Impact of an Ergonomics Program Featuring Adjustable Chairs on Upper Extremity Musculoskeletal Symptoms Among Garment Workers

R.Herbert¹, J.Dropkin¹, D.Sivin² ,
J.Doucette¹, L.Kellogg³, J.Bardin⁴,
N.Warren⁴,D. Kass², S.Zoloth²

¹Mount Sinai Medical Center, Department of Community Medicine, New York, USA
²Hunter College, Center for Occupational and Environmental Health, New York, USA
³Union of Needletrades,Industrial, and Textile Employees, Department of Health and Safety,New York, USA
⁴University of Massachusetts Lowell, Department of Work Environment, Massachusetts, USA

ABSTRACT

This study sought to evaluate the effect of an ergonomics program, which included introduction of adjustable chairs and an education program in the use of these chairs, on the prevalence and intensity of symptoms of work-related musculoskeletal disorders (WRMDs) among garment workers whose work as spoolers required them to repetitively turn a manual, waist-height crank with their right arm while simultaneously performing finger movements with their left hand. Standardized medical, occupational, and symptom histories were obtained from 36 participants three months before and six months after introduction of adjustable chairs. A quantitative pre-and post-intervention exposure assessment was performed among a subgroup (n=19) utilizing Keyserling and Rapid Upper Limb Assessment (RULA) methods. The population consisted of 36 females of mean age 47.9+ 9.3 years and mean years spooling of 7.2+5.4. At baseline, the proportion of examinees with pain in the right shoulder (66%), right elbow (31%), right forearm (29%), right wrist (25%) and right hand (40%) was consistently greater than the corresponding proportion for the same anatomic sites on the left side (36%, 28%, 26%, 17%, and 36%, respectively). Following the introduction of the adjustable chair there was a decrease in the proportion of examinees reporting pain in all anatomic sites. There was a significant reduction (p<.05) in both the prevalence and severity of pain in the right shoulder, left elbow and left forearm. Analysis of exposure data showed that there was a significant decline (p<.05) in exposure to awkward postures in the left wrist after introduction and training in the use of adjustable chairs, with reductions in awkward postures also observed at other sites. This study suggests that implementation of an ergonomics program focused on education and introduction of an adjustable chair designed to increase neutral joint positions may diminish musculoskeletal symptomatology.

INTRODUCTION

Musculoskeletal disorders of occupational origin, also know an ergonomic overuse syndromes, have become the most prevalent occupational diseases in the United States [1]. In the apparel industry, the incidence of work-related musculoskeletal disorders (WRMDs) increased 348% between 1988 and 1992 [2], and the prevalence of these disorders in garment workers in 1991 was five times that of industry overall [3]. The repetitive nature of work in the apparel industry, coupled with rapid piece-rate production speeds and awkward working postures, combine to place garment workers at risk for the development of ergonomic overuse syndromes. While workplace ergonomic interventions are viewed as important in preventing these disorders, there are limited data demonstrating the effect of ergonomic interventions on health. Therefore, this study sought to evaluate the impact of an ergonomic program consisting of introduction of adjustable chairs and training in the use of these chairs on joint position and prevalence of work-related upper extremity musculoskeletal symptoms. In 1992, a garment worker attending the Union Health Center-Occupational Medicine Clinic in New York City, whose job required her to continuously turning a crank with her right arm, was diagnosed with right rotator cuff tendonitis. This index case led the International Ladies Garment Workers Union's (ILGWU, now UNITE, or Union of Needletrades, Industrial, and Textile Employees) Health & Safety specialist to conduct an evaluation of the patient's workplace. The inspection revealed that 80 female workers were performing similar work. The ILGWU Health & Safety Department was able to persuade the employer, Quality Braid (now Sequins International), to participate in a joint labor-management fund, sponsored by the Council on American Fashion Ergonomics Fund, which provides matching funds for workplace ergonomic modifications that are designed to prevent occupational musculoskeletal disorders. As a result of this fund, adjustable ergonomic chairs were introduced in the plant. The employer installed adjustable chairs while the Health and Safety specialist instructed participants in all ergonomic aspects of the chair.

POPULATION

This study was conducted at a sequin manufacturing facility in Queens, New York. The baseline population consisted of female immigrants, 48 (75%) of whom were Hispanic, 15 (23%) of whom were Asian Indian, and 1 (1.6%) consisted of "other." The baseline study population was female spoolers (n=64) whose primary task was to measure and perform quality control on threaded sequins, using either manual (n=55), automatic (n=6), or manual and automatic (n=3) spoolers. The adjustable chair intervention subgroup comprised 36 manual spoolers.

