Association between physical activity, fiber intake, and other lifestyle variables and constipation in a study of women

Association Between Physical Activity, Fiber Intake,and Other Lifestyle Variables and Constipation in aStudy of WomenLaurent Dukas, M.D., Walter C. Willett, M.D., and Edward L. Giovannucci, M.D., Sc.D.
Departments of Nutrition and Epidemiology, Harvard School of Public Health, Boston, Massachusetts;Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital and Harvard MedicalSchool, Boston, Massachusetts; and Geriatric Department, University Hospital, Basel, Switzerland OBJECTIVE: Because the risk factors for constipation in the
2.5 million physician visits annually The greatest general population are poorly defined, we examined the number of physician visits for constipation occurs in association between age, body mass index, and selected adults 65 yr of age and older Patients tend to define lifestyle and dietary factors with the prevalence of consti- constipation in terms of function, such as straining, in- pation in the Nurses’ Health Study.
complete defecation, and hard stool consistency Incontrast, physicians and investigators usually define con- METHODS: A total of 62,036 women, aged 36 – 61 yr and
stipation by stool frequencies twice weekly or less. Ac- free of cancer, responded to mailed questionnaires in 1980and 1982, which assessed bowel movement frequency, di- cording to an international classification, individuals with etary, and lifestyle factors. Constipation was defined as two fewer than three bowel movements per week may be considered to be constipated Depending on thedefinition of constipation, the reported prevalence is as RESULTS: A total of 3327 women (5.4%) were classified as
high as 34% in older women (over age 65 yr) and having constipation. Age and body mass index were in- versely associated with constipation (p for trend Ͻ 0.0001).
Despite the high prevalence, morbidity, and substantial In multivariate analysis, women who reported daily physical cost as a result of physician visits and over-the-counter activity had a lower prevalence of constipation (prevalence laxatives, the etiology of constipation or low bowel move- ratio [PR] ϭ 0.56, 95% CI ϭ 0.44 – 0.70). Women in the ment frequency among adults in the general population has highest quintile of dietary fiber intake (median intake 20 received little systematic investigation. Numerous possible g/day) were less likely to experience constipation (PR ϭ etiological factors have been proposed for idiopathic con- 0.64, 95% CI ϭ 0.57– 0.73) than women in the lowest stipation, but their evaluation has frequently been from quintile (median intake 7 g/day). The PR for constipation for uncontrolled studies, short-term interventions, or anecdotal women who both reported daily physical activity and were reports. Data from large-scale epidemiological studies are in the highest quintile of fiber intake was 0.32 (95% CI ϭ sparse. Using data from the first National Health and Nu- 0.20 – 0.54) compared with those with physical activity less trition Examination Survey, constipation was found to be than once weekly and those who were in the lowest quintile more frequent in women, those with daily inactivity, low of fiber intake. Higher frequency of aspirin intake was income, poor education, higher consumption of coffee and associated with an increasing prevalence of constipation tea, and with some dietary variables, including lower fruit (PR ϭ 1.38, 95% CI ϭ 1.17–1.62). Current smoking and alcohol use were inversely associated with constipation.
Because of the paucity of relevant data, we investi- CONCLUSIONS: Our data suggest that moderate physical ac-
gated the associations between age, body mass index tivity and increasing fiber intake are associated with sub- (BMI), physical activity, dietary fiber intake, and other stantial reduction in the prevalence of constipation in selected lifestyle variables (suspected or known risk fac- women. (Am J Gastroenterol 2003;98:1790 –1796. 2003 tors) with prevalence of constipation, defined as two or fewer bowel movements weekly, in a large cohort ofwomen. Because of the detailed information on multipledemographic, dietary, and lifestyle data, and the large INTRODUCTION
sample size, we were able to control for many factors Constipation represents the most common outpatient diges- simultaneously and to determine the independent influ- tive complaint in the United States and accounts for about Lifestyle Variables and Constipation in Women
MATERIALS AND METHODS
the alternative analysis that included laxative users, laxativeuse was analyzed as a categorical variable (daily/weekly/ Study Population and Assessment of Variables
monthly/less than monthly laxative use).
