Retrospective Data Quality Audits of the Harvard Six Cities and American Cancer Society Studies

• Published in: Journal of toxicology and environmental health. Part A Vol. 66; no. 16; p. 1553
• Main Authors: Hoover, BKristin, Foliart, Donna E, White, Warren H, Cohen, Aaron J, Calisti, Linda J, Krewski, Daniel, Goldberg, Mark S
• Format: Journal Article
• Language: English
• Published: 10.10.2003
• ISSN: 1528-7394
• DOI: 10.1080/15287390390211496

Two longitudinal cohort mortality studies, the Harvard Six Cities Study and the American Cancer Society Study, provided data showing an association between increased mortality and higher ambient levels of fine particulates and sulfate. Intense scientific and political debate ensued over the studies, and the original investigators eventually agreed to an independent reanalysis of the data. Results are presented from retrospective data quality audits of the two studies, focusing on the questionnaires and death certificates from each study and comparing a preselected random sample of all air-pollution measurements in each study against the most raw form of the data. The results show that both studies had been constructed and documented well, and the minor errors found in the data did not impact the original conclusions materially.