Cover Image
close this bookWomen Encounter Technology: Changing Patterns of Employment in the Third World (UNU, 1995, 356 pages)
close this folder13. Gender perspectives on health and safety in information processing
View the document(introductory text...)
View the documentInternational trends in information processing employment
View the documentReconstructing women as 'cheap' labour: New technology employment or the same old story?
View the documentHealth hazards of work with computers and keyboards: The experience from Australia and Europe
View the documentThe relationship between RSI and technology in the workplace
View the documentLDC experience
View the documentLearning from international experience
View the documentNotes
View the documentReferences

(introductory text...)

Learning from international experience

Ruth Pearson

International trends in information processing employment

The nature of employment in information processing is extremely varied. Women's employment reflects the full range of jobs from the low-skilled, low-paid data entry jobs through to the high-skilled, high-status professional jobs of systems analysts and computer programmers, though most are concentrated towards the low-skilled end of the spectrum (Pearson and Mitter, 1993). The dynamics of the adoption and dissemination of computer technology are also varied, reflecting the very different socioeconomic realities in different sectors and countries. In developed countries computer technology has been introduced into most offices in all sectors. The diffusion of information technology in developing countries is much more piecemeal. In many cases diffusion has reflected the pattern found in developed countries, where public sector entities and private corporations have followed a similar pattern of computerization, starting with mainframe computers in the 1970s and progressively adopting microcomputers and PCs during the 1980s.

The bulk of women's employment in computer-related occupations in both developed and developing countries has been concerned with the entry and manipulation of data via computer keyboards. This has long been recognized within developed countries. The fact that telecommunications operate internationally implies that information processing services can be relocated to low wage economies. There is a well-established, if limited, trend to relocate or subcontract office services to low-wage developing countries, where women are employed to enter data at a fraction of the cost of comparable labour in developed countries (Davis and Stack, 1993; ILO, 1990; OTA, 1985; R. Pearson, 1993; Posthuma, 1987). Conversely, highly-skilled software development work is purchased from developing countries' computer professionals, either in situ or by importing the professionals on short-term contracts, a process known as 'body-shopping' (Heeks, 1989; Mitter and Pearson, 1992; Schware, 1992).

Studies of these processes have focused on the opportunities they present for less developed countries (LDCs), in terms of employment for school leavers and earning foreign exchange, and on the unsubstantiated expectation that such relocation of work offers the opportunity to 'leapfrog' into a new technology age in which small LDCs could have a knowledge-based comparative advantage (Girvan, 1989; Hanna, 1991).

The discussion of the diffusion of computer technology in developing countries, either through domestic computerization or through the internationalization of data entry or software development, has almost totally ignored the contractual position, wages, training and promotion and health and safety of new technology workers. Given the amount of interest in these issues within developed countries, particularly as regards health and safety, this is a telling omission.

Reconstructing women as 'cheap' labour: New technology employment or the same old story?

The few case studies available which detail women's employment in international data entry are based on research in the Caribbean, which has become a major centre for external sourcing, particularly of commercial time-sensitive data for North American corporations (Posthuma, 1987; Freeman, n.d.; R. Pearson, 1993). One of the catalysts for this research was the publication of a major US Congress study on Office Automation (OTA, 1985), which considered 'Offshore Office Work' as an alternative cost-reducing strategy to 'Home-Based Automated Office Work', particularly for data input by well-educated Caribbean workers who earned less than one sixth of their American counterparts. This study makes one passing reference to the fact that 'women comprise 99 per cent of the workers' (ibid.: p. 217). OTA focuses on the cheapness1 of LDC labour compared with comparable workers in the United States, and draws a parallel with the concept of 'cheap labour' in analyses of offshore manufacturing during the previous decades. The report also notes the availability of high quality and productive labour (ibid.: p. 223). Both host governments and employers have tended to see the relocation of work as an opportunity for developing countries, since it is a labour-intensive activity involving very little capital outlay, which offers numerous benefits to the host countries:

It is a 'clean' industry, without the heavy equipment, large space requirements, and pollution often associated with other industrial enterprises. The industry provides, at the very least, rudimentary training in computer use - a rare opportunity for workers in less developed countries. finally, it establishes a foundation on which funkier advancement in computer-related industries, such as software development technical services and data transmission, may grow.

(ibid.: p. 225)

Other issues which have been discussed include the security of data transmitted internationally, the lack of a regulatory framework to control piracy, national security, the desirability or practicality of imposing tariffs on the value added to transmitted data (Davis and Stack, 1993; B. Harris, 1989; Locksley, 1986), and the regulation of trade in services (Ryan, 1990). Both radical commentators and voices from orthodox financial institutions such as the World Bank continue to see this form of internationalization as a cost-free, win-win scenario, in which foreign exchange is generated, unemployed workers are given training and jobs, and the seeds of a new place in the international division of labour are planted, to flower in the form of a new comparative advantage in the twenty-first century.

Some criticism of the relocation of computer-based services has been voiced by those who speak from the standpoint of US labour - organized or not. The AFL-CIO (American Federation of Labor-Confederation of Industrial Organizations) has maintained a protectionist position regarding offshore sourcing in the manufacturing sector, and has extended this to data entry operations (OTA, 1985: p. 224). More recently the extension of international relocation to what has been called - provocatively 'intelligent office work', as well as the contracting out of standard software packaging and the continuing practice of 'body shopping', has attracted the interest of professional associations in the West (Heeks, 1989; Miner and Pearson, 1992; Schware, 1992).

The OTA report dismisses talk of disadvantages of off-shore data processing:

The development expert claims that negative reactions, if any, in developing countries usually come from the more educated sectors and the labour movement or, as he put it 'people who are not worried about having a job'. Those citizens of developing countries who view US investment as exploitative are those who are generally of higher economic status and are not in need of the job opportunities that result from such investment.

(ibid.: p. 226)

It is significant that there is no discussion of health and safety aspects of offshore office work, given that the same report devotes a whole shiny page chapter (Chapter 5) to 'Office Automation and the Quality of Worklife', which details the risks of stress, and the results of investigation and research on problems with eyesight, musculoskeletal systems, and reproductive systems. These problems are raised in relation to the increasing numbers of home workers who work with computers and visual display units in the United States (OTA, 1985; Di Martino and Wirth, 1990).

It would be fair to say then, that the literature on information technology, computers and employment in developing countries has not been concerned with the health and safety aspects of such employment, even when there is some awareness of these issues in the context of domestic employment.2

Since women workers form the vast majority of the workforce in LDCs associated with computer technology (Mister and Pearson, 1992; Ng, 1991) it can be concluded that once again women are regarded as a cheap and flexible workforce, able to adapt to the new work practices (Standing, 1989; Elson, 1991) which are at the cutting edge of labour relations (Glucksmann, 1986 and 1990).

In this sense, the absence of discussion about the risks to workers' health in LDCs is not surprising, and reflects the very similar positions taken as regards the expansion of women's employment in manufacturing for export in the 1970s and 1980s, when any discussion of the constraints and contradictions of such employment was attacked as self-seeking and unrealistic.3 Whilst the technology may be new, it would appear that women are in danger of being recreated in a familiar role - that of cheap and abundant labour.