The participants were a subset of women of the Nurses’ For age, BMI, postmenopausal status and hormone use, Health Study, an ongoing prospective study among 121,700 smoking history, and aspirin use, we used the status of 1982, female registered nurses in the United States. At enrollment and for physical activity, alcohol, coffee, and fiber intake, in 1976, participants, who were 30 –55 yr old and married, we used the covariate status of 1980, the year the food- completed a mailed questionnaire that inquired about known frequency questionnaire was first administered. BMI or suspected risk factors for cancer and cardiovascular dis- physical activity fiber alcohol and coffee ease, anthropometric measures, physical activity, reproduc- have been shown previously to be reported with mod- tive factors, other lifestyle exposures, family history of erate-to-high accuracy using our methodology.
related diseases, and medical history. Beginning in 1980, a Prevalence ratios (PRs) for constipation were estimated semiquantitative food-frequency questionnaire was admin- by the ORs and 95% CIs for each category of dietary and istered. Physical activity was assessed through the following other variables. In the multivariate analysis, the trends were questions: “At least once a week, do you engage in any tested using a single ordinal variable in a logistic regression regular activity similar to brisk walking, jogging, bicycling, model to represent the ordered categories of a prediction etc, long enough to work up a sweat? If yes, how many variable. The p values for the trends are two sided.
times per week?” Details of the design and follow-up of thiscohort have been described previously Women provided information about their bowel move- ment frequency on the 1982 Nurses’ Health Study question- For all variables considered, results for the age-adjusted and naire that was mailed to the entire cohort. Possible answers for bowel movement frequency were more than once daily/daily/every other day/every 3– 4 days/every 5– 6 days/once Age
a week, or less. On the 1982 questionnaire, we also queried In the multivariate analysis, age was inversely associated participants about use of laxatives, including softeners, with constipation (p for trend Ͻ 0.0001). Compared with 35 bulking agents, and suppositories. Possible answers were to 39-yr-old women, women aged 60 yr and older had a laxative use daily/at least once a week/one to four times a multivariate-adjusted PR of constipation of 0.41 (95% CI ϭ month/less than once a month/never. The reasons for laxa- tive use and type of laxatives were not assessed.
BMI and Physical Activity
The analyses for this study were limited to women who BMI was inversely related to constipation (p for trend Ͻ responded to the bowel movement frequency question in 0.0001); women with a BMI Ͼ 29 kg/m2 had a multivariate- 1982 (n ϭ 92,106). We excluded women who reported a adjusted PR of 0.48 (95% CI ϭ 0.39 – 0.58) compared with diagnosis of ulcerative colitis before 1982 (n ϭ 651) or who women with a BMI Ͻ 21 kg/m2. There was no association reported a cancer diagnosis other than nonmelanoma skin up to a BMI of 26 kg/m2; then, the risk decreased with cancer preceding and including 1982 (n ϭ 3,015). Because increasing BMI. Increased physical activity was associated laxatives influence bowel movement frequency, we ex- with a reduced prevalence of constipation (p for trend Ͻ cluded laxative users from the main analysis (n ϭ 26,404), 0.0001). Women who reported physical activity two to three but included them in an alternative analysis. A total of times per week had a multivariate-adjusted PR of 0.65 (95% 62,036 women were included in the main analysis; of these, CI ϭ 0.59 – 0.71) and women who reported daily physical 3,327 reported bowel movement frequency every third day activity had a multivariate adjusted PR of 0.56 (95% CI ϭ 0.44 – 0.70) compared with women with physical activityless than once weekly.
