Page 85

4


Epidemiologic Evidence

The panel's effort to evaluate the scientific basis for a relationship between work factors and musculoskeletal disorders of the back and upper extremities required comprehensive reviews of the epidemiologic literature. For each of the two anatomical regions, reviews of the physical and the psychosocial factors were undertaken. Referring back to Figure 1.2, the review of the epidemiologic evidence addresses several components. The workplace factors considered include all three main elements and their relationship to the person. The person is considered in terms of the several outcomes reported in these studies, while adjusting or stratifying for the individual factors that are relevant.

METHODS

Criteria for Selection and Review of Articles

In planning for this process, the panel set a number of criteria specific to the task of selecting articles for the epidemiology review:

  • Both the exposed and the nonexposed (or comparison) populations are clearly defined with explicit inclusion and exclusion criteria. It is evident why subjects who were studied were eligible and why those not studied were ineligible.

  • The participation rate was 70 percent or more.

  • Health outcomes relate to musculoskeletal disorders of the low back, neck, and upper extremities and were measured by well-defined criteria determined before the study. The health outcomes studied are carefully defined so that it is evident how an independent investigator



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 85
Page 85 4 Epidemiologic Evidence The panel's effort to evaluate the scientific basis for a relationship between work factors and musculoskeletal disorders of the back and upper extremities required comprehensive reviews of the epidemiologic literature. For each of the two anatomical regions, reviews of the physical and the psychosocial factors were undertaken. Referring back to Figure 1.2, the review of the epidemiologic evidence addresses several components. The workplace factors considered include all three main elements and their relationship to the person. The person is considered in terms of the several outcomes reported in these studies, while adjusting or stratifying for the individual factors that are relevant. METHODS Criteria for Selection and Review of Articles In planning for this process, the panel set a number of criteria specific to the task of selecting articles for the epidemiology review: Both the exposed and the nonexposed (or comparison) populations are clearly defined with explicit inclusion and exclusion criteria. It is evident why subjects who were studied were eligible and why those not studied were ineligible. The participation rate was 70 percent or more. Health outcomes relate to musculoskeletal disorders of the low back, neck, and upper extremities and were measured by well-defined criteria determined before the study. The health outcomes studied are carefully defined so that it is evident how an independent investigator

OCR for page 85
Page 86 could identify the same outcome in a different study population. Outcomes are measured either by objective means or by self-report. For self-reported outcomes, however, there are explicit criteria for how the data were collected and evidence that the collection method would permit another investigator to repeat the study in another population. The exposure measures are well defined. Self-report of exposure is acceptable so long as the method of collecting self-reports was well specified and there was evidence that the self-reports were reliable reflections of exposures. Job titles as surrogates for exposure were acceptable when the exposure of interest was inherent in the job (e.g., vibration exposure for those operating pneumatic chipping hammers). The article was published in English. The article was peer reviewed. The study was done within the last 20 years (preferably). No specific limitations were placed on study designs acceptable for consideration. The advantages of prospective studies, however, were recognized. For example, there were sufficient prospective studies of low back pain to examine these separately among the studies of physical factors and exclusively among the studies of psychosocial factors. Literature Search Methods The literature reviews were conducted using computer-based bibliographic databases, with MEDLINE (National Library of Medicine, United States of America) a component of all searches. Additional databases included: NIOSHTIC (National Institute for Occupational Safety and Health, United States of America), HSELINE (Health and Safety Executive, United Kingdom), CISDOC (International Labour Organization, Switzerland), Ergoweb (Internet site of the University of Utah), Psychinfo, Oshrom, Ergonomics Abstracts, and ArbLine (National Institute for Working Life, Sweden). The bibliographies of articles (particularly review articles) and the NIOSH comprehensive review (Bernard, 1997b) were examined to identify additional relevant articles. Using these sources, a candidate list of articles was established and then systematically screened to determine which ones met the strict criteria, described above, for inclusion in the review. Each process reduced the list substantially. For physical work factors studied in association with back disorders, 255 studies were initially identified as relevant and 41 met the selection criteria and were reviewed. For psychophysical factors and back disorders, the search resulted in 975 references, which were then reduced to 21 work-related risk factor studies and 29 individual risk factor studies. For work-related physical factors and upper extremity disor-