Health hazards of work with computers and keyboards: The experience from Australia and Europe

Groups of workers who have suffered from conditions attributed to using new technology have brought the issues into public and political forums by their attempts to gain recognition and compensation for their conditions. An article in the International Labour Review accurately reports:

VDUs have aroused intense debate and often controversy over the implications for workers, particularly the risks to health. Although relatively few countries have passed laws or regulations referring specifically to VDUs, several have recommended codes of practice or guidelines for their use. These cover a variety of issues, such as advance notice, consultation and negotiation procedures with workers or their represent atives, training, job design and job security, rest breaks, maximum VDU use, protection during pregnancy, eye care and machine and workstation specifications.

(Di Martino and Wirth, 1990: p. 542)

This gives the impression that there is general agreement in 'several' countries, recognizing the health risks associated with working with computers and related technology, and that there are agreed procedures about prevention and worker protection. It might lead the reader to suppose that these health risks have been defined as industrial injuries for which insured and protected workers are entitled to some degree of compensation.

The reality is quite different. There is an enormous amount of controversy over the medical and legal issues relating to information technology and health risks. The lack of agreement has meant that recent legislation and recommendations within the European Union (EU), have provided much less worker protection than had been expected. It will be worthwhile to survey the medical controversies, and the differing legislative and political debates over repetitive strain injury (RSI) in Australia and the UK. This information will provide a basis for an analysis of the health and safety risks faced by new technology workers in developing countries, from a gender perspective.

Health risks and computer work

Although there is no medical and legal agreement of the degree of risk and vulnerability to various health conditions, the literature has established five types of health hazards which have been attributed to work with computers, and more specifically to the work situation of inputting or manipulating text or data using a visual display unit (VDU) or visual display terminal (VDT):4

1 Musculoskeletal; this includes a range of named disorders of the neck, upper limbs, shoulders and back, including tenosynovitis, tendinitis, peritendinitis, bursitis, epicondylitis, carpal tunnel syndrome, dupuytren's contracture, writers' cramp, ganglions, and cervicobrachial disorders (Putz-Anderson, 1988(a); Huws, 1987). These conditions are not identical with the list of thirty separate diseases in the International Classification of Disease Codes, which includes carpal tunnel syndrome, cervicobrachial syndrome, tenosynovitis, and ganglions (cysts), but which makes no mention of repetitive strain injury (Putz-Anderson, 1988(b): p. 601).

2 Deterioration of and problems with visual capacity, including eyestrain and fatigue, loss of focus and mobility, reduction in capacity to dilate pupils. and cataracts (ibid.). The symptoms linked to eye strain include blurred and double vision, irritability, headaches and migraines, nausea, and discomfort with contact lenses (DeMatteo, 1985).

3 Stress and fatigue; symptoms include short-term problems of fatigue, irritability, depression, headaches, migraine, insomnia, menstrual problems, and accidents, and long-term problems of heart disease, high blood pressure, depression, anxiety, dermatitis, ulcers and fertility problems.

4 Skin complaints including rosacea, acne, dermatitis, telangiectases, pustolosis, urticaria, ostitis and other unspecified changes (ibid.:, and Berg et al., 1990).

5 Reproductive hazards; miscarriages, congenital deformities and fertility problems associated with stress and with emissions of ionising and nonionising radiation from cathode ray tubes fitted in now obsolete computer monitors (Labour Research, 1984; DeMatteo, 1985; Brandt and Nielsen, 1990).

The literature on reproductive hazards remains deeply contested, and is based on reported or observed clusters of adverse pregnancy outcomes including miscarriages, stillbirths and abnormalities. The 1992 UK Health and Safety Regulations state that:

There has been considerable public concern among some groups of visual display unit workers in particular due to electromagnetic radiation. Many scientific studies have been carried out but taken as a whole they do not show any link between miscarriages or birth defects and working with VDUs.

(HSE, 1992: pp. 42-43)

We will confine our discussion to the first category, which has been labelled as RSI (repetitive strain injury), since these injuries can lead to a total inability to carry out many tasks (R.M. Pearson, 1990). It is worth noting however that figures issued by the Data Management Association in North America indicate that eyestrain is in fact the most commonly experienced health problem of VDU workers (Bodek, 1987). This finding is supported by data from Malaysia (Ng and Othman, 1991) and Japan (Shiga, 1987).

'If I can't see it doesn't exist': medical controversies over computer-related repetitive strain injuries

Musculoskeletal disorders cause the most severe incapacity amongst sufferers. Although back and shoulder ache feature very frequently amongst surveys of work-related problems of computer clerical workers, the term RSI applies particularly to problems with hands and wrists, and necks. The range of musculo-skeletal conditions which have been reported illustrates that the precise medical diagnosis can fall within a wide range of conditions, and of course can comprise a composite of two or more of these conditions (Bammer, 1990). Despite the frequency with which symptoms are reported amongst office workers, medical opinion remains divided over:

1. whether a term such as RSI has any meaning or usefulness;

2. the medical diagnosis of the symptoms;

3. whether there is in fact a physiological basis, rather than a psychological basis, for the reported symptoms; and

4. whether there is any evidence of causation which would link particular kinds of work with these symptoms.

It is beyond the scope of this paper to present an exhaustive review of the medical evidence and disputes which swirl around this issue, but it is useful to present some of the arguments, since the controversy over the existence and causation of RSI is important in contextualising the link between the gender bias of the new work categories and work organization associated with computers and attitudes to the health and safety of workers in LDCs.

What do we mean by repetitive strain injury? (It depends on who we are.)

In spite of the ubiquity of the symptoms of RSI there is no agreement about definitions or terminology. In Australia, the conditions of tenosynovitis, tendinitis and bursitis have been recorded as compensatable industrial conditions since the 1960s, and 'from the end of the 1970s newly established workers' health centres began to use the term 'teno' to cover all of the repetition injuries they were finding amongst their predominantly blue collar clientele' (Hopkins, 1990: p. 367, emphasis added). When the incidence spread to the newly established and expanding white collar segment of the workforce, the term 'repetition injury' began to be widely used by organizations such as the Australian Public Service Association, the National Health and Medical Research Council and the National Occupational Health and Safety Commission. This term was extensively adopted by the media and by many doctors who were not specialists but were required to respond to the increasing number of people presenting with symptoms consistent with what was widely understood by the term.

In the UK however, whilst the term 'repetitive strain injury' is widely used by the media, by sufferers (who have formed an RSI association), and by some doctors and specialists, it has been studiously avoided by the Health and Safety Executive who prefer to refer to 'work-related upper limb disorders. These range from temporary fatigue or soreness in the limb to chronic soft tissue disorders like peritendinitis or carpal tunnel syndrome. Some keyboard operators have suffered occupational cramp' (HSE, 1992: p. 41).

In the United States the term 'cumulative trauma disorder (CTD) has been used within the occupational health and safety literature. This is a much less transparent term for the general public. Although some agencies do use what is known as the Australian term (repetition strain injury), a lot of occupational medical literature still uses the generic 'musculo-skeletal problems', which is totally neutral in terms of causation' (Putz-Anderson, 1988b). In the USA there is not - as yet - a widely accepted acronym. It has been claimed that the absence of an acronym contributes to the lower rate of recognition and discussion of the problem in that country.