Statistical Analysis
We used logistic regression models to control simulta-
Medication, Including Aspirin Intake and
neously for multiple variables We included in the Postmenopausal Hormone Use
models variables that are suspected or established risk fac- Total number of medications was not associated with PR of tors for constipation, defined as two or fewer bowel move- constipation (p for trend ϭ 0.09). Women who used three or ments weekly. Variables included age (six 5-yr categories), more drugs per day (aspirin and hormone replacement not BMI (10 categories), nonoccupational physical activity (fre- included, but controlled for) had a PR of constipation of quency of exposure; five categories), smoking history (nev- 1.00 (95% CI ϭ 0.75–1.34) compared with women who er, past, current), aspirin use frequency (five categories), took no drugs. Increasing frequency of aspirin intake was number of medications (five categories), and intake of al- associated with a higher prevalence of constipation (p for cohol (six categories), coffee (six categories), and dietary trend Ͻ 0.0001). Women who used 15 or more aspirin fiber (quintiles). We also considered menopausal status and tablets per week had a multivariate adjusted PR of 1.38 postmenopausal hormone replacement use as covariates. In (95% CI ϭ 1.17–1.62) for constipation compared with Dukas et al.
Table 1. Association Between Physical Activity, Fiber Intake, and Other Lifestyle Factors and Prevalence of Constipation Among 60,036
Women
Ͻ21 21–21.9 22–22.9 23–23.9 24–24.9 25–25.9 26–26.9 27–27.9 28–28.9 Ref. 0.95–1.28 0.92–1.23 0.87–1.17 0.86–1.17 0.88–1.20 0.73–0.99 0.69–0.98 0.50–0.73 0.39–0.58 * Multivariate PR is from a logistic regression model containing age, BMI, smoking status, postmenopausal hormone use and status, physical activity, aspirin use, number ofmedications, alcohol, coffee, and fiber intake.
Lifestyle Variables and Constipation in Women
Figure 1. Combination of physical activity and fiber intake and prevalence of constipation.
women who never used aspirin. Postmenopausal hormone daily fiber intake was relatively low for the whole cohort (12 use was not associated with constipation (multivariate-ad- g daily), and only 1% had a fiber intake of more than 30 g justed PR ϭ 1.07, 95% CI ϭ 0.95–1.23) for women with current hormone use compared with women who never used In a further analysis we classified women simul- taneously by both physical activity and dietary fiber intake.
In the multivariate analysis, the PR for constipation for Smoking, Alcohol, and Coffee Consumption
women who both reported daily physical activity and were Relative to never smoking, current smoking was associated in the highest quintile of fiber intake was 0.32 (95% CI ϭ with a slightly reduced multivariate-adjusted risk (PR ϭ 0.20 – 0.54) compared with those with physical activity less 0.81, 95% CI ϭ 0.74 – 0.89), whereas past smoking had no than once weekly and those who were in the lowest quintile association with constipation (PR ϭ 1.04, 95% CI ϭ 0.95– of fiber intake. For women who reported physical activity 1.13). Daily alcohol consumption was inversely associated two to three times per week and were in the highest quintile with constipation (p for trend Ͻ 0.0001), though at least one of fiber intake, the PR of constipation was 0.41 (95% CI ϭ drink (approximately 12 g of alcohol) per day was required 0.33– 0.50) compared with those with physical activity less to observe the lower prevalence. Women with a median than once weekly and those who were in the lowest quintile daily alcohol intake of 30.4 g had a multivariate adjusted PR of fiber intake. The alternative analysis including laxative of 0.66 (95% CI ϭ 0.57– 0.77) compared with women who users showed similar results as the main analysis (data not did not drink alcohol. Coffee intake seemed to have a nonlinear relation with constipation. Women who drank sixor more cups of coffee daily had a slightly increased mul-tivariate-adjusted PR of 1.17 (95% CI ϭ 1.02–1.34) for DISCUSSION
constipation compared with women who did not drink cof-fee. Low coffee intake (one cup/day or less), however, was Constipation is common in Western countries possibly inversely associated with constipation (PR for one cup/day related to diet and lifestyle factors. Our study found that ϭ 0.79, 95% CI ϭ 0.69–0.91) compared with women who regular physical exercise and higher fiber intake were asso- ciated with reduced risk of constipation even after control-ling for numerous factors. Women who were physically Dietary Fiber
active daily and had approximately 20 g daily of fiber had a Higher dietary fiber intake was associated with a decreased 3-fold lower prevalence of constipation compared with prevalence of constipation (p for trend Ͻ 0.0001), even women who rarely exercised and had about 7 g daily of though the median fiber intake for women in the highest quintile was 20 g daily, which is below the recommended Some authors suggest that low bowel movement fiber intake of 30 g/day. Women in the highest quintile of frequency is associated with functional abnormalities of the fiber intake had a multivariate adjusted PR of 0.64 (95% CI colon, which usually tend to worsen with aging. However, ϭ 0.57–0.73) compared with women in the lowest quintile in our study, as well as in other studies age was of fiber intake (median daily fiber intake of 7 g). The median inversely associated with bowel movement frequency. Co- Dukas et al.
