Abstract
Background
As defined by the International Ergonomics Association (IEA) Council “ergonomics (or human factors) is the scientific discipline concerned with the understanding of interactions among humans and other elements of a system, and the profession that applies theory, principles, data and methods to design in order to optimize human well-being and overall system performance” [1]. Work-related musculoskeletal disorders (WMSDs) are defined as a disabling or painful injury to the muscles, nerves or tendons that is caused by work or aggravated by it. These disorders are a leading cause of loss of productivity [2, 3]. There are some studies that confirm the association between working in packaging units in various industries and the pain in shoulder, elbow and wrist, but also they show controversies about the possible risk factors including posture [4, 5]. Some studies suggest that packaging is considered a low risk job for musculoskeletal disorders [6] but performing repetitive tasks by packaging unit personnel is a known risk factor for carpal tunnel syndrome which itself is one of the prevalent painful disorders in the hand and wrist regions [7].
The Nordic Musculoskeletal Questionnaire consists of standardized questions about symptoms at different anatomical regions like the neck, back and also shoulder, hand, and wrist and is widely used in medical fields for this purpose and it considers the pain in different intervals [8]. Its validity and reliability have been checked in previous studies [9, 10].
For the assessment of posture, there are many methods available and one commonly used method is rapid upper limb assessment (RULA) which is utilized in industrial settings where WMSDs are prevalent [11]. This method scores individuals by measuring angular deviation from the neutral position in different anatomical regions of the body. Its performance does not disturb workers. It is quickly performed so a great number of the workers can be assessed during a short time. This method was previously used to assess ergonomic risk factors [12]. It analyzes external load and muscular function and also the posture of the upper limbs, neck, and trunk. It is a useful tool for assessment of musculoskeletal risks in more sedentary jobs [13]. It was realized that a small time spent for training is enough to score the posture using RULA and it is not dependent on the level of experience of the examiner [14].
In this study, we wished to define the possible risk factors for the occurrence of shoulder and wrist pain among the workers of packaging units of pharmaceutical companies in Tehran and use these risk factors for screening purposes and prioritizing high risk workers.
Methods
Participants
Our study population consisted of pharmaceutical packaging workers in Tehran. At the time of the study (2012-2013), there were 30 pharmaceutical companies in Tehran with a total of 1046 workers. We used clustered sampling and randomly chose 9 of the 30 companies and considered all of the packaging workers as our sample population which consisted of 461 workers. They filled out the Nordic Musculoskeletal Questionnaire and also answered some general questions covering demographic data in addition to the questions about work history (work experience), shift-work, exercise frequency, smoking, the level of education, rest breaks during work, secondary jobs, and absence from work.
The employees’ tasks were all considered light tasks and consisted of packaging pharmaceuticals (maximum weight: 5 kilograms). They did it in a repetitive manner and the tasks were mostly monotonous and the level of exertion was not to the point of perspiration. These employees included operation supervisors, packaging operators (fillers and cappers, labelers, cartoner operators, case sealers), technicians, quality control staff, etc. Some employees performed their tasks mostly in a sitting or standing position or they did it while alternating between these two. Some employees had to use a computer for variable amounts of time.
Before conducting the study, the information about the purpose of the study was given to the employees and informed consent form was filled. We ensured the employees that the results of this study will remain anonymous and the gathered data will not influence their employment status or their salary by any means (personal data and scores remained confidential). The employees could refuse to fill the questionnaire, leave any question empty or leave with the questionnaire unfilled. Partially filled questionnaires were also included in the study. The ergonomists who filled the sheets did not know the names or any data from the questionnaires. Furthermore, this study was approved by the Ethics Committee of Iran University of Medical Sciences.
Demographics and Job-related questionnaire
In this study, work history was defined as the number of years which an employee performed the same task as the current one. Shiftwork was defined as any working hours out of the normal 8 a.m. to 4 p.m. in this study. Those workers who were considered shift-workers were later subdivided into fixed shifts or rotational shifts. Fixed shifts were considered to have the same working hours throughout the year and the rotational shift was supposed a schedule alternating between different working hours.
In Iran, academic education starts after completing 12 years of study and we grouped the level of education into two categories, in one hand those who completed these 12 years or gone further and on the other hand, those who did not finish these 12 years. Smoking was inclusive of any kind of tobacco use in the past or present. Marital status was divided into married and others (consisting of single, divorced and widowed) groups. Working posture was also asked subjectively. It was divided into two groups, one with mostly sitting or standing tasks and the other group alternating between these two. For the alternating group, a minimum of 2 hours alternating between the usual postures (sitting or standing) wasrequired.
