Contents   Search publications                     Previous chapter   Next chapter

4 Reproductive hazards

In the chapter on reproductive hazards in the first edition of the VDU Hazards Handbook (London Hazards Centre 1987) we emphasised the importance of heeding VDU workers' concerns about this issue, and the need for substantial experimental and epidemiological research. Since 1987, several hundred papers have been published on this subject and yet the clear evidence that VDU work does not harm reproductive health has yet to be produced.

Indeed, during the last six years, for nearly every study which has concluded that there is no link between VDU work and reproductive health, there has been a study which suggests that VDU work may be responsible for a number of different adverse effects.

In examining some of these effects in more detail, and in continuing to believe that there is cause for concern, we diverge significantly from the view of the Health and Safety Executive, the institution responsible for enforcing health and safety law in the UK. In Annex B to the DSE Regulations the HSE confidently concludes about the reproductive research studies:

The Hazards Centre takes issue with the statement 'taken as a whole', as if a series of studies with negative results could cancel out a series of studies with positive results. The HSE, employers' organisations and computer manufacturers have been quick to point out methodological defects in studies which have shown a link between VDU work and adverse reproductive effects, yet eager to rely on studies showing no link, to which similar criticism could equally be applied.

Although increased risk of miscarriage and birth defects are the effects most commonly referred to, adverse reproductive effects include a wide range of problems:

• menstrual disorders
• sexual difficulties
• problems with conception
• male fertility problems
• problems with pregnancy
• spontaneous abortion or miscarriage
• birth defects

Another reason for condemning official complacency over reproductive hazards is that research to date has concentrated on incidents of miscarriage, or spontaneous abortion. However, as the list above shows, there are many other adverse reproductive effects. There are several ways in which these effects could be overlooked: The failure of GPs to consider occupational exposure in the causation of reproductive ill-health; the (understandable) reluctance of sufferers to discuss reproductive difficulties with employers or researchers; the lack of experimental and epidemiological research in these areas; the problem of distinguishing an occupational reproductive effect from the high background rate of reproductive problems; the difficulty of obtaining sufficient, accurate, statistically significant data to achieve proof of an effect.

We can illustrate this problem using the following example: Several studies have shown that there may be an increased risk of congenital heart disease in babies born to VDU operators. One VDU operator, working for a small organisation becomes pregnant. Her pregnancy is normal, but the baby is born with a minor heart defect - a small VSD (ventricular septal defect, ie. a small hole in the heart). The fact that the mother had worked at a VDU throughout her pregnancy is not recorded either by the hospital or the GP. The woman, who has not heard of the possible link between congenital heart defects and VDU work doesn't mention her work either.

Of course, it may be that there was no connection whatsoever between exposure to VDUs during the mother's pregnancy and her child's VSD, but, given the lack of resources to make hospital record-keeping nationally consistent, or to correlate abnormal pregnancy outcomes with the parents' occupation, this is likely to remain an area of great uncertainty. In view of this uncertainty, we condemn official complacency and refute assurances that women can protect themselves merely by being given the opportunity to

Furthermore, examination of the last few decades shows that belief in official assurances about the safety of numerous occupational materials and processes has been frequently unfounded. A safety rep, disabled by his mother's exposure to thalidomide, summarises the views of many people concerned about VDUs and reproductive health in his letter to Hazards, a magazine for safety reps:

He goes on to remark about the HSE's dismissal of the problem:

Menstrual disorders

In the VDU Factpack that was the precursor to the first edition of the VDU Hazards Handbook (London Hazards Centre (1987) we included a questionnaire on various aspects of VDU work and health which readers were invited to complete and return to the Hazards Centre. Although the questionnaire did not mention menstrual problems, many of the women returning the questionnaire, particularly those who used VDUs for most of their working day, reported menstrual problems which included heavy, painful and irregular bleeding.

