Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Jul 12.
Published in final edited form as: Eur J Cancer Prev. 2008 Jun;17(3):201–208. doi: 10.1097/CEJ.0b013e3282b6fd78

OCCUPATIONAL PHYSICAL ACTIVITY AND RISK FOR CANCER OF THE COLON AND RECTUM IN SWEDEN AMONG MEN AND WOMEN BY ANATOMIC SUB-SITE

Tahereh Moradi 1, Gloria Gridley 2, Jan Björk 3, Mustafa Dosemeci 2, Bu-Tian Ji 2, Hans J Berkel 4, Stanley Lemeshow 5
PMCID: PMC2901913  NIHMSID: NIHMS211476  PMID: 18414190

Abstract

BACKGROUND

The inverse association between physical activity and colon cancer is well established, at least in men. We investigated the association of occupational physical activity with subsite- specific colorectal cancer risk.

METHODS

Based on occupational titles from the Swedish nationwide censuses in 1960 and 1970, we defined a cohort of women and men with the same work-related physical demands in 1960 and 1970. The incidence of colon and rectum cancer during 1971–1989 was ascertained through linkages to the Cancer Register. Relative risks were estimated through Poisson regression.

RESULTS

The risk for colon cancer increased with decreasing occupational physical activity. RR among sedentary women and men was 1.2 and 1.3 (p for trend=0.08 and <0.001). For men, the risks for proximal and distal colon increased by 20% and 40% (p for trend=0.005 and <0.001). Inactivity appeared to be particularly associated with descending colon cancer (RR =2.4, p for trend<0.001). In women the inverse association with activity was concentrated to proximal parts of colon; RR for cancer in the proximal and transverse colon among sedentary women was 1.4 and 2.0 (p for trend <0.07 and <0.01). Cancer of the rectum was not associated with activity in either sex.

CONCLUSION

We confirmed the well known inverse relationship between activity and risk of colon but not rectal cancer in both sexes. Data suggest that the physical activity-related variation in risk among women is greatest in the proximal and middle parts of colon, while the corresponding peak in men seems to be more distal. Gender specific anatomical and motility differences of the colon might contribute to this sub-site difference.

Keywords: Colon and rectal neoplasm, sub-site, cohort, occupational physical activity, Sweden

INTRODUCTION

The protective effect of physical activity on colon cancer is well established among men. The inverse association is slightly less striking among women (Friedenreich and Orenstein 2002; Vainio, Kaaks et al. 2002; Slattery 2004; Samad, Taylor et al. 2005). Apart from sufficiently different physiology between men and women that warrants separate analysis (Silvaggio and Mattison 1994), there is evidence that distal and proximal colon cancers have a distinct etiology (Iacopetta 2002), different embryological origin (Glebov, Rodriguez et al. 2003), and have different incidence by gender (Jensen 1984; Breivik, Lothe et al. 1997; Cheng, Chen et al. 2001; McCashland, Brand et al. 2001; Takada, Ohsawa et al. 2002; Jubelirer, Wells et al. 2003). Furthermore, the observed proximal shift in the distribution of cancers of the large bowel both in Western countries (Bonithon-Kopp and Benhamiche 1999) and in Japan (Takada, Ohsawa et al. 2002) parallel with the observed even more increasing proximal colon cancer among African Americans (Troisi, Freedman et al. 1999) support distinct pathogenic mechanisms in the carcinogenesis of the proximal and distal colon and thus, etiologic distinction by anatomic sub-site within the colon. Furthermore, life style factors such as major dietary patterns (Kim, Sasaki et al. 2005; Mizoue, Yamaji et al. 2005), carbohydrates (Borugian, Sheps et al. 2002), vegetable and fruit consumption (Voorrips, Goldbohm et al. 2000), red meat consumption (Larsson, Rafter et al. 2005), cigarette smoking and alcohol use (Toyomura, Yamaguchi et al. 2004), and reproductive factors (Yoo, Tajima et al. 1999) are differently associated with proximal and distal colon and with gender. Hence, they deserve distinct risk factor evaluation to enhance both prevention and diagnostic strategies. Several studies of the relationship between physical activity and colorectal cancer have differentiated between anatomic sub-sites (Garabrant, Peters et al. 1984; Vena, Graham et al. 1985; Gerhardsson, Norell et al. 1986; Brownson, Zahm et al. 1989; Fredriksson, Bengtsson et al. 1989; Peters, Garabrant et al. 1989; Severson, Nomura et al. 1989; Gerhardsson de Verdier, Steineck et al. 1990; Kato, Tominaga et al. 1990; Arbman, Axelson et al. 1993; Fraser and Pearce 1993; Chow, Malker et al. 1994; Marcus, Newcomb et al. 1994; Giovannucci, Ascherio et al. 1995; Longnecker, Gerhardsson le Verdier et al. 1995; Thune and Lund 1996; Martinez, Giovannucci et al. 1997; Tavani, Braga et al. 1999; Colbert, Hartman et al. 2001; Slattery, Edwards et al. 2003). However, interpretation is often hampered by small sample size or case-control design. Furthermore, to our knowledge, no study has conclusively addressed possible gender differences with regard to the site where the physical activity-related variation in risk has its maximum.

To improve our understanding of how physical activity affects colorectal carcinogenesis by sub site and gender, we have investigated this in a large cohort of Swedish men and women, whose occupational activity was uniform at two assessments, ten years apart. Moreover, because of the size of our cohort and the long follow-up time, we had enough power to disentangle the effect of physical activity on ascending, transverse, descending, sigmoid colon and rectum separately for the first time among women.

MATERIALS AND METHODS

Census Data

The Swedish Census of the Population and Housing collected information about demographic, occupational (including employment status, job title, industry and work address), and socioeconomic factors for every Swedish resident during one week in October 1960 and again in 1970. The censuses are more than 99 percent complete, since participation was mandatory by law, and great efforts were made to recruit non-respondents.

