Skip Navigation


American Journal of Epidemiology Advance Access originally published online on May 7, 2007
American Journal of Epidemiology 2007 166(2):170-180; doi:10.1093/aje/kwm067
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
166/2/170    most recent
kwm067v2
kwm067v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Park, Y.
Right arrow Articles by Schatzkin, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Park, Y.
Right arrow Articles by Schatzkin, A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Published by the Johns Hopkins Bloomberg School of Public Health 2007.

ORIGINAL CONTRIBUTIONS

Fruit and Vegetable Intakes and Risk of Colorectal Cancer in the NIH–AARP Diet and Health Study

Yikyung Park1, Amy F. Subar2, Victor Kipnis3, Frances E. Thompson2, Traci Mouw1, Albert Hollenbeck4, Michael F. Leitzmann1 and Arthur Schatzkin1

1 Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD
2 Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD
3 Division of Cancer Prevention, National Cancer Institute, Bethesda, MD
4 AARP, Washington, DC

Correspondence to Dr. Yikyung Park, Division of Cancer Epidemiology and Genetics, 6120 Executive Boulevard, Bethesda, MD 20852 (e-mail: parkyik{at}mail.nih.gov).

Received for publication September 15, 2006. Accepted for publication January 24, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The authors examined the associations between fruit and vegetable intakes and risk of colorectal cancer in the NIH–AARP Diet and Health Study. Diet was assessed with a food frequency questionnaire at baseline. Relative risks and 95% confidence intervals were estimated by using the Cox proportional hazards model. During 5-year follow-up of 488,043 men and women aged 50–71 years, 2,972 incident colorectal cancer cases were identified. The respective 10th and 90th percentiles of total fruit and vegetable intake (servings/1,000 kcal per day) were 1.4 and 5.2 for men and 1.8 and 6.5 for women. Compared with that for the lowest quintile of vegetable intake, the multivariate relative risk for the highest quintile was 0.82 (95% confidence interval: 0.71, 0.94) for men and 1.12 (95% confidence interval: 0.90, 1.38) for women. Increased risk of colorectal cancer was observed for very low intake of total fruits and vegetables by men (multivariate relative risk for <1 vs. ≥2.0 servings/1,000 kcal per day = 1.26, 95% confidence interval: 1.03, 1.54). Among subgroups of vegetables, green leafy vegetables were associated with a lower risk of colorectal cancer for men (multivariate relative risk for the highest quintile vs. the lowest = 0.86, 95% confidence interval: 0.74, 0.99). Intake of fruits was not related to risk of colorectal cancer in men or women.

cohort studies; colorectal neoplasms; fruit; vegetables

Abbreviations: CI, confidence interval; FFQ, food frequency questionnaire; RR, relative risk


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Fruits and vegetables, rich in potentially anticarcinogenic constituents such as vitamin C, carotenoids, folate, dietary fiber, flavonoids, plant sterols, phenolic acids, and other phytochemicals, have been hypothesized to reduce risk of colorectal cancer (1). Animal and human feeding studies support the hypothesized protective effect of fruits and vegetables on cancer by providing plausible biologic mechanisms including reduction of oxidative damage to lipids and DNA, induction of phase I and II enzymes, and stimulation of DNA repair and apoptosis (2, 3).

However, evidence from observational studies of fruit and vegetable intakes and risk of colorectal cancer has been inconsistent. The 1997 World Cancer Research Fund report (4), reviewing 22 case-control studies and four prospective cohort studies, concluded that convincing evidence of decreased risk of colorectal cancer existed for vegetable intake but not for fruit intake. In contrast, prospective cohort studies published after the World Cancer Research Fund report have generally found no associations or statistically nonsignificant, weak inverse associations of fruit, vegetable, and total fruit and vegetable intakes with risk of colorectal cancer. A recent review of 11 prospective cohort studies published through early 2003 (5) found that fruit and vegetable intakes were not related to risk of colorectal cancer. Recently, the Women's Health Initiative Randomized Controlled Dietary Modification Trial (6) reported that daily intake of at least five servings of fruits and vegetables along with a low-fat diet did not lower risk of colorectal cancer in postmenopausal women.