This anecdotal evidence has recently been born out by a major study of 3,500 women working in the Inland Revenue, which showed a clear link between stress at work and menstrual disorders, with VDU workers being the worst sufferers (Bramwell and Davidson 1990). Work overload, stagnation (lack of stimulus, control, variety and involvement), the home/work interface and workstation ergonomics were all important stress factors.

Conception and infertility

Statistics on infertility are very imprecise. In the UK the current estimate is that about 10 per cent of couples trying to conceive will not succeed. However, little is known about whether this is due to male infertility, female infertility, or what might be attributed to other factors. The vast majority of research on reproductive hazards relates to women, yet considering the known vulnerability of sperm to agents such as ionising radiation, this focus may mask some of the effects on fertility and pregnancy outcome.

It was hoped that a long-term study by the US National Institute for Occupational Safety and Health (NIOSH) of 2,340 women over three years would include an assessment of the effect of VDU work on fertility, but unfortunately funding restrictions limited the scope of the study. However, the study revealed the unexpected finding that VDU operators were likely to have less subsequent live births after a miscarriage than those in the control group, indicating a possible effect on fertility or very early pregnancy loss (VDT News May/June 1991).

Miscarriage

Miscarriage, or spontaneous abortion, is defined as the expulsion of an embryo or foetus from the womb at a stage of pregnancy when it is incapable of independent survival. This medical definition does not describe the physical, psychological or emotional impact of losing a wanted pregnancy.

The suggestion that VDU work could increase the risk of miscarriage first came to light in the 1970's, with a number of 'clusters' of miscarriages being reported in the UK, Canada, USA and Denmark (London Hazards Centre 1987). It is not the intention of this book to reproduce all the reports and research that have been carried out subsequently, rather we will review some of the more important evidence that has emerged during the last few years and examine some of the difficulties of interpreting the scientific reports and some of the theories that account for the problems in pregnancy.

The scientific community continues to disagree about the role of VDUs in miscarriage clusters, with many dismissing them as chance occurrences. It is interesting to note, however, that if these clusters were unrelated to VDU use then we might expect to see a number of similar 'chance' clusters in offices where there are no VDUs, but as far as we know, none have been reported.

We find the claim that the link between VDUs and increased miscarriage rates is coincidence as unconvincing as the claim that the high rate of leukaemia in the children of men who have worked at the Sellafield nuclear installation is pure coincidence. Furthermore, during the first National Conference on Clustering of Health Events held in Atlanta in February 1989, it was reported that previous researchers may have over-estimated the expected number of clusters ie. the likelihood of adverse pregnancy clusters appearing purely by chance is even less than had been previously assumed (VDT News March/April 1989).

Another difficulty of assessing the information on miscarriages linked to VDU work is that even when increased rates are found in epidemiological studies, the findings are often discounted if no explanation for the results can be found.

An example of how definition of miscarriage risk can depend not just on the results of a study, but also on the particular interpretation of those results is given by the study of Inland Revenue staff already referred to in the section on menstrual disorders (Bramwell and Davidson 1990). In this study the significance of the observation of three separate links between VDU work and miscarriage was discounted on the grounds that one group was too small, that the miscarriages were self-reported, and that although women working more days at a VDU had higher miscarriage rates than those working fewer days, this was probably a random result.

Although 26 out of 122 ie. 21 per cent of VDU users in the first phase of the study miscarried in the first three months of their pregnancy, the overall results of the research were presented as evidence that VDUs do not increase the risk of miscarriage.

The reports that have caused continuing concern include both experimental (laboratory) studies - mainly on chick embryos, mice and rats, and epidemiological studies (of large groups of people).

Epidemiological studies

In 1988, researchers from the Northern California-based Kaiser Permanente Medical Care Programme reported on a case-control study of nearly 1,600 pregnant women over a two year period (Goldhaber et al 1988). They found that women who used VDUs for more than 20 hours a week during the first three months (first trimester) of their pregnancy were more than twice as likely to miscarry as those doing other types of office work. Dr Irving Selikoff, who, until his death in 1992, was one of the world's most respected occupational health specialists, said that the Kaiser Permanente study results added 'substantial authority' to concerns about VDU-related reproductive hazards (VDT News July/August 1988).

