Skip Navigation


American Journal of Epidemiology Advance Access originally published online on February 12, 2008
American Journal of Epidemiology 2008 167(5):586-597; doi:10.1093/aje/kwm361
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
167/5/586    most recent
kwm361v1
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 (15)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Stolzenberg-Solomon, R. Z.
Right arrow Articles by Silverman, D. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stolzenberg-Solomon, R. Z.
Right arrow Articles by Silverman, D. T.
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 2008.

ORIGINAL CONTRIBUTIONS

Adiposity, Physical Activity, and Pancreatic Cancer in the National Institutes of Health–AARP Diet and Health Cohort

Rachael Z. Stolzenberg-Solomon1, Kenneth Adams1, Michael Leitzmann1, Catherine Schairer2, Dominique S. Michaud3, Albert Hollenbeck4, Arthur Schatzkin1 and Debra T. Silverman5

1 Nutritional Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Rockville, MD
2 Biostatistics Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Rockville, MD
3 Department of Epidemiology, Harvard School of Public Health, Boston, MA
4 Environmental Analysis Department, AARP, Washington, DC
5 Occupational Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Rockville, MD

Correspondence to Dr. Rachael Stolzenberg-Solomon, 6120 Executive Boulevard, Suite 320, Rockville, MD 20852 (e-mail: rs221z{at}nih.gov).

Received for publication March 23, 2007. Accepted for publication November 5, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Obesity and lack of physical activity have been inconsistently associated with pancreatic cancer. Using data from a self-administered baseline questionnaire (1995–1996), the authors investigated the association between adiposity and physical activity and pancreatic cancer in 495,035 participants of the National Institutes of Health–AARP Diet and Health Study who were aged 50–71 years. To avoid the influence of subclinical disease, follow-up time started 1 year after baseline, and subjects with a body mass index (BMI) of <18.5 kg/m2 were excluded. A subcohort (n = 302,060) completed a second questionnaire with information about physical activity and waist and hip circumference. During follow-up though 2000, 654 pancreatic cancer cases were identified. The authors used Cox proportional hazard models to generate adjusted hazard ratios and 95% confidence intervals. Compared with those with a BMI of 18.5–<25, those with a BMI of ≥35 had a 45% greater pancreatic cancer risk (95% confidence interval (CI): 1.04, 2.02; ptrend = 0.02). Significant positive associations for BMI were observed among nonsmokers (for BMI ≥35: hazard ratio = 1.70, 95% CI: 1.14, 2.53; ptrend = 0.004) but not recent smokers (pinteraction = 0.08). Waist circumference was positively associated with pancreatic cancer (fourth vs. first quartile: hazard ratio = 2.53, 95% CI: 1.13, 5.65; ptrend = 0.04) in women but not men. The authors observed no association with physical activity. Their results suggest a positive association between adiposity and pancreatic cancer.

adenocarcinoma; adiposity; body mass index; cohort studies; exercise; pancreatic neoplasms; smoking; waist-hip ratio

Abbreviations: BMI, body mass index; CI, confidence interval; HR, hazard ratio; MET, metabolic equivalent; NIH, National Institutes of Health


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Pancreatic cancer ranks fourth for cancer mortality for men and women, respectively, in the United States and has a 5-year survival rate of 4.3 percent (1). In addition to cigarette smoking (2), diabetes mellitus and glucose intolerance are among the few consistent and potentially modifiable risk factors for pancreatic cancer (3). One biologically plausible mechanism whereby type 2 diabetes may be related to pancreatic carcinogenesis is through the growth-promoting effects of insulin (4). This hypothesized mechanism is supported indirectly by positive associations between obesity (5–11) and lack of physical activity (12, 13) and pancreatic cancer in some but not all epidemiologic studies (6, 13–24). Obesity, particularly central or abdominal adiposity, may lead to metabolic changes including higher glucose and insulin concentrations, insulin resistance, and type 2 diabetes (25).

The evidence from cohort and case-control studies more consistently supports obesity as a risk factor for pancreatic cancer (5–7, 9–11, 24) than physical activity (12–14, 18–24, 26). Recent meta-analyses reported weak but significant positive associations between increasing body mass index (BMI) and pancreatic cancer, with associations being slightly stronger for cohort than case-control studies and men than women (10, 19). Five cohort studies have examined some aspect of abdominal obesity (e.g., waist circumference or waist/hip ratio, central torso weight gain) and pancreatic cancer (14, 15, 18, 19, 23) with four showing positive associations (14, 15, 19, 23). Although most studies examining physical activity and pancreatic cancer have not shown significant associations (7, 12–14, 18–24, 26), three studies have shown significant inverse associations with greater physical activity in substrata of subjects (12, 13, 22). None of the pancreatic cancer studies that evaluated interactions of BMI and physical activity reported statistically significant interactions (7, 12, 14, 15, 19, 22, 24); however, one study showed that overweight and obese subjects with low physical activity had greater pancreatic cancer risks compared with those with more physical activity and normal weight (12). The lack of consistent associations between these factors and pancreatic cancer may reflect methodological difficulties including reverse causation, recall, and proxy reporting biases from data collected in case-control studies and the small number of incident cases in many cohort studies, resulting in limited power to observe associations. Age-related variation in lean mass (muscle and skeletal mass) may reduce the validity of body mass index (weight (kg)/height (m)2) as a measure of body fatness, and waist circumference may more accurately reflect body fatness in older persons (27). The positive relation between BMI and pancreatic cancer risk may be stronger in nonsmokers (15, 26) than in smokers (22). However, few studies have considered effect modification of BMI by smoking status (7, 14, 15, 19, 24, 26).