The Cancer and Cancer-Environment Register

The national Swedish Cancer Register, established in 1958, includes more than 98 percent of all diagnosed cancer cases in the country {Mattsson, 1984 #79}. Malignant diseases are coded according to the International Classification of Disease, seventh revision (ICD-7) during the entire period of study. The proportion of cancers that were histologically verified was 98% and 97% for women and men respectively for colon cancer (ICD-7 153 or 153.0–153-4) and 99% and 98% for women and men respectively for rectum cancer (ICD-7 154) in 1991 {CENTER FOR EPIDEMIOLOGY, 1991. #80}. Proximal colon cancers in this study included tumours of the caecum and ascending colon (coded together in the cancer register 153.0) and tumors of transverse and flexures (coded together in the cancer register 153.1). Distal colon cancers included tumours in the descending (code 153.2) and sigmoid colon (code 153.3). The proportion of colon cancers that were unspecified sub-sites was around 17% in 1991{CENTER FOR EPIDEMIOLOGY, 1991. #80}. The Cancer Register contains demographic and tumor data and it is annually linked to the Swedish Register of Causes of Death.

The Cancer–Environment Register III (CERIII) merges data from the Cancer Register of 1971–1989 with the census of 1960 {Official Statistics of Sweden, 1965 #82}and 1970 {Official Statistics of Sweden, 1974 #83}. The national registration number - a unique 10-digit personal identifier assigned to all Swedish residents – allows linkage between these two registers. The CERIII includes only cancer cases. Those included were cancer patients who had resided in Sweden both in 1960 and in 1970, and thus were recorded in both censuses. Missing data (individuals found in the cancer register but not in the censuses) was estimated to be less than 0.9% (EPIDEMIOLOGISKT CENTRUM 1994). To allow calculation of person-years at risk and expected numbers of detected cancers, we established a background register, encompassing all individuals who were registered in both the 1960 and the 1970 censuses. After record linkages, the national registration numbers were removed to ensure confidentiality.

Classification of occupational physical activity and covariates

Occupational physical activity was used as a proxy for total physical activity (or energy expenditure). The occupations reported in the census questionnaires were coded into 245 categories in the 1960 and 248 categories in the 1970 census, using a three-digit-coding scheme, devised by the National Labor Market Board in Sweden. We classified each occupation as demanding very high and high (e.g. charworkers, cleaners and related workers, nursing personnel not elsewhere classified, general farmworkers, farm machinery operators, livestock workers, dockers and freight workers), moderate (e.g. maids and related housekeeping service workers, waitress, bartenders, professional nurses, cooks), light (e.g. sewing machine operators, elementary school teachers, hairdressers and barbers, wholesalers, cashiers), or sedentary (e.g. bookkeepers, secretaries, telephone operators, tellers, workers in telephone-answering services) physical activity. Assessments were done independently by three Swedish specialists in occupational medicine with extensive experience of job classification. To reduce misclassification, we only considered occupations consistently classified by the three experts. Absolute agreement between at least two of the experts was required, while the third was allowed to diverge by no more than one category. A total of 202 occupations were thus unequivocally classified. The two highest categories of physical activity were merged, to gain statistical power, since few were classified as having very high occupational physical activity. A validation study in a historic cohort of twins showed good agreement between the experts’ scoring and self-reports of occupational physical activity (Moradi, Adami et al. 1999).

Socioeconomic status was categorized into four levels (unskilled blue-collar, skilled blue-collar, less educated white-collar, educated white-collar occupations), based on the occupational title, as described in detail elsewhere (Statistics Sweden 1995). Based on the residential municipality we categorized the place of residence into two categories: Urban: the capital, Stockholm, and the second and third largest cities in Sweden, Gothenburg and Malmo; and Non-urban: the other large municipalities, southern and central Sweden (except the above cities and large municipalities) and northern densely populated areas and northern sparsely populated areas.

Study Cohorts

1,343,696 men and 699,519 women reported employment in 1960 census (but not necessarily also in the 1970 census) in a job we could unequivocally classify with regard to physical activity. There were 1,373,770 men and 982,226 women with such jobs reported in the 1970 census (but not necessarily also in the 1960 census). We further identified the 670,066 men and 252,200 women who had jobs classified as demanding the same physical activity level both in 1960 and 1970 censuses.

Follow-up

Person-years of follow-up were calculated from January 1, 1971 until the diagnosis of any malignant tumor, death, or end of follow-up (December 31, 1989), whichever occurred first.

Analyses

We analyzed data in a grouped form. Attained age (age at follow-up) was divided into 14 5-year categories (<25, 25–29, …, 80–84, 85+). The nineteen calendar years of follow-up were divided into three 5-year intervals and one 4-year (January 1, 1971 – December 31, 1974 – December 31, 1975–1979, ……, January 1, 1985 – December 31, 1989). We estimated the risk of colon cancer in relation to physical activity by performing internal comparisons between exposure strata within the cohorts, using rate ratios as the measure of relative risk (RR). The “exposed” group (with sedentary jobs) was compared to the “reference” group (very high/high occupational activity). Adjustment was made for available co-variates such as age at follow-up, calendar year of follow-up, place of residence and socioeconomic status using multivariate Poisson models, estimated by the maximum likelihood method (Preston 1995). In an attempt for the analyses of men and women who changed their occupational activity between the assessments, we did not find enough power to meaningfully estimate the risks for those who had switched their occupation from physical activity level high/very high to sedentary.

RESULTS

During 19 years of follow-up, we observed 21,367 men and 8,729 women with a diagnosis of colon and rectal cancer who reported employment in 1960 census and who could be classified into one of the physical activity levels. These numbers were 15,471 for men and 8,319 for women in the 1970 census. We further observed 10,106 male and 3,122 female colorectal cancers among those who had jobs classified as demanding the same physical activity level in both the 1960 and 1970 censuses.

Tables I and II show the risk for cancer of colon and rectum by sub-sites among men and women, correspondingly, in relation to occupational physical activity during different calendar periods. In simple age-adjusted models we found an increase in risk for colon cancer as well as most sub sites with decreasing physical-activity level in both sexes. Among men the RR for colon cancer was 1.4 (95%CI 1.4–1.5) and 1.3 (95%CI 1.2–1.4) with the estimated physical activity level in 1960 and 1970 respectively for sedentary as compared with high/very high activity level. Women with sedentary level of occupational physical activity had a 10% increased in risk for cancer of the colon compared with high/very high activity level in both 1960 and 1970. Further adjustment for socio-economic status, place of residence and calendar year of follow-up changed the results only slightly. Cancer of the rectum was unaffected by the level of physical activity at work.

Table I.

Relative risk (RR) with 95 percent confidence interval (CI) for cancer of the colon and rectum among men by sub site and by estimated physical activity in 1960 and 1970. Results obtained by Poisson regression.