Performing most of the tasks in the company needed some manual skills and workers were questioned whether they had been taught the necessary skills or not (by supervisors, colleagues, etc.). The second job was defined as having another job outside of the company. Sickness absence, the total number of days they were absent from work due to sickness, were also questioned. Moreover, the existence of a sickness absence more than 3 days was asked. A question was included concerning medical exams in the previous 12 months by an occupational physician.
Sports activity was divided into three groups, the first group consisted of the participants with no regular exercise, the second group, those with less than 3 scheduled sports activities and the third group, the participants with 3 or more scheduled sports activities in the week. Resting time was defined as the total amount of time spent in the company in scheduled time during which that the person does not work. It also included lunch breaks.
Musculoskeletal questionnaire
For assessing of the prevalence of the pain in the previous 12-months in shoulder and wrist, the Nordic musculoskeletal questionnaire was used [8]. It is one of the most widely used tools for assessment of the WMSDs [15]. The validity and reliability of its Persian form have been evaluated in previous studies [9]. This questionnaire is a tool which is applied for determining the frequency of musculoskeletal disorders. It addresses many questions about key anatomical regions in the body (like neck, back, shoulders, and wrists) [16]. It has two parts: the first part is actually a general questionnaire and inquires about the occurrence of pain in different body regions. The second part focuses more deeply on these painful regions of the body [17]. We consider only the pain in the shoulders and wrists (included in the upper arm region of the body) in this article.
Postural assessment
In our study, RULA was utilized for assessment of the posture and we made use of a direct visual manner for assessment (no photographs were taken and no software analysis was performed). While the workers were performing their usual work, ergonomists observed their activity and filled the provided RULA sheets at random times and gave a score of 1 to 9 to each arm. The ergonomists were three graduates in occupational hygiene who were specially trained in ergonomic assessment methods (including RULA). All of them were involved in the assessments in each company almost equally. There were three tables in the sheets. This method actually considers several parts of the body (such as arms, forearms, wrists, neck, trunk and legs) and then adds muscular activity (load) and force for a final score to each side of the body (ranging from 1 to 9) [18]. The lower the number, the lower is the risk for cumulativeinjury.
In this study, we represent the scores as a raw score for each side of the body (left and right sides) and then as the action level of each side separately. By action level 1 (consisting of RULA scores of 1 and 2) it is meant that if the posture is not held statically for a prolonged time, it is in the acceptable range. In action level 2 (consisting of RULA scores of 3 and 4), it would be necessary to study the posture in more detail and ergonomic interventions are probable. In action level 3 (consisting of RULA scores 5 and 6), a more detailed study is warranted and ergonomic interventions should be scheduled in the near future. In action level 4 (consisting of RULA scores of 7 and more), an in-depth study and immediate ergonomic interventions are obligatory [11]. Since there were not enough workers in the action level 1 category, it was combined with the action level 2 category.
Statistical analysis
After all the data were gathered, they were entered into the computer and analyzed using statistical package for social sciences software (SPSS for Windows, Version 16.0. Chicago, SPSS Inc.) Data were analyzed by an independent analyzer to assess the association of wrist and shoulder symptoms with possible risk factors. The p-values less than 0.05 were considered statistically significant.
For quantitative data, averages (means) in both groups were compared using independent sample t-test and for qualitative data, cross tabs were drawn and analyzed with fisher’s exact test (dichotomous variables) and Chi-square test. Afterwards, the variables that were significantly associated with the occurrence of pain in the shoulder and wrist were separately put into a binary logistic regression model.
Results
Employees with a history of musculoskeletal disorders due to accidents were excluded from the analysis. Of the 461 employees, 396 filled the questionnaire (response rate 85.9%) in a way that it could be analyzed. There were some cases of partially filled questionnaires that we also included in our analysis. The average age of the workers was 34.86 years with a standard deviation of 9.18 and the average work history (work experience) was 7.88 years with a standard deviation of 6.9 years. There were 159 females (40.2%) and 233 males (59.8%). Positive history of wrist pain and shoulder pain in the employees were 109 (27.9%) and 93 (23.7%) respectively.
Table 1 provides the classification of the quantitative data in wrist and shoulder pain with corresponding p-values after the comparison. Independent sample t-test was used to calculate the p-values in this table. None of the quantitative values were significantly different between the groups (with or without pain).
Data for the qualitative variables are provided in Table 2. Some quantitative data were added with a certain cut point as qualitative data. As you can see from the table, in the case of wrist pain, having a shiftwork schedule for work and also the type of shiftwork (rotational or fixed), number of missed days in the previous year with the cut point of 7 days, and age with the cut point of 40 were significantly different between the groups (with and without wrist pain). In the case of shoulder pain, smoking, the level of education, age with the cut point of 40 years, and also division of working history into 10-year intervals were significantly different in both groups (with and without shoulder pain).