We conducted an analysis in a large cohort, the National Institutes of Health (NIH)–AARP Diet and Health Study, to examine the association between adiposity, as reflected by BMI, waist circumference, and waist/hip ratio and pancreatic cancer. We also examined physical activity as a risk factor for pancreatic cancer. The NIH–AARP cohort has a larger number of incident pancreatic cancer cases compared with most previous cohort studies (10), which allowed us to examine sex-specific associations and interactions by smoking status.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study population
The NIH–AARP Diet and Health Study is a large prospective study of AARP members established in 1995–1996 (28). Self-administered questionnaires eliciting information on demographic characteristics, dietary intake, and numerous health-related behaviors were mailed to AARP members aged 50 and 71 years, who resided in six US states (California, Florida, Louisiana, New Jersey, North Carolina, and Pennsylvania) and two metropolitan areas (Atlanta, Georgia, and Detroit, Michigan). The questionnaire was returned by 617,119 members, and 567,169 subjects completed the questionnaire satisfactorily (28). The study was approved by the National Cancer Institute Special Studies Institutional Review Board, and informed consent was obtained from all participants.

We excluded subjects with duplicate representation (n = 179), who moved out of the eight areas included in our study before returning the baseline questionnaire (n = 321), who died before study entry (n = 261), or who withdrew (n = 1). We also excluded subjects who had questionnaires completed by proxy respondents (n = 15,760), prevalent cancers as determined by the cancer registry data (n = 8,552), extreme energy intake outside the normal distribution of the cohort by sex (n = 4,793), missing height or weight (n = 12,927), and missing smoking data (n = 16,810). We further excluded subjects who had a BMI of <18.5 kg/m2 (n = 5,601) and those censored during the first year of follow-up (n = 6,929; number of pancreatic cancer cases = 127) to avoid reverse causation bias. Our final baseline analytical cohort consisted of 495,035 individuals (293,562 men, 201,473 women).

Six months after the baseline questionnaire was sent, baseline respondents were sent a second questionnaire that queried information about physical activity during the past 10 years and measurement of waist and hip (28). Among subjects in our baseline analytical cohort, 302,060 subjects completed the second questionnaire. We included all subjects with complete physical activity (n = 294,609) data or who provided waist, hip, or both waist and hip measurements (n = 228,080, 217,930, and 217,119, respectively); therefore, the number of subjects in each subcohort analysis varies slightly.

Cohort follow-up and case ascertainment
Cancer cases were identified by linking cohort members to state cancer registries and to the US National Death Index between 1995 through 2000 and are estimated to be about 90 percent complete (29). The vital status of cohort participants was also ascertained by linkage to the Social Security Administration Death Master File. For these analyses, we included incident primary adenocarcinoma of the exocrine pancreas (International Classification of Diseases for Oncology, Third Edition (ICD-O-3), codes C250–C259). Our case definition excluded endocrine pancreatic tumors (histology types 8150, 8151, 8153, 8155, and 8240), because the etiology of these cancers is thought to be different. A total of 654 incident pancreatic cancer cases (429 men and 225 women) were identified among the baseline analytical cohort and included in the BMI analyses. In the subcohort, 399 (252 men and 147 women) cases had complete physical activity data, and 312 (212 men and 100 women) cases or 290 (191 men and 99 women) cases had complete waist or hip data, respectively.

Assessment of height, weight, diet, and other risk factors
At baseline, study subjects completed a self-administered questionnaire that included questions on current weight and height, diet, demographic factors, medical history, and other health-related behaviors (28). The dietary questionnaire assessed the usual frequency of consumption and portion size of 124 food items and included 21 questions on low-fat, high-fiber foods and food preparation over the previous 12 months. Details of the questionnaire have been published elsewhere (30–32). BMI was calculated from self-reported height and weight (kg/m2).

In the second questionnaire, subjects were queried "how often (never, rarely, weekly: <1, 1–3, 4–7, or >7 hours) did you participate in light or moderate and vigorous activities" during the past 10 years. Examples were provided for "light" (i.e., bowling, golf riding in a cart, table tennis, slow walking/slow dancing, light calisthenics, light gardening, fishing, horseshoes/croquet, and light housework) and "moderate/vigorous" (i.e., tennis, golf walking, biking, swimming, heavy gardening, weight lifting, basketball/baseball, football/soccer, cheerleading/drill team, handball/racquetball) activities. Validation studies of similar physical activity questionnaires suggest that the reliability and validity of our questionnaire are comparable to those used in other studies (33). With detailed instructions, subjects were asked to record their waist and hip measurements.

We developed a physical activity score using literature-based relative metabolic equivalent (MET) values for moderate and vigorous activities (34). MET values were calculated by averaging the 11 examples of light (3.5 METs) and 13 examples of moderate/vigorous (6.0 METs) activities described in the physical activity question, respectively, and multiplying these two averages by the number of hours that individuals reported participating in each type of activity (never = 0; rarely = 0.33; weekly: <1 = 0.67, 1–3 = 2, 4–7 = 5.5, or >7 hours = 8 hours, respectively).

