Results

DISTRIBUTION AND RETURNS OF QUESTIONNAIRES

Altogether, 22 415 subjects were selected for study, but 221 were excluded on their general practitioners' advice (because of bereavement or terminal illness), so that 22 194 questionnaires were posted. Usable responses were obtained from 12 907 subjects, including 9084 people who been at work in the week preceding completion of the questionnaire. The response patterns are described elsewhere.9

FREQUENCY OF EXPOSURE TO WHOLE BODY VIBRATION

A total of 3081 men (56.1% of those in employment) reported having been occupationally exposed to WBV in the week preceding completion of the questionnaire. The prevalence of exposure was similar in those who did and did not require a letter of reminder. The estimated 1 week prevalence of exposure among men of working age in the national population was 35.1% (95% CI 34.3 to 36.0%) overall, or 54.6% (95% CI 53.3 to 55.9%) of those at work.

This corresponded to a total of 7 197 000 (95%CI 7 026 000 to 7 369 000) men exposed. The occupations that contributed the largest estimated numbers of exposed men were: drivers of road goods vehicles (424 000 exposed, 95%CI 412 000 to 436 000), managers and proprietors of service industries (259 000, 95%CI 215 000 to 304 000), metal working, production and maintenance fitters, (246 000 95%CI 216 000 to 275 000), production workers and maintenance managers (229 000 95%CI 190 000 to 269 000), technical and wholesale sales representatives (170 000, 95% CI 149 000 to 191 000), and motor mechanics and auto engineers (174 000, 95%CI 161 000 to 187 000). The industries which contributed the largest estimated numbers of exposed men were: construction (965 000 men exposed, 95%CI 892 000 to 1 038 000), public administration and defence (523 000, 95%CI 473 000 to 573 000), and land transport (494 000, 95%CI 459 000 to 529 000).

Among women, 753 (19.4% of employed women in the sample) had been exposed in the past week, and again the prevalence differed little according to the need or otherwise for a reminder. The 1 week prevalence of exposure in the national population was estimated as 7.9% (95% CI 7.4 to 8.4%) of all women and 17.2% (95% CI 16.1 to 18.3%) of those in work.

The 1 week prevalence of occupational exposure to WBV in women corresponded to an estimated 1 783 000 (95% CI 1 667 000 to 1 899 000) exposed nationally. The occupations with the largest estimated number of exposed women were nurses (116 000 exposed, 95% CI 90 000 to 143 000), sales assistants (108 000, 95% CI 77 000 to 140 000), secretaries, personal assistants, and word processing operators (62 000, 95% CI 37 000 to 88 000), and managers and proprietors in service industries (55 000, 95%CI 29 000 to 82 000). The principal industries in which women were exposed were: health and social work (418 000, 95%CI 361 000 to 474 000) and the retail trade (200 000, 95%CI 155 000 to 245 000).

MOST COMMON OCCUPATIONAL SOURCES OF WHOLE BODY VIBRATION

Table 1 lists the sources of exposure to WBV in the past week that respondents reported most often. Among men, the most common exposures were car (31.5% of respondents), van (17.2%), forklift truck (9.3%), lorry (6.9%), tractor (4.6%), bus (4.3%), and loader (2.7%). The national estimates for these sources indicate that there are >4 000 000 occupational users of cars, >2 000 000 occupational users of vans, and >1 000 000 drivers of forklift trucks. Among women, cars, buses, and vans were the sources of exposure reported most often, with around 1.3 million women journeying in a car while at work during the previous week.

In men, lorries were driven for prolonged periods (median time 10 hours in the past week). Other lengthy weekly exposures arose from the use of cars (median time 6 hours) and vans, excavators, off road vehicles, armoured vehicles, and aircraft (median times 4–4.5 hours). In women, the median time for car journeys during work over the past week was 4 hours.

PERSONAL LEVELS OF EXPOSURE

Among 2301 exposed men from the sample who supplied full information, 836 (15.3% of male respondents) indicated exposures calculated to correspond with a daily average >8.5 ms^-1.75—the lower boundary of the health guidance caution zone suggested in International Standards Organisation (ISO) 2631, 1997.²⁰ These included 114 men (2.1% of male respondents) whose estimated exposure exceeded the action level of 15 ms^-1.75 which is proposed in British standard 6841.¹⁹ Among 616 exposed women providing full information, the corresponding figures were 42 (exposed >8.5 ms^-1.75) and two (>15 ms^-1.75).

