic low back pain: what works? Pain. 2004;107:176–190.

44. Kool J, de Bie R, Oesch P, Knusel O, van den Brandt P, Bachmann S. Exercise reduces sick leave in patients with non-acute non-specific low back pain: a meta-analysis. J Rehabil Med. 2004;36:49–62.

45. Mariconda M, Galasso O, Imbimbo L, Lotti G, Milano C. Relationship between alterations of the lumbar spine, visualized with magnetic resonance imaging, and occupational variables. Eur Spine J. 2007;16:255–266.

46. Boos N, Rieder R, Schade V, Spratt KF, Semmer N, Aebi M. 1995 Volvo Award in clinical sciences. The diagnostic accuracy of magnetic resonance imaging, work perception, and psychosocial factors in identifying symptomatic disc herniations. Spine. 1995;20:2613–2625.

47. Sternfeld B, Ainsworth BE, Quesenberry CP. Physical activity patterns in a diverse population of women. Prev Med. 1999;28:313–323.

48. Karmisholt K, Gotzsche PC. Physical activity for secondary prevention of disease. Systematic reviews of randomised clinical trials. Dan Med Bull. 2005;52:90–94.

49. American College of Sports Medicine Position Stand. Exercise and physical activity for older adults. Med Sci Sports Exerc. 1998;30:992–1008.
50. Bauman A, Owen N. Physical activity of adult Australians: epidemiological evidence and potential strategies for health gain. J Sci Med Sport. 1999;2:30–41.

51. Van Gaal LF, Mertens IL, De Block CE. Mechanisms linking obesity with cardiovascular disease. Nature. 2006;444:875–880.

52. Morso L, Hartvigsen J, Puggaard L, Manniche C. Nordic walking and chronic low back pain: design of a randomized clinical trial. BMC Musculoskelet Disord. 2006;7:77.

53. Blair SN, Kohl HW III, Barlow CE, Paffenbarger RS Jr, Gibbons LW, Macera CA. Changes in physical fitness and all-cause mortality. A prospective study of healthy and unhealthy men. JAMA. 1995;273:1093–1098.

54. Treiber FA, Baranowski T, Braden DS, Strong WB, Levy M, Knox W. Social support for exercise: relationship to physical activity in young adults. Prev Med. 1991;20:737–750.

55. Hagen KB, Hilde G, Jamtvedt G, Winnem MF. The Cochrane review of bed rest for acute low back pain and sciatica. Spine. 2000;25:2932–2939.

Cited By:

This article has been cited 1 time(s).

Bmc Musculoskeletal Disorders
Physical activities at work and risk of musculoskeletal pain and its consequences: protocol for a study with objective field measures among blue-collar workers
Jorgensen, MB; Korshoj, M; Lagersted-Olsen, J; Villumsen, M; Mortensen, OS; Skotte, J; Sogaard, K; Madeleine, P; Thomsen, BL; Holtermann, A
Bmc Musculoskeletal Disorders, 14(): -.
ARTN 213
CrossRef
Back to Top | Article Outline

©2011The American College of Occupational and Environmental Medicine

Login

Article Tools

Images

Share