Although the researchers suggest that women with adverse pregnancy outcomes may have over-reported their VDU exposure, they acknowledge that self-reported pesticide exposure, which they were also studying, was not higher among women with adverse pregnancy outcomes. They point out that if recall bias was operating, they would expect other self-reported exposures to be associated with adverse pregnancy outcome. This is a very important point as in several subsequent epidemiological studies increased miscarriage rates have been attributed to recall bias.

Even where results have been ambiguous, or where explanations for increased miscarriage rates have been unclear, most scientists agree that absolute conclusions cannot yet be drawn and that more research may need to be done before the role of VDUs in adverse pregnancy outcome is established. The researchers involved in a recent UK study, largely sponsored by the HSE were, however, untroubled by such doubts and concluded that 'pregnant women who work with VDUs are not at increased risk of clinically diagnosed spontaneous abortion. ' (Roman et al 1992). On close examination it appears that of the relatively small number of women involved in the study, 447, only 250 had actually used VDUs in their work. Of these, only 52 had used VDUs for 21 or more hours a week. As discussed early, the Kaiser Permanente study, which had involved 1,600 women, found that women who worked 20 hours or more per week at a VDU had the highest risk of miscarriage. Furthermore, the UK study only considered cases of clinically confirmed miscarriage at around the eleventh week of pregnancy. Studies which have included self-reported miscarriage have shown that a high proportion occur very early on in pregnancy, before the pregnancy has been clinically confirmed, and may never be reported to the GP or hospital. In view of these concerns, it would seem foolhardy and premature to rely on the results of the Roman study for assurance of the safety of VDU use for pregnant women.

During the same period that the HSE-sponsored UK research was being conducted, researchers at the California Department of Health Services were carrying out a case-control study of 1,900 women over a two year period. (VDT News 1991). This study confirmed a 'consistent pattern of a possibly slightly increased risk' of miscarriage. The researchers calculated that if the detected risk was genuine, nine per cent of all causes of miscarriage among working women would be due to VDT use, assuming a 50 per cent exposure rate.

A recent epidemiological study, which has provided partial confirmation of some of the experimental studies discussed on page 00, found that women exposed to VDUs which produce high magnetic fields averaging more than 0.3 micro Tesla (µT) had more than three times the miscarriage rate of those who were exposed to fields below this level (Lindbohm et al 1992). (See Chapter 7 for an explanation of magnetic field measurement units). The researchers concluded that exposure to a high level of extremely low frequency (ELF) magnetic fields from VDUs in early pregnancy is related to an increased risk of miscarriage, for a small, but significant proportion of VDU users. In a subsequent discussion of their findings during the Work with Display Units '92 conference the researchers revealed that in Sweden, which has the most stringent requirements on VDU emissions of any European country, more than 25 per cent of the VDUs in use expose operators to more than 0.3µT and 50 per cent to more than 0.2µT (VDT News November/December 1992).

Experimental studies on spontaneous abortion

In a review of experimental work on the subject, a Finnish environmental scientist examined the evidence from studies of several mammalian and non-mammalian species for the effect on pregnancy of exposure to certain types of electromagnetic radiation from VDUs (Juutilainen 1991). In particular, the effects of low-frequency magnetic fields were considered. (See Chapter 7 for more about VDUs and electromagnetic radiation).

The review confirmed that the type of magnetic fields which are associated with VDUs are capable of having an effect on the embryos of fruit flies, sea urchins, toads, fish, chicks, mice and rats. The range of effects include foetal loss (or abortion) and development of abnormalities or defects. In several of the studies reviewed, the effects of the magnetic fields were greatest during the very early stages of development. The researcher concludes firstly, that the mechanism by which such effects are brought about is unclear and secondly, the experimental evidence is not yet sufficiently complete to rule out the possibility that low-frequency magnetic fields affect human prenatal development.