lonic transit may slow with aging but is highly variable was associated with a 2-fold risk of constipation Ex- Although the majority of older people have stool frequen- ercise increases propulsive movements in the large intestine cies in the normal range, laxative use increases with aging and hormonal changes and mechanical effects during The study of Whitehead et al. showed that the use exercise may alter GI function. Other mechanisms for the of laxatives is affected more by the desire to have a bowel influence of exercise are possible. For example, higher en- movement than by actual frequency Furthermore, the ergy expenditure increases energy intake requirements, increase in laxative use with aging may result from which may influence the total amount of dietary fiber con- the elderly having more defecation problems such as strain- sumed. However, in our analyses, controlling for fiber and ing or suffering from hard stool consistency further for energy intake did not appreciably change the Another possible explanation for the inverse association results for physical activity. As shown in the between age and low bowel movement frequency in women associations for physical activity and fiber and bowel move- may be the decreasing estrogen plasma levels after meno- ment frequency were independent. The lowest risk was in pause. Increasing estrogen in the luteal phase of the men- the most active women with the highest fiber intake. Thus, strual cycle is associated with prolonged intestinal transit the combination of a high-fiber diet and a high level of time and diseases related to increased estrogen plasma physical activity would tend to lead to the lowest risk for levels have been shown to be associated with low bowel movement frequency (27–29). Data are not available about The linearly increasing risk of constipation with increas- bowel movement frequency and aging in men in our study.
ing aspirin intake was also seen in two cross-sectional However, most studies show that men are less likely to studies and in several large case– control studies suffer from constipation than women (4, 6, 7, 30). In our and trials Constipation as a side effect is also study, current postmenopausal hormone use was associated known for other nonsteroidal anti-inflammatory drugs with a slight, though not statistically significant, increased The mechanism by which nonsteroidal anti-inflamma- risk of low bowel movement frequency.
tory drugs reduce bowel movement frequency remains un- Increasing BMI was inversely associated with low bowel movement frequency. In other studies BMI had Our finding that smoking is inversely associated with no influence on colonic transit times or bowel movement constipation is supported by two other studies In a frequency. The reason for this inverse association is not study by Cummings et al. smoking withdrawal in- clear. However, within the normal range of BMI (Ͻ26 duced constipation, and in a study by Rausch et al. kg/m2), we observed no relation with constipation.
transdermal application of nicotine increased rectosigmoidal We found an inverse association between physical activ- emptying, suggesting that nicotinic acid may influence ity and low bowel movement frequency. In our study, phys- ical activity only two to six times a week was associated Alcohol intake exceeding 12 g/day was inversely associ- with a 35% lower risk of constipation and daily activity with ated with constipation. Alcohol withdrawal has been asso- a 44% lower risk relative to less than once daily. A review ciated with increasing intestinal transit time and heavy of studies of the influence of exercise on the GI tract alcohol consumption is associated with diarrhea be- indicates that the urge to defecate and bowel movements are cause of inflammation of the colonic mucosa and neurotox- frequently reported during and after exercise However, icity observed with alcohol However, we observed results from other studies on the association between low an inverse association at moderate levels that are unlikely to bowel movement frequency and physical activity are incon- be associated with these adverse effects of heavy drinking.
