ASSOCIATION BETWEEN WHOLE BODY VIBRATION AND LOW BACK DISORDERS IN FARMERS: A SYSTEMATIC REVIEW AND A PROSPECTIVE COHORT STUDY

Abstract

Low back disorders (LBDs) are the most common musculoskeletal problem among farmers, with higher prevalence rates than in other occupations. Operators of tractors and other farm machinery such as combines and all terrain vehicles (ATV) can have considerable accumulation of exposure to whole body vibration (WBV). The causal relationship between LBDs and WBV is not fully clear; however, it may be different among farmers as their work context and exposure is unique. Objectives: The objectives of the two studies which form two manuscripts or chapters in this thesis were to: 1) investigate the associations between WBV and LBDs among farmers using a) systematic review (manuscript 1) and cohort studies (manuscript 2). Methods: Objective 1: Nine databases were searched using groups of terms for two concepts: ‘farming’ and ‘low back disorder’. Screening, data extraction and quality assessment was performed by two reviewers independently. The population was adult farmers or agricultural workers globally irrespective of sex. The intervention was considered to be WBV exposure, such tractor, combine and ATV use. The control was no exposure or low exposure to WBV and the outcome was low back disorders. No limits in date of publication and type of study design were applied in the literature search, and only full text, English language studies were considered. Objective 2: The data source was the Saskatchewan Farm Injury Cohort Study. In 2007, baseline data were collected on accumulated yearly tractor, combine, ATV operation, as well as several biopsychosocial covariates thought to be associated with LBDs. Follow-up data on LBDs and related symptoms were collected during 2013 (6 year follow-up) and 2014 (1-year). This resulted in two datasets for each of two cohorts: 1) the first cohort with 1,149 farm people who had been followed for six years, and 2) the second with 605 participants who had been followed for one year. Generalized estimating equation-modified Poisson regressions were performed with low back and hip symptoms as the outcome. Results: Objective 1: After 276 full texts screened, we found 12 articles which analyzed WBV as a risk factor for LBD. Three were case-control, 6 cross-sectional and 3 retrospective cohorts. Four studies showed no association between WBV and LBDs, 4 studies showed a positive association and for the remaining 4 studies, results were mixed depending on the exposure or the outcome measure. Objective 2: The adjusted model in cohort 1 found LBDs to be associated to tractor operation for 1-150 hrs/year (RR=1.23, 95%CI 1.05-1.44), 151-400 hrs/year (RR=1.32, 95%CI 1.14-1.54) and 401+ hrs/year (RR=1.34, 95%CI 1.15-1.56). In addition, tractor operation for 151-400 hrs/year (RR=1.95, 95%CI 1.45-2.62) and 401+ hrs/year (RR=1.79, 95%CI 1.32-2.45) was also found to be related to hip symptoms. Although combine operation ≥ 61 hrs/year and ATV operation 81+ days/year was related to LBD in the bivariate analysis in cohort 1, this association did not persist after adjustment for confounders. Due to limited power, no significant bivariate association was found between WBV and either LBDs and hip symptoms in cohort 2. Conclusions: Objective 1: A firm conclusion is difficult due to heterogeneity in statistical strategy, LBDs definition, type of farm commodity, and study design. Direct comparisons and synthesis were not possible. Although retrospective cohort studies tended to show a relationship, future studies with a prospective cohort design can help clarify this association further. Objective 2: Although duration of tractor operation and older age showed with both LBDs and hip symptoms in farmers in cohort 1, the true prospective cohort 2 found no significant association between WBV and LBDs.