Statistical analysis
Within and across the BMI categories by sex in table 1, we calculated means for the continuous-population-characteristic variables and frequency proportions for dichotomous characteristics. To avoid the influence of subclinical disease, we started follow-up time 1 year after the date of receipt of the baseline questionnaire through December 2000, diagnosis of pancreatic cancer, or death. Cox proportional hazard models, with age as the underlying time metric, were used to generate hazard ratios and 95 percent confidence intervals. Entry time was defined as the subjects' age in days at the return of the questionnaire plus 1 year. Exit time was defined as the subjects' age in days at cancer diagnosis or censoring.


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

  		TABLE 1. Selected age-adjusted characteristics of the National Institutes of Health–AARP Diet and Health Study cohort (n = 495,035) by body mass index category means and proportions, United States, 1995–2000*

 
Body mass index was categorized to be consistent with the World Health Organization obesity classifications or 18.5–<25 (normal), 25–<30 (overweight), 30–<35 (moderate obesity), and ≥35 kg/m2 (severe obesity) (35). Waist and hip measurements (centimeters) and the waist/hip ratio were categorized on the basis of the cohort sex-specific distribution with quintiles and quartiles for men and women, respectively. We also calculated the continuous hazard ratios for BMI for our results. Metabolic equivalent units were categorized on the basis of sex-specific and cohort distributions. Trend tests were calculated using a score variable for each quantile or category. The dietary variables were energy adjusted by use of the density method. We created a compound smoking variable to control for confounding, on the basis of risk estimates from our data, that integrated never, former (including time since having quit smoking), and current smoking, as well as smoking dose (never, quit ≥10 years ago, quit 5–9 years ago, quit 1–4 years ago, quit <1 year ago or current and smoked ≤20 or >20 cigarettes/day).

We evaluated confounding by individually entering potential confounding variables into the model. Variables remained in the model if they changed the risk estimate ≥10 percent or were putative risk factors for pancreatic cancer. The variables investigated and included in the multivariable models for BMI and abdominal obesity were smoking; race (Caucasian, Black, Hispanic, Asian, Pacific Islander or American Indian/Alaskan Native, and missing); total caloric intake (kcal/day); and total fat intake (g/1,000 kcal/day), self-reported diabetes, and sex (sex combined models). Self-reported diabetes (yes/no), a putative pancreatic cancer risk factor, is potentially on the causal pathway between BMI and pancreatic cancer; therefore, we show the BMI models with and without the inclusion of diabetes. For waist, hip, and the waist/hip ratio, we additionally adjusted for height to control for measurement error and for BMI to evaluate associations independent of fat-free mass. The physical activity models were adjusted for smoking, race, BMI, diabetes, and sex (sex-combined models). Physical activity, alcohol consumption, and folate and protein intakes did not confound any association and were not included in our final regression models.

We evaluated interactions by smoking status, diabetes, physical activity, and BMI by including cross-product terms in multivariable models using the trend score variables and joint analysis. To provide an adequate number of cases in strata to evaluate smoking interactions, we combined never smokers and former smokers who had quit ≥10 years ago as pancreatic cancer risk is similar in these two groups (2, 36–45). Previous studies have demonstrated a rapid reduction in pancreatic cancer risk among former smokers with risks approaching that of never smokers within 5–15 years after smoking cessation (36, 37). Effect modification by length of follow-up was tested using a time-dependent interaction term (<2 years and ≥2 years) and analyses stratified by follow-up time. All statistical analyses were performed with Statistical Analysis Systems (version 8.2; SAS Institute, Inc., Cary, North Carolina) software, and the p values for statistical tests were two tailed.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
For both men and women (table 1), weight, energy and total and saturated fat intakes, and the proportions of subjects who were former smokers, are African American, or had a history of diabetes mellitus were directly related to greater BMI. In contrast, age, height, alcohol consumption, and the proportions of subjects who are current smokers, have a college or postgraduate education, report multivitamin use, or engage in heavy leisure or work physical activity ≥5 times per week were inversely associated with BMI. The proportion of subjects who were ever smokers was positively associated with BMI among men; however, it was inversely associated with BMI among women.

Compared with normal weight, severe obesity was significantly and positively associated with pancreatic cancer in men but not in women (table 2). Among men and women combined, overweight and severe obesity (ptrend = 0.02) were significantly associated with increased pancreatic cancer risk. Weak, positive, borderline-significant associations were observed per unit increase in BMI, and the hazard ratios per 5-unit BMI increase were 1.08 (95 percent confidence interval (CI): 0.97, 1.19), 1.08 (95 percent CI: 1.00, 1.16), and 1.08 (95 percent CI: 1.01, 1.15) for men, women, and men and women combined, respectively (data not shown). These associations were slightly attenuated (7–8 percent risk reduction) with adjustment for history of diabetes. There were no significant interactions of the BMI associations by sex, history of diabetes, or follow-up time (pinteraction > 0.40). Height was not associated with pancreatic cancer among men or women in any models (smoking-, BMI-, race-adjusted model: men ≥183 cm compared with <170 cm: hazard ratio (HR) = 1.14, 95 percent CI: 0.82, 1.60; ptrend = 0.46; women ≥168 cm compared with <157 cm: HR = 0.77, 95 percent CI: 0.52, 1.12; ptrend = 0.24; data not shown).