sistent. One crossover trial found a significant decrease Coffee intake of more than six cups per day was associ- in whole gut transit time during the exercise period, but ated with a slightly increased risk of low bowel movement defecation frequency did not change. Another trial in frequency, whereas low-to-moderate consumption was as- healthy young men and one trial in young healthy sociated with a reduced risk of constipation. In the study by subjects found no association between moderate physical Sandler et al. constipated subjects reported higher con- activity and bowel transit All three trials included sumption of coffee compared with controls. Coffee is small numbers of subjects, and none included reports of the known to induce an increase in colonic motility limited to subjects’ bowel habits. In another study two groups, the rectosigmoid region within 4 min of ingestion and last- one of which trained for 30 min at 70 – 80% of age-predicted ing at least 3 min This influence of coffee on rectosig- maximal heart rate three times weekly for 6 wk, were tested moid responses appeared primarily in men and women who before and after the training program. Although the running claimed that coffee induced a desire to defecate (53% of group’s average transit time decreased from 35 to 24 h, the women and 19% of men). The speed of the response indi- control group remained unchanged at 45 h Contrary to cated that coffee may induce a gastrocolonic response, pos- most trials, several other cross-sectional studies found an sibly by acting on receptors in the stomach or small bowel inverse association between physical activity and constipa- and mediated by neural mechanisms or by GI hormones.
Our observed association between large amounts of coffee Nutrition Examination Survey, low physical activity level and low constipation might be a result of coffee-induced Lifestyle Variables and Constipation in Women
diuresis leading to dehydration, which has been associated ACKNOWLEDGMENTS
with constipation or to other unknown mechanisms.
Associations between dietary factors and constipation The work reported in this manuscript was supported by should be interpreted with caution because the reported diet CA40356, the main Nurses’ Health Study grant. The study may reflect changes in food decision caused by constipation.
was last reviewed and approved by the Human Research In our study, fiber intake was inversely associated with low Committee on July 18, 2002 (Protocol # 1999-P-011114/7).
bowel movement frequency. In intervention trials, addingfiber to specific diets increased bowel movement frequency Reprint requests and correspondence: Edward Giovannucci,
The positive relationship between fiber intake and M.D., Sc.D., Harvard School of Public Health, 665 Huntington stool weight is well established Other studies found a weak or null association between fiber intake and bowel Received Nov. 21, 2002; accepted Feb. 24, 2003. movement frequency possibly because ofsmall sample size or a tendency for people who sufferedfrom constipation to increase their fiber intake. In our study, REFERENCES
the combination of higher fiber intake and regular physical 1. Sonnenberg A, Koch TR. Physician visits in the United States activity was associated with a large decrease in prevalence for constipation: 1958 to 1986. Dig Dis Sci 1989;34:606 –11.
of constipation. We are not aware of any study that exam- 2. Sandler RS, Drossman DA. Bowel habits in young adults not ined the combination of these variables.
seeking health care. Dig Dis Sci 1987;32:841–5.
The large size of our study (62,036 women) allowed us to 3. Preston DM, Lennard-Jones JE. Severe chronic constipation of young women: ‘idiopathic slow transit constipation.’ Gut examine several demographic and dietary characteristics for a more complete description of low bowel movement fre- 4. Connell AM, Hilton C, Irvine G, et al. Variation of bowel habit quency in U.S. women. A major strength of our study is the in two population samples. Br Med J 1965;5470:1095–9.
ability to control for known or suspected risk factors for low 5. Stewart RB, Moore MT, Marks RG, et al. Correlates of con- bowel movement frequency. One limitation of our study is stipation in an ambulatory elderly population. Am J Gastro-enterol 1992;87:859 –64.
that the question about bowel movement frequency was 6. Hale WE, Perkins LL, May FE, et al. Symptom prevalence in based strictly on self-report. However, nurses in this popu- the elderly. An evaluation of age, sex, disease, and medication lation have been shown to report a variety of medical factors use. J Am Geriatr Soc 1986;34:333–40.
with high accuracy. Although there are alternative ways to 7. Whitehead WE, Drinkwater D, Cheskin LJ, et al. Constipation assess constipation, bowel movement frequency can be used in the elderly living at home. Definition, prevalence, and most feasibly in a large population. However, bowel move- relationship to lifestyle and health status. J Am Geriatr Soc1989;37:423–9.