Although aggregate results from prospective cohort studies suggest no significant association between fruit and vegetable intakes and risk of colorectal cancer, the results should be carefully interpreted because of limitations in those studies. First, in the meta-analysis (5, 7), considerable between-study heterogeneity was observed in the estimation of summarized relative risks but the sources of heterogeneity were not identified, suggesting that the summary of relative risks should be interpreted with caution. Second, null or statistically nonsignificant weak associations observed in most prospective cohort studies may be due to methodological limitations in observational studies. These cohort studies used a food frequency questionnaire (FFQ) that measured individuals' usual intakes of fruits and vegetables with error (8, 9). The measurement error in intake estimation combined with errors from covariates modeled in multivariate models may attenuate true relative risks and lead to obscured null associations between intakes of fruits and vegetables and risk of colorectal cancer. Other potential limitations in cohort studies include limited ranges of fruit and vegetable intakes and relatively small numbers of cases in some studies, resulting in a lack of statistical power to detect a true modest association.

Two recent studies have suggested a possible threshold effect of fruit and vegetables on colorectal cancer (10, 11); very low intakes of fruits and vegetables were associated with increased risk of colorectal cancer, but moderate to high intake did not lower this risk. Because our large cohort of men and women reported a wide range of fruit and vegetable intakes, we were able to investigate the associations of very low and high fruit and vegetable intakes with risk of colorectal cancer. In addition, we were able to examine specific subgroups of fruits and vegetables in relation to risk of colorectal cancer.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study population
The NIH–AAPR Diet and Health Study, a collaboration between the National Institutes of Health (NIH) and AARP (formerly the American Association of Retired Persons), has been described previously (12). Briefly, it was initiated in 1995–1996 with the mailing of a self-administrated questionnaire to 3.5 million AARP members aged 50–71 years from six US states (California, Florida, Louisiana, New Jersey, North Carolina, and Pennsylvania) and two metropolitan areas (Atlanta, Georgia; and Detroit, Michigan). The study was approved by the National Cancer Institute Institutional Review Board. Of 567,169 participants who returned questionnaires with satisfactory dietary data, we excluded individuals with duplicate questionnaires (n = 179), who requested to be withdrawn (n = 1), who had moved out of the study area or died at baseline (n = 589) or were diagnosed with colorectal cancer at baseline (n = 26), who indicated they were proxies for the intended respondents (n = 15,760), who had any self-reported prevalent cancer except nonmelanoma skin cancer at baseline (n = 51,207), who had self-reported end-stage renal disease at baseline (n = 997), and who had a cancer cause of death record and no cancer registry record (n = 1,660). In addition, we excluded individuals who reported extreme intakes (beyond two times the interquartile range of sex-specific Box-Cox log-transformed intake) of total energy (n = 4,401) and of fruits and vegetables (n = 4,306). After all exclusions, the analytic cohort consisted of 291,094 men and 196,949 women.

Cancer ascertainment
We identified incident cases of colorectal cancer that occurred during the follow-up through December 31, 2000. The cases were identified through probabilistic linkage with eight state cancer registry databases certified by the North American Association of Central Cancer Registries as being 95 percent complete within 2 years of cancer occurrence. Information on colorectal cancer tumor site and histology was also obtained through linkage with state cancer registries. The case ascertainment method used in the study detected 90 percent of all cancer cases in our cohort (13).

We defined incident cases of colorectal cancer to be both invasive and, if multiple cancers were diagnosed in the same participant, the first malignancy diagnosed during the follow-up period (International Classification of Diseases for Oncology, Third Edition, codes C180–C189, C260, C199, C209). Cases diagnosed with cancer in both the colon and rectum on the same day (n = 24) were considered cases for both sites. We further classified colorectal cancer by tumor site: proximal (codes C180–C184), distal (codes C185–C187), and rectum (codes C199, C209).

Dietary assessment
At baseline, dietary intakes were assessed with a self-administered, 124-item FFQ that was an earlier grid-based version of the Diet History Questionnaire developed at the National Cancer Institute (14). Participants were asked to report their usual frequency of intake and portion size over the last 12 months, using 10 predefined frequency categories ranging from "never" to "≥6 times per day" for beverages and from "never" to "≥2 times per day" for solid foods and three categories of portion size. The food items, portion sizes, and nutrient database were constructed on the basis of Subar et al.'s method (15) using the US Department of Agriculture's 1994–1996 Continuing Survey of Food Intake by Individuals (16). In addition, food groups and their serving sizes were defined by the Pyramid Servings Database corresponding to the 1994–1996 Continuing Survey of Food Intake by Individuals, which uses a recipe file to disaggregate food mixtures into their component ingredients and assigns them to food groups. A serving of fruit was defined as one medium-sized piece of fresh fruit, 1/2 cup of cut fruit, or 6 ounces of juice (1 cup = 237 ml, 1 ounce = 29.6 ml). A serving of vegetables was defined as 1 cup of leafy vegetables, 1/2 cup of other vegetables, or 6 ounces of juice (17). We excluded all potatoes except sweet potatoes from the vegetable group. The FFQ also asked about frequency and type of vitamin supplements used over the last 12 months. We also collected information on demographic characteristics, medical history, and lifestyle at baseline.

