|Home | About | Journals | Submit | Contact Us | Français|
This was a population-based survey conducted in 2008 in a northern area of China.
To investigate back pain prevalence and to examine the associations between potential risk factors and back pain among Chinese farmers.
Few studies have investigated back pain and its associated risk factors among farmers in low-income and developing nations.
Farmers ages 15 years and older were chosen from 800 families in Heilongjiang Province of People’s Republic of China using cluster sampling methods. Complete survey data were obtained from 2045 farmers. The prevalence of self-reported back pain during the previous three-months was reported. Associations between back pain and potential risk factors were examined in logistic regression models which included age, gender, education levels, perceived stress, main farm activities, smoking, and drinking status.
A total of 786 (38.4%) farmers reported back pain. Two-thirds of those with back pain (66.0%) reported that back pain affected work quantity and quality. The adjusted odds ratios of reporting back pain increased with advancing age. Females and farmers who experienced stress regularly were also more likely to report back pain.
Back pain is a common problem among Chinese farmers and is reported more frequently by females. Significant positive associations of gender, age, and perceived stress with back pain warrant additional study.
Among working adults, low back pain (LBP) is one of the principal causes of disability in industrialized nations.1 Occupational risk factors contribute to 37% of LBP worldwide.2 Moreover, back pain leads to substantial losses in financial revenue and to reductions in quality of life. Farming has become increasingly recognized as an occupation that poses a high risk for back pain. The risk of back and joint pain has been found to be higher among farmers when compared with workers from other occupational sectors, including manual workers.3
A number of contributing factors have been associated with back pain among farmers. Potential risk factors include physical occupational demands such as manual handling, repetitive motion, and whole body vibration and mechanical shock from farming equipment.4-6 Additionally, several studies in industrialized nations have shown that people with mental distress and people who smoke are more likely to suffer from back pain.7-9
While there has been research in the United States 4,10 and other developed countries,5,11 less is known about the factors associated with back pain among farmers in low-income and developing nations. In the past three decades, the still-developing People’s Republic of China (PRC) has emerged as a prevailing force on the world market. Producing 30% of the world’s rice, 20% of the world’s corn, 25% of the world’s cotton, an estimated 37% of the world’s fruit and vegetables, and half of the world’s pork, China has become the largest agricultural producer and consumer.12 Not only does China export to the rest of the world, it must also sustain its own population of 1.3 billion. It is thus imperative to investigate and to address the issue of back pain among China’s vast agricultural workforce.
The objectives of this study were to: 1) describe the prevalence of back pain among Chinese farmers by selected demographic characteristics and 2) examine the associations between potential risk factors (perceived stress, smoking and alcohol drinking) and back pain among Chinese farmers.
We analyzed data from a population-based survey conducted in 2008 that was a collaborative effort between the Center for Injury Research and Policy (CIRP) of the Research Institute at Nationwide Children’s Hospital, the Colorado Injury Control Research Center (CICRC) of the Psychology Department at Colorado State University, and a research team from the School of Public Health of Qiqihar Medical University. These data were previously used by Wang et al. in a study investigating the association between alcohol consumption and work-related injuries among Chinese farmers. The survey procedures and study methods were described in detail in their study.13
Qiqihar is a population center located in China’s northernmost province, Heilongjiang Province. Approximately 59% of Qiqihar’s population lives in surrounding agricultural areas. Information about residents was obtained from Qiqihar’s mandatory registration of residents. The following procedures were utilized to select the sample for this study. First, 800 families were selected through systematic sampling from 9 villages with similar soil and farming techniques. Using cluster survey sampling methods, rural residents 15 years of age and older were surveyed.
With help from CIRP and CICRC, the research team at the School of Public Health of Qiqihar Medical University developed the survey questionnaire. After obtaining approval from the Colorado State University Institutional Review Board and the Scientific Research Committee of the Qiqihar Medical University, the study was officially launched.
Trained in May 2008 by the principal investigator, 25 students from the School of Public Health of Qiqihar Medical University served as interviewers. Student interviewers filled in all questionnaires by conducting face-to-face interviews. Data collection began on May 7 and was completed by May 25 in 2008. The principal investigator at the university randomly chose 25 questionnaires following completion of the initial survey and met with the respondents to verify the consistency of the data collected. Data agreement was greater than 90%, and so the previously collected data were considered reliable.
A specific question about back pain was asked in the survey: “During the past three months, that is, February to April, 2008, did you have back pain that lasted for a full day or more?” In this study, a positive response (yes) to this question was defined as back pain. To those who self-reported back pain, another question was asked: “Did any of the back pain you had during the past three months affect your work quality or quantity?” Respondents who answered “yes” were defined as having back pain that affected work.
