Abstract
Cancer is a major public health problem for Thai people. The incidence of cancer is still
high and increasing in some cancer sites. This research aimed to study risk factors for the most
common cancers i.e. cholangiocarcinoma, cervix uteri, breast, colorectum, lung, oralcavity, and
stomach cancer. This study was a cohort study, using data from the Khon Kaen cohort study
which established and recruited the cohort group during 1990-2001. The analysis was performed
as a nested case-control study design during March, 2008 to May 2010. Cohort data was linked
with the Khon Kaen cancer registry to identify cancer cases. There were 245 cholangiocarcinoma,
61 cervix uteri, 10 breast cancer, 22 colorectal cancer, 26 lung cancer, 11 oral cancer and 8
stomach cancer, 297 all sites of cancers. Those who were free from cancer and still alive until
November 2009 were selected as controls; 17,449 (5,201 males, 12,248 females).
Results from univariate and multivariate analyses were:
Cholangiocarcinoma
From the multiple conditional logistic regression and prediction model, there were 4
factors associated with cholangiocarcinoma i.e. body mass index (BMI), sex, age, and OV eggs in
stools. That is the decreasing of each 1 kg./m.2 the risk of cholangiocarcinoma is 1.05 (1/OR =
1/0.95 = 1.05) (95%CI= 1.11-1.01), male is at higher risk than female (OR= 2.45; (95% CI=1.74-
3.45), the increasing of each year of age, the risk is 1.11 (95% CI= 1.08-1.11), those who had OV
eggs in stool were at higher risk of cholangiocarcinoma (OR=1.64; 95% CI=1.16-2.34).
Considering the accuracy of prediction model of area under ROC curve; AUC is 68.70 %.
Cervix uteri cancer
From the multiple conditional logistic regression and prediction model, there were 2
factors associated with cervix cancer i.e. body mass index (BMI) and age at first sexual
intercourse. That is the increasing of each 1 kg./m.2 the risk of cervix cancer is 1.09 (95% CI=
0.99 - 1.18), each year of younger age at first sexual intercourse, the risk of cervix cancer is 1.05 (1/OR = 1/0.95 = 1.05) (95% CI = 0.94 - 1.15). Considering the accurary of prediction model of
area under ROC curve; AUC is 54.93%.
Lung cancer From the univariate analysis, risk factors for lung cancer were sex (male)
(OR=9.89; 95% CI= 3.73-26.2), age (increasing age) (p<0.001), smoking (OR=6.73; 95%
CI=2.88-15.73), and alcohol consumption. BMI 23.5 kg./m.2 or more are at lower risk of lung
cancer.
Colorectal cancer From the univariate analysis, risk factors for colorectal cancer were
sex (male) (OR=2.35; 95% CI=1.02-5.43), age (increasing age) (p<0.05), the history of having
cancer in the family (OR=2.66; 95% CI=1.13-6.27). Food consumption and BMI 23.5 kg./m.2 or
more are at lower risk of colorectal cancer.
Oral cavity cancer From the univariate analysis, risk factors for oral cavity cancer were
betel nut chewing (OR=8.88; 95% CI=1.48-53.2), age 60 year-old or more and sex (male). Male
were at lower risk of oral cavity cancer.
Breast cancer From the univariate analysis, risk factor for breast cancer was the
increasing of BMI; BMI 23.5 kg./m.2 or more compared to lower BMI (OR=1.87; 95% CI=0.36-
9.67).
Stomach cancer From the univariate analysis, the factors associated with the risk of
stomach cancer were sex (male), (OR=1.41; 95% CI=0.34-5.91), and the increasing age (p<0.05).
In conclusion; the findings in this study suggested that public health personnel should
consider and take into account that OV infestation, BMI, smoking, alcohol consumption, food
consumption for the cancer prevention and control programme.