In Sweden, where discussion and remedial policy is well advanced, the terms adopted are ergonomically-related injuries (ERI) or occupational cervicobrachial disorder (OCD), collectively referred to as work-related musculoskeletal disorders. These terms reflect the consensus that occupational labour processes are implicated, but place the emphasis on ergonomic rather than other factors.

These terms are themselves being challenged or refined, not least by medical specialists who are engaged in a fierce dispute about the medical validity of the terminology utilized. Recent (equally disputed) additions to the nomenclature are 'overuse syndrome' (Fry, 1993) and 'refractory cervicobrachial pain' or RCBP (Quintner and Elvey, 1991, cited in Cohen et al., 1992).

There is disagreement as to the meaning of the terms utilized to describe the symptomology (which is not itself disputed), with much specialist medical opinion decrying the 'unscientific' nature of the generic terms employed. Without wishing to enter into this acrimonious dispute, the following extracts convey some of the flavour of the discourse:

Conventional medicine is not prepared to accept that a physical injury or disease process occurred in an upper limb, or indeed elsewhere, unless there are convincing and reproducible physical signs. The accepted signs of an injury and associated healing are those of tenderness, loss of function and associated histological findings of inflammation . . . until such independent support is forthcoming, the overuse concept should remain an eccentric and unproven hypothesis.

(Semple, 1993: p.25)

The self-generating term RSI is misleading for there is no scientific evidence proving that repetitive work causes either tissue strain or injury.... The scientific basis of modern medicine demands that disease is caused by a pathological process which, if not identifiable, has a rational hypothesis, thus enabling formulation and designation of appropriate management.... RSI now bears the hall mark of a sociopolitical phenomenon, rather than a medical condition, which on historical precedent, will decline when this basis of RSI is generally accepted.

(Ireland, 1988: p. 5)

Medical explanations of RSI: the doctors' dilemmas

The dispute as to whether the symptoms known as RSI are related to physical injuries, and if so of which type, continues without any sign of a resolution. The position of Semple, that without discernible and replicable physiological signs the patients' symptoms should be dismissed, represents one extreme of the debate. But amongst those who accept that the symptoms have a physiological cause, opinion is widely divided as to whether the explanatory physiological base lies in damage and malfunction of the central or peripheral nervous system (Quintner and Elvey, 1991; Cohen et al., 1992), or in muscle overuse (Fry, 1993), and whether its treatment belongs within the specialism of rheumatology, orthopaedic surgery, physiotherapy, or some combination of these (R.M. Pearson, 1993). These disagreements partly arise from competition between specialisms, and the tendency to work in narrower and narrower specialism rather than to view problems, diagnosis and treatment from a more holistic perspective. It is important to recognize that these problems have only been presented in large numbers within the last ten to fifteen years. In the absence of any sustained and reliable epidemiological studies, the current battles reflect the inability of conventional medicine to respond adequately to a changing pathology of occupationally-related injuries.

'If we can't find it in the body, it must be all in the mind'

As well as denying the existence of RSI, and the dispute over its physiological nature, there is also a broad body of medical opinion which asserts that the symptoms have a psychological cause. This argument asserts that the pain and other symptoms presented by sufferers result from unresolved psychological conflict or emotional disturbance (Bammer and Martin, 1988). This argument is most strongly proposed by Lucire, a female Australian psychiatrist who argues that RSI is a form of conversion disorder or mass hysteria in which patients exhibit neurotic reactions to keyboards and movements which have become symbols of danger to the vulnerable defined by Lucire as 'eggshell personalities, usually compulsive or dependent people who are powerless and dependent and who cannot otherwise express their righteous rage at their supervisors, employers and spouses, so resort to the use of their exquisitely symbolic pain as a mode of communication of their distress' (Lucire, 1986: p. 325). Needless to say those who support the view that the condition has a psychological basis are also aware that the majority of sufferers are women:

The condition commonly affects young to middle-aged and predominantly female employees engaged in low paying, monotonous low prestige occupations. The symptoms fail to respond to any form of treatment other than psychological counselling.... The treatment of RSI is unrewarding as it requires the patient's acceptance of the psychological basis of their condition.... It is surprising how often unsatisfactory social, family, marital and economic circumstances are expressed as job dissatisfaction.

(Ireland, 1988: p. 9)

Some commentators even suggest that sufferers fake physical symptoms which they have learned will earn them an appropriate diagnosis, or present symptoms consistent with hysteria:

[they] describe 1,000 patients, mostly female workers in offices and factories, and a very typical attitude of the affected upper limb, with half flexion of the elbow, wrist and fingers, without evidence of muscle atrophy.... Nowhere do they suggest that this posture is learned, whether consciously or not. Its pattern of incidence is like complaints of koro in South East Asian Chinese (sudden anxiety about recession of the penis into the surrounding skin), hysterical overbreathing in teenage girls at pop concerts, or several (other) conditions.

(Patkin, 1993: p. 11)

The last writer goes on to compare 'outbreaks' of ERS with outbreaks of computer-related 'facial dermatitis' in Sweden, angina in computer operators in North Carolina, and mass hysteria at a workplace in Singapore. It should give us pause for thought that the assumption here is that all these events were hysterical manifestations, rather than having a physiological explanation linked to the technological and organizational nature of the labour processes. This view of the nature of RSI cannot be understood without an appreciation of the trajectory of RSI in Australia, which is described below.

RSI: the Australian disease

The rapid growth of manifestations, diagnoses, compensation and dispute about RSI in the mid 1980s resulted in RSI being called the Australian disease, the Australian epidemic, or 'kangaroo paw'. This designation refers both to the emergence of the condition and public debate about it within Australia, and the suggestion, often made quite explicit, that RSI was a condition which was only manifest in Australia, because of the incentive for workers to produce symptoms which would enable them to get compensation.

A number of published studies contain succinct descriptions of the 'Australian disease' (Hall and Morrow, 1988; Hopkins, 1990; Reid et al., 1991; Bammer and Martin, 1988; Meekosha and Jakubowicz, 1991). One version is as follows:

In Australia between 1983 and 1987 there was an epidemic of upper limb regional pain which was concentrated among workers in occupations which involved either repetitive movement, or the adoption of constrained postures for lengthy periods of time (e.g. process workers and keyboard operators). Although the phenomenon of upper limb pain was observed among process operators in the late 1960s and early 1970s it only achieved epidemic status in 1983 when the first claims began to be made under workers' compensation. The rate of claims increased dramatically during 1984 and 1985, persisted through 1986, and then equally dramatically declined in 1987. The consensus of informed opinion is that the worst of the epidemic has passed.

(Hall and Morrow, 1988: p. 645)

Other accounts of the same phenomena are more cautious about making such an unproblematic link between the availability of workers' compensation and the rise and fall of RSI in Australia. Certainly the 1980s saw a rise in the number of cases recorded, in the number of people seeking compensation, and the number of computer keyboard operatives, particularly in the public service. The number of new notified cases has now diminished. However the assumption of a causal relationship between the availability of compensation and the incidence of RSI is rather suspect.