Anatomic subsite (ICDO-7) Occupational PA Census 1960 Census 1970
Cases RRa 95 % CI RRb 95 % CI Cases RRa 95 % CI RRb 95 % CI
Colon (153) Very high/high 4415 1.0 Ref. 1.0 Ref. 2668 1.0 Ref. 1.0 Ref.
Moderate 2294 1.2 1.1–1.3 1.2 1.1–1.2 1728 1.1 1.0–1.2 1.0 1.0–1.2
Light 4161 1.4 1.3–1.4 1.3 1.2–1.3 3424 1.3 1.2–1.3 1.2 1.1–1.3
Sedentary 1668 1.4 1.4–1.5 1.3 1.2–1.4 1150 1.3 1.2–1.4 1.2 1.1–1.3
p<0.001 p<0.001 p<0.001 p<0.001
Proximal (153.0-.1) Very high/high 2064 1.0 Ref. 1.0 Ref. 1240 1.0 Ref. 1.0 Ref.
Moderate 1094 1.2 1.1–1.3 1.1 1.0–1.2 802 1.1 1.0–1.2 1.1 1.0–1.2
Light 1910 1.3 1.2–1.4 1.2 1.1–1.4 1588 1.3 1.2–1.4 1.2 1.1–1.4
Sedentary 766 1.4 1.3–1.5 1.3 1.1–1.4 524 1.3 1.2–1.4 1.2 1.1–1.4
p<0.001 p<0.001 p<0.001 p=0.0042
 Ascending+ Cecum (153.0) Very high/high 1467 1.0 Ref. 1.00 Ref. 880 1.0 Ref. 1.00 Ref.
Moderate 747 1.2 1.1–1.3 1.1 1.0–1.2 579 1.1 1.0–1.3 1.1 1.0–1.2
Light 1398 1.4 1.3–1.5 1.3 1.1–1.4 1155 1.3 1.2–1.5 1.3 1.1–1.5
Sedentary 567 1.5 1.3–1.6 1.3 1.1–1.5 376 1.3 1.2–1.5 1.3 1.1–1.5
p<0.001 p=0.002 p<0.001 p=0.0038
 Transverse+ Flexures (153.1) Very high/high 597 1.0 Ref. 1.0 Ref. 360 1.0 Ref. 1.0 Ref.
Moderate 302 1.2 1.0–1.3 1.1 1.0–1.2 223 1.1 0.9–1.2 1.0 0.8–1.2
Light 512 1.2 1.1–1.4 1.2 1.0–1.4 433 1.2 1.0–1.3 1.0 0.9–1.3
Sedentary 199 1.3 1.1–1.5 1.2 1.0–1.5 148 1.2 1.0–1.5 1.1 0.9–1.5
p<0.001 p=0.044 p=0.010 p=ns
Distal (153.2–.3) Very high/high 1718 1.0 Ref. 1.0 Ref. 1039 1.0 Ref. 1.0 Ref.
Moderate 902 1.2 1.1–1.3 1.2 1.1–1.3 668 1.1 1.0–1.2 1.1 1.0–1.2
Light 1663 1.4 1.3–1.5 1.2 1.1–1.4 1341 1.3 1.2–1.4 1.1 1.0–1.3
Sedentary 666 1.5 1.4–1.6 1.3 1.2–1.5 465 1.4 1.2–1.5 1.2 1.0–1.4
p<0.001 p<0.001 p<0.001 p=0.016
 Descending (153.2) Very high/high 214 1.0 Ref. 1.0 Ref. 133 1.0 Ref. 1.0 Ref.
Moderate 120 1.3 1.0–1.6 1.1 0.9–1.5 85 1.1 0.8–1.4 1.0 0.7–1.3
Light 223 1.4 1.2–1.8 1.2 0.9–1.6 192 1.4 1.1–1.7 1.1 0.8–1.5
Sedentary 115 2.0 1.6–2.5 1.7 1.2–2.3 75 1.7 1.3–2.2 1.3 0.9–1.9
p<0.001 p<0.001 p<0.001 p=0.0089
 Sigmoid (153.3) Very high/high 1504 1.0 Ref. 1.0 Ref. 906 1.0 Ref. 1.0 Ref.
Moderate 782 1.2 1.1–1.3 1.2 1.1–1.3 583 1.1 1.0–1.2 1.1 1.0–1.2
Light 1440 1.3 1.2–1.5 1.3 1.2–1.5 1149 1.2 1.0–1.4 1.1 1.0–1.3
Sedentary 551 1.3 1.2–1.5 1.3 1.1–1.5 390 1.2 1.0–1.5 1.1 1.0–1.4
p<0.001 p<0.001 p<0.001 p=0.12
Rectum (154) Very high/high 3492 1.0 Ref. 1.0 Ref. 2151 1.0 Ref. 1.0 Ref.
Moderate 1608 1.1 1.0–1.1 1.0 1.0–1.1 1353 1.1 1.0–1.1 1.0 1.0–1.1
Light 2674 1.1 1.0–1.1 1.0 1.0–1.1 2259 1.0 1.0–1.1 1.0 0.9–1.1
Sedentary 1055 1.1 1.1–1.2 1.1 1.0–1.2 738 1.0 1.0–1.1 1.1 0.9–1.2
p<0.001 p=ns p=ns p=ns
Open in a new tab
a

Relative risk adjusted for age by 5 years interval, and calendar year of follow-up by year.

b

Relative risk adjusted for age by 5 years interval, calendar year of follow-up by year, place of residence and socioeconomic status.

Table II.

Relative risk (RR) with 95 percent confidence interval (CI) for cancer of the colon and rectum among women by sub site and by estimated physical activity in 1960 and 1970. Results obtained by Poisson regression.