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

  		TABLE 2. Age- and multivariable-adjusted hazard ratios and 95% confidence intervals of baseline body mass index and according to smoking history in the National Institutes of Health–AARP Diet and Health Study cohort, United States, 1995–2000*

 
The association between BMI and pancreatic cancer was not significantly different according to smoking status among men (table 2); however, in women there were significant positive risks and a trend with increasing BMI among nonsmokers but not among recent smokers (pinteraction = 0.02). A pattern similar to that observed in women was observed in sex-combined models. Among never smokers, positive associations with increasing BMI were observed only in women (refer to supporting documentation). In joint analyses of BMI and smoking that used normal BMI among nonsmokers as the referent category, pancreatic cancer risks were increased for all BMI categories within recent smokers (figure 1).


Figure 1
View larger version (24K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIGURE 1. Hazard ratios for body mass index and pancreatic cancer by smoking status among men and woman combined, United States, 1995–2000. pinteraction = 0.09. Compared with the referent, all hazard ratios were statistically significant; the confidence interval does not include 1.00. "Nonsmoker" refers to a never/former smoker who quit ≥10 years ago, and "recent smoker" refers to a current/former smoker who quit <10 years ago. All models should be considered adjusted for age because age is the time metric. Models were additionally adjusted for smoking within smoking strata, race (Caucasian, Black, Hispanic, Asian, Pacific Islander or American Indian/Alaskan Native, and missing), energy, energy-adjusted total fat (continuous), and sex.

 
Waist, hip, or the waist/hip ratio was not associated with pancreatic cancer among men (table 3). Among women, increasing waist circumference was significantly associated with pancreatic cancer (fourth vs. first quartile: HR = 2.53, 95 percent CI: 1.13, 5.65; ptrend = 0.04). Significant risk associated with greater hip circumference was attenuated and nonsignificant with adjustment for BMI, and the waist/hip ratio was not associated with pancreatic cancer. No significant associations were observed between low or high intensity physical activity or METs and pancreatic cancer by sex or in sex-combined models (table 4). The waist, hip, waist/hip ratio, and physical activity associations were not modified by smoking, history of diabetes, BMI, or follow-up time (pinteraction > 0.05).


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

  		TABLE 3. Multivariable-adjusted hazard ratios and 95% confidence intervals of waist, hip, and waist/hip ratio and pancreatic cancer by gender in the National Institutes of Health–AARP Diet and Health Study cohort, United States, 1995–2000

 

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

  		TABLE 4. Hazard ratios and 95% confidence intervals for low intensity and vigorous activity and metabolic equivalents in the National Institutes of Health–AARP Diet and Health Study cohort, United States, 1995–2000

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
We observed significant increased pancreatic cancer risks among NIH–AARP Diet and Health Study cohort members with severe obesity (BMI: ≥35 kg/m2), compared with those with normal weight, which was more apparent in men than women. Among nonsmokers, however, stronger risks were observed with increasing BMI, particularly among women, but no significant associations were evident among current smokers or those who quit <10 years before. Greater waist circumference was associated with a greater than twofold pancreatic cancer risk in women, but no association was observed in men. Physical activity was not associated with pancreatic cancer in our study.

Our BMI results are consistent with and similar in magnitude to two recent meta-analyses of obesity and pancreatic cancer (6, 10). The first meta-analysis based on six case-control studies and eight cohort studies showed a weak but statistically significant summary relative risk of 1.02 (95 percent CI: 1.01, 1.03) per unit increase of BMI and estimated a 19 percent (95 percent CI: 1.10, 1.29) increased pancreatic cancer risk for obese compared with nonobese subjects (6). The second meta-analysis based on 21 cohort studies had a summary relative risk of 1.12 (95 percent CI: 1.06, 1.17) per 5-unit BMI increase. Among all the prospective studies (5, 10, 11, 23, 26, 46), 13 studies have shown significant associations for pancreatic cancer, with relative risks ranging from 1.25 to 3.10 for high or >30-kg/m2 BMI among men (5, 11, 14, 15, 24, 26, 46–50), among women (11, 12, 14, 50), or among men and women combined (12, 14, 15, 42, 46, 50).

We found that waist circumference was positively associated with pancreatic cancer in women but not in men. The 2.5-fold risk in women was independent of BMI, suggesting that adiposity independent of fat-free body mass may account for our observed association. Four cohort studies examined waist circumference or the waist/hip ratio (15, 18, 19, 23), with three reporting positive associations for increasing waist circumference (15, 19, 23) or the waist/hip ratio (19) in men and women and two being statistically significant (19, 23). The American Cancer Society cohort demonstrated a marginally significant positive association for pancreatic cancer with the self-reported tendency for abdominal compared with peripheral weight gain (14).

The magnitude of our positive association between higher BMI and pancreatic cancer was slightly attenuated (7–8 percent risk reduction) with adjustment for history of diabetes. There is evidence to suggest that type 2 diabetes is etiologically involved in pancreatic carcinogenesis (3, 51), as well as the result of subclinical malignancy (52–54). The attenuation of our BMI association after adjustment for diabetes may be due to the inability to completely remove diabetes because of latent pancreatic cancer, despite the fact that we designed our study to minimize reverse causation. Prospective cohort studies with extended follow-up have reported positive associations between higher glucose concentrations, biochemically defined diabetes, and pancreatic cancer (47, 55–59). A recent epidemiologic study also showed a positive pancreatic cancer association with increasing glucose and insulin concentrations that was stronger among subjects with follow-up of >10 years (58). Experimental studies demonstrate that insulin has mitogenic effects on pancreatic cancer cell lines (60), and peripheral insulin resistance promotes ductal pancreatic carcinogenesis in animals (61–64). A history of diabetes is associated with higher BMI in our cohort, and subjects with type 2 diabetes exhibit higher insulin concentrations and insulin resistance during the early stages of their disease (25); therefore, the reduction in pancreatic cancer risk with adjustment for diabetes may be related to type 2 diabetes being in the causal pathway of the obesity and pancreatic cancer association.