Part of the story lies in the fact that from the early 1980s the Australian government explicitly recorded the incidence of RSI, and accepted compensation claims on the basis of certification by the claimant's own doctor, who needed only to state that the claimant was suffering from RSI and required a specific period of rest (ibid.: p.367). In the United States and Britain at that time there was no available source of national information on the incidence of such conditions (Putz-Anderson, 1988b: pp.604- 605).

The absence of public recognition of RSI as an occupational and therefore compensatable disease in other countries meant there was less public awareness of the issues, and this reinforced the view in Australia that there was no comparable condition elsewhere, in spite of extensive, if not systematic, data relating to its prevalence in the United States, Scandinavia, and Japan (Hopkins, 1990: pp.366-367; English et al., 1989; Bammer and Martin, 1988; Polakoff, 1991)

The reported decline in notified RSI diagnosis in Australia was not the result of a reduction in the incidence of the symptoms of the condition. It was directly related to changes in the compensation procedure, rehabilitation programmes and medical and legal delegitimation of the condition (Meekosha and Jakubowicz, 1991). In many states the system was modified to reduce statutory employer liability and to introduce mandatory rehabilitation programmes. Instead of institutionalizing employer liability, the state has joined forces with those seeking to devalidate claimants' cases by arguing that the condition is psychological in origin. As the medical controversies rage on, workers who have participated in mandatory rehabilitation programmes report a lack of professional support and the exacerbation of their condition by intensive testing procedures. The rehabilitation therapy, officially designed to assist sufferers in regaining fitness to work, has been transformed into an official routine to deny people's perception of their own pain and mobility:

A major issue for interviewees is the pain involved during and after the key test.... Many expressed shock at receiving an assessment that they were fit for work when they had felt sure that the pain they had reported during the test would have indicated that they were not fully recovered.

(ACT RSI Support Group, 1991: pp. 4-5)

In spite of the apparently positive attitude towards health hazards in the 1980s, the acrimonious debate has altered public and professional perceptions of RSI. The standard explanation that RSI in Australia is the result of the increased pace and duration of keyboard utilization, in ergonomically unsound workplaces, has been undermined (Bammer and Martin, 1998). Alternative hypotheses are that people with RSI are malingerers who don't want to work, who suffer from a form of compensation neurosis (whereby symptoms disappear in the absence of financial gain), that it is a form of psychological hysteria or conversion neurosis (as discussed above), or that RSI is a manifestation of normal fatigue experienced by all sectors of the population with no underlying injury or pathology. These counter hypotheses have a distinct gender bias, as was apparent in the earlier citations referring to dependent personalities, women in repetitive unskilled jobs, and various forms of neurotic dispositions.

Reid et al. ( 1991) have described the experiences of women sufferers of RSI in the increasingly hostile climate of Australia's legal, medical and compensatory systems as a 'pilgrimage of pain' in which women's encounters with the system 'contributed to the chronicity, unemployment, bewilderment and despair reported by so many' (ibid.: p. 602). Their research indicates that the 'polarized environment characterised by doubt, derision and debate' (ibid.) in which sufferers sought advice and treatment, created a situation in which judgements were made about their situation which were directly linked to their gender, family circumstances, body shape and emotional distress. Professionals and experts revealed prejudices about semi-skilled women workers which reflected class and gender conflicts in the wider society.

This is underlined by the recent publication of a number of studies in the medical journals which report on research on 'overuse syndrome' amongst musicians (Fry. 1986, 1988 and 1993; Dennett and Fry, 1988; Lippmann, 1991). These studies have examined the similarities between musicians' 'overuse syndrome' and repetitive strain injury of keyboard and process workers (Bammer, 1993). Although mainly addressing medical analysis, the implications in terms of the prejudices of the experts dealing with RSI are not lost on all writers:

Much of the heat in the debate over whether occupational overuse exists as a clinical entity has centred on whether the examining physician believes the patient. The problems lie in the lack of objective clinical signs and repeatable investigative tests. Excluding clear cut tendinitis, synovitis and carpal tunnel syndrome, other entities, even as diagnostically straightforward as epicondylitis lack objectivity. When one moves proximally towards the shoulder and neck the problems are greatly increased.... The two schools of 'real illness' and 'malingerers' shape up to each other in the courtroom, provide a field day for the lawyers, but do little to help the patients.

How can we get closer to resolving these issues? After all, many of the 'sufferers' are poorly paid manual workers undertaking soul destroying repetitive tasks. They have much to gain from compensation and a respite from occupational boredom. They are also likely to slip inexorably into chronicity.

The search for truth in the patients' symptoms is thus frequently dogged by the knowledge that they would be advantaged by stopping work and indeed their intellectual powers may preclude a good history in the first place. The position of professional and student musicians is in marked contrast. Here are a group of intelligent, highly motivated individuals who have everything to lose if they develop a disabling pain, and yet their 'work' involves repetitive movements and abnormal postures.

. . . musicians develop symptom-sign complexes of overuse syndrome indistinguishable from those of less gifted and less motivated workers undertaking work practices which involve comparable hand-arm movements. Ballet dancers develop equally disabling pain and tenderness but in the lower limbs. Again here is another group for whom changing employment would be a disaster.

If we 'believe' the disease to be genuine in dedicated artists, should we not approach the lowly manual worker with more open mindedness? After all, the overuse is genuine and obvious in both groups. The belief in the effects is being driven by our preconceptions of patients and their motivations. Dr Fry has done us all a service by his work into the occupational hazards of the performing arts. It does not provide the incontrovertible proof we all seek, but should make those who espouse the malingering theory to sleep less easily in their beds.

(Harrington, 1993)

Because the 'lowly manual workers' are in fact women keyboard operators, the assumption is that they do not need to work. This is the familiar gendered notion that women's wages are not central for her household, and that paid work is not central to the identity and self image of women in the same way as it is for men. As another medical 'expert' commented, most of the people who consulted him with RSI symptoms were characterized by 'short periods of involvement in the workforce, impulse resignations and unsatisfactory relationships with other workers . . . like the depressed subjects, the women who have conversion reactions frequently reported difficulty with their employers' (Black, 1987, cited in Meekosha and Jacubowicz, 1991).

It is true that women are the majority of RSI sufferers. In 1984, women accounted for 2,800 of the 3,022 cases reported amongst civil servants in New South Wales (Meekosha and Jakubowicz, 1986).

Whilst the debate about the genuineness of the symptoms continues, it is not totally clear that the 'epidemic' in Australia has subsided. The basis for recording the figures has been changed and researchers report that it is now very difficult to obtain data on RSI incidence. Moreover with more public consciousness of the problem, employers are carrying out pre-employment checks which are aimed at eliminating stereotypically RSI prone applicants, which also has the effect of intimidating other employees from making complaints or taking action. It may also be that the apparent decline in incidence relates to better keyboards and changes in the kinds of work being carried out (Meekosha and Jakubowicz, 1991).