OCR for page 85
Page 87ders, the initial list of 265 references was reduced to 13 that provided direct and 29 that provided indirect measures of exposure. For psychophysical factors and upper extremity disorders, the initial 120 references were reduced to 28. Analysis of Study Results Definition of Measures: Relative Risks In epidemiology, the relative risk is a measure of the strength of an association, here meaning the relationship between the frequency of an exposure and the occurrence of an outcome (e.g., amount of vibration and incidence of back pain). Because human populations typically have a variety of exposures occurring in near proximity, relative risk is typically measured as the incidence of disease in the exposed (e.g., helicopter pilots who experience vibration) and the incidence of disease in the unexposed (similar people, like ground crews, who are considered to share nearly the same other exposures as the exposed, such as recreational activities, diet, and living conditions). The ratio of incidence provides a measure of association, and the higher this ratio of incidences (the relative risk), the stronger the association, the more confidence we can place in a conclusion that the association is meaningful. Because incidence is a rate calculated by following people over time, and many studies are cross-sectional or retrospective (case-control), other measures, such as the prevalence ratio and the odds ratio, have been developed to summarize the association between exposure and outcomes for these other study designs. Our analysis focused on associations expressed by such risk estimates as the odds ratio and the relative risk. These estimates were retrieved from the original article or calculated when sufficient raw data were presented. Definition of Measure: Attributable Risk The attributable risk is another measure used to help generate inferences. In its simplest form, it is the difference between the incidence in those exposed and those unexposed—a risk difference. This risk difference is thought of as the attributable risk in that, in theory, removing this exposure entirely would reduce the frequency of the outcome to the level of those who are unexposed. Rotham and Greenland (1998a, 1998b) discuss some of the limitations of this simple assumption. Attributable risk is often calculated as a ratio rather than a difference: risk in the exposed is divided by risk in the unexposed, producing an attributable fraction. The attributable fraction is the proportion by which the rate of the outcome

OCR for page 85
Page 88among the exposed would be reduced if the exposure were eliminated. This fraction is calculated as the ratio of (RR − 1)/RR, where RR is the relative risk or the prevalence ratio of risk in the exposed compared with the unexposed: AF e = (RR − 1)/RR The attributable fraction helps scientists and policy makers recognize that in many cases a variety of factors contribute to the total incidence of a disease or other outcome, so that removal of an exposure typically does not reduce the outcome rate to zero. However, in its simplest form, the attributable risk is a measure that suggests that if the offending exposure were removed (by intervention or regulation), then the amount of disease outcomes would be estimated to be reduced by the calculated amount. As is noted below, this simple summary is enmeshed in caveats. It is important to recognize in this calculation that the result depends on what is included. That is, if one considers a calculation of one factor as it relates to an outcome and then performs a separate calculation for another factor for the same outcome, there is overlapping (correlation) between factors that could make the sum of the two separate factors sum to more than 100 percent. Attributable fraction, then, represents a crude but important estimation of the impact of control of risk factors. An estimate of the attributable fraction for a multifactorial disease such as a musculoskeletal disorder provides only an estimate of the relative importance of the various factors studied. It is not, and cannot be, considered a direct estimate of the proportion of the disease in the population that would be eliminated if only this single factor were removed (Rotham and Greenland, 1998a). Rather it provides guidance to the relative importance of exposure reduction in those settings in which the exposure under study is prevalent. Consequently, we have not attempted to rank or further interpret the findings for attributable fractions and have chosen only to report them as a rough guide to the relative importance of the factors in the study settings in which they have been examined. In this review, the relative risk in longitudinal studies and the prevalence or odds ratio in cross-sectional surveys were used to calculate the attributable fraction for the risk factors studied. For example, if workers exposed to frequent bending and twisting have a prevalence of low back pain that is 3 times that of those not exposed, then among the exposed the attributable fraction will be: AF e = (3 − 1)/3 = 0.67 By this hypothetical calculation, 67 percent of low back pain in the exposed group could be prevented by eliminating work that requires bending and twisting.