ment frequency represents only one component of the spec- 8. Sandler RS, Jordan MC, Shelton BJ. Demographic and dietary trum of constipation, and whether the same relationships we determinants of constipation in the US population. Am J observed would hold for other clinical subtypes of consti- 9. Giovannucci E, Colditz GA, Stampfer MJ, et al. A prospective Another limitation is that we may have underestimated study of cigarette smoking and risk of colorectal adenoma and absolute fiber intake because our food-frequency question- colorectal cancer in U.S. women. J Nat Cancer Inst 1994;86:192–9.
naire in 1980 did not assess all sources of fiber. However, 10. Platz EA, Martinez ME, Grodstein F, et al. Parity and other this diet questionnaire adequately ranked individuals by reproductive factors and risk of adenomatous polyps of the quintile of fiber intake, and in a subset of participants who distal colorectum (United States). Cancer Causes Control completed detailed daily dietary records, the absolute in- takes of fiber assessed by questionnaire and by dietary 11. Cupples LA, D’Agostino RB, Anderson K, et al. Comparison of baseline and repeated measure covariate techniques in the records were comparable We do not know if higher Framingham Heart Study. Stat Med 1988;7:205–22.
intakes of fiber would further reduce risk of constipation.
12. Rimm EB, Stampfer MJ, Colditz GA, et al. Validity of self- Another limitation is also that we did not assess fluid intake reported waist and hip circumferences in men and women.
in the 1980 food-frequency questionnaire, so we could not exclude residual confounding by water intake.
13. Chasan-Taber S, Rimm EB, Stampfer MJ, et al. Reproducibil- A variety of lifestyle factors may influence the risk of ity and validity of a self-administered physical activity ques-tionnaire for male health professionals. Epidemiology 1996; constipation. Some of these factors, particularly regular physical activity and higher fiber intake, seem to be bene- 14. Hu FB, Rimm E, Smith-Warner SA, et al. Reproducibility and ficial regarding constipation. Combining regular physical validity of dietary patterns assessed with a food-frequency activity and increasing fiber intake may represent an impor- questionnaire. Am J Clin Nutr 1999;69:243–9.
tant, inexpensive, and feasible therapeutic measure for this 15. Giovannucci E, Colditz G, Stampfer MJ, et al. The assessment of alcohol consumption by a simple self-administered ques- widespread digestive complaint. The possible influence of tionnaire. Am J Epidemiol 1991;133:810 –7.
other factors, including BMI, aspirin use, alcohol, and cof- 16. Grobbee DE, Rimm EB, Giovannucci E, et al. Coffee, caffeine fee, on bowel movement frequency need to be considered in and cardiovascular disease in men. N Engl J Med 1990;323: relation to their other effects on overall health status.
Dukas et al.
17. Burkitt DP. Epidemiology of cancer of the colon and rectum.
running program on bowel transit time. J Sports Med Phys 18. Waller SL, Misiewicz JJ. Colonic motility in constipation and 37. Campbell AJ, Busby WJ, Horwath CC. Factors associated diarrhea. Scand J Gastroenterol 1972;7:93–6.
with constipation in a community based sample of people aged 19. Meunier P, Rochas A, Lambert R. Motor activity of the 70 years and over. J Epidemiol Comm Health 1993;47:23–6.
sigmoid colon in chronic constipation: Comparative study 38. Holdstock DJ, Misiewicz JJ, Smith T, et al. Propulsion (mass with normal subjects. Gut 1979;20:1095–101.
movements) in the human colon and its relationship to meals 20. Frieri G, Parisi F, Corazziari E, et al. Colonic electromyogra- and somatic activity. Gut 1970;11:91–9.
phy in chronic constipation. Gastroenterology 1983;84:737– 39. Talley NJ, Weaver AL, Zinsmeister AR, et al. Functional constipation and outlet delay: A population-based study. Gas- 21. Bueno L, Fioramonti J, Ruckebusch Y, et al. Evaluation of colonic myoelectrical activity in health and functional disor- 40. Meade TW, Roderick PJ, Brennan PJ, et al. Extra-cranial bleeding and other symptoms due to low dose aspirin and low 22. Milne JS, Williamson J. Bowel habit in older people. Gerontol intensity oral anticoagulation. Throm Haemostas 1992;68:1–6.