In our study, BMI appeared to be more clearly associated with pancreatic cancer risk among subjects that were never smokers/quitters for >10 years. Although we do not observe significant interactions by smoking status in men, we observe significant interactions by smoking in women. Some studies have shown stronger BMI-associated pancreatic cancer risk in never smokers compared with that observed in the overall population (5, 49, 50). Five of six cohort studies that evaluated effect modification of BMI and pancreatic cancer risk by smoking status (14, 15, 19, 22, 24) did not report significant interactions by smoker status, which contrasts one study in a Japanese cohort that reported an elevated risk with increase in BMI among current smokers but not nonsmokers (26). Similar to our study (figure 1), a pooled analysis of Swedish women and men showed positive associations for BMI and pancreatic cancer, independent of smoking; however, all associations in current smokers were significantly higher than that of the nonsmokers (15). Our study and the latter (15) suggest that the carcinogenic effect of cigarette smoke on the pancreas appears stronger than that of BMI and may mask the BMI–pancreatic cancer association among the current smokers or recent quitters.

Although we hypothesized a protective relation, we did not observe an association between greater physical activity and pancreatic cancer, which is consistent with most studies (12, 14, 18–21, 24, 26). Three previous studies found some evidence for a protective effect with greater physical activity. One case-control study observed an inverse association with a composite variable for greater moderate and strenuous activity in men but not in women (13). A pooled analysis of health professionals and nurses observed an inverse association between greater metabolic equivalents of moderate activity but no association with greater total or vigorous activity (12). In a cohort study of male smokers, men with sedentary leisure activity had significant protective associations with greater occupational activity, although no association was observed overall for occupational or leisure physical activity (22). We did not observe a statistically significant interaction between BMI and physical activity, which is consistent with the other studies (7, 12, 14, 15, 19, 22, 24). Physical activity is a complex behavior that encompasses multiple domains (e.g., occupation, leisure, household, transportation) and characteristics (frequency, intensity, and duration) and is not precisely measured in most epidemiologic studies. Measurement imprecision may contribute to the lack of association observed in most studies. Alternatively, the lack of association between physical activity and pancreatic cancer may possibly suggest an etiologic mechanism related to adiposity, not influenced by physical activity that contributes to pancreatic cancer.

The strength of our study is its large prospective nature with body weight and physical activity being assessed prior to cancer diagnosis, thereby reducing biases and the influence of reverse causality. It also has a larger number of cases compared with many previous prospective studies (6, 14–19, 48, 65), providing greater power to detect differences in risk and the ability to stratify analyses by sex and by smoking status. Our study has internal validity, as the cases arose from the cohort that includes the noncases and therefore does not have control selection bias, has relatively good follow-up of members and outcome assessment, and includes adjustment for confounders. The NIH–AARP cohort includes both sexes and never, former, and current smokers; therefore, results may be generalizable to many older adults. Our cohort is limited because of a relatively short follow-up time (up to 5 years), and our associations may become stronger with extended follow-up. Although most cohort studies with longer follow-up time than our cohort show positive associations between BMI and pancreatic cancer, the majority are not statistically significant and a third have less magnitude (10). Our case definition includes nonmicroscopically confirmed cancer, which could contribute to misclassification of case status and attenuated risk estimates (11). Measurement error related to self-reported measurements (66, 67) and physical activity assessment is likely present and could contribute to spurious or attenuated associations.

In conclusion, our results from this large prospective study support the hypothesis that adiposity may increase the risk of developing pancreatic cancer in older persons. Our results, as well as those of others, may have important implications for cancer prevention particularly related to the avoidance of obesity.


   ACKNOWLEDGMENTS
 
This research was supported by the Intramural Research Program of the Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Department of Health and Human Services.