The relationship between RSI and technology in the workplace

Some analysis has been carried out to identify the extent to which the incidence of RSI can be related directly to changes in technology and the organization of work. Bammer (1987) argues that there are four causes of repetitive motion injury (an alternative designation to RSI) which can act alone or in combination: repetitive movements; less frequent, but more forceful movements; static loading and awkward body postures. Introduction of VDTs into offices has increased three of these risk factors rapid repetitive movements, static loading and awkward body movements. Soft touch keyboards allow users to attain much higher keystroke rates, and the elimination of other actions such as carriage returns, the application of correction fluid etc. increases the static loading through the prolonged maintenance of keying positions. Badly positioned and unadjustable screens contribute to damaging postures. Additional function keys increase the need to adopt awkward positions - especially of the wrist joints and fifth fingers. Ironically the absence of the fourth risk factor, forceful movements, lessens the risk of tenosynovitis, which is a notifiable industrial injury in the UK and elsewhere.

These findings have encouraged the view that the ergonomic design of workplaces and equipment will eliminate RSI, and reinforce the view that better design is a major reason for the demise of RSI in Australia. However the research also indicates that there are other factors involved, both in the construction of risk and in the predisposition of some people, or people carrying out specific tasks, to RSI. Workers who are in a competitive situation, who are very self motivated and/or workers who are at the end of a rigid hierarchy which leaves them little room for autonomy over the pace of their work, the timing or rest breaks, or decisions over which work should have highest priority are likely to be particularly vulnerable. Women in subordinate data entry jobs have limited autonomy not only in terms of the allocation of their time, but also in terms of their space and their physical mobility (Meekosha and Jakubowicz, 1986).5

It is too soon to ascertain whether improved workplace and equipment design, combined with consciousness of the importance of task diversity and breaks from continuous keying, will eliminate the risk factor even for those workers with the predispositions described above.

LDC experience

To my knowledge there are relatively few studies on which to base any reliable picture of the incidence or prevalence of RSI or other computer-related health risks in developing countries. Many writers make the probably correct assumption that computer operators are vulnerable to the same kind of health risks which have been documented in the advanced countries. Barnes reported that:

While data entry firms in the UK, Canada and the US are being forced into an awareness of [these] risks by women's organizations, unions and the female workers themselves, the industry in Jamaica still has to catch up. No special attention is paid to the type of seating used, the frequency with which operators are able to get up and walk about, lighting, or anything else.

(Barnes, 1989)

My own study of offshore data entry work in Jamaica confirms the view that there is very little public discussion of health and safety aspects of keyboarding within the sector (R. Pearson, 1993). But, as I argued at the beginning of this paper, given the attention paid to RSI and other health and safety issues in advanced countries, it is surprising that this has not arisen as an issue for wider public discussion. Although there is virtually no literature on the working conditions and health outcomes of offshore contract data entry or data processing workers in LDCs, it is interesting to note that an American company operating in the Philippines boasted to potential customers that 10 per cent of its 800 non-union employees 'are deaf mutes (for the best accuracy)' (N. Cohen, 1992). It may be that the inability to complain about health and safety aspects is as prized as the implied conscientiousness of these handicapped employees.

Ng, whose research is concerned with the introduction of computers into the domestic telecommunications industry, asserts that 'the effects (of information technology) on the health and safety of its users is a major and less controversial issue' (Ng, 1991, emphasis added), but she reports no evidence on RSI or other musculo-skeletal problems of data entry or other workers, beyond the statement that 'complaints related to the eyes, hands and wrists, shoulders, neck and back, and general health problems, are commonly registered by these VDU users (ibid.: p. 45). Data about abnormal pregnancy outcomes is also difficult to assess since it is presented without any standard of comparison with a similar cohort of women in different occupational situations (ibid.: pp. 45-47).

There is one study (Soares, 1991) of data entry clerks in a public sector data processing centre in Brazil. This found evidence of problems in arms, upper back, and legs. A significant number of workers were certified as having tenosynovitis, with a larger number displaying the symptoms indicating this condition. Soares notes that the work of these keyboard operators is extremely controlled, with talking and mobility prohibited. Although chair heights were adjustable, the back rests were not. Although the screens were not adjustable, few workers reported problems with their vision, which the researcher suggested may be because they spend less than half of their time watching the screen. Soares also suggests that the high levels of stress, irritability and exhaustion are linked to the facts that women have to do domestic work at home rather than rest and recover from their work-related strains, that they routinely do overtime, and that they spend considerable time commuting to work.

Learning from international experience

Clearly more research is necessary to assess the extent of health problems among keyboard operatives in LDCs, and particular attention must be paid to the specifics of their working routines, tasks and allied problems. In the light of the Australian literature and other studies it is possible to suggest what the parameters of the problems may be.

First, activities in which standards of productivity and speed are the basis for payment, job security, and promotion are likely to be those in which the risk of RSI is greater. However, offshore data entry facilities may in fact be less pressurised than large workplaces in the public or corporate sector, if a continuous flow of work cannot be guaranteed.6 This is an important observation, since it means that the location and nature of the activity cannot in themselves be used to predict the risk of the workforce, without an accurate assessment of the particular labour process.

Second, we need to take into account that in many countries the widespread introduction of new technology in large corporations coincides with privatization, the deprotection of workers, and the introduction of new forms of work organization (see, e.g. Ng, 1991: pp. 3740). It is necessary to distinguish between the nature of the technology, the nature of the task and the ways in which the workers are required to carry out the task.

The context in which information technology is introduced will make a difference to the degree to which women have any leverage over how their work is organized and controlled, and the implications of such employment cannot be predicted a priori. There are three possibilities: the intensification; decomposition; or recomposition of existing gender and other social relations (Elson and Pearson, 1984). To give just one example, there are plans to promote computers in Saudi Arabia to increase the home-based employment opportunities of tertiary educated women, in order to maintain their seclusion (Kingdom of Saudi Arabia, n.d.).

Third, the extent to which formal worker protection by trade unions is a useful predictor of protection from the health risks of new technology is debatable. There is evidence from the Australian literature that, having been extremely helpful in the earlier phases of the Australian epidemic, the unions failed to combat attempts to psychologize the condition, and have not taken up the long-term issue of what happens to expelled and injured workers (Meekosha and Jakubowicz, 1991). The study carried out by the RSI sufferers reports that 60 per cent found their union to be unhelpful (ACT RSI, 1992). Ng (1991: pp. 36-7) also reports that the union did not see the health and safety protection of women workers as a priority.

But we should also be clear that information technology workers cannot necessarily be expected to take direct action on health hazards. L. Harris (1989) reports that women workers in an American chemical factory in County Mayo, Ireland, refused to take industrial action although they knew that the raw material they worked with had been banned elsewhere because of its supposed carcinogenic qualities. These workers had taken action on other grievances, such as equal pay, but they recognized that the location of the company in Ireland, and thus their employment, was the direct result of the health hazard they were being exposed to. Many information technology workers are in a similar position: they may or may not be informed about the risks of such employment, and may or may not have some bargaining power to address their concerns, but their response is mediated by a complex set of objectives including the need to earn an income and the benefits they obtain from employment in terms of association, socializing and status.