Exposure information was incomplete for 780 men and 137 women exposed to WBV (either because of missing durations of exposure or missing data on representative vibration magnitudes), but in workers who were exposed to several sources of WBV, sufficient information was provided to suggest that at least a further 79 men and a further four women had a minimum eVDV >15 ms^-1.75. On this basis, the minimum number of men in the national population exceeding this action level was estimated to be at least 374 000 (95% CI 322 000 to 426 000), and the estimated number of women was 9000 (95% CI 2000 to 17 000). Figure 1 presents further details in the form of a cumulative frequency curve of the estimated numbers in the population exceeding different minimum values of eVDV in the past week.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Minimum estimated numbers of men and women in Great Britain whose daily equivalent estimated dose of vibration (eVDV) from all occupational sources in the past week exceeded the values indicated.

The proportion of workers by occupation and industry with significant levels of exposure is illustrated in figures 2 and 3. Among men, the occupations in which minimum estimated dose most often exceeded 15 ms^-1.75 were forklift truck and mechanical truck drivers (42%), farm owners and managers (31%), farm workers (27%), non-commissioned officers in the United Kingdom armed forces (11%), and drivers of road goods vehicles (9%). Exposures were most often estimated to exceed 8.5 ms^-1.75 in drivers of buses and coaches (95%), forklift trucks (91%), and roads goods vehicles (86%), and in farm owners (78%) and farm workers (76%).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Occupations in which significant exposures to whole body vibration most commonly arose in the past week among employed men.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Industries in which significant exposures to whole body vibration most commonly arose in the past week among employed men.

Table 2 presents estimates of the minimum numbers of men by occupation nationally with exposures >15 ms^-1.75 in the past week. Farm owners and managers (52 000), drivers of roads goods vehicles (39 000), and forklift truck drivers (34 000) contributed the greatest numbers by occupation; whereas workers in agriculture (67 000), construction (55 000), and land transport (34 000) contributed most by industry.

Table 3 presents the occupations in men with the highest median occupational estimated dose values. To examine the relative importance of leisure time exposures, the separate contribution to total eVDV from common non-occupational exposures was also estimated. The median eVDV from all sources combined was 6.4 ms^-1.75 (interquartile range (IQR) 5.1–8.4). The median leisure time value (5.4 ms^-1.75) exceeded the corresponding occupational estimate (2.7 ms^-1.75), but among men counted as having occupational exposure to WBV in the past week the total eVDV was higher (median 8.2 ms^-1.75, IQR 6.7–10.6), and the median occupational exposure (7.4 ms^-1.75) exceeded the median leisure time value (5.4 ms^-1.75). The highest median occupational values were found in men driving forklift trucks, buses and coaches, and road goods vehicles, and among technical and wholesale sales representatives. A wide spread of values was estimated for several of these occupations. In women, the median occupational eVDVs were zero, and 92% of respondents were estimated to incur greater exposure outside work than in employment, as were 53% of the women who reported occupational exposure in the past week.

REPRESENTATIVENESS OF EXPOSURES TO WHOLE BODY VIBRATION IN THE PAST WEEK

To determine whether there were seasonal differences in exposure, a comparison was made between the summer time and winter time responses. The overall prevalence of exposure was similar in the two periods (in men 43.1% v 43.6%, and in women 11.7% v 12.9%). Men from the winter posting were somewhat more likely to have an eVDV >15 ms^-1.75 (2.6% v 1.6%), but only minor differences were found among women.

Nine per cent of those reporting exposure (280 men and 70 women) stated that the past week had not been typical of their job. Among 141 men who supplied further particulars, 89 indicated that the exposure had been greater than normal, and 52 that it had been less. Among 22 women who provided more information, 13 reported that exposure was normally less and nine said it was usually more.

Discussion

Our findings indicate that in Great Britain some 9 000 000 people are exposed on a weekly basis to occupational sources of WBV, and highlight the principal occupations in which exposures occur, the major sources of exposure, and the contribution of leisure time activities to total dose of eVDV.

These estimates need to be considered in the light of possible biases and errors. We have argued elsewhere9 that the study sample is likely to have been adequately representative, given the large sample size, the wide geographical distribution and the completeness of family doctors' registers; and, moreover, a process of standardisation was used to compensate for differences in sampling and response rates by occupation; and major response bias seems unlikely, given the similarity of estimates of exposure in people who responded with and without a reminder.