The FFQ used in the study was evaluated using two nonconsecutive 24-hour recalls among 1,953 AARP participants (F. E. Thompson, National Cancer Institute, personal communication, 2006). The energy-adjusted correlation coefficients between intake from the FFQ and from 24-hour recalls were 0.72 (men) and 0.61 (women) for total fruit and vegetable intake, 0.70 (men) and 0.65 (women) for vitamin C from foods, 0.64 (men) and 0.69 (women) for folate from foods, and 0.72 (men) and 0.66 (women) for dietary fiber from foods.

Statistical analysis
Relative risks and two-sided 95 percent confidence intervals were estimated with the Cox proportional hazards model (18) by using age as an underlying time metric and the SAS PROC PHREG procedure (19). Person-years of follow-up time were calculated from the date of the baseline questionnaire until the date of colorectal cancer diagnosis, death, move from the registry areas, or end of follow-up, whichever occurred first. The proportional hazards assumption was evaluated by modeling interaction terms of time and intakes of fruits and vegetables; no statistically significant interactions were found, confirming the assumption. The relative risks of colorectal cancer were estimated according to sex-specific quintiles of intake as well as categories defined by prespecified cutpoints of intake. The test for trend across categories of intake was performed by assigning participants the median value of their categories and entering it as a continuous term in a regression model.

We used a nutrient density model, in which daily intakes of fruits and vegetables were expressed as number of pyramid servings per 1,000 kcal of total energy. In multivariate models, we adjusted for education (less than high school, high school graduate, some college, and college graduate/postgraduate), physical activity (at work or home, including exercise, sports, and activities such as carrying heavy loads, which lasted 20 minutes or more and caused either increases in breathing or heart rate or working up a sweat; never/rarely, <3 times/month, 1–2, 3–4, and ≥5 times/week), smoking (never, ≤20 cigarettes/day in the past, >20 cigarettes/day in the past, currently ≤20 cigarettes/day, and currently >20 cigarettes/day), alcohol consumption (0, <5, 5–<15, 15–<30, and ≥30 g/day), and intakes of red meat (quintiles), dietary calcium (quintiles), and total energy (continuous). An indicator variable for missing responses for each covariate was created. We also examined interactions between fruit and vegetable intakes and each of the covariates listed above. In additional analyses, we further adjusted for race; body mass index; family history of colorectal cancer; colorectal cancer screening; use of multivitamins, nonsteroidal antiinflammatory drugs, and menopausal hormone therapy in women; and vitamin D intake, and we found that results were similar to those from the parsimonious models. We also analyzed data by using a standard multivariate model, in which intakes of fruits and vegetables were expressed as number of pyramid servings per day and total energy and other covariates were included in a model, and we observed results similar to those from the multivariate nutrient density method.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
During 2,121,664 person-years of follow-up (mean follow-up, 4.3 years), 2,972 incident cases of colorectal cancer (2,048 in men and 924 in women) were identified. The 10th and 90th percentiles of total fruit and vegetable intake were 1.4 and 5.2 servings/1,000 kcal per day (2.4 and 10.5 servings/day) for men and 1.8 and 6.5 servings/1,000 kcal per day (2.4 and 10.5 servings/day) for women. Compared with individuals with low total fruit and vegetable intake, individuals with higher total fruit and vegetable intake were more educated, more physically active, and less likely to smoke. In addition, individuals with high total fruit and vegetable intake consumed less alcohol and red meat but more dietary calcium than those with low total fruit and vegetable intake (table 1).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Characteristics of study participants by quintiles of total fruit and vegetable intake in the NIH–AARP Diet and Health Study, United States, 1995/1996–2000

