Summary A previous case-control study which utilised the occupational information available on the New Zealand Cancer Registry found an increased risk of multiple myeloma in agricultural workers consistent with previous findings in the United States. The findings are now presented for the second phase of the study which involved interviewing 76 cases of multiple myeloma (who had been included in the previous study) together with 315 controls with other types of cancer. The previous finding on an excess of farmers in the case group was confirmed by the interview data (odds ratio= 1.7, 95% confidence limits 1.0-2.9, P=0.04). There were no significant differences between cases and controls regarding potential exposure to phenoxy herbicides or chlorophenols. There were also no significant differences regarding activities involving potential exposure to other agricultural chemicals, although the odds ratio for fencing work, which may involve exposure to arsenic and sodium pentachlorophenate, was 1.6 (95% confidence limits 0.9-2.7, P=0.l1). The odds ratios were significantly elevated for sheep farming (odds ratio= 1.9, 95% confidence limits 1.0-3.6, P=0.04) and exposure to beef cattle (odds ratio=1.7, 95% conficence limits 1.0-2.9, P=0.05). The odds ratio was also elevated for persons reporting a history of hay fever (odds ratio= 1.9, 95% confidence limits 1.0-3.5, P=0.05). Overall, these findings suggest that the search for the causes of elevated mortality in farmers from multiple myeloma should be directed to potential causes other than pesticide exposure.

A previous case-control study which used the occupational information available on the New Zealand Cancer Registry found that agricultural workers were at increased risk of developing non-Hodgkin's lymphoma. The findings are now presented for the second phase of the study which entailed interviewing 83 cases of non-Hodgkin's lymphoma registered under code 202 of the International Classification of Diseases together with 168 controls with other types of cancer and 228 general population controls. The findings for the two control groups were similar, and there-were no significant differences between cases and controls regarding potential exposure to phenoxyherbicides (odds ratio = 1-4, 90% confidence limits 0-7-2-5, p = 0-26) or chlorophenols (odds ratio = 13, 90% confidence limits 0-6-2-7, p = 0-39). The odds ratio for fencing work, necessitating exposure to several potential risk factors including arsenic and sodium pentachlorophenate was 2-0

Occupational and social class differences in cancer mortality among New Zealand males aged 15-64 are examined for the period 1974-78. Age-standardized cancer mortality rates are presented for the Registrar General's social classes as well as for each of six occupational orders and 79 occupational groups. The rates for specific cancer sites are also presented for each social class and for those occupational groups with significantly elevated relative risks. The findings of the social class analyses were generally consistent with those of recent British studies with mortality from cancer of the liver, larynx, lung, buccal cavity and stomach being particularly high in the lower social classes and mortality from multiple myeloma, malignant melanoma and lymphatic leukaemia being particularly high in the upper social classes. The findings of the occupational group analyses were also generally in line with those of recent British studies and those associations which have been found in British studies and were also present in the New Zealand data are discussed. It is concluded that there are a number of associations which warrant further investigation including: large bowel cancer in woodworkers and printers; bladder cancer in hairdressers and beauticians; and malignant lymphoma in farmers.

The findings are reported from a New Zealand Cancer Registry-based case-control study involving 427 male patients with testicular cancer registered during the period 1979-1983 and aged 20 years or more at time of registration. Controls were also males chosen from the Cancer Registry with two controls per case, matched on age and year of registration. It was found that, as in other countries, persons in the upper social class groupings were at increased risk of testicular cancer. Persons in professional occupations were also at increased risk, but the odds ratio of 1.09 was much smaller than found in other studies. The previously reported excess risks for farmers, food and beverage workers, forestry workers, and pulp and paper workers were not supported by the New Zealand data. On the other hand, the previously reported excess risk for sales and service workers including members of the armed forces was supported, to some extent, by the New Zealand data with odds ratios of 1.38 (95% confidence limits 0.98-1.93) and 2.15 (95% confidence limits 0.80-5.79), respectively. Other groups with elevated risk include: physicians (odds ratio = 6.50, 95% confidence limits 1.29-32.6); production supervisors (odds ratio = 2.85,95% confidence limits 1.00-8.13); and motor vehicle mechanics (odds ratio = 2.02, 95% confidence limits 0.93-4.42). However, the New Zealand data generally does not suggest that occupational factors (or lifestyle factors associated with occupation) are of major direct importance in the etiology of testicular cancer. The incidence of testicular cancer has a bimodal age distribution in New Zealand and has risen markedly during the period 1948-1979. The New Zealand data differed from patterns observed in other countries in that the relative increase was approximately uniform across age groups rather than being stronger in the younger age groups.

Mortality studies have indicated that workers in agriculture and forestry may be at increased risk of leukemia. Findings are reported from a New Zealand Cancer Registry-based case-control study involving 546 male leukemia patients registered during 1979-1983 and aged 20 years or more at time of registration. Controls were also males chosen from the Cancer Registry with four controls per case, matched on age and year of registration. The case group contained an excess of the occupational category involving agriculture and forestry (odds ratio (OR) = 1.24, 95% confidence interval (CI) = 0.95-1.61) with the greatest relative risk being for livestock farmers (OR = 3.00, 95% CI = 1.23-7.32). There was also an excess of electrical workers (OR = 1.72, 95% CI = 0.92-3.20). The agricultural excess was greatest in patients aged 65 years or more at time of registration (OR = 1.29, 95% CI = 0.94-1.78), particularly in patients with acute myeloid leukemia (OR = 1.55, 95% CI = 0.90-2.67) or acute monocytic leukemia OR = 10.38, 95% CI = 1.99-54.29), although the latter excess only involved five cases. Acute myeloid leukemia was also elevated in meat workers (OR = 2.51, 95% CI = 1.19-5.30).