OCR for page 85
Page 89 Confounding None of the musculoskeletal disorders examined in this report is uniquely caused by work exposures. They are what the World Health Organization calls work-related conditions. “Work-related diseases may be partially caused by adverse working conditions. They may be aggravated, accelerated, or exacerbated by workplace exposures, and they may impair working capacity. Personal characteristics and other environmental and socio-cultural factors usually play a role as risk factors in work-related diseases, which are often more common than occupational diseases” (World Health Organization, 1985). In Chapter 3 we note that the epidemiologic study of causes related to health outcomes such as musculoskeletal disorders requires careful attention to the several factors associated with the outcome. The objective of a study will determine which factor or factors are the focus and which factors might “confound” the association. In the case of musculoskeletal disorders, a study may have as its objective the investigation of individual risk factors. Such a study, however, cannot evaluate individual risk factors effectively if it does not also consider relevant work exposures; the work exposures are potential confounders of the association with individual risk factors. Conversely, a study that evaluates work exposures cannot effectively evaluate these factors if it does not also consider relevant individual risk factors; the individual risk factors are potential confounders of the association with work exposures. Therefore, when studying the relationship of musculoskeletal disorders to work, it is necessary to consider the other known factors that cause or modify the likelihood that the disorder will occur, such as individual factors and nonwork exposures. For example, the frequency of many musculoskeletal disorders is a function of age, so age has to be taken into account before attributing a musculoskeletal disorder to a work exposure. Another common concern is whether a recreational exposure accounts for an outcome that otherwise might be attributed to work. In every epidemiologic study, confounders need to be measured and, when relevant, included in the data analysis. The confounders selected for consideration in the analysis of data from a specific study depend on the types of exposures studied, the types of outcomes measured, and the detail on potential confounders that can be accurately collected on a sufficient number of the study subjects. As a consequence, our approach to reviewing epidemiologic studies of work and musculoskeletal disorders documented the attention given to a wide range of potential confounders (see the panel's abstract form in Box 4.1). No study can measure every possible confounder; however, the papers included by the panel were judged to have given adequate attention to the primary individual factors

OCR for page 85
Page 90 BOX 4.1 Individual Factors Considered in Analyses Form Used in Describing Studies Included in the Review Described Used in Analysis Does Not Vary Age Gender Body mass index Weight Height Smoking Marital status Income Educational status Comorbid states Hormone-related conditions (e.g., pregnancy) Strength or capacity Race Workers' compensation policies Nonoccupational exposure factors Methods used to control confounding: Matching Stratification Standardization None Regression Other: —— Consideration of interactions: Interaction between different types of work exposures Interaction between work exposures and nonwork exposures/cofactors that might have confounded the work exposures under study. These include in particular age and gender, as well as, when necessary and possible, such factors as obesity, cigarette smoking, and comorbid states. The role of potential confounders in epidemiologic studies and their proper management is often confusing to the nonepidemiologist. The difficulty stems from the fact that the potential confounder is often known to be associated with the disease, in this case musculoskeletal disorders. The association of a risk factor such as age with the disease, however, does not make it a true confounder of the study's examination of a separate risk

OCR for page 85
Page 91factor such as work exposures. True confounding occurs only when, for example, both the risk factors being studied (age and work exposures) are associated with the outcome (musculoskeletal disorders) and the two risk factors are also correlated (for example, those with more work exposure are also older). Fortunately, as noted in Chapter 3, there are statistical methods available to manage confounding that provide a way to “separate,” in this example, the effects of the work exposure from the effects of age. The panel recognizes that a number of nonwork factors are associated with or also cause the musculoskeletal disorders under study. These were not separately studied, but they were considered, as necessary, to evaluate the significance of the work factors that were studied. In our judgment, it is evident that confounding alone is highly unlikely to explain the associations of musculoskeletal disorders with work that are noted. More detailed consideration of confounding in future studies, however, should further improve the precision and accuracy of risk estimates. Measures of Workplace Exposures Physical Exposures The measures of physical exposures investigated include force, repetition, posture, vibration, and temperature. Available approaches to estimating exposure to these physical stressors include worker self-report, bioinstrumentation, and direct observation. The optimal choice among methods depends on characteristics of the methods as well as of the jobs under study. Job exposure can be considered a weighted sum of the different task-specific exposures that make up the job, with weights coming from task distributions (Winkel and Mathiassen, 1994). Each of two components—exposures in each task and the relative frequency of each task—must be estimated. Workers with the same job title may have different exposure levels because of between-worker variability in either the duration and distribution of tasks within jobs or the exposures within tasks. Furthermore, job title may indicate homogenous exposure groups for some stressors, such as repetitiveness and force demands, while other features such as posture may vary widely among workers in the same job (e.g., Punnett and Keyserling, 1987; Silverstein, Fine, and Armstrong, 1987). In highly routinized or cyclical work, such as that at a machine-paced assembly line, without job rotation there is only one task, the short duration and regularity of which make the exposure determination a relatively simple problem. In contrast, in nonroutinized work, such as construction and maintenance, determination of task distributions over an extended period of time may be a more difficult undertaking. As jobs

