Experimental studies have shown that metronidazole is carcinogenic in rodents and mutagenic in bacteria. In 771 women given metronidazole for the treatment of vaginal trichomoniasis, more cancers developed than had been expected after exclusion of carcinoma of one uterine cervix (observed, 24 cases; expected on the basis of the Connecticut Tumor Registry, 21.7; expected on the basis of the Third National Cancer Survey, 18.4). However, the excess was not statistically significant (P greater than 0.05). The observed and expected numbers of breast-cancer cases were the same, but four lung-cancer cases were observed, whereas 0.6 would have been expected. This finding is confounded by the fact that all four lung cancers developed in women who were smokers. Overall, we observed no appreciable increase in cases of cancer.

Rochester, Minnesota, has over many decades the highest reported incidence rates of primary intracranial neoplasms. This occurs in the absence of any readily identifiable risk factors. A case-control study of neuroepithelial tumors has been undertaken utilizing the records-linkage system at the Mayo Clinic. This study failed to yield any significant findings for the factors examined.

We compared the point prevalence rates of dementing illnesses in the population of Rochester, Minnesota, on January 1, 1980, with the previously published rates of January 1, 1975, by using extensive health care records available from all sources of care in the community, in order to identify and classify cases with standard diagnostic criteria derived from DSM-III. The overall age- and sex-adjusted prevalence rate per 100,000 population for dementia on January 1, 1980, was 402.5 compared with 388.4 on January 1, 1975; for Alzheimer's disease it was 259.8 in 1980 and 259.5 in 1975. The stability of the rates suggests no change in the incidence and mortality associated with these conditions. The prevalence rates for Rochester, Minnesota, are similar to many other rates estimated in other North American communities, but they differ from two other population-based estimates from East Boston and California. Case definitions and methods of assessment probably account for the differences in estimated prevalence.

No abstract