Anatomic subsite (ICDO-7) Occupational PA Census 1960 Census 1970
Cases RRa 95 % CI RRb 95 % CI Cases RRa 95 % CI RRb 95 % CI
Colon (153) Very high/high 1073 1.0 Ref. 1.0 Ref. 1442 1.0 Ref. 1.0 Ref.
Moderate 1759 1.0 0.9–1.1 1.0 0.9–1.1 1812 1.0 1.0–1.1 1.1 1.0–1.1
Light 1876 1.1 1.0–1.1 1.1 1.0–1.2 1372 1.1 1.0–1.1 1.1 1.0–1.2
Sedentary 1033 1.1 1.0–1.2 1.1 1.1–1.3 732 1.1 1.0–1.2 1.2 1.1–1.4
p=0.014 p=0.0055 p=0.10 p<0.001
Proximal (153.0–.1) Very high/high 510 1.0 Ref. 1.0 Ref. 663 1.0 Ref. 1.0 Ref.
Moderate 850 1.0 0.9–1.1 1.1 0.9–1.2 810 1.0 0.9–1.1 1.0 0.9–1.1
Light 884 1.1 0.9–1.2 1.1 1.0–1.2 606 1.0 0.9–1.1 1.1 1.0–1.2
Sedentary 497 1.2 1.0–1.3 1.2 1.0–1.5 344 1.2 1.0–1.3 1.2 1.0–1.5
p=0.024 p=0.011 p=0.056 p=0.0032
 Ascending + Cecum (153.0) Very high/high 357 1.0 Ref. 1.0 Ref. 489 1.0 Ref. 1.0 Ref.
Moderate 633 1.1 1.0–1.2 1.1 1.0–1.3 594 1.0 0.9–1.1 1.0 0.9–1.2
Light 635 1.1 1.0–1.2 1.1 1.0–1.3 425 1.0 0.9–1.1 1.0 0.9–1.2
Sedentary 350 1.2 1.0–1.4 1.3 1.0–1.6 238 1.1 0.9–1.3 1.2 0.9–1.4
p=0.062 p=0.0084 p=ns p=0.096
 Transverse+Flexures (153.1) Very high/high 153 1.0 Ref. 1.0 Ref. 174 1.0 Ref. 1.0 Ref.
Moderate 217 0.9 0.7–1.1 0.9 0.7–1.1 216 1.0 0.8–1.3 1.0 0.8–1.3
Light 249 1.0 0.8–1.2 1.0 0.8–1.3 181 1.2 0.9–1.4 1.3 1.0–1.6
Sedentary 147 1.1 0.9–1.4 1.1 0.8–1.5 106 1.3 1.0–1.7 1.5 1.1–2.2
p=ns p=ns p=0.013 p=0.0037
Distal (153.2–.3) Very high/high 410 1.0 Ref. 1.0 Ref 563 1.0 Ref. 1.0 Ref.
Moderate 641 1.0 0.8–1.1 1.0 0.9–1.1 722 1.1 0.9–1.2 1.1 1.0–1.2
Light 714 1.0 0.9–1.2 1.1 0.9–1.2 526 1.0 0.9–1.2 1.1 1.0–1.3
Sedentary 383 1.0 0.9–1.2 1.2 1.0–1.4 272 1.0 0.9–1.2 1.1 0.9–1.4
p=ns p=0.022 p=ns p=0.045
 Descending (153.2) Very high/high 66 1.0 Ref. 1.0 Ref. 80 1.0 Ref. 1.0 Ref.
Moderate 80 0.7 0.5–1.0 0.7 0.5–1.0 103 1.1 0.8–1.4 1.0 0.7–1.4
Light 108 1.0 0.7–1.3 1.0 0.7–1.4 73 1.0 0.7–1.4 0.9 0.6–1.3
Sedentary 53 0.9 0.6–1.3 1.0 0.6–1.7 41 1.1 0.7–1.6 1.0 0.6–1.7
p=ns p=ns p=ns p=ns
 Sigmoid (153.3) Very high/high 344 1.0 Ref. 1.0 Ref. 483 1.0 Ref. 1.0 Ref.
Moderate 561 1.0 0.9–1.1 1.0 0.9–1.2 619 1.1 0.9–1.2 1.1 1.0–1.2
Light 606 1.1 0.9–1.2 1.1 1.0–1.3 453 1.0 0.9–1.2 1.1 1.0–1.3
Sedentary 330 1.1 0.9–1.3 1.2 1.0–1.5 231 1.0 0.9–1.2 1.2 0.9–1.4
p=ns p=0.039 p=ns p=0.070
Rectum (154) Very high/high 539 1.0 Ref. 1.0 Ref. 778 1.0 Ref. 1.0 Ref.
Moderate 937 1.1 1.0–1.2 1.1 1.0–1.2 1012 1.1 1.0–1.2 1.1 1.0–1.2
Light 986 1.1 1.0–1.2 1.1 1.0–1.3 768 1.1 1.0–1.2 1.1 1.0–1.2
Sedentary 526 1.1 0.9–1.2 1.1 0.9–1.3 403 1.1 1.0–1.2 1.0 0.9–1.2
p=ns p=ns p=ns p=ns
Open in a new tab
a

Relative risk adjusted for age by 5 years interval, and calendar year of follow-up by year.

b

Relative risk adjusted for age by 5 years interval, calendar year of follow-up by year, place of residence and socioeconomic status.

Table III shows the relative risks associated with occupational physical activity for cancer of the colon and rectum and by sub-sites and by gender adjusted for age, calendar year of follow-up, place of residence and socioeconomic status. The risks were estimated among those who had the same level of occupational physical activity. The relative risk for colon cancer was 1.3 (95% CI 1.2 to 1.5) among men and 1.2 (95% CI 1.0 to 1.5) among women classified as holding sedentary jobs both in 1960 and in 1970, relative to those estimated as having the physically most demanding jobs.

Table III.

Relative risk (RR) with 95 percent confidence interval (CI) for cancer of the colon and rectum by sub-site and by occupational physical activity level in 1960 and 1970.