OCR for page 85
Page 92become less routinized, i.e., less predictably structured, valid estimation of both task distributions and task-specific exposures becomes increasingly challenging. Typically, both observational and direct measurement techniques generate highly detailed, accurate exposure analyses for a relatively short period of elapsed time in each job. Most protocols for these methods assume that the work is cyclical, with little variability over time, so that it is reasonable to measure exposures for a short period and extrapolate them to the long term. But many jobs do not fit this model: they are not comprised of work cycles, or the cycles are highly variable in their total duration or content (the number or sequence of steps that comprise each cycle) and do not account for all of the work performed by an individual with any given job title. For these jobs, it would be infeasible to undertake continuous measurements for entire cycles as an exposure assessment strategy, because either there are no cycles, or a very large number of (long) cycles would have to be recorded in order to quantify accurately the total and average duration of exposures. With short measuring times, the data collected are of uncertain representativeness because these time periods do not match the duration of exposures that are thought to be relevant to musculoskeletal disorder development. A versatile alternative for estimating physical exposures is the use of data collected directly from workers. Such reports may address both task-specific exposures within jobs and the distributions of tasks performed by each worker. In addition to being time-efficient, self-reports permit assessment of exposures in the past as well as the present and may be structured with task-specific questions or organized to cover the job as a whole. Some researchers have explicitly recommended a composite approach to the analysis of nonroutine jobs, in which task-specific exposures are measured directly and the temporal distribution (frequency and duration) of each task is obtained from self-report. Self-reported data can take various forms, including duration, frequency, and intensity of exposure. In some studies, absolute ratings have agreed well with observations or direct measurements of the corresponding exposures, while others have diverged significantly, especially with use of continuous estimates or responses that required choices among a large number of categories (e.g., Burdorf and Laan, 1991; Faucett and Rempel, 1996; Lindström, Öhlund, and Nachemson, 1994; Rossignol and Baetz, 1987; Torgén et al., 1999; Viikari-Juntura, 1996; Wiktorin et al., 1993). Retrospective recall of occupational exposures has been frequently employed in studies of musculoskeletal disorders, but there are few data on the reproducibility of such information. Three studies have examined the potential for differential error (i.e., information bias) in self-reported exposure with respect to musculoskeletal disorders with mixed results;

OCR for page 85
Page 93some risk estimates were biased away from the null value, some toward it, and others not at all (Torgén et al., 1999; Viikari-Juntura, 1996; Wiktorin et al., 1993). In the REBUS1 study follow-up population, Toomingas et al. (1997a) found no evidence that individual subjects systematically over-rated or underrated either exposures or symptoms in the same direction. Self-reported exposures have promise, but their validity depends on the specific design of the questions and response categories. A variety of instrumentation methods exist for direct measurement of such dimensions as muscle force exertion (electromyography), joint angles and motion frequency (e.g., electrogoniometry), and vibration (accelerometers). For example, the goniometer has been used in a variety of studies of wrist posture, including field assessments of ergonomic risk factors (Moore, Wells, and Ranney, 1991; Wells et al., 1994), comparisons of keyboard designs (Smutz, Serina, and Rempel, 1994), and clinical trials (Ojima et al., 1991). Hansson et al. (1996) evaluated the goniometer for use in epidemiologic studies, and Marras developed a device for measuring the complex motion of the spine (Marras, 1992). While many consider these methods to represent collectively the standard for specific exposures, each instrument measures only one exposure, and usually only at one body part. When multiple exposures are present simultaneously and must be assessed at multiple body parts, the time required to perform instrumented analyses on each subject may limit their applicability to epidemiologic research (Kilbom, 1994). Another practical concern is the potential invasiveness that may interfere with job performance, alter work practices, or reduce worker cooperation. Thus, there is a trade-off between the precision of bioinstrumentation and the time efficiency and flexibility of visual observation and worker self-report. As discussed in Chapter 6, gross categorical exposure measures (e.g., >10 kg versus < 10 kg) used in epidemiologic studies may limit the possibility of observing an exposure-risk relationship; a continuous measure based on bioinstrumentation might make such a relationship more apparent. Thus, their high accuracy (for the period of measurement) gives these methods utility for validating other methods on population subsets and added value when they can be applied in epidemiologic studies. A large number of observational methods for ergonomic job analysis have been proposed in the last two decades (see Kilbom, 1994). These 1In the original REBUS study conducted in 1969, participants were asked to complete a questionnaire regarding health status—all selected were given a medical examination. A diagnosis of musculoskeletal disorder required signs and symptoms. The follow-up study, conducted in 1993, asked the younger participants in the original REBUS study to participate in a reexamination.