Managerial strategies can exploit these contradictions. In Jamaica, companies followed explicit strategies to prevent the unionization of data entry workers, lest they emulated the well-known and feared militancy of garment factory workers in the free trade zones (R. Pearson, 1993). There has also been a conscious attempt to promote the notion that keyboard operators were different, and superior to, assembly-line production workers (Freeman), similar to the differentiation made between electronics and garments workers in the Mexican maquiladoras (Pearson, 1991b). The 'clean' image of information technology work, in contrast to the environmentally hazardous emissions of process manufacturing, can also be utilized to diminish the perceived health hazards associated with computer work.

Such considerations may be central in understanding the responses of low paid semi-skilled workers, but not of professional women working with computers. It is well recognized that many women (and men) who have a much greater autonomy over the ways in which they carry out their work - such as academics, journalists and writers - consistently utilize computers in ways which entail severe risks of developing RSI and other conditions, even though they have access to very up-to-date knowledge of the risks and prevention strategies.8 Externally imposed pressure and deadlines, as well as an increasingly competitive environment, provide a partial explanation, but self motivation and exploitation must also be factors.

Nor is it axiomatic that the interests of women workers, particularly in LDCs, are antithetic to the strategies of development planners who see the adoption and dissemination of information technology within their economies as the route to long-term economic viability and growth. Sen's model of 'cooperative conflict' (See, 1990), which encompasses a situation in which an individual's utility is both in conflict with, and totally inextricable from, that of the wider group might also prove a useful model for conceptualizing the contradictory relationship women have to the risks and benefits from information technology. Such a framework would seem infinitely more realistic than either a totally negative response, which suggests that all risk-bearing employment for women should be prohibited, or one which argues that health and safety issues should be totally subordinated to wider issues of economic growth, employment, and foreign exchange generation.8

The experience of advanced countries, particularly of Australia, indicates that new times as well as new technologies require new analyses and new strategies. There were institutional innovations by women in response to the 'epidemic', including support groups such as WRIST (Women's Repetitive Strain Injury Support Team), in Victoria, which not only have provided support and referral advice to sufferers but have played a central monitoring and lobbying role in relation to changes in the institutional and legislative framework. Workplace groups have also organized health and safety circles in an attempt to promote collective endeavours to devise working practices that are more rational, and have devised exercise routines which are both helpful and practicable.

The Australian experience is in a sense a microcosm of the contradictions that information technology presents - to producers, workers and users. The often irrational way in which medical and other professionals have sought to refute the links between technology, work practices and health hazards illustrates most directly the ways in which gendered notions of legitimacy are at the heart of the contradictions which this technology catalyses. It is the role of gender analysis to probe the nature of these contradictions and their gendered manifestations and responses, in order to work towards useful strategic responses.

Notes

1 The term 'cheap labour' has been deconstructed by Elson and Pearson, 1984; and Pearson, 1988.

2 It is also true that the importance of health and safety aspects of new technology employment in advanced countries is still not universally recognized, outside of a specialized literature. Webster's excellent Office Automation: The Labour Process and Women's Work in Britain, 1990, contains no references at all to health and safety issues. Crompton and Jones' pioneering study White Collar Proletariat: Deskilling and Gender in Clerical Work, in 1984, was also silent on health and safety. A volume concerned with the social issues relating to information technology (Finnegan et al., 1987) again has no discussion.

3 For instance Linda Lim, who has published a number of reports based mainly on secondary data, in her own name and for UN organizations, on the nature and condition of women's employment in export manufacturing by multinationals has recently insisted that 'feminists who see patriarchy and gender subordination as crucial underpinnings and inevitable consequences of all capitalism refuse to recognize any benefits to women in the Third World from employment in export factories, insisting that such employment intensifies rather than alleviates their gender subordination. The works of Elson and Pearson are popular with this group' (Lim, 1990: p. 116).

4 See OTA, 1985, especially Chapter 7; Putz-Anderson, 1988b; De Matteo, 1985; Shiga, 1987; ILO, 1990; Craig, 1991; English et al., 1989; Bammer, 1987, 1990.

5 Women's lack of physical mobility in the workplace, compared to men, is one of the features of the gendering of work widely reported in studies of industrial and other workers (Humphrey, 1987).

6 My field work in Jamaica indicated that many companies had quite irregular workloads. During slack periods the women keyboard operators had considerable freedom. The situation in the Philippines appears to be rather different.

7 For instance, I suffered from debilitating back ache whilst preparing this paper. I personally know of four women academics and researchers of my generation in the UK who have or have had RSI.

8 These issues are explored in the context of the Jamaican data entry sector in R. Pearson, 1993.

References

ACT RSI Support Group Inc. (1991), Consumers' Perspective of Rehabilitation in RSI Cases; Stage 1 Report, Canberra, Australia

ACT RSI Support Group Inc. (1992), Consumers' Perspective of Rehabilitation in RSI Cases; Stage 2 Report, Canberra, Australia

Bammer, G. (1987), 'How Technologic Change can Increase the Risk of Repetitive Motions Injuries', Seminars in Occupational Medicine, Vol. 2, No.1, March: pp. 25-30

Bammer, G. (1988), 'The Prevalence of Work-related Neck and Upper Limb Disorders Among Office Workers in 7 Countries - a Pilot Study', Reprints of International Conference on Ergonomics, Occupational Safety and Health and the Environment, Vol. 1, Beijing, October

Bammer, G. (1990), 'Review of Current Knowledge - Musculoskeletal Problems', in L. Berlinguet and D. Berthelette (eds), Work with Display Units, Amsterdam, Elsevier Science Publishers

Bammer, G. (ed.) (1993), Discussion Papers on the Pathology of Work-Related Neck and Upper Limb Disorders and the Implications for Diagnosis and Treatment, Working Paper No. 32, National Centre for Epidemiology and Population Health, Australian National University, Canberra

Bammer, G., and Martin, B. (1988), 'The Arguments About RSI: an Examination', Community Health Studies, Vol. XII, No. 3: pp. 348-358

Barnes, C. (1989), 'Data Entry Demands', Sistren Magazine, Vol. 11, No. 3: pp. 1821

Bawa, J. (1993), 'Ergonomics in the Office', Guardian, Special Report - Computers at Work: p. 17, 22 April

Berg, M. et al. (1990), 'Facial Skin Complaints and Work at Visual Display Units', Journal of the American Academy of Dermatology, Vol. 22, No. 4: pp. 621-625

Black, P. (1987), 'Psychiatric Aspects of Regional Pain Syndrome', Medical Journal of Australia, Vol. 147, 7 September: p. 257

Bodek, N. (ed.) (1987), The Best of DEMA: Data Entry Management Handbook, USA, DEMA (Data Entry Management Association)

Brandt, L. and Nielsen, C. (1990), 'Congenital Malformation Among Children of Women Working with Video Display Terminals', Scan J Work Environ Health, Vol. 16: pp. 329-333

Castells, M. (1989), The Informational City: Information Technology, Economic Restructuring, and the Urban-Regional Process, Blackwell, Oxford