Anatomic sub-site (ICD-7) Occupational Physical Activity Female Male
Cases RRa 95 % CI Cases RRa 95 % CI
Colon (153) Very high/high 359 1.0 Ref. 1989 1.0 Ref.
Moderate 569 1.0 0.9–1.1 894 1.2 1.1–1.3
Light 706 1.1 0.9–1.2 2314 1.3 1.1–1.4
Sedentary 366 1.2 1.0–1.5 703 1.3 1.2–1.5
p-value for trend p=0.045 p<0.001
Proximal Colon (153.0-1) Very high/high 174 1.0 Ref. 937 1.0 Ref.
Moderate 269 1.0 0.8–1.2 416 1.2 1.0–1.3
Light 325 1.1 0.9–1.3 1092 1.3 1.1–1.5
Sedentary 188 1.4 1.1–2.0 319 1.2 1.0–1.5
p-value for trend p=0.029 p=0.0036
 Ascending + Cecum (153.0) Very high/high 136 1.0 Ref. 660 1.0 Ref
Moderate 201 1.0 0.8–1.2 297 1.2 1.0–1.4
Light 228 1.0 0.8–1.2 806 1.4 1.2–1.6
Sedentary 132 1.3 0.9–1.8 238 1.3 1.1–1.6
p-value for trend p=ns p=0.0024
 Transverse+ Flexures (153.1) Very high/high 38 1.0 Ref. 277 1.0 Ref.
Moderate 68 1.1 0.7–1.6 119 1.1 0.8–1.4
Light 97 1.4 0.9–2.2 286 1.0 0.8–1.4
Sedentary 56 2.0 1.1–3.5 81 1.0 0.7–1.5
p-value for trend P=0.044 p=ns
Distal Colon (153.2-3) Very high/high 141 1.0 Ref. 773 1.0 Ref.
Moderate 215 1.0 0.8–1.2 352 1.2 1.1–1.4
Light 274 1.1 0.8–1.4 899 1.3 1.1–1.5
Sedentary 135 1.2 0.8–1.7 285 1.4 1.2–1.7
p-value for trend P=ns p<0.001
 Descending (153.2) Very high/high 23 1.0 Ref. 97 1.0 Ref.
Moderate 32 0.8 0.4–1.4 48 1.4 1.0–2.0
Light 40 0.9 0.5–1.6 120 1.6 1.1–2.4
Sedentary 19 1.1 0.4–2.5 49 2.4 1.4–3.9
p-value for trend P=ns p<0.001
 Sigmoid (153.3) Very high/high 118 1.0 Ref. 676 1.0 Ref.
Moderate 183 1.0 0.8–1.3 304 1.2 1.0–1.4
Light 234 1.1 0.9–1.4 779 1.3 1.1–1.5
Sedentary 116 1.2 0.8–1.8 236 1.3 1.0–1.6
p-value for trend P=ns p=0.020
Rectum (154) Very high/high 194 1.0 Ref. 1581 1.0 Ref.
Moderate 315 1.0 0.9–1.2 667 1.1 1.0–1.2
Light 395 1.1 0.9–1.3 1515 1.0 0.9–1.2
Sedentary 218 1.0 0.8–1.3 443 1.1 0.9–1.2
p-value for trend P=ns p=ns
Open in a new tab
a

Relative risk adjusted for 5- year age and calendar-year intervals of follow-up, place of residence, socioeconomic status

In men, both distal- and proximal colon cancer risk were elevated among those with sedentary jobs compared with men having very high/high job activities. Cancer of the descending colon was most affected (RR=2.4; 95% CI 1.4 to 3.9) by sedentary jobs followed by cancers of the ascending and sigmoid colon, with 30% higher risk for sedentary jobs relative to men having the physically active jobs. Risk for cancers of the transverse colon and rectum with sedentary jobs in both 1960 and 1970 was the same as in men with physically most challenging jobs. In women only cancer of proximal colon was affected by sedentary jobs. Cancer of the transverse colon was inversely significantly associated with physical activity where cancer of ascending colon was not associated with level of activity at work. Women with sustained sedentary jobs in both 1960 and 1970 had twice the risk (95% CI 1.1 to 3.5) of transverse colon cancer of women with physically active jobs. Cancer of the rectum remained unaffected by occupational physical activity in both sexes.

DISCUSSION

This study further confirms the prior consistent observations that physical activity decreases the risk of colon but not rectal cancer in both sexes. Among men we found an inverse association between estimated work-place activity and risk for ascending colon, descending colon, and sigmoid colon cancer. Descending colon showed the strongest inverse association. Our study supports evidence of a dose-dependent inverse association between estimated work place activity and risk for transverse colon cancer among women. The finding is novel and has not been explored in as much detail before in women.

The unique characteristics of our study includes the long and complete follow-up of colorectal cancer occurrence, the large size of the cohort, and the exposure assessment on two occasions 10 years apart, not based on self-report. The large number of colorectal cancer allowed us to evaluate colon cancer risk among men and women separately by subsite. Misclassification of exposure through job titles is likely. However, since the reporting of occupation occurred long before occurrence of any outcome, any exposure misclassification would be non-differential. Although we lacked information on other potential factors associated with the risk of colon cancer, an inverse association with physical activity observed in other studies was not confounded by body mass index, diet, smoking, aspirin use, education, family history of colorectal cancer, menopausal hormone use, alcohol consumption and sun exposure (Gerhardsson de Verdier, Steineck et al. 1990; Giovannucci, Ascherio et al. 1995; Le Marchand, Wilkens et al. 1997; Martinez, Giovannucci et al. 1997; Slattery, Potter et al. 1997; Steindorf, Jedrychowski et al. 2005) {Calton, 2006 #77}. Random error is of concern in subsite analyses where the analyses could for example be hampered by lack of statistical power. Misclassification of subsite categories is another concern since it is based on routine diagnosis that has not been confirmed. Our finding of different subsite effects among men and women could therefore be subject to these errors and thus warrant further investigation. We lacked information on leisure time physical activity. It is reasonable to assume that such exercise has effects in the same direction as exercise at work. Our inability to take leisure time physical activity into account may bias our risk estimates in an unpredictable direction. The true association would be attenuated if women with sedentary jobs are more- and overestimated if they are less- physically active after work than women with more strenuous occupations. The former of these alternatives may be more likely, but reliable population-based data on leisure time activity in different occupational groups is lacking. In any case, it is highly unlikely that work time-leisure time differences in physical activity could create a spurious association between physical activity and cancer risk.

Our risk estimates based on maintenance of the same occupational physical activity at both censuses would be biased if maintenance of sedentary occupations is related to worst health. The observed estimates would be exaggerated if the prevalence of conditions such as cholecystectomy or ulcerative colitis were related to maintenance of sedentary occupations. However, the estimates for those who maintained sedentary occupation at two censuses were in the same direction as for those with sedentary occupation at each census. Therefore, any major bias due to differences in health status appears unlikely.

Our finding of no association between occupational physical activity and risk for rectal cancer in either sexes is in agreement with the result of the meta-analyses of studies published through 2001 {Samad, 2005 #43} and one recent study not included in the meta-analyses {Larsson, 2006 #76}. Another recent study has showed a significant inverse association between vigorous occupational physical activity and risk for rectal cancer among women and no significant association among men {Slattery, 2003 #17}.

