Modifiable Factors and Colon Cancer Risk in Thai Population

To demonstrate the possible impact of modifiable factors on colon cancer development in Thai population, we conducted this case-control study from June 2016 until June 2017. The study was conducted in 11 Thai provincial hospitals. The hospitals in this study were selected by stratification by regions. Patients included 504 ones who were newly diagnosed with colon cancer within 1 month. In the control group, 997 health individuals were enrolled. Both case and control were adjusted by age. The results of this study showed that age and socioeconomic factors were associated with colon cancer risk. In addition, it was found that family history of colon cancer had very high association with colon cancer risk. Behavioral factors, including smoking, inadequate physical exercise, and salty food consumption were associated with colon cancer. We detected no association between obesity, alcohol consumption, and colon cancer. The results suggested that colon cancer might have higher association with genetic factors than behavioral factors among Thai patients. Based on the results of this study, stop smoking and promote adequate physical activity are suggested to reduce the incidence of colon cancrr among Thai patients.


Modifiable Factors and Colon Cancer Risk in Thai Population
prolong survival of the patients; nevertheless, the mortality and morbidity of the disease are high. Moreover, colonic cancer treatment costs become higher than in the past due to development of new-targeted therapies (Faivre, 1994;Lerdkiattikorn et al., 2015). Prevention can be one of the most cost-effective measures for low resources countries such as Thailand (Khuhaprema and Srivatanakul, 2008;Lepage et al., 2010).
To prevent all NCDs, WHO and United Nations proposed a prevention and control model for NCDs called a 4 × 4 model (WHO, 2018). The 4 × 4 model comprises of four risk behaviors, namely tobacco use, physical inactivity, unhealthy diet, and the harmful use of alcohol and four abnormal physiology markers, including hyperglycemia, hyperlipidemia, hypertension, and overweight as well as obesity (Demirbas, 2015;Slattery et al., 1990;WHO, 2017WHO, , 2018Wolin et al., 2009). For meta-analyses found that many inflammatory bowel diseases and heredity factor had the highest association with development of colon cancer (Bye et al., 2017). For modifiable risks are obesity, smoking, exercise, and eating red meat (Khuhaprema and Srivatanakul, 2008;Wheat et al., 2016). To demonstrate the impact of modifiable risks on colon cancer in Thai population, we developed a survey questionnaire comprising of three parts, namely subject profile, standard questionnaire for smoking behaviors; excessive alcohol consumption and physical exercise; and cognitive-behavioral factors. The questionnaire was tested and validated in Samuprakarn province, Thailand. Our study aimed to show the odd ratio of modifiable risks of colonic cancer, including tobacco use, physical inactivity, and unhealthy diet in Thai population according to 4 x 4 model proposed by WHO.

Materials and Methods
We conducted this case-control study from June 2016 until June 2017. The study was done in 11 Thai provincial hospitals. The hospitals in this study were stratifically selected by regions. The sample size was calculated using odd ratio. For this purpose, we set the significant odd ratio at 1.5 (n=990 per group). The formula for odd ratio was as follows: We enrolled all newly diagnosed colon cancer within 1 month from outpatient clinics and hospital wards (n=504). For control group, we recruited individuals who came to the hospital for health check up. They were all healthy (n=997). Inclusion criteria were being Thai and age above 35 years old. Exclusion criteria were being non-Thai and critically ill. Case record form was developed from standard NCDs questionnaires, namely Cage test for cigarette smoking (Williams, 2014), Fagerstrom for nicotine dependence (Ebbert et al., 2006;Uysal et al., 2004), AUDIT for alcoholic dependence (Conigraveet al., 1995;Nordqvist et al., 2004), and WHO STEP wise for physical activity (WHO, 2017). The questionnaire was tested for reliability and validity in Samut Prakarn province, Thailand (Chamroonsawasdi et al., 2017). Content validity of the questionnaire was determined by the mean of the item content validity index (I-CVI) from five experts. The mean I-CVI ranged from 0.90-1.00 and for difficulty of knowledge ranged from 0.3-0.9. The reliability test of knowledge by KR-20 ranged from 0.648-0.799. The data were analyzed using t-test, Chi-square, and multiple logistic regression analysis. P-value < 0.05 was set as statistically significant.
All patients who enrolled in the study completed a written informed consent. In addition, these patients received full information and explanation about the study objectives and rights that a patient had before enrolling in the study.  Hospital EC no. 32.102/1090, Pharhol Phonphayuhasana Hospital EC. No. 2016-06, Prathumtani Hospital EC. No. 32.203.3/1962. This research was carried out in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki).

Demographic characteristics of the participants
In this study, it was found that older participants were more likely to have colon cancer than the younger. Age group of 60 years and over had 12.22 times (95% CI = 5.20-28.69) higher chance for developing colon cancer. Age group of 40-59 years had 8.21 times (95% CI = 6.24-10.80) chance for developing colon cancer. Other socioeconomic factors found to be associated with colon cancer were education, incomes, occupation, etc.
Participants with other chronic non-communicable diseases had higher chance for having colon cancer, especially those with family history of colon cancer. These participants had 312 times higher chance for being diagnosed with colon cancer (95% CI = 190.54-513.63) ( Table 1).
The results of the study showed that the socioeconomic factors of age, occupation, and incomes were significantly associated with risk of colon cancer.