Abnormalities and defects

The publication of reports on abnormalities and defects in babies born to VDU workers has followed a similar pattern to that for the link between VDU work and miscarriage.

In particular, there have been several reports showing increased rates of underweight babies and of congenital malformations, especially of the heart.

Concern has been fuelled by a number of experimental studies which have shown that ELF magnetic fields of the type emitted by VDUs do affect the development of chick embryos. Sceptics have argued that these results cannot be extrapolated to human embryos. More cautious researchers have pointed out that any material or substance capable of a biological effect on embryo development in one species may also be capable of affecting embryo development in another species. In view of the epidemiological evidence showing a possible link between VDU use, particularly during early pregnancy, and birth defects, it would be unwise at this stage to discount the experimental evidence.

As well as research on the effects of magnetic fields, researchers have looked at other factors thought likely to affect pregnancy outcome, such as stress. A Danish study of 2,252 women found that the babies of women who work at VDUs and who have high levels of job stress had a non-statistically significant 80 per cent increase in the incidence of congenital malformations (VDT News January/February 1991). The risk of water on the brain (hydrocephalus) was found to be 12 times higher - a statistically significant result, but which was based on a small number of cases. In one Canadian study, the risk of kidney defects was raised, but this finding was dismissed because it had never been reported before! (On this basis, it is hard to see how newly revealed effects could be investigated).

Mechanisms for effects

In addition to the problems already outlined of determining whether or not VDU work increases the risk of reproductive problems, there are also major difficulties in establishing how the effects are produced. Current theories include the following:

• chance
• stress
• ergonomics
• electromagnetic radiation
• psychosocial factors

Chance

As discussed on page 00, the theory that all the documented clusters of problems in pregnancy could be caused by chance should be viewed with some scepticism. If chance were the explanation, we would expect to see similar clusters in work places where VDUs are not in use.

Stress and psychosocial factors

During the last ten years, there have been numerous research studies, both experimental and epidemiological, which have shown that stress in pregnancy can increase the risk of stillbirths, low birth weights and spontaneous abortion. In particular, a recent USA study showed that women in jobs with high demands and low control were at greatest risk for these adverse pregnancy outcomes (Brandt and Nielsen 1992). As is often the case when adverse pregnancy results are recorded, the authors caution that the results may have been due to recall bias. However, another group of scientists found that 93 per cent of women completing a self-administered questionnaire on pregnancy outcome did so with sufficient accuracy to obtain valid data (Eskenazi and Pearson 1988).

As discussed in Chapter 6, many VDU jobs are associated with high levels of stress. During the Berlin Work with Display Units '92 conference the issue of stress was the focus of many discussions, with a recognition that there has been a increase in reported levels of stress among many groups of VDU workers. As one researcher put it: 'In designing VDT jobs, employers have lost all of the research work from factories that shows the importance of cohesiveness for the work group' (VDT News November/December 1992).

In the 1990 study of Inland Revenue staff mentioned earlier in this chapter, VDU users experienced significantly more symptoms of stress including loss of appetite, indigestion, anxiety, and depression compared to non-VDU users. In particular, VDU workers found the competing demands of home and work very difficult to resolve. As one women described it:

Reproductive outcome may also be affected by the indirect consequence of high stress levels - the increased consumption of caffeine, alcohol, cigarette and tranquilliser consumption. Cigarettes in particular are known to damage health including pregnancy and birth weight.

Ergonomics and physical stress

Ergonomics is the term used to describe the engineering aspects of the relationship between workers and their working environment. Poor ergonomics, for example poorly designed computer hardware, software and work stations can contribute to physical stress and in turn, may affect pregnancy outcome. (Ergonomic aspects of VDU work are discussed in more detail in Chapter 9). Long working hours, unrealistic productivity targets, rapid work pace, lack of control over the workload, and long hours in a fixed or cramped position can all add to physical stress.

