Combined life-style factors and all-causes and cancer-specific mortality in Korea

Abstract

Objectives Lifestyle factors such as obesity, smoking habit and alcohol drinking habit are known to be associated with increased risk of all-cause mortality in Korea. However, little has examined the association of combined life-style factors with the risk of mortality.

Methods The study population of 15,740 participants over 20 years old was selected from the Korean Multi-center Cancer Cohort (KMCC) study. Information on lifestyle factors was obtained through structured questionnaire interview. The total number of persons who completed the follow-up through the Death Certificate database of the National Statistical Office was 145,218.4 person-years by the end of December 31, 2008. The total number of death during the study period was 1,094 persons, and 474 deaths were due to cancer and 229 persons were due to cardio-vascular diseases. Each risk factor was dichotomized based on the previous study as follows

body mass index (BMI) under 22.6 or over 27.5 kg/m2 versus 22.6~27.5 kg/m2

current or past smokers versus non-smokers

non-drinkers or drinkers with alcohol amount of 90 gram per week or more versus drinkers less than 90 grams per week. Combined effect of three risk factors was estimated in two different ways

according to the number of risk factors qualitatively, and to the combined risk scores ranged from 0 to 14 quantitatively. The hazard ratios (HRs) and 95% confidence intervals (CI) for mortality were estimated by Coxs proportional hazard regression model adjusting for age, sex, locality, education, marriage, and the past history of chronic diseases.

Results The risk of dying was 1.47 times (95% CI = 1.27~1.69) in BMI, 1.50 times (95% CI=1.25~1.79) in smoking habit, and 1.36 times (95% CI=1.08~1.72) higher in alcohol drinking habit than the baseline hazard. The HRs for cancer death was 1.35 (95% CI=1.10~1.66) in BMI, and 1.75 (95% CI=1.32~2.33) in smoking habit, but not statistically significant in alcohol drinking habit. The HRs for both all-causes mortality and cancer mortality showed linearly increasing pattern with statistical significant according to the number of risk factor (p<0.001). Overall, the combined effect of lifestyle factors was more prominent in female. Particularly noteworthy was that the risk of dying increases according to the combined risk score increasing (p<0.001). Subjects with highest score (1`0~14) had significantly at the greatest risk of deaths (HR=3.46, 95% CI=1.89-6.32). The risk of dying was more prominent in female and participants under age<60.

Conclusions This study confirms abnormal BMI, smoking habit and alcohol drinking habit are major risk factor of premature deaths, as well as cancer death in Korea. Combined effect of lifestyle factors would additively be associated with marked increased risks of both all-causes and cancer-specific death in Korea.