OCR for page 85
Page 94include checklists and similar qualitative approaches to identify peak stressors (e.g., Keyserling et al., 1993; Stetson et al., 1991). The limitation with checklists is that they provide little information beyond the presence or absence of an exposure, with a possibly crude estimate of the exposure duration. The qualitative approaches are not likely to provide sufficient detail to effectively assess exposure for epidemiologic studies. The most common observational techniques used to characterize ergonomic exposures are based on either time study or work sampling. Both of these techniques require a trained observer to characterize the ergonomic stressors. Methods based on time study (e.g., Armstrong et al., 1982; Keyserling, 1986) are usually used to create a continuous or semicontinuous description of posture and, occasionally, force level. Therefore, changes in the exposure level, as well as the proportion of time a worker is at a given level, may be estimated. Because methods based on time study tend to be very time intensive, they are better suited to work with fairly short and easily definable work cycles. A different approach, work sampling, involves observation of worker(s) at either random or fixed, usually infrequent, time intervals and is more appropriate for nonrepetitive work (e.g., Karhu, Hansi, and Kuorinka, 1977; Buchholz et al., 1996). Observations during work sampling provide estimates of the proportion of time that workers are exposed to various stressors, although the sequence of events is lost. Though less time intensive than time study, work sampling still requires too much time for use in an epidemiologic study, especially one that employs individual measures of exposure. There are also a few highly detailed, easily used observational analyses for use as an exposure assessment tool in an epidemiologic study. These methods employ subjective ratings made by expert observers. For example, Rodgers (1988, 1992) has developed methods based on physiological limits of exposure that rate effort level, duration, and frequency. The method developed by Moore and Garg (1995) employs ratings similar to those of Rodgers and adds posture and speed of work ratings. Moore and Garg's strain index is designed to estimate strain for the distal upper extremity. It is the weighted product of six factors placed on a common five-point scale (subjective ratings of force, hand/wrist posture, and speed of work and measurement of duration of exertion, frequency of exertion, and duration of task per day). The strain index is a single priority score designed to represent risk for upper extremity musculoskeletal disorders and is conceptually similar to the lift index for low back disorders. The lift index was developed as part of the revised NIOSH lifting equation (Waters et al., 1993) and is the ratio of the load lifted and the recommended weight limit. Recently, Latko et al. (1997) developed a method employing visual analog scales for expert rating of hand activity level (called HAL). The