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 Cancer Surveillance Section of the California Department of Health Services. 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. 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 views expressed herein are solely those of the authors and do not necessarily reflect those of the contractor or the Florida Department of Health. The Pennsylvania Department of Health specifically disclaims responsibility for any analyses, interpretations, or conclusions.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Edwards BK, Brown ML, Wingo PA, et al. Annual report to the nation on the status of cancer, 1975 –2002, featuring population-based trends in cancer treatment. J Natl Cancer Inst (2005) 97:1407–27.[Abstract/Free Full Text]
  2. Anderson KE, Mack TM, Silverman D. Pancreatic cancer. In: Schottenfeld D, Fraumeni JF, eds. Cancer epidemiology and prevention. (2006) New York, NY: Oxford University Press.
  3. Huxley R, Ansary-Moghaddam A, Berrington de González A, et al. Type-II diabetes and pancreatic cancer: a meta-analysis of 36 studies. Br J Cancer (2005) 92:2076–83.[CrossRef][Web of Science][Medline]
  4. Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer (2004) 4:579–91.[CrossRef][Web of Science][Medline]
  5. Samanic C, Gridley G, Chow WH, et al. Obesity and cancer risk among white and black United States veterans. Cancer Causes Control (2004) 15:35–43.[CrossRef][Web of Science][Medline]
  6. Berrington A, Sweetland S, Spencer E. A meta-analysis of obesity and the risk of pancreatic cancer. Br J Cancer (2003) 89:519–23.[CrossRef][Web of Science][Medline]
  7. Eberle CA, Bracci PM, Holly EA. Anthropometric factors and pancreatic cancer in a population-based case-control study in the San Francisco Bay area. Cancer Causes Control (2005) 16:1235–44.[CrossRef][Web of Science][Medline]
  8. Pan SY, Mao Y, Ugnat AM. Physical activity, obesity, energy intake, and the risk of non-Hodgkin's lymphoma: a population-based case-control study. Am J Epidemiol (2005) 162:1162–73.[Abstract/Free Full Text]
  9. Fryzek JP, Schenk M, Kinnard M, et al. The association of body mass index and pancreatic cancer in residents of southeastern Michigan, 1996 –1999. Am J Epidemiol (2005) 162:222–8.[Abstract/Free Full Text]
  10. Larsson SC, Orsini N, Wolk A. Body mass index and pancreatic cancer risk: a meta-analysis of prospective studies. Int J Cancer (2007) 120:1993–8.[CrossRef][Web of Science][Medline]
  11. Verhage BA, Schouten LJ, Goldbohm RA, et al. Anthropometry and pancreatic cancer risk: an illustration of the importance of microscopic verification. Cancer Epidemiol Biomarkers Prev (2007) 16:1449–54.[Abstract/Free Full Text]
  12. Michaud DS, Giovannucci E, Willett WC, et al. Physical activity, obesity, height, and the risk of pancreatic cancer. JAMA (2001) 286:921–9.[Abstract/Free Full Text]
  13. Hanley AJ, Johnson KC, Villeneuve PJ, et al. Physical activity, anthropometric factors and risk of pancreatic cancer: results from the Canadian enhanced cancer surveillance system. Int J Cancer (2001) 94:140–7.[CrossRef][Web of Science][Medline]
  14. Patel AV, Rodriguez C, Bernstein L, et al. Obesity, recreational physical activity, and risk of pancreatic cancer in a large U.S. cohort. Cancer Epidemiol Biomarkers Prev (2005) 14:459–66.[Abstract/Free Full Text]
  15. Larsson SC, Permert J, Hakansson N, et al. Overall obesity, abdominal adiposity, diabetes and cigarette smoking in relation to the risk of pancreatic cancer in two Swedish population-based cohorts. Br J Cancer (2005) 93:1310–15.[CrossRef][Web of Science][Medline]
  16. Batty GD, Shipley MJ, Jarrett RJ, et al. Obesity and overweight in relation to organ-specific cancer mortality in London (UK): findings from the original Whitehall study. Int J Obes (Lond) (2005) 29:1267–74.[CrossRef][Medline]
  17. Lukanova A, Bjor O, Kaaks R, et al. Body mass index and cancer: results from the Northern Sweden Health and Disease Cohort. Int J Cancer (2006) 118:458–66.[CrossRef][Web of Science][Medline]
  18. Sinner PJ, Schmitz KH, Anderson KE, et al. Lack of association of physical activity and obesity with incident pancreatic cancer in elderly women. Cancer Epidemiol Biomarkers Prev (2005) 14:1571–3.[Free Full Text]
  19. Berrington de González A, Spencer EA, Bueno-de-Mesquita HB, et al. Anthropometry, physical activity, and the risk of pancreatic cancer in the European Prospective Investigation into Cancer and Nutrition. Cancer Epidemiol Biomarkers Prev (2006) 15:879–85.[Abstract/Free Full Text]
  20. Lee IM, Sesso HD, Oguma Y, et al. Physical activity, body weight, and pancreatic cancer mortality. Br J Cancer (2003) 88:679–83.[CrossRef][Web of Science][Medline]
  21. Lee IM, Paffenbarger RS Jr. Physical activity and its relation to cancer risk: a prospective study of college alumni. Med Sci Sports Exerc (1994) 26:831–7.
  22. Stolzenberg-Solomon RZ, Pietinen P, Taylor PR, et al. A prospective study of medical conditions, anthropometry, physical activity, and pancreatic cancer in male smokers (Finland). Cancer Causes Control (2002) 13:417–26.[CrossRef][Web of Science][Medline]
  23. Ansary-Moghaddam A, Huxley R, Barzi F, et al. The effect of modifiable risk factors on pancreatic cancer mortality in populations of the Asia-Pacific region. Cancer Epidemiol Biomarkers Prev (2006) 15:2435–40.[Abstract/Free Full Text]
  24. Nothlings U, Wilkens LR, Murphy SP, et al. Body mass index and physical activity as risk factors for pancreatic cancer: the Multiethnic Cohort Study. Cancer Causes Control (2007) 18:165–75.[CrossRef][Web of Science][Medline]
  25. Kahn SE, Hull RL, Utzschneider KM. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature (2006) 444:840–6.[CrossRef][Medline]
  26. Lin Y, Kikuchi S, Tamakoshi A, et al. Obesity, physical activity and the risk of pancreatic cancer in a large Japanese cohort. Int J Cancer (2007) 120:2665–71.[CrossRef][Web of Science][Medline]
  27. Willett W. Anthropometric measures and body composition. In: Nutritional epidemiology—Willett W, ed. (1998) New York, NY: Oxford University Press. 244–72.
  28. 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]
  29. Michaud DS, Midthune D, Hermansen S, 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 Regist Manage (2005) 32:70–5.
  30. Subar AF, Thompson FE, Smith AF, et al. Improving food frequency questionnaires: a qualitative approach using cognitive interviewing. J Am Diet Assoc (1995) 95:781–8.[CrossRef][Web of Science][Medline]
  31. Thompson FE, Subar AF, Brown CC, et al. Cognitive research enhances accuracy of food frequency questionnaire reports: results of an experimental validation study. J Am Diet Assoc (2002) 102:212–25.[CrossRef][Web of Science][Medline]
  32. 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]
