Asbestos In Drinking Water And Cancer In The San Francisco Bay Area: 1969-1974 Incidence

Our plain-English summary of Asbestos In Drinking Water And Cancer In The San Francisco Bay Area: 1969-1974 Incidence, by P.M. Conforti, M.S. Kanarek, L.A. Jackson, et al (1981).

The study

• Authors: P.M. Conforti, M.S. Kanarek, L.A. Jackson, R.C. Cooper, J.C. Murchio
• Published: Journal Of Chronic Diseases 34(5):211-224 (1981)
• Type of article: Journal article
• Our copy: from Johns-Manville's own medical library, J-M no. 000731. The article is still under copyright, so we describe it here rather than post it.

In plain English

This geographic correlation study extended an earlier study of 1969-1971 San Francisco-Oakland data. The authors compared 1969-1974 cancer incidence (51,314 cases) across 410 'super tracts' (groups of census tracts) with chrysotile asbestos counts in each area's drinking water, up to 36,000,000 fibers per liter. Results were reported for the white population. Across three statistical tests, all digestive cancers combined and esophageal cancer were consistently and positively associated with asbestos level in both white men and white women. In men, so were all cancers combined and cancers of the digestive tract, stomach, colon and digestive related organs. In women, so was retroperitoneal cancer. Female reproductive and corpus uteri cancers showed a negative association that the authors could not explain. For selected cancers, the pattern appeared at every income and education level. In the 307 most stable tracts, results were broadly similar, but some associations lost significance, including all digestive cancers in women. The authors noted that their models explained little of the variation and that many tests were run. They said better data on smoking, occupation and alcohol were needed. They concluded the results point to a statistical association and called for a case-control study.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Geographic correlation study comparing cancer incidence between areas with differing measured asbestos levels in drinking water. It extended an earlier study of 1969-1971 data.
• Who was studied: 410 super tracts (groupings of 1960 and 1970 census tracts) of the San Francisco-Oakland Standard Metropolitan Statistical Area, which covers Alameda, Contra Costa, Marin, San Francisco and San Mateo counties. The 1970 population was slightly over 3 million. There were 51,314 incident cancer cases from 1969 to 1974, and results were reported for the white population.
• Exposure: Chrysotile asbestos in drinking water. The water supplies have varying contact with naturally occurring serpentine. Fibers were counted by transmission electron microscopy in 353 water samples from the original study plus 19 more. Each super tract was assigned the mean value for its water service area. Counts ranged from below the detection limit to 3.6 x 10^7 (36,000,000) fibers per liter.
• Main results:
  • In chi-square slope tests, all sites, all digestive and digestive tract cancers showed highly significant (p less than 0.00001) positive gradients with asbestos level in white males.
  • In white females, female reproductive and corpus uteri cancers showed highly significant negative gradients: lower cancer rates went with higher asbestos levels.
  • Significant (p less than 0.05) positive gradients were found in both sexes for all digestive, digestive tract, esophagus and colon cancers.
  • For the cancer sites with significant (p less than 0.01) positive chi-square slope results, tracts were cross-classified by asbestos level and by education or income. The authors reported that the relationship of asbestos to cancer appeared to be observed at each socioeconomic level.
  • In regression analyses that took income, education, marital status and asbestos-industry workers into account, the most highly significant (p less than 0.001) positive asbestos results were for all digestive, digestive related organs and pancreas cancers in white males. In white females they were for all respiratory and trachea, bronchus and lung cancers.
  • Across all three tests (chi-square, regression and correlation), all digestive and esophageal cancers were consistently, significantly and positively related to asbestos in both white males and white females.
  • Also significant across all three tests in white males were all sites, digestive tract, stomach, colon and digestive related organs. In white females, retroperitoneum was also significant across all three tests.
  • When the analysis was limited to the 307 'high stability' super tracts, the authors reported close correspondence with the analysis of all 410 tracts.
  • Some associations were not significant in the high stability tracts. In white males, all sites lost significance under regression. In white females, all sites, all digestive and breast lost significance under regression. In white females, all digestive, stomach, rectum, gall bladder, pleura and corpus uteri lost significance under correlation testing.
• The authors' conclusion: The authors concluded that the results point to a statistical association between the incidence of certain cancers, particularly digestive system cancers, and increasing amounts of asbestos in the Bay Area domestic water supply. They said a carefully designed case-control study is needed to clarify this.
• Limitations the authors state:
  • The R squared values were low (always less than 0.172 and usually between 0.02 and 0.05), so the variables in the model explained only a small proportion of the variation in cancer incidence. The authors said the regression data must be viewed with caution.
  • As many as 210 regression equations were computed, so the usual significance levels cannot be taken at face value. The authors said the consistency across procedures reduces this concern.
  • More accurate information on smoking, occupation and alcohol consumption is needed.
  • The study was based on place of residence and did not account for daily mobility such as place of work.
  • The negative relationship for white female reproductive cancers is contrary to the hypothesis, and the authors had no obvious explanation for it.

Topics

Medical terms in this study, explained in our Lexicon: Chrysotile, Esophageal Cancer, Case-Control Study.

Browse return to medical research.