OCR for page 85
Page 95method has also been generalized to assessment of other physical stressors, including force, posture, and contact stress (Latko et al., 1997, 1999). The HAL employs five verbal anchors, so that observers can rate the stressors reliably. In an evaluation, a team of expert observers comes to a consensus on ratings for individual jobs. These ratings correlated well with two quantitative measures, recovery time/cycle and exertions/second, and are found to be reliable when compared with ratings of the same jobs 1.5 to 2 years later (Latko et al., 1997). In sum, there are many methods for assessment of ergonomic exposures. The challenge for ergonomists and epidemiologists is to determine a method of characterizing level of exposure that is efficient enough to permit analysis of intersubject and intrasubject variability across hundreds of subjects and that can also produce exposure data at the level of detail needed to examine etiologic relationships with musculoskeletal disease. The HAL, as developed by Latko, is easy to apply and has proven to be predictive of the prevalence of upper extremity musculoskeletal disorders in cross-sectional studies. Psychosocial Exposures Measures of psychosocial exposures reported in the literature are obtained through the use of various self-report surveys. These surveys are typically presented to subjects in a paper format in which the subject is requested to complete a series of questions. These survey tools typically comprise multiple scales used to assess psychosocial risk factors. Many of these measures assess the construct of interest using a continuous scale of measurement, by which it is possible to provide a measure of exposure in terms of degree, and not simply whether it was present or absent. Response items vary depending on the scale and typically range from 0-5, 0-7, or 0-10, with options anchored so that the respondent has a frame of reference for various responses. Some measures are standardized, well-developed, self-report tools whose psychometric properties (reliability and validity) have been established based on past research, while other items or scales were developed for the purposes of a single study. Currently, all scales used are self-report. Depending on the length of the survey, the time to completion can range from 10 minutes to several hours. It is rare that the perceptions reported by the respondent are corroborated by an independent assessment tool or process (e.g., supervisor or coworker evaluations or direct observation of a workplace). Although it can be helpful to assess such independently collected information to support workers' reports of their sense or opinions of their environments, perceptions are, by their nature, best collected through self-report.

OCR for page 85
Page 173 Poor job control (reliability Q referenced): Control over amount of work, availability of materials, policies and procedures, pace, quality, and scheduled hours 1.6 1.0-2.7 37% Marcus and Gerr, 1996 449 female office workers of 40 years or younger Pain or soreness of the neck or shoulder at least once per week of at least moderate intensity during the month preceding completion of the questionnaire. Prevalence neck or shoulder symptoms 63% Low job security (likely to lose job) 2.23 1.3-3.7 54% High perceived job stress: High job stress previous 2 weeks 2.47 1.2-5.1 60% Pain or soreness in the finger, hands, wrists, forearms or elbows, or numbness or tingling of the fingers of at least moderate intensity during the month preceding completion of the questionnaire High perceived job stress: High job stress previous 2 weeks 2.04 1.0-4.0 43% Prevalence arm symptoms is 34% continues

OCR for page 85
Page 174 APPENDIX TABLE 4.9 Continued Author Study Population Outcomes Psychosocial Risk Factors Risk 95% Confidence Interval Attributable Fractiona Pickett and Lees, 1991 79 data entry office workers in 5 different offices of the same company Self-reported work-related symptoms of shoulder, arm or hand/wrist; precise question not presented; shoulder symptoms prevalence 76% Rest breaks in task Not presented - - Perceived stress: Occupational stress: single item, portion of time at work operators perceived themselves to be under emotional or mental stress (rarely, sometimes, almost, always) Not presented - - Hand-wrist symptoms: Prevalence 52% Occupational stress Not presented - - Arm symptoms: Prevalence 53% Occupational stress Not presented - -

OCR for page 85
Page 175 Pocekay et al., 1995 3,175 semi-conductor workers from 8 manufacturing companies Several health outcomes; relationships with risk factors not separately presented: Any distal upper-extremity symptom = physician diagnosed carpal tunnel syndrome within the past year, questionnaire (self-report) diagnosed CTS, hand/wrist pain daily for 1 week within past year; elbow/forearm pain daily for 1 week within the past year Medical diagnosed CTS in past year Hand/wrist pain daily for 1 week within past year Elbow/forearm pain daily for 1 week within the past year Epidemiologic CTS in past year Daily shoulder pain for 1 week in past year Medical diagnosed tendinitis in past year Perceived stress: Job stress index: job is very demanding; job is very tiring; job is very stressful Range 1.1-1.5 - 9-33% Non-work stress: Somatization index (symptoms over the most recent 4 weeks: feeling tired, tingling in fingers or toes, heart palpitations, feeling irritable, light headedness, lack of muscle strength, chest tightness) Range 1.4-1.8 - 28-44% 1.4 - 28% continues