41. Harker LA, Boissel JP, Pilgrim AJ, et al. Comparative safety 23. Dent OF, Goulston KJ, Zubrzycki J, et al. Bowel symptoms in and tolerability of clopidogrel and aspirin: Results from CA- an apparently well population. Dis Colon Rectum 1986;29: PRIE. CAPRIE Steering Committee and Investigators. Clopi- dogrel versus aspirin in patients at risk of ischaemic events.
24. Everhart JE, Go VL, Johannes RS, et al. A longitudinal survey of self-reported bowel habits in the United States. Dig Dis Sci 42. Wright MG. Micro-encapsulated aspirin (Levius) compared with aloxiprin (Palaprin Forte) in the treatment of rheumatoid arthritis. J Int Med Res 1976;4:272–5.
25. Madsen JL. Effects of gender, age, and body mass index on 43. Peto R, Gray R, Collins R, et al. Randomised trial of prophy- gastrointestinal transit times. Dig Dis Sci 1992;37:1548 –53.
lactic daily aspirin in British male doctors. BMJ 1988;296: 26. Wald A, Van Thiel DH, Hoechstetter L, et al. Gastrointestinal transit: The effect of the menstrual cycle. Gastroenterology 44. Cummings KM, Giovino G, Jaen CR, et al. Reports of smok- ing withdrawal symptoms over a 21 day period of abstinence.
27. Corraziari E, Bausano G, Torsoli A, et al. Italian cooperative study on chronic constipation. Ital J Gastroenterol 1981;13: 45. Rausch T, Beglinger C, Alam N, et al. Effect of transdermal application of nicotine on colonic transit in healthy nonsmok- 28. Preston DM, Rees LH, Lennard-Jones JE. Gynaecological ing volunteers. Neurogastroenterol Motil 1998;10:263–70.
disorders and hyperprolactinaemia in chronic constipation.
46. Bouchoucha M, Nalpas B, Berger M, et al. Recovery from disturbed colonic transit time after alcohol withdrawal. Dis 29. Watier A, Devroede G, Duranceau A, et al. Constipation with colonic inertia. A manifestation of systemic disease? Dig Dis 47. Papa A, Tursi A, Cammarota G, et al. Effect of moderate and heavy alcohol consumption on intestinal transit time. Panmin- 30. Jones RH, Tait CL. Gastrointestinal side-effects of NSAIDs in the community. Br J Clin Pract 1995;49:67–70.
48. Brown SR, Cann PA, Read NW. Effect of coffee on distal 31. Towers AL, Burgio KL, Locher JL, et al. Constipation in the colon function. Gut 1990;31:450 –3.
elderly: Influence of dietary, psychological, and physiological 49. Davies L, Holdsworth MD, MacFarlane D. Dietary fibre in- factors. J Am Geriatr Soc 1994;42:701–6.
takes in the United Kingdom before and after retirement from 32. Moses FM. The effect of exercise on the gastrointestinal tract.
work. Hum Nutr Appl Nutr 1986;40:431–9.
50. Astrup A, Vrist E, Quaade F. Dietary fibre added to very low 33. Oettle GJ. Effect of moderate exercise on bowel habit. Gut calorie diet reduces hunger and alleviates constipation. Int J 34. Coenen C, Wegener M, Wedmann B, et al. Does physical 51. Rigaud D, Ryttig KR, Leeds AR, et al. Effects of a moderate exercise influence bowel transit time in healthy young men? dietary fibre supplement on hunger rating, energy input and faecal energy output in young, healthy volunteers. A random- 35. Bingham SA, Cummings JH. Effect of exercise and physical ized, double-blind, cross-over trial. Int J Obes 1987;11:73–8.
fitness on large intestinal function. Gastroenterology 1989;97: 52. Chen HL, Haack VS, Janecky CW, et al. Mechanisms by which wheat bran and oat bran increase stool weight in hu- 36. Cordain L, Latin RW, Behnke JJ. The effects of an aerobic mans. Am J Clin Nutr 1998;68:711–9.

Source: http://diacom.com.br/manuaisprodutos/referenciasbemvitalmultifibras/4_Dukas_2003_Association_between_physical_activity_fiber_intake_and_constipation.pdf