"This study showed that modifiable risk factors measured in 1992 were predictors of medical care costs in 1998," said the study's lead author, Dr. Sun Ha Jee of the Graduate School of Health Science and Management at Yonsei University in Seoul, Korea.

This study is the first on the relationship between modifiable lifestyle risk factors and medical care costs in Korea, where the field of health promotion is still fairly new. Several similar studies have been conducted in the United States.

As the standard of living has improved in Korea in recent decades, more deaths now relate to lifestyle factors than to infectious diseases. Nearly 70 percent of men smoke, diets are spicy and high in salt, and an increasingly sedentary population is consuming more high-fat foods, researchers noted. Other factors taking their toll include higher rates of binge drinking, automobile and industrial accidents and stress, they said.

When Jee and associates examined insurance company data on nearly 130,000 men and women, they found modifiable risk factors accounted for about 23 percent of men's medical costs in 1998 and nearly 9 percent of women's medical costs that year.

"If these values hold up under further scrutiny, they provide excellent justification for additional investment in programs to prevent and reduce these risk factors through behavioral and medical interventions," Jee said.

In men, modifiable factors that increased medical costs included positive urinary glucose (which is associated with diabetes), high blood pressure and cholesterol, and being a former smoker. Current male smokers may avoid seeking medical care until they develop health problems that force them to quit smoking, according to the study, published in the "American Journal of Health Promotion."

In women, modifiable factors that increased medical costs included positive urinary glucose, high cholesterol, lack of exercise and smoking. In general, the proportion of modifiable risk factors was much lower in women than men, the study showed.

The Korean study did have some limitations, according to its researchers: Although the study was large, it consisted primarily of teachers and civil servants, and results may not be transferrable to the general population. In addition, there were no measures for nutrition, stress and depression.

"If additional risk factors had been measured, it is likely that a greater portion of total medical cost would have been accounted for," Jee said.