OCR for page 85
Page 176 APPENDIX TABLE 4.9 Continued Author Study Population Outcomes Psychosocial Risk Factors Risk 95% Confidence Interval Attributable Fractiona Silverstein, Fine, and Armstrong, 1987 136 workers more than 4 years employed in an investment casting plant Carefully defined and well described (additional paper) outcome assessment of hand/wrist CTD by interview and physical examination disorders, nerve entrapment, non specific included as clear pattern shown in: Interview: Pain, numbness, tingling Lasting > 1 week or > 20 times last year No acute traumatic onset Onset not before 1983 study job Physical Examination: Characterize signs and endpoints Exclude referred symptoms Prevalence: Low exposed: 5% High exposed (repetition and force): 13% Poor job satisfaction, assessed by interview: How often do you find your work satisfying? Very often/fairly often/sometimes/rarely Little variation in job dissatisfaction was presented In general the workers reported to be very or fairly satisfied with their work No association - - No stimulus from work, assessed by interview with the following question: How often do you find your work interesting? Very often/fairly often/sometimes/rarely No association - -

OCR for page 85
Page 177 Hoekstra, Hurrell, and Swanson, 1995 108 workers at 2 teleservice centers Any symptoms (pain, numbness, tingling, aching, stiffness, or burning) within the preceding year and all of the following: No preceding acute and non occupational injury Symptoms began after starting the current job Symptoms lasted > 1 week or occurred at least once a month within the past year Symptoms were reported as moderate (midpoint) or worse on 5-point intensity scale Poor job control: Job control Not presented was measured with a NS multi-item scale. Not presented NS - - Variability of workload: Not presented Perceived workload NS variability was measured with a multi-item scale; specified as continually changing workload during the day Although not explicitly stated, high-perceived workload variability is presumed to be the risk full exposure and not low Not presented NS - - Shoulder: prev 35% Elbow prevalence: 20% Poor job control Not presented NS - - Variability of workload Not presented NS - - Hand/wrist: prevalence: 30% Poor job control Not presented NS - - Variability of workload Not presented NS - - continues

OCR for page 85
Page 178 APPENDIX TABLE 4.9 Continued Author Study Population Outcomes Psychosocial Risk Factors Risk 95% Confidence Interval Attributable Fraction a Toomingas et al., 1997b 83 male furniture movers, 89 female medical secretaries, 96 men and 90 women from the working population, resulting in 358 men and women in various occupations, but with large groups of males with heavy work and females with office work Nordic Questionnaire and 24 signs recorded at the physical examination were included concerning neck/shoulder/elbow/hand/wrist. Signs and symptoms were combined in two relevant syndromes for the neck/upper extremities, i.e., tension neck syndrome and tendalgia of the upper extremities. Outcome-variables: Symptoms (Nordic) Signs (24 in total) Syndromes (combined signs and symptoms) High job demands: Multi-item scale JCQ - - - Low control: Multi-item scale: Low decision latitude (low control and little stimulus from work, learning ability, etc.) NS NS - Poor social support Multi-item scale High job strain: Job demands divided by decision latitude Shoulder 3.2 1.3-7.8 68% PR Hand/wrist 1.8 1.1-3.1 44% PR Shoulder 2.2 1.0-5.1 55% PR Hand/wrist 1.5 0.8-2.8 33%PR

OCR for page 85
Page 179 Westgaard, Jensen, and Hansen, 1993 52 female production workers (chocolate plant) and 34 female office workers Symptom score for each region based on intensity and frequency of symptoms in the last 12 months and their whole employment period within the present function Ranged from no symptoms to daily occurrence of severe symptoms: Shoulder/neck Perceived stress Overall psychosocial work-related stress score based on a 14 item questionnaire with items on: Mental stress due to work task, new work tasks Stress due to reorganization Demands because of efficiency, work speed Effect of redundancies Personal development Support by colleagues and supervisor Total score dichotomized, but the way it is dichotomized is unclear Positive association - - Arms Perceived stress Positive association - - continues

