Obesity is a major public health problem and is a risk factor for hypertension, diabetes, heart disease and cancer. In addition, the prevalence of obesity appears to be increasing worldwide, making it important to determine the nature and extent of obesity in populations at risk. The aim of this study was to determine the predictors of body mass index (BMI) in Taiwanese adults 19 through 44 years old. Independent variables from the Nutrition and Health Survey in Taiwan (NAHSIT) included biological factors (age and sex), dietary factors (energy intake and fat intake), environmental factors (educational level, residence area, and nutrition knowledge), and lifestyle factors (leisure time physical activity, smoking habits, and alcohol consumption). Results indicated that nearly 40% of the 1,182 adult survey participants were obese or overweight, by Taiwanese standards. Regression analysis found that significant predictors of BMI were residence area, educational level, and age. Taiwanese adults more likely to have higher BMIs were from the mountainous area, had less education. An increase in ten years of age was related to an average increase in BMI of 1.13 kg/m�� (p<0.0001), controlling for other variables. These three factors accounted for 21% of the variance in all adults (15% of the variance in men, and 28% of the variance in women). In addition, smoking was a predictor of higher BMI in males, and alcohol intake was a predictor for higher BMI in females. Energy intake, fat intake, leisure time physical activity, and nutrition knowledge were not significant predictors for either males or females. The findings from this study will help public health professionals identify target areas and program needs to reduce obesity in Taiwan, particularly in the mountainous areas where the prevalence of obesity/overweight was seventy percent. / Graduation date: 2004

As the leading cause of death in the United States, coronary heart disease (CHD) is a growing public health problem, despite the fact that many risk factors for the disease are preventable, especially if addressed early in life. The purpose of the current study was to examine the effects of loss-framed versus gain-framed versus information-only health messages on both intention to attend and actual attendance at an appointment to get screened for CHD risk factors (i.e., hypertension, diabetes, and dyslipidemia). It was hypothesized that a population of young adults would be more likely to view screening for CHD risk factors as a low-risk, health-affirming behavior as opposed to a risky, illness-detecting behavior and would thus be more strongly influenced by gain-framed messages than loss-framed messages. Additional goals included the exploration of the extensively researched individual health beliefs of perceived threat (as defined by the health belief model) and health locus of control as they relate to message frames. One hundred forty-three undergraduate students were randomly assigned to either the loss-framed, gain-framed, or information-only control conditions. Framing manipulation checks revealed that participants failed to discern differences in the tone and emphasis of the experimental pamphlets. As a result, no tests of framing effects could be conducted. Sixteen (11.2%) of the 143 participants who participated in Part 1 of the experiment participated in Part 2 (i.e., attended a risk factor screening appointment). Multiple regression analysis revealed risk index, age, and powerful others health locus of control as significant predictors of screening intention. Gender was the only demographic or health related variable that was significantly related to screening outcome, such that women were more likely to get screened than men. Limitations and recommendations are discussed.