The potential risk factors examined in this study were gender, age, education, main agricultural activities, perceived life stress, cigarette use, and alcohol consumption. Subjects were asked to evaluate the overall level of stress they felt in relation to external factors, as well as whether they perceived stress in their family and neighbor relationships. Subjects were asked whether they smoked or drank, as well as a detailed history of their use.
EpiData 3.02 was used to collect and store the data.13 Both SPSS 13.0 and SAS 9.1 were used in the data analysis. Our analysis was completed in phases. First, the prevalence of self-reported back pain was compared across gender, age, education, and main agricultural activities using Pearson Chi-square tests.. Similarly, comparisons in back pain prevalence were considered across stress, smoking, and alcohol consumption variables. Finally, logistic regression models were fitted with back pain and back pain that affected work as the outcome variables and gender, age, education , main farm activities, perceived stress, smoking, and alcohol use status as independent variables. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported. We also evaluated the use of the Generalized Estimating Equation (GEE) to account for non-independence because more than one adult could participate from the each family (using the SAS GENMOD procedure). The original survey did not include a cluster variable, but we could identify clusters using phone numbers for just over 43% of our sample. On average, 2 individuals from each family (i.e. with the same phone number) participated in our survey. Our statistical results from the standard logistic regression analysis and the GEE that accounted for clustering effects were similar, and the conclusions were the same. Therefore, we choose to present the results of logistic regression models, so we could utilize the full dataset.
Of the 2264 respondents initially selected, only 15 (0.7%) declined to participate in the survey, and another 51 (2.3%) provided incomplete information. After data collection, we found that 148 persons actually spent little time performing farm work so they were excluded from the study. Therefore, 2045 farmers were included in the final analyses. There were 786 (38.4%) farmers who reported having experienced general back pain and 518 (25.4%) who reported having experienced back pain that affected work within the 3 months preceding the survey. Two-thirds of those with back pain (66.0%) reported that back pain affected work quantity and quality. Because patterns and potential risk factors were similar for back pain and back pain that affected work, the results and discussion in this article focus primarily on general back pain.
As shown in Table 1, 40.7% of females reported back pain compared with 36.3% of males (P=0.041). Additionally, farmers ages 25 and older reported significantly higher rates of back pain relative to younger farmers (15-24 year olds) with those over age 35 reporting back pain over 40% more often than the younger farmers. Farmers whose main farm activity was crop cultivation were more likely to report back pain than farmers whose main activity was breeding livestock, although this difference was not statistically significant (42.1% vs. 37.9%, P=0.092).
Of the 584 farmers who reported experiencing stress “regularly,” 53.9% reported back pain compared with 31.3% of farmers who reported back pain among those that said they never experienced stress (P<0.001) (Table 2). Overall, less than 2% of respondents reported stress in their familial and neighbor relationships, but those with familial stress reported back pain more often than those without that stress (P=0.003). Back pain was self reported more often among smokers than among non-smokers (P=0.008). Current and former drinkers reported back pain more often than those abstaining from alcohol.
Gender, age, and stress were statistically significant in the multivariate model (Table 3). The associations of smoking and drinking with back pain were not statistically significant after controlling for other variables in the multivariate model. Females were more likely to experience back pain than males (OR 1.22; 95% CI: 1.09-1.36). The odds ratios of back pain increased with increasing age among farmers. Farmers with regularly perceived stress had an increased odds ratio of reported back pain (OR 2.51; 95% CI: 2.00-3.16) compared with farmers who “never or rarely” experienced stress.
