Bowel Motility, Laxative Use, and Risk of Colorectal Cancer
Bowel Motility, Laxative Use, and Risk of Colorectal Cancer
Selected characteristics of VITamins And Lifestyle participants as well as age- and sex-adjusted ORs and 95% CIs for the association between these characteristics and fiber and non-fiber laxative use are given in Table 1. Increasing age, female sex, history of colonoscopy, history of polyp removal, aspirin/nonsteroidal anti-inflammatory drug use, and hormone replacement therapy use were positively associated with both fiber and non-fiber laxative use. Education, moderate/vigorous physical activity, dietary fiber intake, and fruit and vegetable consumption were positively associated with fiber laxative use and inversely associated with non-fiber laxative use. Alcohol consumption was less common in those who used non-fiber and fiber laxatives. Bowel movement frequency was inversely related to non-fiber laxative use. Although constipation was positively associated with both fiber and non-fiber laxative use, the association was more pronounced for non-fiber laxative use.
Participants were followed up for a total of 494,902 person-years. Cancer cases contributed 1,948 person-years, whereas controls contributed 492,954 person-years.
Associations of bowel habits and laxative use with CRC risk are presented in Table 2. Constipation and bowel movement frequency were not statistically significantly associated with CRC risk in either the multivariate model adjusted for covariates or in the fully adjusted model, which included covariates as well as the four interrelated main exposures. Results for non-fiber and fiber laxative use were similar in the multivariate and fully adjusted models. In the full, mutually adjusted model, both low (1–4 times per year on average over the past 10 years) and high (≥5 times per year) non-fiber laxative use over the prior 10 years was associated with 43–49% increase in CRC risk relative to use less than once a year (HR=1.49, 95% CI: 1.04–2.14; HR=1.43, 95% CI: 0.82–2.48, respectively); although this is not a monotonic trend, there was a statistically significant trend across the three groups (Ptrend=0.05). High fiber laxative use over the past 10 years (≥4 days per week and ≥4 years) vs. none was associated with a statistically significant decrease in CRC risk (HR=0.44, 95% CI: 0.21–0.95); however, there was no risk reduction for low fiber laxative use (HR=1.17, 95% CI: 0.82–1.68) and the test for trend was nonsignificant (Ptrend=0.19).
Table 3, Table 4 and Table 5 present analyses of the findings related to the two types of laxative use stratified by sex, anatomic subsite, and cancer stage at diagnosis, respectively. The association between high non-fiber laxative use in the 10 years before baseline and increased cancer risk was more apparent for men than for women, for colon cancer (vs. rectal) and for local (vs. regional/distant) CRC, although the differences were not statistically significant (all Pinteraction>0.10). The inverse association between fiber laxative use and CRC risk was similar for men and women, colon and rectum, and local and regional/distant disease, with no evident heterogeneity (all Pinteraction>0.47). Finally, the risk reduction associated with high fiber laxative use was similar among those in the lower half of dietary fiber intake (HR=0.54; 95% CI: 0.20–1.47) and those in the upper half (HR=0.36, 95% CI: 0.11–1.15; Pinteraction=0.26) (data not shown).
In a sensitivity analysis, the associations between bowel habit, laxative use, and CRC risk remained virtually unchanged after excluding cases diagnosed within 1 year of follow-up.
Results
Selected characteristics of VITamins And Lifestyle participants as well as age- and sex-adjusted ORs and 95% CIs for the association between these characteristics and fiber and non-fiber laxative use are given in Table 1. Increasing age, female sex, history of colonoscopy, history of polyp removal, aspirin/nonsteroidal anti-inflammatory drug use, and hormone replacement therapy use were positively associated with both fiber and non-fiber laxative use. Education, moderate/vigorous physical activity, dietary fiber intake, and fruit and vegetable consumption were positively associated with fiber laxative use and inversely associated with non-fiber laxative use. Alcohol consumption was less common in those who used non-fiber and fiber laxatives. Bowel movement frequency was inversely related to non-fiber laxative use. Although constipation was positively associated with both fiber and non-fiber laxative use, the association was more pronounced for non-fiber laxative use.
Participants were followed up for a total of 494,902 person-years. Cancer cases contributed 1,948 person-years, whereas controls contributed 492,954 person-years.
Associations of bowel habits and laxative use with CRC risk are presented in Table 2. Constipation and bowel movement frequency were not statistically significantly associated with CRC risk in either the multivariate model adjusted for covariates or in the fully adjusted model, which included covariates as well as the four interrelated main exposures. Results for non-fiber and fiber laxative use were similar in the multivariate and fully adjusted models. In the full, mutually adjusted model, both low (1–4 times per year on average over the past 10 years) and high (≥5 times per year) non-fiber laxative use over the prior 10 years was associated with 43–49% increase in CRC risk relative to use less than once a year (HR=1.49, 95% CI: 1.04–2.14; HR=1.43, 95% CI: 0.82–2.48, respectively); although this is not a monotonic trend, there was a statistically significant trend across the three groups (Ptrend=0.05). High fiber laxative use over the past 10 years (≥4 days per week and ≥4 years) vs. none was associated with a statistically significant decrease in CRC risk (HR=0.44, 95% CI: 0.21–0.95); however, there was no risk reduction for low fiber laxative use (HR=1.17, 95% CI: 0.82–1.68) and the test for trend was nonsignificant (Ptrend=0.19).
Table 3, Table 4 and Table 5 present analyses of the findings related to the two types of laxative use stratified by sex, anatomic subsite, and cancer stage at diagnosis, respectively. The association between high non-fiber laxative use in the 10 years before baseline and increased cancer risk was more apparent for men than for women, for colon cancer (vs. rectal) and for local (vs. regional/distant) CRC, although the differences were not statistically significant (all Pinteraction>0.10). The inverse association between fiber laxative use and CRC risk was similar for men and women, colon and rectum, and local and regional/distant disease, with no evident heterogeneity (all Pinteraction>0.47). Finally, the risk reduction associated with high fiber laxative use was similar among those in the lower half of dietary fiber intake (HR=0.54; 95% CI: 0.20–1.47) and those in the upper half (HR=0.36, 95% CI: 0.11–1.15; Pinteraction=0.26) (data not shown).
In a sensitivity analysis, the associations between bowel habit, laxative use, and CRC risk remained virtually unchanged after excluding cases diagnosed within 1 year of follow-up.
Source...