METHODS

Baseline (round 1) symptom questionnaires and upper extremity physical examinations were administered to 64 of 76 currently working spoolers in June, 1993. The symptom questionnaire was a modification of a questionnaire used by NIOSH [4]. It was administered in English and Spanish. It obtained information on demographics, medical history, occupational history (including ergonomic exposures), and non-occupational risk factors for the development of upper extremity pain, stiffness, numbness, or tingling in the previous month. Symptoms were graded on a scale from 1 to 5, where 1 was equivalent to "no pain," while 5 was equivalent to "worst pain in life."

Baseline upper extremity physical examinations were conducted by physicians from the Mount Sinai Medical Center. Standardized examination and recording protocols were utilized. In September, 1993, the adjustable chair subgroup (n=36) received education about cumulative trauma disorders from the ILGWU Health and Safety specialist and were provided with Biofit adjustable, padded chairs by the employer. Features of the chair included pneumatic height adjustable, forward locking and tilting seat pan; adjustable height and depth back rest; and foot rest. The participants were also instructed and trained in adjustment and use of the chair. Additionally, the plant manager, supervisors, and spoolers received general education from the IlGWU Health and Safety specialist about ergonomics and WRMDs.

Of the intial cohort of 64, due to layoffs and attrition only 47 participants remained as spoolers in February, 1994. Modified symptom questionnaires were administered to 36 of these participants who had received adjustable, padded chairs and who had continued to perform manual spooling operations. The remaining 11 baseline participants worked on a partially modified spooling machine without an adjustable chair.

Quantitative ergonomic exposure measures were obtained before (July, 1993) and after (February, 1994) the installation of adjustable chairs among a randomly selected subgroup (n=19) of the 36 participants. All 36 participants were videotaped at baseline (round 1) while performing primary work elements (spooling) and secondary work elements (cutting sequins, checking sequins, restocking). Ergonomic exposures after the installation of adjustable stools were reassessed (round 2) by revideotaping the subgroup at work.

An ergonomist and physical therapist utilized Keyserling [5] and Rapid Upper Limb Assessment (RULA) [6] methods of work analysis to collect and assess information on these participants. Using a video, data was first collected using the Keyserling's "Postural Analysis in Simulated Real Time" computer program, designed to obtain information on range of motion and posture of the neck, shoulders, elbows, forearms, wrists and hands, percent of time in that posture relative to the task, and frequency (cycles/second) of the task. When the study began, we initially chose to categorize and quantify exposure directly from subjects using this method. However, the RULA method of work analysis, published in 1993, seemed to offer a more concise way to quantifying exposure. Thus, based on the results of Keyserling's program, RULA was used to attach posture scores, ranging from 1 to 4, to the neck, upper arm, lower arm, wrist and wrist twist.

Muscle and force scores, usually a part of RULA, were not used because all participants has identical values for muscle score and force score. Workers all received 1 on the muscle score because their activity was either repeated for more than four times per minute (e.g., right shoulder), or because their upper limb was held for more than one minute in an awkward, static position (e.g., left elbow). Workers all received 0 on the force score because their tasks required negligible resistance.

DATA ANALYSIS

Comparison of those baseline subjects receiving an adjustable chair (n=36) to those subjects not in the intervention cohort (n=28) with respect to demographic and other characteristics was done using two-sample tests for continuous measures and chi-square tests for categorical factors. Within the intervention cohort, McNemar's test was utilized for assessing changes in both the prevalence and severity of reported pain, as well as changes in exposure to awkward postures, at each anatomic site. In the case of severity of pain, subjects' responses at the two time points (for a given site) were first classified as either improved (lower severity after the introduction of the chair), worsened (higher severity) or unchanged. Similarly for exposure assessment, RULA scores before and after introduction of the chair were classified as improved, worsened, or unchanged. McNemar's test uses chi-square distribution to assess whether significant changes have occurred by comparing the relative number of subjects who exhibit a change in one direction to the number who change in the other direction. Thus, subjects who were unchanged between the two time points did not contribute to the analysis for these comparisons.

RESULTS

Demograpics

The intervention cohort (n=36) was 100% female with a mean age of 47.9+9.3 and mean years worked as a spooler of 7.2+5.4. Seventy-three percent were Hispanic, 25% were Asian Indian and 2% were "other." Seventy-two percent employed Spanish as their primary language and the remaining 28% were English speaking.