The overall prevalence of self-reported back pain in the preceding 3 months among Chinese farmers in Qiqihar was 38.4%; back pain affected the quantity and quality of work for over a quarter of the farmers. Though comparisons to other studies face limitations because of differences in reference periods, study population, and back pain definitions, the results of a U.S. study and several other Chinese studies provide context for our results. In the United States, a study using a large nationally representative dataset, reported a previous 3 month prevalence of LBP (with and without neck pain) of 26.3%.14 Studies of LBP among farmers, mostly males in the U.S. have sometimes reported higher levels of prevalence.4,15 Several studies have described self-reported LBP prevalence rates in Chinese populations. Nearly 67% of farmers in eastern China self-reported back pain in the previous 12 months, and the prevalence rate among women was consistently higher than that among men until age 60.16 Jin, et al. reported an annual LBP prevalence rate of 50% for 3 types of workers in Shanghai: teachers, battery/ kiln factory workers, and garment workers.17 A cross-sectional study of villages in Tibet found a point prevalence rate of LBP of 34.1% and a 12-month prevalence of 41.9%. 18 In our multivariate modeling, sex, age, and perceived stress were statistically associated with back pain while controlling for education, farm activity, smoking, and alcohol use, but gender had only a weak association in the multivariate logistic model. Greater back pain reported among women has been seen in other large population based studies.14,19 Of the 4 identified studies about back pain among Chinese general populations or workers, 16-18,20 only one study of a rural population in Tibet reported no gender-related difference in prevalence of back pain.18 The three other previous studies and our study suggest that females have slightly higher prevalence of back pain.16-18,20 This higher prevalence of back pain among Chinese females may be due to reporting bias (i.e., females may be more likely to report back pain than males) or pain among females due to menstruation.16,17
In the multivariate modeling, advancing age and perceived stress were more strongly associated with back pain than gender. Close to half of Chinese farmers older than 45 years of age reported back pain, and those farmers with regular stress reported back pain more often than those who never or rarely perceived stress. There are a number of literature reviews regarding back pain and psychosocial factors at work21,22 and in private life.23,24 More recently, a review of review studies has been published.25 While there appears to be value in including psychosocial factors in studies of back pain,25,26 associations should be viewed cautiously.22 It is not possible to determine the direction of causality between pain and stress in this cross-sectional survey.
In the univariate analysis, we observed that smokers were significantly more likely to have self-reported general back pain, and individuals with 20 or more years of smoking cigarettes had a significantly higher prevalence of general back pain. In addition, former drinkers were significantly more likely to report general back pain and back pain that affected work quality and quantity. The prevalence of self-reported back pain that affected work also increased with years of alcohol drinking. Our results further suggested that years of alcohol drinking has a larger impact on back pain that affected work than on general back pain. These results could be caused by the confounding effects of age. We found a strong association between age and both general back pain and back pain that affected work in this study. Older individuals are more likely to be former alcohol drinkers and smokers than young individuals in China. Cigarette smoking and alcohol drinking were not associated with back pain in the multivariate model controlling for confounding effects of age. These results suggest that association between years of cigarette smoking/alcohol drinking and self-reported back pain could be explained by the confounding effects of age. According to a systematic literature review of original research reports published between 1987 and 1995, there is no positive association between alcohol consumption and low back pain.27 However, one recent prospective study suggests that moderate alcohol consumption had a protective effect on back pain.28 A meta-analysis of 40 studies published prior to February 2009 indicates that the association between smoking and low back pain was fairly modest.29
This study has several limitations. First, because of the variability of estimates and their high sensitivity to various factors, such as inquiry phrasing, setting and cultural differences, comparison of prevalence rates of back pain among various populations and countries can be problematic.3,30 In future studies, standard definitions of back pain for use in epidemiological studies developed and agreed upon by a group of international experts should be used.31 Their use will facilitate comparisons among back pain studies internationally.32 Second, results were based on self-reports of back pain and the potential risk factors and may have been affected by recall bias. Also, due to cultural stigmatization regarding alcohol consumption and cigarette usage, it is possible that participants underreported actual usage. Third, although the study questionnaire asked a general question about whether subjects “perceived life stresses,” the more detailed questions focused only on familial and neighbor relationships, not allowing assessment of other sources of stress, such as job, economic and other life pressures. In addition, the psychometric properties of the stress questions used in the study have not been validated. Finally, because our study was cross-sectional, causality cannot be evaluated.
This study found a high prevalence of self-reported back pain among farmers in this northern area of China. Because two-thirds of those with back pain (66.0%) reported that back pain affected work quantity and quality, back pain is an important public health issue among agricultural workers in China. Future research should address remaining important questions regarding prevalence of back pain, risk factors, and effective interventions.
We investigated back pain and back pain that affected work among Chinese farmers in a northern area of China. Both were self-reported in face-to-face interviews in this population-based survey. High prevalence of back pain was found among farmers, especially among females, older farmers, and those reporting perceived stress.
Dr Limin Wang was a trainee of the USA-China Agricultural Injury Research Training project. Grant sponsor: National Institutes of Health Fogarty International Center (PIs: L Stallones and H Xiang; Grant #: 1D43TW007257-01A2). The findings and conclusions in this report are those of authors and do not necessarily represent the views of the funding agency.
We thank researchers and students at the School of Public Health, Qiqihar Medical University for data collection.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.