The association between cigarette smoking and colorectal cancer over five decades
Peppone, Luke Joseph
MetadataShow full item record
Objective. To examine colorectal cancer risk by smoking status using data from patient admissions who were treated at Roswell Park Cancer Institute between 1957 and 1997 with a variety of epidemiological study design. Background. Earlier studies (1950s-1960s) examining smoking and colorectal cancer risk failed to find an association. Studies in the 1980s consistently observed an increased risk for colorectal adenomas among smokers, a colorectal cancer precursor. The majority of subsequent, well-designed studies observed an increase in colorectal cancer risk among the heaviest smokers, while a smaller number of studies still did not find an increased risk. Methods. This dissertation used 4 separate data aims to examine the association between smoking and colorectal cancer. Specific Aim #1 (1957-1965) extended the original sampling frame of a prior hospital-based case-control analysis and recalculated risks by smoking exposure. Specific Aim #2 (1957-2002) used a historical cohort design to analyze colorectal cancer risk from smoking. Specific Aim #3 (1965-1975) used a hospital-based case-control design to examine colorectal cancer risk by smoking. Specific Aim #4 (1982-1997) used a hospital-base case-control to calculate colorectal cancer risk by smoking. The last aim, Specific Aim #5 analyzed the changes in cigarettes and its components over the period of the dissertation using peer-reviewed literature and industry documents. Results. Specific Aim #1 failed to find an association between smoking and colorectal cancer (>42 Pack-Years: OR=0.99, 95%CI=0.75-1.31). Specific Aim #2 did not find a positive association between heavy smoking and colorectal cancer (> 19 Pack-Years: RR=1.19, 95%CI=0.85-1.67). Specific Aim #3 did not find a positive association between smoking and colorectal cancer (>41 Pack-Years: OR=0.69, 95%CI=0.45-0.89). Specific Aim #4 found a small association between smoking and colorectal cancer (>40 Pack-Years: OR=1.23, 95%CI=1.02-1.48) and the risk was greater when those with second-hand smoke exposure were removed (>40 Pack-Years: OR=2.12, 95%CI=1.42-3.15). Specific Aim #5 found that over the previous 5 decades, levels of heterocyclic amines have risen significantly, possibly accounting for the increase in risk over time. Conclusion. The results from this dissertation are in agreement with previous literature. Future research is needed to determine if the association between smoking and colorectal cancer is causal.