Electromagnetic radiation

In the early days of VDU use, there was some concern about possible leakage of ionising radiation, such as X-rays from cathode ray tubes (see also Chapter 7 for more about screen technology). However, measurements of modern equipment have shown that X-ray leakage, which is undetectable from most machines, is unlikely to be the cause of reproductive problems. Over the last ten years attention has focused on a particular type of non-ionising electromagnetic radiation - very low frequency (VLF) and extremely low frequency (ELF) radiation.

As discussed in the sections on miscarriage and birth defects, there have been a number of major experimental studies which have shown that exposure to VLF and ELF of the type emitted from VDUs can affect pregnancy outcome in several animal species. There has been some recent evidence that exposure to high levels of ELF during early pregnancy increases the risk of miscarriage (Lindbohm et al 1992).

The study of the biological effects of exposure to electromagnetic radiation has been the subject of several very useful books (see Resources). In particular, Currents of Death (Brodeur 1989) explores in detail the attempts over the years of bodies with vested interests to discount and dismiss findings which show there may be a real danger from low-frequency electromagnetic radiation from VDUs. Clearly, much more work needs to be done before firm conclusions about their effects can be drawn. In the meantime, it would seem to make sense to accelerate research into alternative forms of visual display, which do not emit electromagnetic radiation. (Alternatives to cathode ray tube VDUs are discussed in Chapter 7). Perhaps now that a possible increased risk of brain tumours from electromagnetic field (EMF) exposure in the users of mobile phones has been suggested, we will see greater pressure for the development of non-emitting equipment.

The Schedule to the DSE Regulations states that:

This clause might go some way to protecting VDU users from the effects of exposure to VLF and ELF radiation were it not for the qualification which it receives in the Guidance to the Regulations. Here, it is stated that:

Information from around the world showing that widely varying, and potentially harmful amounts of non-ionising radiation can be emitted from different types of VDU seems not to have come to light during the preparation of this guidance. For example, in 1990 the USA computer journal Macworld reported that measurements of magnetic fields of up to 1.6 µT were taken at the sides of some major manufacturers' monitors. This is five times the level found by Finnish researchers to produce a threefold increased risk of miscarriage (Lindbohm et al 1992).

The difference, which appears to have escaped the authors of this Guidance, is that pregnant women do not usually spend up to eight hours a day sitting a couple of feet away from a working food processor or hair dryer! Indeed, important research on the use of electric blankets which emit magnetic fields of the same frequency as most VDUs, and which pregnant women may be exposed to for eight hours or more, showed that women who use these in pregnancy may be at increased risk of miscarriage (Brodeur 1989). In other words, far from offering assurance, this analogy demonstrates not that emissions from VDUs are safe, but emissions from some domestic appliances could be hazardous if pregnant women were exposed to them for several hours every day.

Combined effects

One possibility rarely considered by those investigating possible VDU-related reproductive hazards is that more than one effect could be operating at the same time.

It is to be hoped that one of the largest prospective studies on the reproductive effect of VDUs, involving more than 10,000 workers, being conducted by the US Mount Sinai Medical Center and the research and campaign group 9to5, which is due to report in 1994, will provide sufficiently clear results to enable VDU workers to be certain of whether or not they face serious reproductive risks from doing their job. In the meantime, we must continue to assume that VDU work may increase the risk of adverse reproductive outcome.

In the next section we look at how new laws may help to protect pregnant women from health hazards at work, and on agreements which have been negotiated by unions to help protect their members.

Pregnancy and the law

Clearly, since the DSE Regulations discount any link between VDU work and pregnancy problems, we cannot expect to find much in the way of improved protection for women in this piece of legislation.

