Materials and methods

A total of 42 rubber factories in the United Kingdom, all members of the British Rubber Manufacturers' Association (BRMA), supplied identifying particulars, work histories, and information on smoking for those 9031 male and female workers first employed in the factory environment of the participating factories during the 10 year period 1982–91. All study subjects were employed for a minimum period of 12 months. Each period of working was classified by one of 27 product sectors—for example, new tyres, retread, footwear—and one of 55 job categories—for example, weighing of compounding ingredients, extruding, painting of finished goods—the 55 job categories relate to 11 broader job types. Information on cigarette smoking at the time of first employment was available for 4821 study subjects, but only in terms of current smoker (n=2108), ex-smoker (n=372), and lifelong non-smoker (n=2341).

The Office for National Statistics (ONS) provided information on the vital status of each study subject up to the current closing date of the survey (31 December 1998): 8329 (96.3%) subjects were alive, 27 (0.3%) subjects had emigrated, 136 (1.6%) were dead, and 159 (1.8%) subjects were untraced. Flagging is not available in Northern Ireland, and 380 workers from Northern Ireland are not included in the present analysis of 8651 study subjects. The underlying cause of death was coded according to the 9th revision of the international classification of diseases (ICD-9). The ONS also supplied information on cancer registrations (incident cancers) for the period 1971–94.

Expected numbers of deaths were calculated by applying sex, age, and period specific mortalities for England and Wales to corresponding person-years at risk (pyr). Each study subject contributed pyr from the end of the first 12 months of employment to the closing date of the study, death, emigration, or date last known alive, whichever was the earliest. Standardised mortality ratios (SMRs) were calculated as the ratio of observed deaths to expected deaths, expressed as a percentage. These procedures were carried out with the PERSONYEARS programme. Standardised registration ratios were calculated in the same way for the cancer morbidity data, except that the closing date of the study was 31 December 1994. The mean age at the start of employment was 28.5 years and the mean period of follow up (mortality analysis) was 11.9 years.

Results

There was evidence of a strong healthy worker effect in the data with mortality from all causes (men and women combined) being particularly low in the period after first employment (1–4 y: SMR 43, obs 23; 5–9 y: SMR 74, obs 63; 10–19 y: SMR 75, obs 50). Observed and expected numbers of deaths and cancer registrations are shown in the table for several different cancers, together with observed and expected numbers of deaths for the larger non-cancer groupings. Marked deficits in mortality are shown in male workers both for diseases of the circulatory system and all causes. The only significant excess is shown for mortality from cancer of the testis (obs 3, exp 0.51, SMR 589, 95% confidence interval (95% CI) 122 to 1722), although findings for incident cancers of the testis were unexceptional (obs 5, exp 5.13, standardised registration ratio (SRR) 97 (95% CI 32 to 227).

There was a significant gradient (p<0.05) of all causes mortality relative to cigarette smoking (non-smokers: SMR 44, obs 20; ex-smokers: SMR 63, obs 7; current smokers: SMR 93, obs 52; men and women combined, not shown in table). This gradient was most marked for diseases of the circulatory system (non-smokers: SMR 8, obs 1; ex-smokers: SMR 26, obs 1; current smokers: SMR 97, obs 18).

Discussion

Although at a relatively early stage of follow up, the new study of recent entrants to the United Kingdom industry has failed to find a significant excess of cancer morbidity for any type of cancer and only found a significant excess mortality for a single cause of death, cancer of the testis. The absence of any excess, significant or otherwise, in the morbidity findings for cancer of the testis argues strongly against working in the rubber industry being involved. (Interestingly, there was a significant deficit for mortality from testicular cancer in the earlier BRMA study.6) It is clear that having data on cancer incidence as well as cancer mortality makes confident interpretation easier. The unexceptional findings for cancers of the stomach and lung will be welcome news for United Kingdom rubber workers, albeit they are associated with wide 95% CIs; the position needs to be monitored over the next 5–10 years. In the course of time, there will be a sufficient number of events in this new cohort study to make use of the detailed work history information and the information on smoking in a single analysis.