OCR for page 85
Page 180 APPENDIX TABLE 4.9 Continued Author Study Population Outcomes Psychosocial Risk Factors Risk 95% Confidence Interval Attributable Fractiona Westgaard and Jansen, 1992a, 1992b 210 production workers, mainly sewing machine operators in several plants of a garment industry and 35 office workers employed at this industry Symptoms score based on frequency and intensity Shoulder/neck Nonwork distress Scoring of psychological problems by the interviewer in low, intermediate, high after a worker interview (based on interviewer's impression) High indicates recurring depression or anxiety No association - - Arms Nonwork distress No association - - Zetterberg et al., 1997 564 car assembly workers (440 men and 124 women) Symptoms: Nordic Questionnaire Extensive physical examination; 114 signs established (76 hand/wrist), including: myalgia, impingement, epicondylitis, nerve entrapment, tendinitis, joint signs symptoms/signs shoulder Perceived stress Work satisfaction 5-item work APGAR but questions taken apart in support, satisfaction, and stress Positive association - - Low social support Positive association - - Poor job satisfaction Positive association - -

OCR for page 85
Page 181 Symptoms/signs wrist hand Perceived stress Positive association - - Low social support - - - Poor job-satisfaction - - - Magnavita et al., 1999 2,041 physician sonographers Self-reported question with a 21-item symptom list. Divided in syndromes by factor analysis Hand/wrist concerned 6 items. Three or more of these symptoms of the hand/wrist region is the effect studied. Prevalence: 5.3% Limited rest break opportunities: Long average time executing sonology activities without intermittent rest break is only a proxy for rest break opportunities. This variable measures more physical load than psychosocial load and additional measures of perceived time pressure or job demands were not included 1.50 1.1-2.1 33% a The attributable fraction is calculated with the OR as estimate for the relative risk. This approximation will be fairly accurate when the prevalence of the health effect at study is below 20-30%. With higher prevalences the OR is an overestimation of the relative risk and thus the attributable fraction is overestimated. Attributable fractions were not presented by the article authors and, therefore, were calculated using results available from the data presented in the published studies. NOTE: CTS = carpal tunnel syndrome; JCQ = job content questionnaire; NS = not significant; S = significant; VDU = video display unit.

OCR for page 85
Page 182 APPENDIX TABLE 4.10 Summary Table of Psychosocial Factors and Work-Related Upper Extremity Disorders: Longitudinal Studies Author Study Population Outcomes Psychosocial Risk Factors Risk 95% Confidence Interval Attributable Fraction Bergqvist, 1995 341 visual display unit workers Shoulder-neck discomfort during the last 12 months (Nordic Questionnaire) Prevalence total population at end of follow up: 44% Increased perceived monotony - - - Prevalence unexposed not presented Elbow-shoulder discomfort during the last 12 months (Nordic Questionnaire) Prevalence total population at end of follow up: 27% Increased perceived monotony - - - Hand/wrist discomfort during the last 12 months (Nordic Questionnaire) Prevalence total population at end of follow up: 16% Increased perceived monotony 1.7 0.6-4.4 41% 3.1 1.2-7.8 68%

OCR for page 85
Page 183 Ferreira, De Souza Conceição, and Hilário Nascimento Salvida, 1997 (retrospective cohort) 106 bank employees (telephone tasks) History in medical records of one or more periods of upper extremity symptoms with time away from work confirmed by at least two medical specialists (who based the diagnosis on recurrent pain with or without clinical evidence from physical examination of tendon or tendon sheath impairment or nerve entrapment based) Time pressure increase (i.e., shorter processing time task) Change in management (new administrative procedures) Registered overtime work Rest break opportunities Limited rest breaks and relatively high time pressure were associated with WRUED - - Roquelaure et al., 1997 65 (55 women, 10 men) cases with CTS matched with 65 controls (55 women, 10 men) Cases and controls recruited from television manufacturing plant Case: blue-collar worker, 18-59, with medical history of carpal tunnel syndrome (CTS) between 1/1/1990 and 12/30/1992. Subjects with a history of CTS problems, diabetes, thyroid or musculoskeletal dysfunction, malignancies, rheumatic diseases before 1990 excluded. Referent: blue-collar, same gender, same year of birth, free of CTS or musculoskeletal disorders of the upper limb from 1984 to 1992. Work organization factors: No job rotation between different work stations No association for: Autonomy: possibility to choose the way the work is done Rest break opportunities: duration and number of breaks 6.3 2.1-19.3 NA NOTE: CTS = carpal tunnel syndrome; WRUED = work-related upper extremity disorder(s).