  33. Pereira MA, FitzerGerald SJ, Gregg EW, et al. A collection of physical activity questionnaires for health-related research. Med Sci Sports Exerc (1997) 29(suppl):S1–205.
  34. Ainsworth BE, Haskell WL, Leon AS, et al. Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc (1993) 25:71–80.
  35. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. (1995) 854. World Health Organ Tech Rep Ser. 1–452.
  36. Boyle P, Maisonneuve P, Bueno de Mesquita B, et al. Cigarette smoking and pancreas cancer: a case control study of the search programme of the IARC. Int J Cancer (1996) 67:63–71.[CrossRef][Web of Science][Medline]
  37. Fuchs CS, Colditz GA, Stampfer MJ, et al. A prospective study of cigarette smoking and the risk of pancreatic cancer. Arch Intern Med (1996) 156:2255–60.[Abstract/Free Full Text]
  38. Silverman DT, Dunn JA, Hoover RN, et al. Cigarette smoking and pancreas cancer: a case-control study based on direct interviews. J Natl Cancer Inst (1994) 86:1510–16.[Abstract/Free Full Text]
  39. Falk RT, Pickle LW, Fontham ET, et al. Life-style risk factors for pancreatic cancer in Louisiana: a case-control study. Am J Epidemiol (1988) 128:324–36.[Abstract/Free Full Text]
  40. Farrow DC, Davis S. Risk of pancreatic cancer in relation to medical history and the use of tobacco, alcohol and coffee. Int J Cancer (1990) 45:816–20.[Web of Science][Medline]
  41. Bueno de Mesquita HB, Maisonneuve P, Moerman CJ, et al. Life-time history of smoking and exocrine carcinoma of the pancreas: a population-based case-control study in the Netherlands. Int J Cancer (1991) 49:816–22.[Web of Science][Medline]
  42. Friedman GD, van den Eeden SK. Risk factors for pancreatic cancer: an exploratory study. Int J Epidemiol (1993) 22:30–7.[Abstract/Free Full Text]
  43. Mack TM, Yu MC, Hanisch R, et al. Pancreas cancer and smoking, beverage consumption, and past medical history. J Natl Cancer Inst (1986) 76:49–60.[Web of Science][Medline]
  44. Harnack LJ, Anderson KE, Zheng W, et al. Smoking, alcohol, coffee, and tea intake and incidence of cancer of the exocrine pancreas: the Iowa Women's Health Study. Cancer Epidemiol Biomarkers Prev (1997) 6:1081–6.[Abstract]
  45. Howe GR, Jain M, Burch JD, et al. Cigarette smoking and cancer of the pancreas: evidence from a population-based case-control study in Toronto, Canada. Int J Cancer (1991) 47:323–8.[Web of Science][Medline]
  46. Wolk A, Gridley G, Svensson M, et al. A prospective study of obesity and cancer risk (Sweden). Cancer Causes Control (2001) 12:13–21.[CrossRef][Web of Science][Medline]
  47. Gapstur SM, Gann PH, Lowe W, et al. Abnormal glucose metabolism and pancreatic cancer mortality. JAMA (2000) 283:2552–8.[Abstract/Free Full Text]
  48. Rapp K, Schroeder J, Klenk J, et al. Obesity and incidence of cancer: a large cohort study of over 145,000 adults in Austria. Br J Cancer (2005) 93:1062–7.[CrossRef][Web of Science][Medline]
  49. Samanic C, Chow WH, Gridley G, et al. Relation of body mass index to cancer risk in 362,552 Swedish men. Cancer Causes Control (2006) 17:901–9.[CrossRef][Web of Science][Medline]
  50. Calle EE, Rodriguez C, Walker-Thurmond K, et al. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med (2003) 348:1625–38.[Abstract/Free Full Text]
  51. Silverman DT, Schiffman M, Everhart J, et al. Diabetes mellitus, other medical conditions and familial history of cancer as risk factors for pancreatic cancer. Br J Cancer (1999) 80:1830–7.[CrossRef][Web of Science][Medline]
  52. Permert J, Ihse I, Jorfeldt L, et al. Pancreatic cancer is associated with impaired glucose metabolism. Eur J Surg (1993) 159:101–7.[Web of Science][Medline]
  53. Permert J, Larsson J, Westermark GT, et al. Islet amyloid polypeptide in patients with pancreatic cancer and diabetes. N Engl J Med (1994) 330:313–18.[Abstract/Free Full Text]
  54. Permert J, Ihse I, Jorfeldt L, et al. Improved glucose metabolism after subtotal pancreatectomy for pancreatic cancer. Br J Surg (1993) 80:1047–50.[Web of Science][Medline]
  55. Jee SH, Ohrr H, Sull JW, et al. Fasting serum glucose level and cancer risk in Korean men and women. JAMA (2005) 293:194–202.[Abstract/Free Full Text]
  56. Batty GD, Shipley MJ, Marmot M, et al. Diabetes status and post-load plasma glucose concentration in relation to site-specific cancer mortality: findings from the original Whitehall study. Cancer Causes Control (2004) 15:873–81.[CrossRef][Web of Science][Medline]
  57. Balkau B, Barrett-Connor E, Eschwege E, et al. Diabetes and pancreatic carcinoma. Diabete Metab (1993) 19:458–62.[Web of Science][Medline]
  58. Stolzenberg-Solomon RZ, Graubard BI, Chari S, et al. Insulin, glucose, insulin resistance, and pancreatic cancer in male smokers. JAMA (2005) 294:2872–8.[Abstract/Free Full Text]
  59. Stattin P, Bjor O, Ferrari P, et al. Prospective study of hyperglycemia and cancer risk. Diabetes Care (2007) 30:561–7.[Abstract/Free Full Text]
  60. Hennig R, Ding XZ, Adrian TE. On the role of the islets of Langerhans in pancreatic cancer. Histol Histopathol (2004) 19:999–1011.[Web of Science][Medline]
  61. Bell RH Jr, McCullough PJ, Pour PM. Influence of diabetes on susceptibility to experimental pancreatic cancer. Am J Surg (1988) 155:159–64.[CrossRef][Web of Science][Medline]
  62. Pour PM, Kazakoff K, Carlson K. Inhibition of streptozotocin-induced islet cell tumors and N-nitrosobis(2-oxopropyl)amine-induced pancreatic exocrine tumors in Syrian hamsters by exogenous insulin. Cancer Res (1990) 50:1634–9.[Abstract/Free Full Text]
  63. Pour PM, Stepan K. Modification of pancreatic carcinogenesis in the hamster model. VIII. Inhibitory effect of exogenous insulin. J Natl Cancer Inst (1984) 72:1205–8.[Web of Science][Medline]
  64. Schneider MB, Matsuzaki H, Haorah J, et al. Prevention of pancreatic cancer induction in hamsters by metformin. Gastroenterology (2001) 120:1263–70.[CrossRef][Web of Science][Medline]
  65. Kuriyama S, Tsubono Y, Hozawa A, et al. Obesity and risk of cancer in Japan. Int J Cancer (2005) 113:148–57.[CrossRef][Web of Science][Medline]
  66. Spencer EA, Appleby PN, Davey GK, et al. Validity of self-reported height and weight in 4808 EPIC-Oxford participants. Public Health Nutr (2002) 5:561–5.[CrossRef][Web of Science][Medline]
  67. Ulijaszek SJ, Kerr DA. Anthropometric measurement error and the assessment of nutritional status. Br J Nutr (1999) 82:165–77.[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
 Cancer Epidemiol. Biomarkers Prev.Home page
L. D. Arnold, A. V. Patel, Y. Yan, E. J. Jacobs, M. J. Thun, E. E. Calle, and G. A. Colditz
Are Racial Disparities in Pancreatic Cancer Explained by Smoking and Overweight/Obesity?
Cancer Epidemiol. Biomarkers Prev., September 1, 2009; 18(9): 2397 - 2405.
[Abstract] [Full Text] [PDF]