Exposure

The task of spooling involved examining, removing and reworking imperfections in the sequins before the product reached the customer. This work required constant right shoulder movement; manual spoolers were required to repetitively turn a waist height crank with their right hand. The ledt hand statically held a string of sequins just prior to being spooled. If an imperfection was felt, the operator would remove the flaw with a pair of scissors and tie the sequin ends together using both hands. Table 1 presents the proportion of subjects exhibiting a change in exposure as measured by RULA at each anatomic site among garment workers for whom a videotape exposure assessment were performed (n=19). The data indicate a significant decline in exposure to awkward posture in the left wrist. While changes at other sites were not statistically significant, we observed a decline in exposure to awkward posture at most sites except for the right elbow and right forearm after introduction of the chair.

Symptomatology

For those 36 garment workers receiving the adjustable chair and present at round 2, Table 2 shows the proportion of workers reporting pain at each anatomic site before and after the intervention.

There was a significant reduction in the prevalence of pain in the right shoulder, left elbow, and left forearm, after the adjustable chair was introduced in the study population. Although not statistically significant, a decreased prevalence of pain at all sites was also seen.

For changes in the reported severity of pain, by anatomic site, following the introduction of an adjustable chair, the proportion of examinees whose pain improved was significantly greater than the proportion of examinees whose pain worsened in the right shoulder (42% improved, 8% worsened), left elbow (25% improved, 3% worsened), and left forearm (25% improved, 0% worsened). Although not achieving statistical significance, 28% or more of the examinees experienced improvement in symptoms of pain in the neck, left shoulder and right elbow.

DISCUSSION

The major finding of this study was of a reduction in the percentage of garment workers reporting symptoms of pain in the upper extremities after the introduction of an ergonomics program that featured the provision of adjustable chairs, and training in the use of those chairs.

Baseline data in Table 2 reveal that, prior to the intervention, the prevalence of pain among garment workers in the right shoulder, right elbow, right forearm, and right wrist was consistently greater than the prevalence of pain in those sites on the left side. There was a significant reduction in the prevalence of examinees reporting pain in the right shoulder, left elbow and left forearm after the introduction of an adjustable chair. The anatomic sites where reported severity of pain significantly improved include the right shoulder, left elbow, and left forearm. As shown in Table 1, the left wrist showed a significant improvement in exposure to awkward posture after introduction and training in use of the adjustable chair.

This study has several limitations. We were unable to identify a suitable comparison group for this intervention study. Because of the absence of a control group, the Hawthorne effect, in which subjects may change their behavior because of the study, cannot be excluded. Psychosocial effects were not accounted for, because no known validated assessment instrument for a population with this demographic makeup existed at the time of the study. A subgroup from the initial cohort was used, and although this group seemed representative, they may not have been. There may not have been adequate statistical power to detect improvements, when applying RULA, at sites other than left wrist using the subgroup of 19 workers. These findings need validation in a larger study.

Despite these limitations, this study suggests that an ergonomics program that includes the provision of adjustable furniture, and training in the use of that furniture in the apparel industry, can reduce musculoskeletal disorders for certain workers who suffer from these illnesses. Specifically, this study suggests that introduction of an adjustable chair designed to increase neutral joint positions may result in diminished musculoskeletal symptomatology for workers performing manual spooling operations.

[1] U.S. Department of Labor, Bureau of Labor Statistics, 1992, Occupational Injuries and Illnesses in the United States by Industry, 1990. pp. 4-6
[2] U.S. Department of Labor, Bureau of Labor Statistics, News: Workplace Injuries and Illnesses, 1992.
[3] U.S. Department of Labor, Bureau of Labor Statistics, News: Workplace Injuries and Illnesses, 1991.
[4] National Institute for Occupational Safety and Health (1993): Health Hazard Evaluation Report: Los Angeles Times. Cincinnati, Ohio: U.S. Department of Health and Human Services,Public Health Service, Centers for Disease Control, NIOSH, NIOSH Report No. HETA 90-013-2277.
[5] WM Keyserling, Postural Analysis of the Trunk and Shoulders in Simulated Real Time.Ergonomics 1986;29:569-83 [6] L.McAtamney et al,RULA: A Survey Method for the Investigation of Work-Related Upper Limb Disorders.Applied Ergonomics 1993;24(2):91-99.

*p<.05 for McNemar's Tests (two-tailed)
1 missing data