Search

Objective: The majority of research on breast cancer risk and socioeconomic status (SES) has been conducted for blacks and whites. This study evaluates the relationship between SES and breast cancer incidence in California for four race/ethnic groups., Methods: Principal component analysis was used to create an SES index using 1990 Census data. Untracted cases were randomly allocated to census block groups within their county of residence. A total of 97,227 female breast cancer cases diagnosed in California between 1988 and 1992 were evaluated. Incidence rates and rate ratios (RRs) were estimated and a chi2 test for trend across SES levels was performed., Results: SES was positively related to breast cancer incidence, and this effect was stronger for Hispanics and Asian/others than for whites and blacks. Adjusting by SES did not eliminate the differences in breast cancer rates among race/ethnic groups. RR differences between the race/ethnic groups were greatest in the lowest SES category and attenuated with increasing SES. An increasing trend over SES was statistically significant for all race/ethnic groups. Including randomly allocated cases affected RR estimates for white women only., Conclusions: Our results are consistent with similar findings for the Los Angeles area but differ from previous results for the San Francisco Bay area.

A probability sample of California workplaces in 1986 estimated the numbers of workers using lead and their coverage by biological and environmental monitoring. Study steps included (1) identifying 505 stand ard industrial classifications (SICs) with possible lead use; (2) creating an employer lis; (3) identifying for oversampling seven single SICs and a group of 165 SICs with documented high ambient air lead; (4) further stratification by workforce size; (5) prediction from National Institute for Occupational Safety and Health surveys the number of lead-using workers for each employer; (6) drawing facilities into the sample with selection probability proportional to the number of workers predicted to use lead; (7) adjusting the sample selection if an employer listing covered multiple facilities; (8) sending a first-stage telephone/mail questionnaire to identify lead users, who were sent a second questionnaire; (10)follow-up which, aided by statutory authority, resulted in completion rates over 90%; (11) ranking lead-using processes by intensity of exposure; and (12) analysis. Sources of error include random sampling error, misleadingly large standard errors because sampling was with replacement, gaps in the employer list, and inaccurate responses. As a measure for the probability proportional to size sampling, total work force count would have been superior to the predicted number of lead users. Restriction to a single hazard makes a questionnaire-based study feasible, but names of workers, rather than counts, should be requested and a subsample should be visited.

An estimated 230,000 workers were reported to work in lead-using activities as a result of a 1986 statistical survey of employers in California. However, 68% of these workers worked in activities of relatively low hazard. This included 114,000 workers engaged in electronic soldering. Over 1800 workers were reported to work in battery manufacture, a relatively high-hazard activity, but with consistent monitoring and protection. In contrast, over 4300 workers were involved in radiator repair activities, a high-hazard activity with inconsistent monitoring and protection. Over 8% of construction workers were reported to work in lead-using processes. This sector is exempt from federal Occupational Safety and Health Administration (Fed/OSHA) and California Occupational Safety and Health (Cal/OSHA) lead standards, yet some of the most hazardous activities (welding, grinding, and cutting, etc.) occurred here. Comparison of projections from the National Occupational Hazard Survey of potential lead exposure in California with these survey results suggests consistent and reliable assessment of the major sources of potential occupational exposure to lead.

Patterns of environmental and biological monitoring for lead exposure were surveyed in lead-using industries in California. Employer self-reporting indicates a large proportion of potentially lead-exposed workers have never participated in a monitoring program. Only 2.6 percent of facilities have done environmental monitoring for lead, and only 1.4 percent have routine biological monitoring programs. Monitoring practices vary by size of facility, with higher proportions in industries in which larger facilities predominate. Almost 80 percent of battery manufacturing employees work in job classifications which have been monitored, versus only 1 percent of radiator-repair workers. These findings suggest that laboratory-based surveillance for occupational lead poisoning may seriously underestimate the true number of lead poisoned workers and raise serious questions regarding compliance with key elements of the OSHA Lead Standard.