Afterwards, the statistically significant variables were put into a binary logistic regression model to see if these variables could predict the occurrence of pain in the specific anatomical regions and also to adjust for confounding factors. For the regression models of the wrist and shoulder pain refer to Tables 3 and 4 respectively. After regression, in the case of wrist pain, only shiftwork and its type (rotational or fixed) remained significant. Regarding shoulder pain, only work history stayed significant and other variables lost their significance.
In this population, those workers who have a shift work schedule have an odds ratio of 2.35 concerning wrist pain and those with fixed shift-work have an odds ratio of 1.97 compared to workers with rotational shifts. In case of shoulder pain, working history has an odds ratio of 14.43 for the group with a history of 10 to 20 years compared to the group with less than 10 year experience and this ratio for those with more than 20 year experience is 32.25.
Discussion
In our study, it was found out that the best predictor for the wrist pain was having shiftwork schedule and for the shoulder pain, it was the working history. Having a fixed shift-work was significantly associated with wrist symptoms yet the companies we considered did not usually work 24 hours a day and it seems that the majority of shift workers were evening workers. The climate control or other factors did not differ in those hours so we can attribute most of the problems to the shift work itself. The highest rate of shoulder pain was seen among employees working between 10 and 20 years. It can be speculated that a minimum amount of 10 years is necessary for the shoulder pain to arise in these workers.
There are many studies that emphasize the association between shoulder pain and work history [19, 20]. The association between shiftwork and the occurrence of wrist pain has also been observed in nurses [21, 22] and in petrochemical industry workers [23, 24] among others. These observations which were mostly performed in developing countries are similar to the results of our study. There are also studies that associate shiftwork with higher levels of smoking, alcohol consumption and drug abuse [25]. In our study the level of smoking was not associated with the wrist pain and we did not search for alcohol consumption or drug abuse because these two cases are punishable by the law in Iran and asking these questions could have negatively influenced the process of filling the questionnaires.
Our study did not show any association between posture (assessed by RULA method) and pain in wrist or shoulder regions. In the literature, there is some discordance between methods used to assess upper limb posture and musculoskeletal disorders attributed to it [13]. In some studies, there was no observed association between musculoskeletal disorders and RULA scores in the upper limb (with a sample population of dentists) [18]. Some other studies demonstrate an association between posture and pain in the upper extremity [26, 27]. These sample populations consisted of jewelry workers, students, and medical sonographers. The results of these studies show that different working populations cannot be equally screened by RULA for the risk of pain in the upper limb. RULA by itself is a screening tool for bad postures but it should not be used as a proxy for musculoskeletal complaints in the upper limb.
In our study, age with the cut point of 40 years was the only variable significantly associated with the pain in wrists and shoulders but its effects did not remain significant after the regression in both models. It has been shown that the age of the worker is associated with higher RULA scores [2]. In our study smoking was associated with shoulder pain and it was also shown in previous studies [28, 29]. In our study, lower degrees of education also turned out to be associated with shoulder pain. However, we should have it in mind that the effects of age, smoking and education did not remain significant after regression models.
There are limitations to our study. The most important is the cross-sectional design of the study which cannot be used for assessment of causation. A study with the longitudinal design is recommended for this purpose. This study was performed in packaging units of the pharmaceutical companies where workers performed only light tasks, so these data might not be valid for other industries with different working conditions. We recommend other studies in other industrial settings to define specific risk factors for those working conditions. Moreover, our sample population is not representative of all of the countries. However, it can somehow be considered representative of the working conditions in developing countries. We did not use random sampling of the whole population and it reduced the expandability of our results. A questionnaire was used to define the risk factors (except RULA) and the occurrence of pain in the upper limb and the workers were not physically examined so our findings were subjective. Even though the workers were ensured of the confidentiality of the data, some may not have declared their problems and our disease rates were probably underestimated.
Our study also provides some advantages. We chose a group of workers that are all light manual workers (as opposed to heavy manual workers) so that we could assess risk factors other than heavy lifting or other common risk factors that are usually considered in industrial settings. We performed RULA for all of the sample population to determine its efficiency in predicting musculoskeletal pain in the upper limb. Also 9 different companies were included so that our sample population would be more representative of the industry in Tehran. All of the risk factors mentioned in this article were also assessed for low back pain and neck pain which were considered separately. This allows for a better comparison between body regions and the definition of specific risk factors for different parts of the body in this industry. Our study population consisted of an industry which has not been assessed for ergonomic risk factors previously.
Conclusion
In this study, RULA was not associated with pain in the upper limb (shoulder and wrist) in the previous 12 months in light working employees of pharmaceutical packaging units. The use of decades of working history and shiftwork (specially the fixed shiftwork type) can be considered for predicting the shoulder and wrist pain respectively. The risk factors considered in this study can only be considered in industries with light manual work but for industries with heavy manual work, further studies are recommended.
Conflict of interest
None to report.