Incidence and prevalence of upper-extremity musculoskeletal disorders. A systematic appraisal of the literature

Upper extremity disorders (UEDs) are a major problem in modern society. Besides the impact on patients themselves, the disorders also form a huge economic burden due to costs for sick leave and health care. UEDs affect people all over the world. In the early 1980's in Australia Hocking [1] even reported an epidemic disease of a disorder he called RSI (repetitive strain injury). Numerous other terms have been used to indicate UEDs such as cumulative trauma disorders, physical overuse syndrome, and occupational cervicobrachial disorders. UEDs comprise various clinically defined (e.g. carpal tunnel syndrome) and undefined conditions of muscles, tendons, or nerves in the upper extremity due to multiple factors. Not only occupational use of the upper limbs, but also psychosocial work characteristics such as high job stress [2], high job demand [3], non-work-related stress [2] and personal characteristics such as coping [4] can cause UEDs. Most UEDs are manifested by pain, discomfort, or tingling in the upper extremity [5].

In medical literature authors repeatedly suggested that during the last decade's data are reported to indicate the extent, and in some cases increase of UEDs over time in Australia, Canada, the USA, France, The Netherlands, and elsewhere [1, 2, 6–11]. For example, based on workers' compensation claims Silverstein et al [12] reported a dramatic increase of UEDs since the early 1980s in the USA affecting workers in virtually every industry. In 1981, 28,6% of the allowed workers' compensation claims in New York State concerned UEDs and by 1986, these numbers were increased by 10,2% [13]. In 1989 the total U.S. workers compensation costs for UEDs was estimated to be $563 million [14]. Also in the early 1990s UEDs have dramatically increased in incidence according the data from the U.S. Bureau of Labor Statistics, 1998a [15].

In Ontario, Canada, UEDs constituted up to 24% of lost-time workers compensation claims in 1992 [16]. In 2000/01, one in ten Canadians aged 20 or older reported an UED that was serious enough to limit their normal activities in the previous 12 months [17]. In 2000 the Health Council of the Netherlands reported that if no distinction is made on the basis of duration or seriousness, the prevalence of UEDs in the Netherlands was between 20 and 40 percent [18].

A systematic appraisal of worldwide incidence and prevalence studies may permit us to gauge the range of incidence and prevalence of UEDs in various countries and, where possible, to pool data. It provides the basis for determining whether these estimates of UEDs are increasing over time. The data are also needed to estimate the size of study populations for experimental and preventive trials. Therefore, a systematic appraisal of the worldwide incidence and prevalence rates of UEDs reported in available studies will be presented here.

In this systematic appraisal worldwide incidence and prevalence rates for UEDs available in scientific literature were collected. No studies were found with regard to the incidence of UEDs that met the inclusion criteria. The estimates of the prevalence rates varied enormously across the 13 included studies. The point prevalence ranged from 1.6–53% and the 12-months prevalence ranged from 2.3–41%. One study reported on the lifetime prevalence (29%). Only Picavet et al [22] studied the prevalence in an open population. The low point prevalence they reported can not be compared with the other studies available, because they all studied a specific (working) population. In addition, Picavet et al [22] reported on the occurrence of 'RSI', while the occurrence of an epicondylitis (around 11%) and a tendonitis or capsulitis (for the whole body they reported a prevalence rate around 16%) were reported separately and therefore not included in 'RSI'.

In this study studies were included that reported incidence and prevalence rates of the whole upper extremity. Studies, which reported incidence or prevalence rates on different regions of the upper extremity separately, but give no estimates for the whole upper extremity, were excluded. Reviews on the prevalence rates of a specific disorder or complaints in one region of the upper extremity have been reported elsewhere. For example, the estimates of the occurrence of the Carpal Tunnel Syndrome in different occupational groups was studied by Hagbert et al [33] and varied between 0.6 and 61%. Luime et al [34] reported on prevalence rates of shoulder pain studied in open population: the point-prevalence ranged from 7 – 27% and the 12-months prevalence ranged from 8.4 – 20%

In general, higher prevalence rates of UEDs were found in women then in men and the estimates of self-reported complaints were higher than those acquired by using (in addition) physical examinations. No evidence of a clear increasing or decreasing pattern over time was found. Although period prevalence can be more biased then point prevalence because of incomplete response or due to recall bias [35], 'firm' conclusions can not been drawn because of the diversity of terms and definitions of UEDs used in the included studies.

To describe the conditions a variation of terms such as 'pain', 'disorders', 'complaints', 'syndrome', 'symptoms', and 'injury' are used in the literature. Because of the different meanings of the terms, it is important to give sound arguments when using certain terms. For example, it you want to describe specific and non-specific cases, using the term disorder is not very clear, because a 'disorder' indicates a specific disease, which can be diagnosed by fixed criteria. All terms used for UEDs in the included studies, except those used by Picavet et al [22] and Fry et al [24] indicated the location of the condition. In our opinion, it is practical and functional to use the localization of the conditions in the term.

Although the term used for UEDs is important because of the perception it causes and the clarity of the medical condition, the definition is even more important. This is not only the case for researchers when they want to compare data of different studies, but also for medical and paramedical staff, so they can speak in an unambiguous way or 'language'. This unambiguous 'language' has to make sure that physicians and other healthcare workers have in mind and speak about the same condition when they discuss the subject or, for example to evaluate the (multidisciplinary) treatment of one of their patients. The case definitions used in the included studies varied enormously. Although studies reporting prevalence rates for UEDs were not included and this appraisal was limited to studies which included 500 cases or more and studies of which the data were published in scientific literature, the diversity of case definitions and classification of UEDs that was found was substantial. This is a general problem and reported in literature by many authors before [36–39].

The diversity in terms and case definitions of EUDs in the included 13 studies prevented any meaningful pooling of data. Drawing comparisons between countries, different working population and assessment of changes in time within a population or country could therefore not be carried out in a quantitative manner.

Different questionnaires and tests used for the physical examinations were presented in the studies; little was said about the validity and reliability of the measurement tools. Developing criteria for classification or diagnosis would be easy if gold-standard diagnostic tests would be available. Unfortunately, no criterion standard for any of the upper extremity soft tissue musculoskeletal conditions is available [37].

If we want to make progress in this field, the first requirement is to agree on unambiguous terminology and classification of EUDs. Physicians and other healthcare workers dealing with patients with these conditions should be involved in such a project. Studies of classification criteria suggest that expert clinicians can more accurately identify cases than most history, physical examination, or laboratory parameters [40]. Furthermore, involving all key disciplines dealing with patients with UEDs will make implementation of the results more successful. Therefore, a multidisciplinary project on national or international level in which all key disciplines cooperate with the intention to achieve multidisciplinary consensus on terminology and classification of UEDs is recommended. When they have agreed about an 'unambiguous' language, the next step is to achieve consensus about valid diagnostic criteria for UEDs and to study the best (multidisciplinary) prevention and/or treatment.

No studies were found with regard to incidence rates of UEDs and there are substantial differences in reported prevalence rates on UEDs. One of the main reasons for this is the absence of a universally accepted way of labelling or defining UEDs. Health professionals and policy makers should be aware of this problem when they estimate the occurrence of the conditions in populations and the necessary demand and related costs for health care.