| 1 Background and ObjectiveColorectal cancer(CRC) is the third most common malignant cancer in theworld, and it ranks second in the developed country. In Europe, it has 370,000 CRCcases, 12% of the total cancer incidence, of which the number of death is about200,000. China is a low-incidence country around the world, but the increasing trendis obvious, especially in some large cities. For example, in Beijing between 1982～1997, the men-standardized incident rates of colon cancer and rectal cancer increasedby 53.30% and 23.8%; the women-standardized incident rates of colon cancer andrectal cancer increased by 16.9% and 15.3%.The incidence of CRC is associated with social environment, life styles, geneticfactors and so on. Many epidemiological studies shows age, histories of bowel polyps, colitis and gallbladder operation increase the risk of CRC. From the epidemiologicalstudies of migration and datum of adjusted dietary factors, CRC is a type ofenvironment-and-life-styles cancer and the main pathogenies are life styles, dietaryand environment.Jiashan is an area with a high incidence rate of CRC in China. In order toinvestigate the relation between dietary factors and CRC, we conduct apopulation-based case-control study in Jiashan County.2 Material and methodsA colorectal cancer screening cohort in Jiashan County, Zhejiang Province with 64693 persons enrolled in which covered 10 towns and was established during 1989to 1990 was treated as the study population. 207prevalent patients of colorectal cancerdiagnosed from May 1990 to May 2005 in the cohort made up of the ease group. Bymeans of simple random sampling method, a control sample with 414 persons, a2-folded sample size of the case group was selected.A constructed questionnaire elicited information on the demographic condition,diet, and history of selected diseases, et al. All subjects were interviewed face-to-faceby trained interviewers. For each food, amounts consumed were estimated accordingto models of the more frequently consumed foods for the accuracy of survey.The amounts (X) of nutrients per 100g-food were calculated with the formula asfollow: X=A~*EP/100. 'A' represented the amounts of nutrients per 100g-food whichwere useful; 'EP' represented the amounts of the useful food per 100g-food. 'EP' and'X'were all cited from Chinese food contents list 2002 and Chinese food contents list2004. Some season food were adjusted with season.The data was input with software of EPIDATA twice, verified with the commandof VALIDATE DUPLICAT FILES and then processed with software of SPSS13.0.The statistical methods included chi-square test, nonparametric test because of biaseddistributions of the data, CHAID (Chi-squared Automatic Interaction Detector), andunconditional logistic regression models.3 Results3.1 General condition and colorectal cancerOne-factor analysis showed that there were statistical associations betweencolon cancer and age, occupation, while the association between rectal cancer andage, occupation had not been found. Furthermore they were not added into theclassified tree model.3.2 Disease history and colorectal cancerOne-factor analysis showed that there were statistical associations between colon cancer and histories of polypus (x~2=19.150, p＜0.01), colorectal inflammation(x~2=7.032, p＜0.01), schistosomiasis(x~2=23.292, p＜0.01). Meanwhile there werestrong associations between rectal cancer and histories of polypus(x~2=26.199,p＜0.01), colorectal inflammation(x~2=7.553, p＜0.01), schistosomiasis(x~2=20.226,p＜0.01). Model showed that rectal cancer positive rate was higher among the peoplewho had schistosomiasis history. Through multi-factors analysis we found thatpeople who had polypus(OR 13.031, 95%CI: 2.411～70.428 ) andschistosomiasis(OR 4.839, 95%CI: 2.647～8.844) history may had higher risk tosuffer from colon cancer. Same results occurred between rectal cancer and historiesofpolypus and schistosomiasis.3.3 Life condition and colorectal cancerIt showed that there were statistical associations between colon cancer anddrinking firewater(x~2=4.529, p＜0.05), drinking year(z=1.965, p＜0.05), passivitysmoking(x~2=4.58, p＜0.05), while rectal cancer only had association with passivitysmoking(x~2=4.84, p＜0.05).Model showed people who never smoking(7.1%) or justsmoking one type