However, some improvements for pregnant women at work are offered in the Trade Union Reform and Employment Rights Bill. This is a far reaching piece of employment legislation, which, as well as containing additional benefits for some groups of workers, contains much which will further undermine some of the most vulnerable sectors of the workforce. At the time of writing, the Bill has several stages to go through before it receives Royal Assent, and becomes an Act, which is likely to happen by October/November 1993.

The Bill's clauses on maternity rights are supposed to implement the provisions of the proposed EC Directive 'concerning the protection at work of pregnant women or women who have recently given birth'. The British Government did its best to derail the Directive and many of its original provisions will not now be implemented (London Hazards Centre 1992).

However, the following improvements are likely to become law:

• all employees, irrespective of their length of service, will be entitled to fourteen weeks maternity leave
• all employees, irrespective of their length of service will have the right not to be dismissed on the grounds of pregnancy or childbirth
• pregnant or breastfeeding employees who would have to be suspended from work for health and safety reasons will have the right to be offered suitable alternative work and to be paid during the period of suspension

In addition to the provisions in the Trade Union Reform and Employment Rights Bill, a new entitlement provided by the Workplace (Health, Safety and Welfare) Regulations 1992 is also of particular importance for women:

As we have concluded in the previous section, the possibility that VDU work does increase the risk of adverse reproductive outcome cannot be excluded on the basis of currently available information. The risks appear to be greater for those working long hours, under stress and with old VDUs or those emitting high levels of ELF or VLF radiation. Therefore, the best protection for pregnant women or those planning to become pregnant, is the right to transfer to alternative work if they wish to do so.

The law as it stands does not provide this right. As far as we know, the only workplaces in which this right has been negotiated are those which are represented by trade unions. The following are examples of negotiated agreements on VDU work and pregnancy:

Negotiated agreements

Inland Revenue Staff Federation (IRSF)

The joint guidance on VDUs and pregnancy agreed between the IRSF and Inland Revenue Management covers transfers to non-VDU work for pregnant staff, transfers to non-VDU work for staff hoping to become pregnant, and protection of the pay levels before transfer. Although the agreement makes clear management's responsibility to deal sympathetically with requests to transfer, the full support of the trade union may still be required to ensure that women who wish to, can take up their entitlements. In their study of women working for the Inland Revenue researchers found that only a small proportion of those entitled exercised this right, not because they did not wish to do so, but because they believed they would not be offered satisfactory alternative work, or they had experienced pressure from their managers to stay with VDU work (Bramwell and Davidson 1990).

National Union of Journalists (NUJ) BBC World Service

This agreement states that:

National Association of Local Government Officers (NALGO)[now UNISON]

The new technology agreements between NALGO branches and several local authorities also include clauses on rights to transfer. For example the following is extracted from the agreement between NALGO and Sheffield City Council:

'The Council has a policy that all pregnant women or women who believe they are in the earliest stages of pregnancy have the right to choose to be temporarily redeployed to other work during their pregnancy without detriment to their pay or conditions of employment.' (Unusually, this agreement also includes the right to transfer for sufferers of photosensitive epilepsy).

Checklist

The following checklist summarises the measures that can be taken, in order of preference, to protect VDU workers who are pregnant, or planning to become pregnant:

• A negotiated agreement between the trade union and employer providing the right to transfer, if requested, to non-VDU work
• Use of LCD monitors or CRT monitors which conform at least to the Swedish MPRII and TCO standards
• Limitation of amount of time per day working at the VDU to a maximum of four hours
• The right to frequent rest breaks away from the VDU - at least 10 minutes in every hour, but preferably whenever required
• No stress-inducing procedures such as machine pacing, electronic monitoring or payment-by results to be used
• Set up VDU workstation so that operators need be no closer than 70cm to their VDUs, and at least 1 metre from other terminals
• Use only fully adjustable comfortable chairs and a workstation set-up that suits individual requirement (see Chapter 9 for more on workstation design)
• Ensure that all VDUs are switched off when not in use.