Source: Dissertation Abstracts International, Volume: 78-01(E), Section: B.

Adviser: Susan T. Mayne.

Access restricted by licensing agreement.

This item is not available from ProQuest Dissertations & Theses.

BACKGROUND Alcoholic beverage consumption is widespread in the US. Although moderate use may confer certain health benefits, alcohol is a known carcinogen at some organ sites and higher use especially has been associated with increased risk for a variety of other important health outcomes. Therefore, the goal of this dissertation was to evaluate the relationship between alcoholic beverage consumption and three health outcomes, with careful consideration of confounding by tobacco, non-linear relationships, and effects at moderate alcohol doses. The outcomes included lung carcinoma, where different types of lung cancer may have different relationships with alcohol; oral health, which is relatively understudied in relation to alcohol; and colorectal cancer and adenoma, using biomarker-based methods to assess alcohol exposures.

METHODS Chapter 1: Alcohol and Lung Carcinoma by Histological Subtype We prospectively evaluated the relationship between self-reported alcohol consumption and lung carcinoma, overall and by histological subtype, in 492,902 participants from the National Institutes of Health-AARP Diet and Health Study. We calculated hazard ratios and 95% confidence intervals using Cox proportional hazards regression models, adjusting for tobacco smoking and other potential confounders.

Chapter 2: Alcohol Consumption and Oral Health Outcomes We used data from the US nationally representative National Health and Nutrition Examination Survey (NHANES) 2009-2012 to evaluate the relationship between self-reported alcohol consumption and the following oral health outcomes: periodontal disease, tooth loss, untreated dental caries, self-reported oral health status, and self-reported gum disease. We calculated odds ratios and 95% confidence intervals using multivariable logistic regression models, adjusting for cigarette smoking, other tobacco use, and other potential confounders.

Chapter 3: Alcohol Metabolites, Colorectal Cancer, and Colorectal Adenoma We examined the association between novel serum biomarkers of alcoholic beverage consumption and risk of colorectal cancer and colorectal adenoma in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO) and the Navy Colon Adenoma Study (Navy Adenoma Study), respectively. We estimated odds ratios and 95% confidence intervals using multivariable logistic regression models, adjusting for tobacco use and other potential confounders, for the following eight alcohol consumption biomarkers: ethyl glucuronide; 4-androstene-3beta,17beta-diol disulfate 1; 5-alpha-androstan- 3beta,l7beta-diol disulfate; 16-hydroxypalmitate; bilirubin (E,Z or Z,E); cyclo (- leu-pro); dihomo-linoleate (20:2n6); and palmitoleate (16:1n7).

RESULTS Chapter 1: Alcohol and Lung Carcinoma by Histological Subtype For total lung carcinoma, when compared to non-drinkers alcohol consumption was associated with a non-linear association, whereas lower risk was observed with modest intake and increased risk was observed with higher intake. The magnitude of each association was modest. The observed associations differed by histological subtype. For squamous cell carcinoma, alcohol consumption was associated with lower risk that became attenuated with higher consumption. However, for adenocarcinoma higher alcohol consumption was modestly associated with increased risk.

Chapter 2: Alcohol Consumption and Oral Health Outcomes We found some evidence for associations between self-reported alcohol consumption and periodontal disease (PD), tooth loss, and untreated dental caries. For PD, although the overall test for an association was statistically significant, none of the individual categories of alcohol consumption were significantly associated, and there was no clear pattern across categories. The results for self-reported gum disease were similarly inconsistent. Although tooth lass and untreated dents! caries were inversely associated with alcohol consumption, there was no dose response, with similar associations found across categories of alcohol intake. There was also no clear association observed for self-reported oral health status. Together, our findings suggest that moderate alcohol consumers are at similar risk to non-drinkers for PD, but may be at a lower risk for tooth loss and untreated dental caries.

Chapter 3: Alcohol Metabolites, Colorectal Cancer, and Colorectal Adenoma We found no clear evidence for a strong association between novel serum biomarkers of alcoholic beverage consumption and either colorectal cancer or colorectal adenoma. There was some evidence for an association between one marker (cyclo (-leu-pro)) and lower risk of colorectal adenoma. However, it is important to interpret these findings cautiously given that we performed multiple comparisons and had a modest sample size.

CONCLUSION In this dissertation, we found no evidence for adverse associations between moderate alcoholic beverage consumption and the endpoints of lung carcinoma, multiple markers of poor oral health, or colorectal carcinoma and adenoma. Instead, we observed some evidence that moderate alcohol drinkers had lower risk of total lung carcinoma, squamous cell carcinoma, tooth loss, and untreated dental caries. Overall, we conclude that in the range of consumption previously linked to a lower risk of total mortality, there are no apparent adverse effects on the endpoints studied that could cause reconsideration of current public health guidelines.