Work-related muscular neck-shoulder pain is a major health risk among computer operators, particularly in women (Hagberg & Wegmann, 1987; Oberg & Astrom, 2000; Punnet & Berquist, 1997; Sauter, Schleifer, & Knutson, 1991). Many risk factors for musculoskeletal disorders have been identified, especially for work using visual display terminals (VDTs; Berquist, Wolgast, Nilsson, & Voss, 1995; Faucett & Rempel, 1994). Factors such as constrained postures, repetitive work tasks, mental demands, and improper workstation ergonomics may contribute to the development of these disorders (Hansson et al., 1992; Jensen, Ryholt, Villadsen, & Christensen, 2002; Takala & Viikari-Juntura, 1991; Veiersted, Westgaard, & Andersen, 1993), but the pathophysiological mechanisms remain unclear. Typically, suboptimal ergonomic solutions lead to increased muscle activity, which normally is demonstrated by surface electromyography (EMG; Aaras, 1994a, 1994b). During VDT work the level of muscle activity is low, and there is no clear evidence of a strong relationship between increased electrical activity of muscles and the development of musculoskeletal disorders (Westgaard, Jansen, & Jensen, 1996).
Although it appears that unfavorable working conditions can increase both musculoskeletal symptoms and muscle activity, a causal link between increased muscle activity and pain has not been proven. Hagg's Cinderella hypothesis (Hagg, 1991) tried to resolve this discrepancy by postulating that overload may be attributable to prolonged, single motor unit (MU) activity not detected by surface EMG. According to this hypothesis, the continuous activity of specific MUs during low-level contraction may metabolically overload specific MUs because of lack of recovery, causing muscle pain and strain. Potential mechanisms of how continuous activity of single MUs might provoke pain were reviewed by Sjogaard, Lundberg, and Kadefors (2000). Thus, to study muscular load, it is necessary to register the activity of single MUs over a prolonged period.
Several investigations have shown that prolonged static muscle contraction is related to head, shoulder, and neck pain (Hagberg & Wegmann, 1987; Kvarnstrom, 1983; Onishi et al., 1976). Secretaries who type during a major part of their workday have a high prevalence of shoulder myalgia (Kamwendo, Linton, & Moritz, 1991). Aaras (1994b) and Veiersted, Westgaard, and Andersen (1990) suggested that long-lasting, low-level, static postures are the cause of those problems. Experimental studies by Kleine, Schumann, Bradl, Grieshaber, and Scholle (1999) found that the increase in trapezius muscle activity during keyboard typing by secretaries was, to a certain extent, related to postural changes. In a real-life study of 72 female supermarket cashiers who were exposed to daily repetitive loads during work, the women with pain had a more pronounced increase in surface EMG activity during work than did those without pain (Lundberg et al., 1999). In a large review of work-related back pain, Devereux, Buckle, and Vlachonikolis (1999) found a cumulative risk for physical and psychosocial factors.
In the majority of the aforementioned studies, surface EMG recordings were used. Despite numerous studies that have shown the potential usefulness of the surface EMG approach, however, this technique still has limitations related to the type of contraction performed and to the sensitivity and specificity of its diagnostic properties. In order to understand the physiological mechanisms behind the development of muscle fatigue, pain, and strain at low-intensity work, one must accurately decompose EMG recordings at the MU level. An increase in surface EMG activity indicates an increased load, but it need not be a risk factor for pain. Following the line of thought expressed by Hagg's (1991) hypothesis, prolonged activity of a single motor unit could be a key factor for cellular overload that would induce pain. …