cigarette(33.3%) had lower colon cancer positive rate than peoplesmoking two types, and all above factors had not been added into multi-factorsanalysis model.3.4 Food habit and colorectal cancerIt had association between rectal cancer and rigidity of food(z=2.360, p＜0.05).Model also showed that people often eat soft(21.8%) or moderate(44%) food hadlower rectal cancer positive rate than people often eat hard food.3.5 Variety of food and eolorectal cancerThere were statistical associations between colon cancer and rice (z=2.564,p＜0.05), potato (z=2.405, p＜0.05), legume (z=2.207, p＜0.05 ), fruit (z=2.567,p＜0.05), watermelon (z=3.262, p＜0.05), dried fruit (z=3.110, p＜0.05 ), whilerectal cancer had statistical association with rice (z=4.335, p＜0.05), fieshwater fish(z=2.207, p＜0.05 ), vegetable (z=5.124, p＜0.05), potato (z=2.455, p＜0.05),legume (z=4.302, p＜0.05), legume production (z=1.588, p＜0.05), fruit (z =2.414, p＜70.05), banana (z=1.976, p＜0.05), watermelon (z=2.885, p＜0.05).Model showed that amount of orange, shallot, edible bacteria can makestatistical influence on colon cancer positive rate when calcium≤4.42 (26302.68mg/y). Frequency of orange eating was an important factor for coloncancer, that is people often eat orange (18%) observably had lower colon cancer ratethan others(38%). People ingested more shallot(7.1%) had lower colon cancer ratethan other(27.3%).Amount of banana(p＜0.05) and egg(p=0.05) may make statisticalinfluence on colon cancer rate when calcium＞4.42.Model showed ingest level of calcium was also an important factor for rectalcancer. Along with the heighten of calcium's ingest level, rectal cancer rate reduce to34.9%(calcium＜4.33),7%(4.33＜calcium≤4.98), 17.3 %(4.98＜calcium≤5.11), 5.3%(calcium＞5.11). Furthermore, different ingest level of calcium made differentinfluence on rectal cancer. Amount of vegetable, potato, dried fruit and rigidity offood made statistical influence on rectal cancer rate when calcium≤4.33(21379.62mg/y). People who ingest more vegetable (27.6%) had lower rectalcancer rate than others (55.1%).Among the crowed with high vegetable ingest level,people often eat soft or moderate food had lower rectal cancer rate than people ofteneat hard food. When 4.33＜calcium≤4.98(95499.26mg/y),apple,peach,schistosomiasis all had statistical association with rectal cancer.The results from multi-factors analysis showed orange and shallot decreased therisk of colon cancer, while banana increased the risk. Vegetable was protective factorof colon cancer, while potato was a risk factor.3.6 Energy, nutriment and colorectal cancerOne-factor analysis showed that there were statistical associations between coloncancer and energy (z=1.962, p＜0.05), food's fibre (z=2.340, p＜0.05), folacin(z=2.089, p＜0.05), calcium (z=5.770, p＜0.05), selenium (z=2.104, p＜0.05).Rectal cancer had statistical association with energy (z=2.406, p＜0.05), folacin(z=2.278, p＜0.05),calcium (z=7.000, p＜0.05),zinc (z=2.499, p＜0.05),selenium(z=2.358, p＜0.05). The classified tree analysis demonstration meals calcium located on the maintrunk of rectal cancer classify tree (classified tree structure first), the low ingest levelof calcium's rectal cancer rate remarkably to be higher than high ingest level;Moreover the multi-factors analysis model also demonstrated the meals calciumtakes in comparatively high to reduce colorectal cancer risk. Therefore the calciumwas the most important protection factor of colorectal cancer.Also, the multi-factors analysis showed that the calcium associated withcolorectal cancer, which clewed the increasing of calcium intake may decrease therisk of colorectal cancer.4 ConclusionsBased on epidemiological questionnaire survey on Jiashan County (colorectalcancer high-incidence area), we know the life styles, variety of food and the situationof nutriment intake in Jiashan and find out the relationship between colorectal cancerand nutriment intake, using case-control study. The results show that the oranges andshallot are protective factors of colon cancer, and banana is its risk factor. Thevegetables is a protective factor of rectal cancer arid potato is its risk factor. In totalthe calcium is an important protective factor of colorectal cancer, while histories ofpolypus and schistosomiasis are risk factors.
|
PDF Full Text Request