The cancer mortality scare. Problems of estimation using monthly data

L. Chiazze Jr, D. T. Silverman and D. L. Levin

Considerable attention has been given recently to an apparent 5.2% rise in cancer mortality, based on data for the first seven months of 1975 as reported to the National Center for Health Statistics (NCHS). Later data disclosed that the rise in crude total cancer mortality, while greater than in any recent year, is less than half of the 5.2% originally reported, and the rise in age-adjusted total cancer mortality, while less than 1%, is also higher than the average annual increase from 1968 to 1974. The Monthly Vital Statistics Report provides death rates not standardized for age, race, and sex, and does not include information for specific cancer sites. Therefore, although information in the Monthly Vital Statistics Report is useful for detecting sudden increases in mortality from acute epidemics, it is not appropriate for forecasting trends in cancer mortality.