A separate supporting study suggests that workers generally report their exposures to WBV accurately.15 However, a concern particular to the assessment of WBV is that recreational exposures might sometimes be reported as if they were incurred at work. The scope for this error is greatest for common means of private and public transport, as the distinction between using these vehicles at work (occupationally) as compared with journeys to and from work (non-occupationally) may be one that subjects fail to draw. In keeping with this, in our supporting study, a smaller proportion of work time exposures to the car in the past week were confirmed as feasible than for other sources of WBV. To explore this further, a check was performed to determine the occupations of people reporting prolonged work time exposure to cars, buses, coaches, trains, and motor cycles. In most cases the occupations were ones in which such levels of exposure were plausible—for example, chauffeurs, taxi drivers, and bus and train drivers—so large scale overreporting of exposure to these sources is considered unlikely.

None the less, the reporting of exposure to public and private transport vehicles has an important impact on the estimated numbers with exposure. To illustrate the size of effect, three alternative definitions of exposure are presented in table 4, the first encompassing all of the vehicles counted in the survey; the second excluding use of cars; and the third confined solely to industrial vehicles and machines. For comparison, information is provided on the reported prevalence of exposure by occupation in the survey of self reported working conditions.8 In professional managers and education and welfare workers, a substantial proportion of those exposed were classified as such solely on the basis of car use, whereas in other groups, such as farming, fishing and forestry, coal mining, road transport, and material moving and storing, few of those with exposure reported cars as the only source. The most restrictive definition resulted in a much lower estimate of frequency of exposure overall (23% v 56% for all men), and in specific occupational groups—for example, in professional and related support managers, 5% v 51%—but the estimated prevalences of exposure were generally higher than those in the survey of self reported working conditions. That community survey differed in several respects from our own, being based on a smaller (but randomly chosen) sample and on interviews rather than a postal questionnaire. It was also undertaken in a different calendar period; but the difference in findings is most likely to be accounted for by a difference in the questions used. In the survey of self reported working conditions exposure was ascertained by the single question: “Does/did your job ever involve you sitting or standing on a vibrating machine or in a vibrating vehicle?”, by contrast with our own checklist of sources. Respondents may not have considered cars, vans, buses, coaches, trains, and motor cycles to be included within the scope of the earlier survey's question, and may have also had difficulties in identifying or recalling relevant, substantive exposures in the absence of a clearer prompt, in which case our estimates may be more reliable.

Estimates of personal exposure are subject to several sources of error.16 The measurement of a_wz vibration was based on a subjective choice, and single values were chosen whereas the true magnitude will vary between vehicles that differ in model type, age, source of manufacture, seating, and tyre characteristics; and between similar vehicles which are used for different tasks, or driven over different surfaces, or at different speeds. To test the appropriateness of our choices, sample measurements were made on common sources of exposure,17 and in general the median measured values were close to those assigned (listed in the ). Only weekly exposure information was available, and so an average estimate of dose was made on the assumption that daily durations of exposure were sufficiently similar for the average to predict the likely risk, but in practice the relative importance of daily exposure pattern as a risk factor for back pain is not clear. Finally, estimates of personal doses of exposure relied on self estimates of duration of exposure, but empirical data from our accompanying paper suggest that this is reasonable, given the good general agreement between reported and observed exposure times.15

The data relate to exposures in the previous week, and this might convey a misleading impression of annual exposure in occupations where levels of exposure vary markedly by season. There will also have been people, especially users of cars, trains, buses, and planes, who were exposed unusually in the past week—for example, people attending occasional meetings. However, as estimates of reported exposure were similar in the winter and summer and few people reported their exposures to be unusual, these factors are not thought to have had a large impact.

Conclusions

Occupational exposures to WBV are common in Great Britain, with an estimated 7 200 000 men and 1 800 000 women exposed in a 1 week period. Only a small proportion of exposures exceed the action level (eVDV >15 ms^-1.75) proposed in British standards, but this represents an estimated minimum of 374 000 men and 9000 women with such levels of exposure nationally. The commonest occupational sources of WBV have been identified, as have the occupations where particularly high exposures arise—providing a basis for targeting future control activities. However, in many occupations, non-occupational sources make a more important contribution to exposure.