Untangling Differences in Cancer Mortality Rates: A Closer Look at Race and Education

Abstract

JNCI Vol. 99, Issue 18 | September 19, 2007 In this issue of the Journal, Albano et al. ( 1 ) draw attention to the association between education and cancer mortality overall and for the most common sites among white men and women and among black men and women. Their effort to cull the information on educational level from death certificates adds to the wealth of descriptive data on racial disparities in cancer mortality in the United States. However, the authors are not able to address with confidence a question that many readers will ask: do differences in education explain racial differences in cancer mortality? Albano et al. ( 1 ) show age-adjusted cancer mortality rates by years of formal education within strata that are defi ned by sex and race for individuals aged 25 – 64 years using information from death certifi cates collected in 47 of the 50 states and the District of Columbia for the year 2001. These rates are broadly representative of those for individuals younger than 65 years in the US population. The age restriction excludes deaths in older people, which, for example, constitute more than 90% of deaths from prostate cancer ( 2 ). Two unrealistic assumptions are required to extrapolate from the data presented by Albano et al. ( 1 ) to a conclusion that education level explains some fraction of the difference in cancer mortality rates between blacks and whites: fi rst, that we can estimate the effect of race on educational achievement, and second, that we can estimate the effect of increasing a child’s educational level on his or her subsequent risk of death from cancer. Estimating these effects is challenging because educational level both refl ects and affects personal and family income in childhood and adulthood, health insurance coverage and the use of preventive measures, early detection of cancer and medical care, occupational choices, and behavioral, occupational, and environmental factors — all of which have an impact on mortality. How might comparisons of education-, sex-, and race-specifi c cancer mortality rates obtained from vital statistics records improve health? Unequal mortality rates between groups can sometimes provide hints about where a successful public health intervention might have the greatest potential impact. The extra detail about educational level provided in the study by Albano et al. ( 1 ) might help in the design of more intensive attempts at reducing cancer Untangling Differences in Cancer Mortality Rates: A Closer Look at Race and Education