Comparison of physiological factors between two groups using univariate analysis
Low BMI, high systolic blood pressure, high FBS, low HDL and, high triglyceride were associated with risk of developing colon cancer. There was no association between other physiological factors and colon cancer ( Table 2).

The multiple variables relationship between physiological factors and colon cancer risk
After adjusting the influence of socioeconomic status, it was found that low BMI, high systolic blood pressure and low HDL were associated with colon cancer risk. There was no association between other physiological factors and colon cancer risk. Participants with BMI < 18.5 had 6.20 times higher chance (95% CI = 3.56-10.81) for developing this cancer compared with those with normal BMI. Participants with abnormal systolic pressure (> 160 mmHg) had 5.20 times higher chance for developing colon cancer (95% CI = 1.02-26.56) compared to those with normal systolic pressure (Table 3).
Using multivariate analysis, it was found that the status of physiological factors could predict the likelihood of developing colon cancer. For instance, Participants with low BMI (<18.5) had OR of 5.26 (95% CI = 1.10-25.14) compared to those with normal BMI (18.5-24.9). Participants with abnormal systolic blood pressure (> 160 mmHg) or more found to have 30.48 times higher chance of developing this cancer (95% CI = 1.85-502.85) compared to those with normal systolic pressure. Participants with HDL less than 40 mg / dL had 8.80 times higher chance for having this cancer (95% CI = 2.62-29.25) (

Association between behavioral factors and colon cancer risk Association between behavior factors and colon cancer risk based on univariate analysis
Smoking, low physical activity, and eating sweet, fatty, and salty food were found to be associated with colon cancer risk (Table 4).

Exploring the association between behavior factors and colon cancer risk using multivariate analysis
After adjusting the socioeconomic status, no association was observed between alcohol consumption and AUDIT score and colon cancer risk, while an association was found between smoking and colon cancer risk. Participants who smoked had higher chance for developing colon cancer revealing dose-response relationship (less than 20 pack a year, OR =1.57 times; 95% CI = 1.00-2.47 and more than 20 pack a year, OR = 3.00 times; 95% CI = 1.19-7.55). In addition, it was found that participants with inadequate physical activity had higher chance for developing colon cancer (OR = 2.86, 95% CI = 1.82-4.48) compared to those who had adequate physical activity. Participants who regularly ate salty foods had higher chance of developing colon cancer (OR = 1.68, 95% CI = 1.25-2.27) ( Table 5).
Multivariate analysis results showed that smoking, inadequate physical activity, and eating salty foods were associated with colon cancer risk. Those who smoked more than 20 packs a years had 3.25 fold chance of developing bowel cancer (OR = 3.25, 95% CI = 1.17-9.03). Those with inadequate physical activity had 3.09 times of having colon disease (OR = 3.09, 95% CI = 1.88-5.07). Participants who regularly ate salty foods had also higher chance of developing colon cancer (OR

Discussion
Many factors are associated with abnormal division and growth of colonic epithelial cells. The abnormal growth will eventually progress to cancer. Patients usually come to hospital with symptoms of colonic obstruction or bleeding. Diagnostic gold standard for bowel cancer is tissue diagnosis. The colon cancer is divided into 3 types based on its etiology, namely sporadic colon cancer, familial colon cancer, and hereditary colon cancer. In patients with sporadic colon cancer, there is no history of colon cancer in the family. About 70-85% of all colon cancer patients are in this group. More than 80% of the patients aged more than 50 years-old. However, patients with familial colon cancer always have a history of colon cancer in their family member. In this type, abnormal growth is due to genetic factors. This group comprises 20-25 % of all colon cancer patients (Claes et al., 2011). The third type is found in about 5-10 % of all colorectal cancer patients. whose inheritance are familial polyposis   (Knudsen et al., 2003). Neverthelse, there are modifiable and unmodifiable factors contributing to development of this cancer.
Modifiable risks factors can be obesity, smoking, heavy alcoholic consumption, and lack of physical activity. Overweight and obesity can cause changes in the body leading to cancer, such as increase in levels of certain hormones and inflammation. There are 13 types of cancer associated with overweight and obesity, including colon cancer (Massetti et al., 2017). Tobacco contains at least 70 carcinogenic chemicals getting into bloodstream and damaging DNA. Damaged DNA can lead to abnormal cells growth and cancer. Alcohol breaks down into acetaldehyde. Acetaldehyde can damage DNA. The damaged DNA can lead to cancer development (Humans, 2010). Inadequate physical activity is associated with many cancers in human. Understanding the biological mechanisms underlying the associations between physical activity and cancer will provide needed insights for reducing cancer risk. For colon cancer, the major hypothesis is that physical activity lowers fecal bile acid   (Clague and Bernstein, 2012;Wolin et al., 2009). The results of this study on Thai patients with colon cancer revealed that age and socioeconmic factors were associated with colon cancer risk. In this study, family history of colon cancer was significantly associated with colon cancer risk. It was also found that physiological abnormalities such as high systolic blood pressure, dyslipidemia (low HDL), and low BMI were associated with colon cancer risk. However, our results may not be reliable due to missing data. The missing laboratory data in both group were due to nature of our study.
Behavioral factors, including smoking, inadequate physical exercise, and salty food consumption, were associated with colon cancer risk. We found no association between obesity and alcohol consumption and colon cancer. The results suggested that genetic factors may have higher association with colon cancer risk among Thai patients than behavioral factors.