Cohen, M. et al. (1992), The Relevance of Concepts of Hyperalgesia to RSI, Working Paper No.31, in G. Bammer (ed.), Discussion Papers on the Pathology of Work-Related Neck and Upper Limb Disorders and the Implications for Diagnosis and Treatment, National Centre for Epidemiology and Population Health, Australian National University, Canberra

Cohen, N. (1992), "'Extraordinary" crime records deal attacked', Independent, 30 June

Craig, M. (1991), Office Workers' Survival Handbook, The Women's Press, London

Crompton, R. and Jones, G. (1984), White-collar Proletariat: Deskilling and Gender in Clerical Work, Macmillan, London

Davis, J. and Stack. M. (1993), 'Knowledge in Production', Race and Class, Vol. 34, No. 3

DeMatteo, B. (1985), Terminal Shock: The Health Hazards of Video Display Terminals, New Canada Publications, Toronto

Dennett, X. and Fry, H.J.H. (1988), 'Overuse Syndrome: a Muscle Biopsy Study', Lancet Vol. 1: pp. 905-908

Di Martino, V. and Wirth, L. (1990), 'Telework: a New Way of Working and Living', International Labour Review, Vol. 129, No. 5: pp. 529-554

Dobyns, J. (1991), 'Cumulative Trauma Disorder of the Upper Limb', Frontiers in Hand Rehabilitation, Vol. 7, No. 3, August: pp. 587-505

Eliasson, G. (1990), 'The Knowledge-based Information Economy', in G. Eliasson et al. (eds) The Knowledge-based Information Economy, Almqvist and Wicksell International, Stockholm

Elson, D. (1991), 'Appraising Recent Developments in the World Market for Nimble Fingers: Accumulation, Regulation Organization', paper presented at an international workshop on Women Organizing in the Process of Industrialization, ISS, The Hague, April

Elson, D. and Pearson, R. (1984), 'The Subordination of Women and the Internationalization of Factory Production', in K. Young, et al. (eds), Of Marriage and the Market: Women's Subordination in International Perspective, London, Routledge

English, CJ. et al. (1989), 'Clinical Epidemiological Study of Relations Between Upper Limb Soft Tissue Disorders and Repetitive Movements at Work', Final Research Report, Institute of Occupational Medicine, Edinburgh

Finnegan, R. et al. (1987), Information Technology: Social Issues, Milton Keynes, Open University Press

Freeman, C. (n.d.), 'High-tech and High Heels: Barbadian Women in the Off-shore Information Industry', (mimeo) paper presented to the Fifteenth Annual Conference in the Caribbean Studies Association, Trinidad

Fry, H.J.H. (1986), 'Overuse Syndrome in Musicians: Prevention and Management', Lancet Vol. 2: pp. 728-731

Fry, H.J.H. (1988), 'The Treatment of Overuse Syndrome in Musicians. Results in 175 patients', Journal of the Royal Society of Medicine, Vol. 81: pp. 572-575

Fry, H.J.H. (1993), Overuse Syndrome and the Overuse Concept, Working Paper No. 32, in G. Bammer (ed.), Discussion Papers on the Pathology of Work-Related Neck and Upper Limb Disorders and the Implications for Diagnosis and Treatment, National Centre for Epidemiology and Population Health, Australian National University, Canberra

Girvan, N. (1989), 'Technological Change and the Caribbean: Formulating Strategic Responses', Social and Economic Studies, Vol. 38, No. 2: pp. 111-135

Glucksmann, M. (1986), 'In a Class of Their Own: Women Workers in the New Industries in Inter-war in Britain', Feminist Review, No. 24, October: pp. 7-37

Glucksmann, M. (1990), Women Assemble: Women Workers and the New Industries in Inter-war Britain, London, Routledge

Hall, W. and Morrow, L. (1988), 'Repetition Strain Injury: an Australian Epidemic of Upper Limb Pain', Social Science and Medicine, Vol. 27, No. 6: pp. 645-649

Hanna, N. (1991), 'Informatics and the Developing World', Finance and Development, December: pp. 45-47

Harrington, M. (1993), 'Response to Hunter Fry', in H.J.H. Fry, Overuse Syndrome and the Overuse Concept, Working Paper No.32, in G. Bammer (ed.), Discussion Papers on the Pathology of Work-related Neck and Upper Limb Disorders and the Implications for Diagnosis and Treatment, National Centre for Epidemiology and Population Health, Australian National University, Canberra: pp. 10-11

Harris, B. (1989), 'Does the Third World Need Computers?', Environment and Planning B: Planning and Design, Vol. 16: pp. 371-376

Harris, L. (1989), 'Women's Response to Multinationals in County Mayo', in D. Elson and R. Pearson (eds), Women's Employment and Multinationals in Europe, Macmillan, London

Heeks, R. (1989), The Impact of New Technologies on the International Division of Labour in the Indian Software Industry, Milton Keynes, Development Policy and Practice Group Working Paper

Hopkins, A. (1990), The Social Recognition of Repetitive Strain Injuries: an Australian/American Comparison, Social Science and Medicine, Vol. 30, No. 3: pp. 365-372

HSE (Health and Safety Executive) (1992), Display Screen Equipment Work. Guidance on Regulations, London, Health and Safety (Display Screen Equipment) Regulations

Huang, D.C. (1990), Size and Growth Trends of the Information Industry: 1975-87, incidental paper, Program for Information Policy Research, Harvard University, Cambridge, Mass.

Humphrey, J. (1987), Gender and Work in the Third World: Sexual Divisions in Brazilian Industry, Tavistock, London

Huws, U. (1987), VDU Hazards Handbook, The London Hazards Centre

ILO (1990), Conditions of Work Digest, Vol. 9, No.1/1990, Part I

Ireland DCR. (1988), 'Psychological and Physical Aspects of Occupational Arm Pain', The Journal of Hand Surgery (Aust.), Vol 13-B(1), February

Kandiyoti D. (1988), 'Bargaining with Patriarchy', Gender and Society Vol. 2, No. 3

Kingdom of Saudi Arabia (n.d.), Fourth Development Plan, 1985-90 AD, Saudi Arabia, Ministry of Planning Press

Kutay, A. (1986), 'Effects of Telecommunications Technology on Office Location', Urban Geography, Vol. 7: pp. 243-257

Labour Research, (1984), 'Health at work: 1. Pregnancy and VDUs', Vol. 79, No. 33: December, pp. 299-300

Lim, L. (1990), 'Women's Work in Export Factories: the Politics of a Cause', in I. Tinker (ed.), Persistent Inequalities: Women and World Development, New York and Oxford, Oxford University Press

Lippmann, H. (1991), 'A Fresh Look at the Overuse Syndrome in Musical Performers: is "Overuse" Overused?', Medical Problems of Performing Artists, Vol. 6, No. 2: pp. 57-60

Locksley, G. (1986), 'Information Technology and Capitalist Development', Capital and Class, Vol. 27

Lucire, Y. (1986), 'Neurosis in the Workplace', The Medical Journal of Australia, Vol. 145, 6 Oct: pp. 323-327