 
In the age-adjusted model for men, risk of colorectal cancer was reduced by approximately 20–30 percent when the highest versus the lowest quintile of fruit, vegetable, and total fruit and vegetable intakes was compared (table 2). However, after adjustment for other colorectal cancer risk factors, the associations for fruit and for total fruit and vegetable intake were attenuated and no longer significant, whereas the association for vegetable intake remained statistically significant (multivariate relative risk (RR) for the highest vs. the lowest quintile = 0.82, 95 percent confidence interval (CI): 0.71, 0.94). Mutual adjustment for intakes of fruits and vegetables did not change the results (data not shown). When we further adjusted for dietary folate intake, the association for vegetable intake was slightly attenuated but remained statistically significant. Adjustment for dietary fiber intake did not change the result. For women, fruit, vegetable, and total fruit and vegetable intakes were not related to risk of colorectal cancer in both the age-adjusted and multivariate models. The associations for fruit, vegetable, and total fruit and vegetable intakes did not differ by physical activity, smoking status, alcohol consumption, multivitamin use, and intakes of red meat and dietary calcium for men and women (data not shown). When we stratified analyses by colorectal cancer screening, we did not find a significant difference in results: among participants who had colorectal cancer screening, the multivariate relative risk for the highest versus the lowest quintile of total fruit and vegetable intake was 0.92 (95 percent CI: 0.68, 1.24) for men and 1.37 (95 percent CI: 0.80, 2.35) for women. Among those who did not have colorectal cancer screening, the multivariate relative risk for the highest versus the lowest quintile was 0.82 (95 percent CI: 0.62, 1.10) for men and 1.20 (95 percent CI: 0.85, 1.68) for women.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Relative risks of colorectal cancer by quintiles* of fruit and vegetable intakes in the NIH–AARP Diet and Health Study, United States, 1995/1996–2000

 
Because we observed similar colorectal cancer relative risks for quintiles 2–5 compared with the lowest quintile of vegetable intake for men, we examined the risk of colorectal cancer by subcategories of low intake (table 3). Very low intakes of fruits and vegetables increased the risk of colorectal cancer for men. Compared with that for ≥1.00 servings/1,000 kcal per day of intake, the multivariate relative risk for the lowest category for men was 1.24 (95 percent CI: 1.03, 1.49 for <0.25 serving/1,000 kcal per day) for fruits and 1.20 (95 percent CI: 0.99, 1.44 for <0.50 serving/1,000 kcal per day) for vegetables. No increased risk of colorectal cancer in the lowest category of fruit and vegetable intakes was observed for women.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Relative risks of colorectal cancer by categories* of low intakes of fruits and vegetables in the NIH–AARP Diet and Health Study, United States, 1995/1996–2000

 
We also examined the associations of intakes of fruits, vegetables, and total fruits and vegetables with colorectal cancer risk by tumor site (table 4). For men, vegetable intake was related to a significantly lower risk of distal colon cancer (multivariate RR for the highest vs. the lowest quintile = 0.76, 95 percent CI: 0.59, 0.98). For women, fruit intake was associated with an increased risk of rectal cancer (multivariate RR for the highest vs. the lowest quintile = 1.59, 95 percent CI: 1.04, 2.44).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Multivariate relative risks* for quintiles{dagger} of fruit and vegetable intakes by tumor site, NIH–AARP Diet and Health Study, United States, 1995/1996–2000

 
Among subgroups of fruits and vegetables, green leafy vegetable intake was associated with lower risk of colorectal cancer for men (multivariate RR for the highest vs. the lowest quintile = 0.86, 95 percent CI: 0.74, 0.99; table 5). Intake of citrus fruit also showed a weak, nonsignificant inverse association with risk of colorectal cancer for men (multivariate RR for the highest vs. the lowest = 0.89, 95 percent CI: 0.77, 1.03). No specific subgroup of fruits and vegetables was related to risk of colorectal cancer for women.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Relative risks of colorectal cancer by quintile of intakes* of fruit and vegetable groups{dagger} in the NIH–AARP Diet and Health Study, United States, 1995/1996–2000

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
The hypothesis of a beneficial effect of fruit and vegetable intakes on risk of colorectal cancer has been supported by findings from case-control studies. A recent review of nine case-control studies (5) summarized that colorectal cancer risk was lowered by 13 percent (odds ratio = 0.87, 95 percent CI: 0.78, 0.97) and 37 percent (odds ratio = 0.63, 95 percent CI: 0.56, 0.70) when the highest versus the lowest category of fruit and vegetable intakes, respectively, were compared. Because of recall and selection bias and the high likelihood of publication of case-control studies with significant results, the validity of these findings from case-control studies has been questioned.