Cancer of the distal colon, a predominant cancer in the large bowel in men (de Jong, Day et al. 1972; Iacopetta 2002), was associated with a stronger effect for the descending than the sigmoid colon in our study, consistent with an early study (Garabrant, Peters et al. 1984) but in contrast with other studies finding the highest risk in the transverse colon (Gerhardsson, Norell et al. 1986), the ascending colon (Vena, Graham et al. 1985) or in all parts of the colon (Kato, Tominaga et al. 1990). Borader subsite analysis in a cohort of male smokers revealed a significant inverse association with occupational activity in the distal but not the proximal colon (Colbert, Hartman et al. 2001) similar to the non-significant inverse association of a recent study {Zhang, 2006 #78} where the contrast was found in another study (Thune and Lund 1996). No association of self-reported occupational physical activity was found in the cohort of Swedish men, where leisure time physical activity was inversely associated with risk of colon cancer in this studyand, similar to our result, stronger for distal colon than for proximal colon {Larsson, 2006 #76}.

Contrary to men, we found a more pronounced inverse association of physical activity in the proximal colon, a leading cancer in the large bowel among women (de Jong, Day et al. 1972; Iacopetta 2002), with a statistically significant association only in transverse colon. Although, to our knowledge there has been no study investigating the possible association between physical activity and cancer of the colon by comparable fine subsites, assessment of the association among women by broader subsite categories in one study has shown a result similar to our study with a stronger inverse association between leisure time physical activity for the proximal than the distal colon (Thune and Lund 1996). Other studies have revealed an inverse association between physical activity and distal colon (Martinez, Giovannucci et al. 1997), and finally no significant decrease in risk was found for either proximal or distal colon in other studies (Marcus, Newcomb et al. 1994) {Calton, 2006 #77} {Zhang, 2006 #78}.

Although the protective effect of physical activity has been hypothesized to influence several factors relevant to colon carcinogenesis such as, the immune system, the bile acid metabolism, insulin, insulin-like growth factor-I (IGF-I), IGF binding proteins (IGFBPs)(Sandhu, Dunger et al. 2002), prostaglandin levels and colon transit time (Oettle 1991), serum cholesterol and gastrointestinal and pancreatic hormone profiles (Quadrilatero and Hoffman-Goetz 2003), there is little empirical data to support them. The observed differences in the effect of physical activity between genders and sub-site specific add complications to an already complicated unknown mechanism. Anatomical and functional differences might contribute to the sub site difference seen in men and women. The female colon is inherently longer despite women’s smaller stature. The length difference is caused by a longer transverse colon which topography has a different rout in women reaching the true pelvis significantly more often than in men (Saunders, Fukumoto et al. 1996). Moreover, colonic transit is slower in women something which is further enhanced in the luteal phase of the menstrual cycle (Jung, Kim et al. 2003; Sadik, Abrahamsson et al. 2003) indicating that hormonal factors influence motility. These observations might explain why the increased risk of transverse colon cancer is seen in women but not in men with sedentary jobs. The transit through transverse colon might therefore be more dependent of physical activity in women than in men. Less physical activity then further enhance the exposure of the colonic mucosa to the fecal content (Lampe, Fredstrom et al. 1993). (might be deleted) There are clear evidence of the demographic and geographic differences in the incidence of distal and proximal colon cancer (Jensen 1984; Bonithon-Kopp and Benhamiche 1999; Troisi, Freedman et al. 1999; McCashland, Brand et al. 2001; Takada, Ohsawa et al. 2002; Jubelirer, Wells et al. 2003). Differences in molecular level such as bacterial composition and bacterial activity (Macfarlane, Gibson et al. 1992), as well as differences in clinic pathologic features (Ward, Meagher et al. 2001) such as cell proliferation and apoptotic activity (Anti, Armuzzi et al. 2001) between distal and proximal colon have been observed. In addition patients with cancer of proximal colon appears to have a survival benefit with chemotherapy compared with patients with cancer of distal colon (Elsaleh, Joseph et al. 2000). Thus, the protective effect of physical activity on colon cancer risk that differs by anatomic sub-sites and gender observed in this study and others, although with unclear mechanism, would be expected. However, additional finding in our study of an inverse association between physical activity level and the risk for descending in distal colon cancer in men and for transverse in proximal colon cancer in women further complicates the issue and remains to be confirmed by other studies.

Acknowledgments

This work was supported by The Intramural Research Program of The National Institutes of Health, National Cancer Institute, Bethesda, MD, USA; MA N02-CP-7100 ”Record Linkage”