Home page
 JAMAHome page
D. Li, J. S. Morris, J. Liu, M. M. Hassan, R. S. Day, M. L. Bondy, and J. L. Abbruzzese
Body Mass Index and Risk, Age of Onset, and Survival in Patients With Pancreatic Cancer
JAMA, June 24, 2009; 301(24): 2553 - 2562.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
J. B. Peel, X. Sui, C. E. Matthews, S. A. Adams, J. R. Hebert, J. W. Hardin, T. S. Church, and S. N. Blair
Cardiorespiratory Fitness and Digestive Cancer Mortality: Findings from the Aerobics Center Longitudinal Study
Cancer Epidemiol. Biomarkers Prev., April 1, 2009; 18(4): 1111 - 1117.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
G. D. Batty, M. Kivimaki, D. Morrison, R. Huxley, G. D. Smith, R. Clarke, M. G. Marmot, and M. J. Shipley
Risk Factors for Pancreatic Cancer Mortality: Extended Follow-up of the Original Whitehall Study
Cancer Epidemiol. Biomarkers Prev., February 1, 2009; 18(2): 673 - 675.
[Abstract] [Full Text] [PDF]

Home page
 Am J EpidemiolHome page
R. Z. Stolzenberg-Solomon, S. Weinstein, M. Pollak, Y. Tao, P. R. Taylor, J. Virtamo, and D. Albanes
Prediagnostic Adiponectin Concentrations and Pancreatic Cancer Risk in Male Smokers
Am. J. Epidemiol., November 1, 2008; 168(9): 1047 - 1055.
[Abstract] [Full Text] [PDF]

Home page
 Integr Cancer TherHome page
S. Kranz
Integrative Tumor Board: Pancreatic Cancer Physical Therapy
Integr Cancer Ther, June 1, 2008; 7(2): 108 - 110.
[PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
167/5/586    most recent
kwm361v1
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 (15)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Stolzenberg-Solomon, R. Z.
Right arrow Articles by Silverman, D. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stolzenberg-Solomon, R. Z.
Right arrow Articles by Silverman, D. T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?