In contrast to case-control studies, prospective cohort studies conducted in the United States (10, 20–25), Europe (11, 26–28), and Japan (29, 30) have reported inconsistent results. Some cohort studies (10, 11, 20, 27, 28) observed an approximately 25–40 percent reduction in colorectal cancer risk for those with high intakes of fruits, vegetables, and total fruits and vegetables, whereas other studies found no associations (21–26, 29, 30). However, some of the cohort studies found that the colorectal cancer risk for fruit and vegetable intakes decreased markedly in the second lowest category, and the reduced risk persisted through the highest category (10, 11, 23, 25, 27). Consistent with those studies, our study found that the multivariate relative risk for the second quintile of vegetable intake was similar through the highest quintile compared with the lowest quintile of men. Furthermore, we observed increased risk of colorectal cancer for men with very low intakes of vegetables and total fruits and vegetables compared with those with low to high intakes. The Cancer Prevention Study II Nutrition Cohort Study (10) also found that very low intake of vegetables significantly increased risk of colorectal cancer for men (multivariate RR for <0.80 vs. ≥1.30 servings/day of vegetables = 1.79, 95 percent CI: 1.22, 2.61). In addition, the Swedish Mammography Screening Cohort Study (11) found that low intake of total fruits and vegetables increased risk of colorectal cancer (multivariate RR for <1.5 vs. >2.5 servings/day = 1.65, 95 percent CI: 1.23, 2.20). This nonlinear relation between fruit and vegetable intakes and risk of colorectal cancer suggests a threshold effect of fruit and vegetable intakes and the importance of consuming a certain minimum amount of fruits and vegetables every day, but high intakes of fruits and vegetables may have little additional benefit in terms of colorectal cancer risk.

Our study, consistent with findings from the Cancer Prevention Study II Nutrition Cohort Study (10), observed that vegetable intake was related to lower risk of colorectal cancer for men but not for women. This gender difference may be partially explained by differential reporting errors. Studies have found that healthful attitudes, beliefs, and dietary habits were more strongly correlated with vegetable intake among women than among men (31, 32) and that women overreported foods perceived as healthy (33, 34). Misreporting of fruit and vegetable intakes by women in our study would have led to exposure misclassification, resulting in an attenuated association.

Although not statistically significant, we found that the effect of vegetable intake in men was somewhat stronger in the distal (RR = 0.76, 95 percent CI: 0.59, 0.98) than in the proximal (RR = 0.90, 95 percent CI: 0.72, 1.14) colon when we compared the highest versus the lowest quintile of intake. Similar results were found in the Netherlands Cohort Study on Diet and Cancer (27): when the highest versus the lowest quintile of vegetable intake was compared, the multivariate relative risk of distal colon cancer was 0.76 (95 percent CI: 0.27, 1.30) for men and 0.64 (95 percent CI: 0.36, 1.17) for women, whereas the multivariate relative risk of proximal colon cancer was 1.03 (95 percent CI: 0.59, 1.81) for men and 0.99 (95 percent CI: 0.57, 1.72) for women. On the other hand, the Swedish Mammography Screening Cohort Study (11) found that vegetable intake was more strongly associated with risk of proximal colon cancer (multivariate RR = 0.72, 95 percent CI: 0.44, 1.20 for the highest vs. the lowest quintile) than that of distal colon cancer (multivariate RR = 1.13, 95 percent CI: 0.66, 1.94 for the highest vs. the lowest quintile).

A positive association between fruit intake and risk of rectal cancer in women was an unexpected finding. In the Netherlands Cohort Study on Diet and Cancer (27), vegetable intake by women was associated with a statistically nonsignificant increased risk of rectal cancer (multivariate RR = 1.78, 95 percent CI: 0.94, 3.38 for the highest vs. the lowest quintile). Although colon and rectal cancer share many etiologic factors, it is plausible that risk factors differ by tumor sites and that certain risk factors have not yet been established for less-studied rectal cancer. On the other hand, our finding for fruit intake and rectal cancer in women may simply be due to residual confounding or chance.

Some studies found that certain subgroups of fruits and vegetables were associated with colorectal cancer risk. In our study, green leafy vegetable intake was associated with a lower risk of colorectal cancer for men. In the Netherlands Cohort Study on Diet and Cancer (27), intakes of cooked leafy vegetables and cruciferous vegetables were inversely associated with risk of colon cancer for women: when the highest versus the lowest quintiles of intake were compared, the multivariate relative risk for cooked leafy vegetables was 0.62 (95 percent CI: 0.40, 0.96) and for cruciferous vegetables was 0.51 (95 percent CI: 0.33, 0.80). The Cancer Prevention Study II Nutrition Cohort Study (10) found that fruits and vegetables high in ß-carotene lowered risk of colorectal cancer for men (multivariate RR = 0.60, 95 percent CI: 0.40, 0.89 comparing the highest vs. the lowest quintile).

Mean fruit and vegetable intakes in our study were higher than those in the US population (35) and other cohort studies (10, 11, 21, 23) but were comparable to those in another cohort study conducted in the United States (25). Moreover, our study had a wider range of fruit and vegetable intakes than that in other studies: more than 10 percent of our study participants reported fewer than 2.5 servings/day of total fruit and vegetables, and 10 percent of participants indicated more than 10 servings/day. This wide intake range enabled us to examine the risk of colorectal cancer at fairly low and high intakes and thereby increased the likelihood of detecting a true association with colorectal cancer.