Meekosha, H. and Jakubowicz, A. (1986), 'Women Suffering RSI: the Hidden Relations of Gender, the Labour Process and Medicine', Journal of Occupational Health and Safety, Australia/New Zealand, Vol. 2, No. 5: pp. 390-401

Meekosha, H. and Jakubowicz, A. (1991), 'Repetition Strain injury: the Rise and Fall of an "Australian" Disease', Critical Social Policy, Vol. 11, No.1, Summer

Mitter, S. and Pearson, R. (1992), Global Information Processing: the Emergence of Software Services and Data Entry Jobs in Selected Developing Countries, Sectoral Activities Programme Working Papers, Geneva, ILO

Mody, A. and Dahlman, C. (1992), 'Performance and Potential of Information Technology: an International Perspective', World Development, Vol. 20, No. 12, December

Moussa, A. and Schware, R. (1992), 'Informatics in Africa: Lessons from World Bank Experience', World Development, Vol. 20, No. 12, December

Nelson, K. (1986), 'Labor Demand, Labor Supply and the Suburbanization of Low-wage Office Work', in A. Scott and M. Storper (eds), Work Production and Territory, Allen and Unwin, London

Ng, C. (1991), Information Technology, Economic Restructuring and their Impact on Women Workers: the Case of the Telecommunications Industry in Malaysia, Working Paper, Institute of Social Studies, The Hague

Ng, C. and Othman, J. (1991), 'Occupational Health and Safety Among Office Workers: a Preliminary Study', in C. Nicholas and A. Wangel (eds), Safety at Work in Malaysia, Kuala Lumpur, University of Malaya

OECD (1986), Trends in the Information Economy, OECD, Paris

OTA (1985), Automation of America's Offices, Washington, Office of Technology Assessment

Patkin, M. (1993), 'Response to Hunter Fry', in H.J.H. Fry, Overuse Syndrome and the Overuse Concept, Working Paper No. 32, in G. Bammer (ed.), Discussion Papers on the Pathology of Work-related Neck and Upper Limb Disorders and the Implications for Diagnosis and Treatment, National Centre for Epidemiology and Population Health, Australian National University, Canberra: pp. 11-12

Pearce, B. (1993), 'Is a Change as Good as a Rest?', Guardian, Special Report Computers at work: p. 20, 22 April

Pearson, R. (1988), 'Female Workers in the First and Third Worlds: the Greening of Women's Labour', in R. Pahl (ed.), On Work. Oxford, Blackwell

Pearson, R. (1991a), New Technology and the Internationalization of Office Work: Prospects and Conditions for Women's Employment in LDCs, Gender Analysis in Development Discussion Paper No. 5, School of Development Studies, University of East Anglia, Norwich, UK

Pearson, R. (1991b), 'Male Bias and Women's Work in Mexico's Border Industries', in D. Elson (ed.) Male Bias in the Development Process, Manchester, Manchester University Press

Pearson, R. (1993), 'Gender and New Technology in the Caribbean: New Work for Women?', in J. Momsen (ed.), Women and Change in the Caribbean, lames Currey, London

Pearson, R.M. (1990), Muscle Overuse Syndrome (RSI), (mimeo), London, Musicians and Keyboard Clinic

Pearson, R.M. (1993), 'Response to Hunter Fry', in H.J.H. Fry, Overuse Syndrome and the Overuse Concept, Working Paper No. 32, in G. Bammer (ed.), Discussion Papers on the Pathology of Work-related Neck and Upper Limb Disorders and the Implications for Diagnosis and Treatment, National Centre for Epidemiology and Population Health, Australian National University, Canberra: p. 13

Pearson, R. and Mitter, S. (1993), 'Employment and Working Conditions of Low-skilled Information Processing Workers in Less Developed Countries', International Labour Review, Vol. 132, No. 1: pp. 49-63

Polakoff, P. (1991), 'Repetitive-motion, Radiation and Eye Concerns Mount at Computer Worksites', Occupational Health & Safety, Vol. 60, Part 4, April, pp. 34 and 50

Posthuma, A. (1987), 'The Internationalization of Clerical Work: a Study of Offshore Office Services in the Caribbean', SPRU Occasional Papers No. 24, University of Sussex, Brighton, UK

Putz-Anderson, V. (1988a), 'Prevention Strategies Adopted by Select Countries for Work-related Musculoskeletal Disorders from Repetitive Trauma', in F. Aghazadeh (ed.), Trends in Ergonomics/Human Factors V, Amsterdam, Elsevier Science Publishers

Putz-Anderson, V. (ed.) (1988b), Cumulative Trauma Disorders: A Manual for Musculoskeletal Diseases of the Upper Limbs, NIOSH, Cincinnati, Ohio, USA

Quintner, J. and Elvey, R. (1991), 'The Neurogenic Hypothesis of RSI', in G. Bammer (ed.), Discussion Papers on the Pathology of Work-related Neck and Upper Limb Disorders and the Implications for Diagnosis and Treatment, NCEPH Working Paper No. 24, Canberra, Australia

Reid, I., Ewans, C. and Lowy, E. (1991), 'Pilgrimage of Pain: the Illness Experiences of Women with Repetitive Strain Injury and the Search for Credibility', Social Science and Medicine, Vol. 32, No. 5: pp. 601-612

Ryan, C. (1990), 'Trade Liberalisation and Financial Services', World Economy, Vol. 13, No. 3, September: pp. 349-366

Schware, R. (1992), 'Software Industry Entry Strategies for Developing Countries: a "walking on two legs" Proposition', World Development, Vol. 20, No. 2: pp. 143164

Semple C. (1991), 'Tenosynovitis, Repetitive Strain Injury, Cumulative Trauma Disorder, and Overuse Syndrome, et cetera', The Journal of Bone and Joint Surgery (Br.), Vol. 73-B: pp. 536-538

Semple, C. (1993), 'Response to Hunter Fry', in H.J.H. Fry Overuse Syndrome and the Overuse Concept, Working Paper No. 32, in G. Bammer (ed.), Discussion Papers on the Pathology of Work-related Neck and Upper Limb Disorders and the Implications for Diagnosis and Treatment, National Centre for Epidemiology and Population Health, Australian National University, Canberra: pp. 24-25

Sen, A.K. (1990), 'Gender and Cooperative Conflicts', in I. Tinker (ed.), Persistent Inequalities: Women and World Development, New York, Oxford, Oxford University Press

Shiga, H. (1987), 'Microelectronics and women in Japan', in C. Ng (ed.), Technology and Gender, Kuala Lumpur, Malaysian Social Science Association

Smith, I. (1993), 'The Health and Safety (Display Screen Equipment) Regulations 1992', in I. Smith, C. Goddard and N. Randall (eds), Health and Safety - the new legal framework, London, Butterworth

Soares, A. (1991), 'The Hard Life of the Unskilled Workers in New Technologies: Data Entry Clerks in Brazil', in H.J. Bullinger (ed.), Human Aspects in Computing, Amsterdam, Elsevier Science Publishers

Standing, G. (1989), 'Global Feminization Through Flexible Labor', World Development, Vol. 17, No. 7

Webster, J. (1990), Office Automation: The Labour Process and Women's Work in Britain, New York and London, Harvester Wheatsheaf