Cancer prevalence and education by cancer site: logistic regression analysis
Johnson, S.; Corsten, M.J.; McDonald, J.T.; Gupta, M.
Journal of Otolaryngology - Head and Neck Surgery 39(5): 555-560
2010
ISSN/ISBN: 1916-0216 PMID: 20828519 Document Number: 640506
Previously, using the American National Health Interview Survey (NHIS) and a logistic regression analysis, we found that upper aerodigestive tract (UADT) cancer is correlated with low socioeconomic status (SES). The objective of this study was to determine if this correlation between low SES and cancer prevalence exists for other cancers. We again used the NHIS and employed education level as our main measure of SES. We controlled for potentially confounding factors, including smoking status and alcohol consumption. We found that only two cancer subsites shared the pattern of increased prevalence with low education level and decreased prevalence with high education level: UADT cancer and cervical cancer. UADT cancer and cervical cancer were the only two cancers identified that had a link between prevalence and lower education level. This raises the possibility that an associated risk factor for the two cancers is causing the relationship between lower education level and prevalence.