We assessed participants' fruit and vegetable intakes with the FFQ. It has been debated whether measurement errors in the FFQ could cause significant distortion of observed associations (8, 9). Studies have shown that fruit and vegetable intakes from the FFQ correlated reasonably well with consumption from other self-reported reference instruments such as food records and multiple 24-hour recalls, and biomarkers such as plasma vitamin C concentration (8, 36). Such correlations, however, do not guarantee that the FFQ is accurate enough to detect a moderate, but important diet-cancer relation. We therefore cannot rule out the possibility that true associations of fruits and vegetables with colorectal cancer have been substantially attenuated.

Dietary changes made during the follow-up period could not be addressed in our study because dietary intakes were measured at baseline only. Nonetheless, Goldbohm et al. (37) reported that fruit and vegetable intakes were consistent through five annual assessments using a FFQ, and another study with several repeated assessments of fruit and vegetable intakes (23) found that results using baseline-only data yielded the same results as those using repeated measurement. In addition, considering the relatively short follow-up period of our study, it is unlikely that a significant number of participants changed their intakes of fruits and vegetables. On the other hand, if fruits and vegetables exert an anticarcinogenic effect only after many years, that is, in the context of a long latency period for colorectal cancer, it is possible that the duration of follow-up in our study was insufficient to detect such an effect.

One concern related to dietary change is food fortification with folate since our study began. Folate, abundant in green leafy vegetables, has been associated with lower risk of colorectal cancer (38). If the beneficial effect of fruits and vegetables was largely due to folate, and folate intake increased as a result of food fortification while intake of fruits and vegetables remained fairly constant during follow-up, the increased folate intake might have diminished the fruits and vegetables–colorectal cancer association. In addition, high prevalence of multivitamin supplement use among participants in our study may obscure the relation of fruits and vegetables to risk of colorectal cancer because multivitamin supplements have a high quantity of potentially anticarcinogenic nutrients also abundant in fruits and vegetables. However, when we examined the associations between intakes of fruits and vegetables and risk of colorectal cancer among multivitamin supplement users, we observed results similar to those for multivitamin supplement nonusers.

In this large, prospective cohort study with 2,972 incident colorectal cancer cases and extensive information on diet and other colorectal cancer risk factors, we observed that vegetable intake was associated with a lower risk of colorectal cancer for men but not for women. The association was stronger among individuals with very low intakes of fruits and vegetables, suggesting a certain minimum amount of daily fruit and vegetable consumption to avoid increased risk of colorectal cancer. Among subgroups of vegetables, green leafy vegetable intake was inversely associated with risk of colorectal cancer for men.


   ACKNOWLEDGMENTS
 
This research was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute.

Cancer incidence data from the Atlanta metropolitan area were collected by the Georgia Center for Cancer Statistics, Department of Epidemiology, Rollins School of Public Health, Emory University. Cancer incidence data from California were collected by the California Department of Health Services, Cancer Surveillance Section. Cancer incidence data from the Detroit metropolitan area were collected by the Michigan Cancer Surveillance Program, Community Health Administration, State of Michigan. The Florida cancer incidence data used in this report were collected by the Florida Cancer Data System under contract to the Department of Health (DOH). The views expressed herein are solely those of the authors and do not necessarily reflect those of the contractor or DOH. Cancer incidence data from Louisiana were collected by the Louisiana Tumor Registry, Louisiana State University Medical Center in New Orleans. Cancer incidence data from New Jersey were collected by the New Jersey State Cancer Registry, Cancer Epidemiology Services, New Jersey State Department of Health and Senior Services. Cancer incidence data from North Carolina were collected by the North Carolina Central Cancer Registry. Cancer incidence data from Pennsylvania were supplied by the Division of Health Statistics and Research, Pennsylvania Department of Health, Harrisburg, Pennsylvania. The Pennsylvania Department of Health specifically disclaims responsibility for any analyses, interpretations, or conclusions.

Conflict of interest: none specified.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Steinmetz KA, Potter JD. Vegetables, fruit, and cancer. II. Mechanisms. Cancer Causes Control (1991) 2:427–42.[CrossRef][Web of Science][Medline]
  2. Dragsted LO, Strube M, Larsen JC. Cancer-protective factors in fruits and vegetables: biochemical and biological background. Pharmacol Toxicol (1993) 72(suppl 1):116–35.[Web of Science][Medline]
  3. Dragsted LO, Krath B, Ravn-Haren G, et al. Biological effects of fruit and vegetables. Proc Nutr Soc (2006) 65:61–7.[CrossRef][Web of Science][Medline]
  4. World Cancer Research Fund, American Institute for Cancer Research Expert Panel. Food, nutrition and the prevention of cancer: a global perspective (1997) Washington, DC: American Institute for Cancer Research.
  5. Fruit and vegetables. IARC handbooks of cancer prevention. Vol 8 (2003) Lyon, France: International Agency for Research on Cancer.
  6. Beresford SA, Johnson KC, Ritenbaugh C, et al. Low-fat dietary pattern and risk of colorectal cancer: the Women's Health Initiative Randomized Controlled Dietary Modification Trial. JAMA (2006) 295:643–54.[Abstract/Free Full Text]
  7. Riboli E, Norat T. Epidemiologic evidence of the protective effect of fruit and vegetables on cancer risk. Am J Clin Nutr (2003) 78:559S–69S.[Abstract/Free Full Text]
  8. Willett W. Nutritional epidemiology (1998) New York, NY: Oxford University Press.
  9. Kipnis V, Midthune D, Freedman LS, et al. Empirical evidence of correlated biases in dietary assessment instruments and its implications. Am J Epidemiol (2001) 153:394–403.[Abstract/Free Full Text]
  10. McCullough ML, Robertson AS, Chao A, et al. A prospective study of whole grains, fruits, vegetables and colon cancer risk. Cancer Causes Control (2003) 14:959–70.[CrossRef][Web of Science][Medline]
  11. Terry P, Giovannucci E, Michels KB, et al. Fruit, vegetables, dietary fiber, and risk of colorectal cancer. J Natl Cancer Inst (2001) 93:525–33.[Abstract/Free Full Text]
  12. Schatzkin A, Subar AF, Thompson FE, et al. Design and serendipity in establishing a large cohort with wide dietary intake distributions: the National Institutes of Health-American Association of Retired Persons Diet and Health Study. Am J Epidemiol (2001) 154:1119–25.[Abstract/Free Full Text]
  13. Michaud DMDHS, Leitzmann M, Harlan L, et al. Comparison of cancer registry case ascertainment with SEER estimates and self-reporting in a subset of the NIH-AARP Diet and Health Study. J Registry Manage (2005) 32:70–5.
  14. National Cancer Institute. Diet History Questionnaire. (http://www.riskfactor.cancer.gov/DHQ).
  15. Subar AF, Midthune D, Kulldorff M, et al. Evaluation of alternative approaches to assign nutrient values to food groups in food frequency questionnaires. Am J Epidemiol (2000) 152:279–86.[Abstract/Free Full Text]
  16. Tippett KS, Cypel YS. Design and operation: the continuing survey of food intakes by individuals and diet and health knowledge survey, 1994 –96. Continuing survey of food intakes by individuals, nationwide food surveys (1997) Washington, DC: US Department of Agriculture, Agricultural Research Service. (NFS report no. 96-1).
  17. US Department of Agriculture, US Department of Health and Human Services. The food guide pyramid (1992) Washington, DC: US Government Printing Office. (Home and garden bulletin no. 252).
  18. Cox DR. Regression models and life tables (with discussion). J R Stat Soc (B) (1972) 34:187–220.
  19. SAS II. SAS/STAT user's guide, version 8 (1999) Cary, NC: SAS Institute, Inc.
  20. Shibata A, Paganini-Hill A, Ross RK, et al. Intake of vegetables, fruits, beta-carotene, vitamin C and vitamin supplements and cancer incidence among the elderly: a prospective study. Br J Cancer (1992) 66:673–9.[Web of Science][Medline]
  21. Steinmetz KA, Kushi LH, Bostick RM, et al. Vegetables, fruit, and colon cancer in the Iowa Women's Health Study. Am J Epidemiol (1994) 139:1–15.[Abstract/Free Full Text]
  22. Kato I, Akhmedkhanov A, Koenig K, et al. Prospective study of diet and female colorectal cancer: The New York University Women's Health Study. Nutr Cancer (1997) 28:276–81.[Web of Science][Medline]
  23. Michels KB, Edward G, Joshipura KJ, et al. Prospective study of fruit and vegetable consumption and incidence of colon and rectal cancers. J Natl Cancer Inst (2000) 92:1740–52.[Abstract/Free Full Text]
  24. Flood A, Velie EM, Chaterjee N, et al. Fruit and vegetable intakes and the risk of colorectal cancer in the Breast Cancer Detection Demonstration Project follow-up cohort. Am J Clin Nutr (2002) 75:936–43.[Abstract/Free Full Text]
