Colorectal cancer is the second most common cause of cancer deaths in Americans; producing approximately 60,000 deaths in Americans each year. Studies of high and low risk populations have suggested a dietary etiology for colorectal cancer, although investigations of specific dietary factors have produced inconsistent results.

Population based studies of California Seventh-day Adventists demonstrate dietary variety regarding factors suspected as important in the etiology of colorectal cancer. These studies document a substantially reduced mortality and incidence from colorectal cancer in Seventh-day Adventists when compared to other California Populations. Although specific reasons for this difference remain obscure, the 'Unique lifestyle and dietary practices of Adventists are of prominent interest This investigation was facilitated by abundant data collected in previous studies of California Seventh-day Adventists.

Colorectal cancer cases and controls were identified from within the Adventist Health Study cohort which has been followed for cancer incidence since 1976. In total, 78 colorectal cancer cases and 198 controls completed a detailed questionnaire measuring dietary and lifestyle practices. Dietary information included daily frequency of use for approximately 85 foods and beverages. Specific attention was given to selected food/nutrient categories.

The purpose of this research was investigation of the association between colorectal cancer in California Seventh-day Adventists and past consumption of specific foods and nutrients. Major focus was placed upon the potential protective roles of vitamin A (retinol and beta-carotene) in colorectal cancer etiology. Attempts to isolate associations with vitamin A components required measurement and elimination of potential confounding produced by several other dietary variables. Extraneous factors included use of meat, coffee, crude fiber, crucifer vegetables, animal fat, vegetable fat, total fat and protease inhibitor rich foods.

Results identified no relationships consonant with positive associations with colorectal cancer for use of coffee, meat, animal fat, total fat or vegetable fat. A weak association between past total vitamin A (retinol and beta-carotene) intake and colorectal cancer was identified. Although the finding was consistent with a potential protective role in the etiology of cancer, it was not statistically significant for tests of independence or gradient (trend) across exposure levels. No differences were identified regarding retinol intake between colorectal cancer cases and controls. The index used to measure beta-carotene showed a statistically significant dose-response association with colorectal cancer which was consistent with a potential protective effect. The association was independent of other dietary factors listed above and persisted after controlling potential effects produced by age, gender, meat, crude fiber and protease inhibitor food use. Nearly identical results were produced using stratified analysis (Mantel-Haenszel summary odds ratios) and linear-logistic regression analysis.

Crude fiber showed a curvilinear association with colorectal cancer with odds ratios for low exposure levels = 1.0, moderate=3.4 and high = 1.9. None differed significantly from the null value. The association was markedly altered by simultaneous and individual adjustment of beta-carotene and protease inhibitor food use. The association with crude fiber did not appear to be direct.

No associations with colorectal cancer were observed for ascorbic acid or crucifer vegetable use. A protease inhibitor food use index based upon usual frequency of use for beans, legumes, seeds, corn and nuts showed a significant, dose-response protective association in females; no association was seen for males. Linear-logistic regression analysis identified a multiplicative interaction between gender and the protease inhibitor food index.

Results suggest possible protective associations for colorectal cancer for beta-carotene (both sexes) and protease inhibitor food use (females only). Potential systematic differences in recall of past diet and other biases acknowledge for case control studies can not be eliminated as a possible source of these associations.

LLU Discipline



Epidemiology and Biostatistics


School of Health

First Advisor

Gary E. Fraser

Second Advisor

Thomas Mack

Third Advisor

James Blankenship

Fourth Advisor

Paul Yahiku

Fifth Advisor

Richard Koobs

Degree Name

Doctor of Public Health (DrPH)

Degree Level


Year Degree Awarded


Date (Title Page)




Library of Congress/MESH Subject Headings

Seventh-day Adventists -- case studies; Vitamin A; Colorectal Neoplasms -- prevention & control



Page Count

x; 316

Digital Format


Digital Publisher

Loma Linda University Libraries

Usage Rights

This title appears here courtesy of the author, who has granted Loma Linda University a limited, non-exclusive right to make this publication available to the public. The author retains all other copyrights.


Loma Linda University Electronic Theses and Dissertations

Collection Website



Loma Linda University. Del E. Webb Memorial Library. University Archives