References

  1. Anti M, Armuzzi A, et al. Severe imbalance of cell proliferation and apoptosis in the left colon and in the rectosigmoid tract in subjects with a history of large adenomas. Gut. 2001;48(2):238–46. doi: 10.1136/gut.48.2.238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arbman G, Axelson O, et al. Do occupational factors influence the risk of colon and rectal cancer in different ways? Cancer. 1993;72(9):2543–9. doi: 10.1002/1097-0142(19931101)72:9<2543::aid-cncr2820720906>3.0.co;2-d. [DOI] [PubMed] [Google Scholar]
  3. Bonithon-Kopp C, Benhamiche AM. Are there several colorectal cancers? Epidemiological data. Eur J Cancer Prev. 1999;8(Suppl 1):S3–12. [PubMed] [Google Scholar]
  4. Borugian MJ, Sheps SB, et al. Carbohydrates and colorectal cancer risk among Chinese in North America. Cancer Epidemiol Biomarkers Prev. 2002;11(2):187–93. [PubMed] [Google Scholar]
  5. Breivik J, Lothe RA, et al. Different genetic pathways to proximal and distal colorectal cancer influenced by sex-related factors. Int J Cancer. 1997;74(6):664–9. doi: 10.1002/(sici)1097-0215(19971219)74:6<664::aid-ijc18>3.0.co;2-5. [DOI] [PubMed] [Google Scholar]
  6. Brownson RC, Zahm SH, et al. Occupational risk of colon cancer. An analysis by anatomic subsite. Am J Epidemiol. 1989;130(4):675–87. doi: 10.1093/oxfordjournals.aje.a115389. [DOI] [PubMed] [Google Scholar]
  7. Cheng X, V, Chen W, et al. Subsite-specific incidence rate and stage of disease in colorectal cancer by race, gender, and age group in the United States, 1992–1997. Cancer. 2001;92(10):2547–54. doi: 10.1002/1097-0142(20011115)92:10<2547::aid-cncr1606>3.0.co;2-k. [DOI] [PubMed] [Google Scholar]
  8. Chow WH, Malker HS, et al. Occupational risks for colon cancer in Sweden. J Occup Med. 1994;36(6):647–51. [PubMed] [Google Scholar]
  9. Colbert LH, Hartman TJ, et al. Physical activity in relation to cancer of the colon and rectum in a cohort of male smokers. Cancer Epidemiol Biomarkers Prev. 2001;10(3):265–8. [PubMed] [Google Scholar]
  10. de Jong UW, Day NE, et al. The distribution of cancer within the large bowel. Int J Cancer. 1972;10(3):463–77. doi: 10.1002/ijc.2910100305. [DOI] [PubMed] [Google Scholar]
  11. Elsaleh H, Joseph D, et al. Association of tumour site and sex with survival benefit from adjuvant chemotherapy in colorectal cancer. Lancet. 2000;355(9217):1745–50. doi: 10.1016/S0140-6736(00)02261-3. [DOI] [PubMed] [Google Scholar]
  12. EPIDEMIOLOGISKT CENTRUM. Rapport 4, Socialstyrelsen, Stockholm, Sweden Cancer Miljöregistret 1960–1970. 1994. [Google Scholar]
  13. Fraser G, Pearce N. Occupational physical activity and risk of cancer of the colon and rectum in New Zealand males. Cancer Causes Control. 1993;4(1):45–50. doi: 10.1007/BF00051713. [DOI] [PubMed] [Google Scholar]
  14. Fredriksson M, Bengtsson NO, et al. Colon cancer, physical activity, and occupational exposures. A case-control study. Cancer. 1989;63(9):1838–42. doi: 10.1002/1097-0142(19900501)63:9<1838::aid-cncr2820630930>3.0.co;2-4. [DOI] [PubMed] [Google Scholar]
  15. Friedenreich CM, Orenstein MR. Physical activity and cancer prevention: etiologic evidence and biological mechanisms. J Nutr. 2002;132(11 Suppl):3456S–3464S. doi: 10.1093/jn/132.11.3456S. [DOI] [PubMed] [Google Scholar]
  16. Garabrant DH, Peters JM, et al. Job activity and colon cancer risk. Am J Epidemiol. 1984;119(6):1005–14. doi: 10.1093/oxfordjournals.aje.a113805. [DOI] [PubMed] [Google Scholar]
  17. Gerhardsson de Verdier M, Steineck G, et al. Physical activity and colon cancer: a case-referent study in Stockholm. Int J Cancer. 1990;46(6):985–9. doi: 10.1002/ijc.2910460606. [DOI] [PubMed] [Google Scholar]
  18. Gerhardsson M, Norell SE, et al. Sedentary jobs and colon cancer. Am J Epidemiol. 1986;123(5):775–80. doi: 10.1093/oxfordjournals.aje.a114306. [DOI] [PubMed] [Google Scholar]
  19. Giovannucci E, Ascherio A, et al. Physical activity, obesity, and risk for colon cancer and adenoma in men. Ann Intern Med. 1995;122(5):327–34. doi: 10.7326/0003-4819-122-5-199503010-00002. [DOI] [PubMed] [Google Scholar]
  20. Glebov OK, Rodriguez LM, et al. Distinguishing right from left colon by the pattern of gene expression. Cancer Epidemiol Biomarkers Prev. 2003;12(8):755–62. [PubMed] [Google Scholar]
  21. Iacopetta B. Are there two sides to colorectal cancer? Int J Cancer. 2002;101(5):403–8. doi: 10.1002/ijc.10635. [DOI] [PubMed] [Google Scholar]
  22. Jensen OM. Different age and sex relationship for cancer of subsites of the large bowel. Br J Cancer. 1984;50(6):825–9. doi: 10.1038/bjc.1984.262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jubelirer SJ, Wells JB, et al. Incidence of colorectal cancer in West Virginia from 1993–1999: an update by gender, age, subsite and stage. W V Med J. 2003;99(5):182–6. [PubMed] [Google Scholar]
  24. Jung HK, Kim DY, et al. Effects of gender and menstrual cycle on colonic transit time in healthy subjects. Korean J Intern Med. 2003;18(3):181–6. doi: 10.3904/kjim.2003.18.3.181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kato I, Tominaga S, et al. A case-control study of male colorectal cancer in Aichi Prefecture, Japan: with special reference to occupational activity level, drinking habits and family history. Jpn J Cancer Res. 1990;81(2):115–21. doi: 10.1111/j.1349-7006.1990.tb02536.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kim MK, Sasaki S, et al. Dietary patterns and subsequent colorectal cancer risk by subsite: a prospective cohort study. Int J Cancer. 2005;115(5):790–8. doi: 10.1002/ijc.20943. [DOI] [PubMed] [Google Scholar]
  27. Lampe JW, Fredstrom SB, et al. Sex differences in colonic function: a randomised trial. Gut. 1993;34(4):531–6. doi: 10.1136/gut.34.4.531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Larsson SC, Rafter J, et al. Red meat consumption and risk of cancers of the proximal colon, distal colon and rectum: the Swedish Mammography Cohort. Int J Cancer. 2005;113(5):829–34. doi: 10.1002/ijc.20658. [DOI] [PubMed] [Google Scholar]
  29. Le Marchand L, Wilkens LR, et al. Associations of sedentary lifestyle, obesity, smoking, alcohol use, and diabetes with the risk of colorectal cancer. Cancer Res. 1997;57(21):4787–94. [PubMed] [Google Scholar]
  30. Longnecker MP, Gerhardsson le Verdier M, et al. A case-control study of physical activity in relation to risk of cancer of the right colon and rectum in men. Int J Epidemiol. 1995;24(1):42–50. doi: 10.1093/ije/24.1.42. [DOI] [PubMed] [Google Scholar]