  25. Lin J, Zhang SM, Cook NR, et al. Dietary intakes of fruit, vegetables, and fiber, and risk of colorectal cancer in a prospective cohort of women (United States). Cancer Causes Control (2005) 16:225–33.[CrossRef][Web of Science][Medline]
  26. Pietinen P, Malila N, Virtanen M, et al. Diet and risk of colorectal cancer in a cohort of Finnish men. Cancer Causes Control (1999) 10:387–96.[CrossRef][Web of Science][Medline]
  27. Voorrips LE, Goldbohm RA, van Poppel G, et al. Vegetable and fruit consumption and risks of colon and rectal cancer in a prospective cohort study: The Netherlands Cohort Study on Diet and Cancer. Am J Epidemiol (2000) 152:1081–92.[Abstract/Free Full Text]
  28. Sanjoaquin MA, Appleby PN, Thorogood M, et al. Nutrition, lifestyle and colorectal cancer incidence: a prospective investigation of 10998 vegetarians and non-vegetarians in the United Kingdom. Br J Cancer (2004) 90:118–21.[CrossRef][Web of Science][Medline]
  29. Tsubono Y, Otani T, Kobayashi M, et al. No association between fruit or vegetable consumption and the risk of colorectal cancer in Japan. Br J Cancer (2005) 92:1782–4.[CrossRef][Web of Science][Medline]
  30. Sato Y, Tsubono Y, Nakaya N, et al. Fruit and vegetable consumption and risk of colorectal cancer in Japan: The Miyagi Cohort Study. Public Health Nutr (2005) 8:309–14.[CrossRef][Web of Science][Medline]
  31. Satia JA, Kristal AR, Patterson RE, et al. Psychosocial factors and dietary habits associated with vegetable consumption. Nutrition (2002) 18:247–54.[CrossRef][Web of Science][Medline]
  32. Trudeau E, Kristal AR, Li S, et al. Demographic and psychosocial predictors of fruit and vegetable intakes differ: implications for dietary interventions. J Am Diet Assoc (1998) 98:1412–17.[CrossRef][Web of Science][Medline]
  33. Subar AF, Kipnis V, Troiano RP, et al. Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study. Am J Epidemiol (2003) 158:1–13.[Abstract/Free Full Text]
  34. Bogers RP, Brug J, van Assema P, et al. Explaining fruit and vegetable consumption: the theory of planned behaviour and misconception of personal intake levels. Appetite (2004) 42:157–66.[CrossRef][Web of Science][Medline]
  35. Guenther PM, Dodd KW, Reedy J, et al. Most Americans eat much less than recommended amounts of fruits and vegetables. J Am Diet Assoc (2006) 106:1371–9.[CrossRef][Web of Science][Medline]
  36. Michels KB, Welch AA, Luben R, et al. Measurement of fruit and vegetable consumption with diet questionnaires and implications for analyses and interpretation. Am J Epidemiol (2005) 161:987–94.[Abstract/Free Full Text]
  37. Goldbohm RA, van't Veer P, van den Brandt PA, et al. Reproducibility of a food frequency questionnaire and stability of dietary habits determined from five annually repeated measurements. Eur J Clin Nutr (1995) 49:420–9.[Web of Science][Medline]
  38. Sanjoaquin MA, Allen N, Couto E, et al. Folate intake and colorectal cancer risk: a meta-analytical approach. Int J Cancer (2005) 113:825–8.[CrossRef][Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Nutr.Home page
S. A. Talegawkar, G. Beretta, K.-J. Yeum, E. J. Johnson, T. C. Carithers, H. A. Taylor Jr, R. M. Russell, and K. L. Tucker
Total Antioxidant Performance Is Associated with Diet and Serum Antioxidants in Participants of the Diet and Physical Activity Substudy of the Jackson Heart Study
J. Nutr., October 1, 2009; 139(10): 1964 - 1971.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
Y. Park, A. N. Freedman, M. H. Gail, D. Pee, A. Hollenbeck, A. Schatzkin, and R. M. Pfeiffer
Validation of a Colorectal Cancer Risk Prediction Model Among White Patients Age 50 Years and Older
J. Clin. Oncol., February 10, 2009; 27(5): 694 - 698.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
A. E Millen, A. F Subar, B. I Graubard, U. Peters, R. B Hayes, J. L Weissfeld, L. A Yokochi, R. G Ziegler, and for the Prostate, Lung, Colorectal, and Ovarian (P
Fruit and vegetable intake and prevalence of colorectal adenoma in a cancer screening trial
Am. J. Clinical Nutrition, December 1, 2007; 86(6): 1754 - 1764.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
166/2/170    most recent
kwm067v2
kwm067v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Park, Y.
Right arrow Articles by Schatzkin, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Park, Y.
Right arrow Articles by Schatzkin, A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?