  31. Macfarlane GT, Gibson GR, et al. Comparison of fermentation reactions in different regions of the human colon. J Appl Bacteriol. 1992;72(1):57–64. doi: 10.1111/j.1365-2672.1992.tb04882.x. [DOI] [PubMed] [Google Scholar]
  32. Marcus PM, Newcomb PA, et al. Early adulthood physical activity and colon cancer risk among Wisconsin women. Cancer Epidemiol Biomarkers Prev. 1994;3(8):641–4. [PubMed] [Google Scholar]
  33. Martinez ME, Giovannucci E, et al. Leisure-time physical activity, body size, and colon cancer in women. Nurses’ Health Study Research Group. J Natl Cancer Inst. 1997;89(13):948–55. doi: 10.1093/jnci/89.13.948. [DOI] [PubMed] [Google Scholar]
  34. McCashland TM, Brand R, et al. Gender differences in colorectal polyps and tumors. Am J Gastroenterol. 2001;96(3):882–6. doi: 10.1111/j.1572-0241.2001.3638_a.x. [DOI] [PubMed] [Google Scholar]
  35. Mizoue T, Yamaji T, et al. Dietary patterns and colorectal adenomas in Japanese men: the Self-Defense Forces Health Study. Am J Epidemiol. 2005;161(4):338–45. doi: 10.1093/aje/kwi049. [DOI] [PubMed] [Google Scholar]
  36. Moradi T, Adami HO, et al. Occupational physical activity and risk for breast cancer in a nationwide cohort study in Sweden. Cancer Causes Control. 1999;10(5):423–30. doi: 10.1023/a:1008922205665. [DOI] [PubMed] [Google Scholar]
  37. Oettle G. Effect of moderate exercise on bowel habit. Gut. 1991;32(8):941–4. doi: 10.1136/gut.32.8.941. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Peters RK, Garabrant DH, et al. A case-control study of occupational and dietary factors in colorectal cancer in young men by subsite. Cancer Res. 1989;49(19):5459–68. [PubMed] [Google Scholar]
  39. Preston DL, Lubin JH, Pierce DA. Epicure: Generalized Regression Models for Epidemiologic Data. Seattle, WA: HiroSoft International Corporation; 1995. [Google Scholar]
  40. Quadrilatero J, Hoffman-Goetz L. Physical activity and colon cancer. A systematic review of potential mechanisms. J Sports Med Phys Fitness. 2003;43(2):121–38. [PubMed] [Google Scholar]
  41. Sadik R, Abrahamsson H, et al. Gender differences in gut transit shown with a newly developed radiological procedure. Scand J Gastroenterol. 2003;38(1):36–42. doi: 10.1080/00365520310000410. [DOI] [PubMed] [Google Scholar]
  42. Samad AKA, Taylor RS, et al. A meta-analysis of the association of physical activity with reduced risk of colorectal cancer. Colorectal Disease. 2005;7(3):204–213. doi: 10.1111/j.1463-1318.2005.00747.x. [DOI] [PubMed] [Google Scholar]
  43. Sandhu MS, Dunger DB, et al. Insulin, insulin-like growth factor-I (IGF-I), IGF binding proteins, their biologic interactions, and colorectal cancer. J Natl Cancer Inst. 2002;94(13):972–80. doi: 10.1093/jnci/94.13.972. [DOI] [PubMed] [Google Scholar]
  44. Saunders BP, Fukumoto M, et al. Why is colonoscopy more difficult in women? Gastrointest Endosc. 1996;43(2 Pt 1):124–6. doi: 10.1016/s0016-5107(06)80113-6. [DOI] [PubMed] [Google Scholar]
  45. Severson RK, Nomura AM, et al. A prospective analysis of physical activity and cancer. Am J Epidemiol. 1989;130(3):522–9. doi: 10.1093/oxfordjournals.aje.a115366. [DOI] [PubMed] [Google Scholar]
  46. Silvaggio T, Mattison DR. Setting occupational health standards: toxicokinetic differences among and between men and women. J Occup Med. 1994;36(8):849–54. [PubMed] [Google Scholar]
  47. Slattery ML. Physical activity and colorectal cancer. Sports Med. 2004;34(4):239–52. doi: 10.2165/00007256-200434040-00004. [DOI] [PubMed] [Google Scholar]
  48. Slattery ML, Edwards S, et al. Physical activity and colorectal cancer. Am J Epidemiol. 2003;158(3):214–24. doi: 10.1093/aje/kwg134. [DOI] [PubMed] [Google Scholar]
  49. Slattery ML, Potter J, et al. Energy balance and colon cancer--beyond physical activity. Cancer Res. 1997;57(1):75–80. [PubMed] [Google Scholar]
  50. Statistics Sweden. Reports on Statistical Coordination 1982. Stockholm, Sweden: 1995. Swedish Socioeconomic Classification; p. 4. [Google Scholar]
  51. Steindorf K, Jedrychowski W, et al. Case-control study of lifetime occupational and recreational physical activity and risks of colon and rectal cancer. Eur J Cancer Prev. 2005;14(4):363–71. doi: 10.1097/00008469-200508000-00009. [DOI] [PubMed] [Google Scholar]
  52. Takada H, Ohsawa T, et al. Changing site distribution of colorectal cancer in Japan. Dis Colon Rectum. 2002;45(9):1249–54. doi: 10.1007/s10350-004-6400-0. [DOI] [PubMed] [Google Scholar]
  53. Tavani A, Braga C, et al. Physical activity and risk of cancers of the colon and rectum: an Italian case-control study. Br J Cancer. 1999;79(11–12):1912–6. doi: 10.1038/sj.bjc.6690304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Thune I, Lund E. Physical activity and risk of colorectal cancer in men and women. Br J Cancer. 1996;73(9):1134–40. doi: 10.1038/bjc.1996.218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Toyomura K, Yamaguchi K, et al. Relation of cigarette smoking and alcohol use to colorectal adenomas by subsite: the self-defense forces health study. Cancer Sci. 2004;95(1):72–6. doi: 10.1111/j.1349-7006.2004.tb03173.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Troisi RJ, Freedman AN, et al. Incidence of colorectal carcinoma in the U.S.: an update of trends by gender, race, age, subsite, and stage, 1975–1994. Cancer. 1999;85(8):1670–6. [PubMed] [Google Scholar]
  57. Vainio H, Kaaks R, et al. Weight control and physical activity in cancer prevention: international evaluation of the evidence. Eur J Cancer Prev. 2002;11(Suppl 2):S94–100. [PubMed] [Google Scholar]
  58. Ward R, Meagher A, et al. Microsatellite instability and the clinicopathological features of sporadic colorectal cancer. Gut. 2001;48(6):821–9. doi: 10.1136/gut.48.6.821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Vena JE, Graham S, et al. Lifetime occupational exercise and colon cancer. Am J Epidemiol. 1985;122(3):357–65. doi: 10.1093/oxfordjournals.aje.a114116. [DOI] [PubMed] [Google Scholar]
  60. Voorrips LE, Goldbohm RA, 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(11):1081–92. doi: 10.1093/aje/152.11.1081. [DOI] [PubMed] [Google Scholar]
  61. Yoo KY, Tajima K, et al. Reproductive factors related to the risk of colorectal cancer by subsite: a case-control analysis. Br J Cancer. 1999;79(11–12):1901–6. doi: 10.1038/sj.bjc.6690302. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES