Background
: TSH is of great use as screening test in health examination. Recently, new sensitive EIA has been developed. We calculated normal range of TSH and positive predictive value of TFT abnormality by this method. Also, this study was carried out for the purpose of presuming the prevalence of TFT abnormality in community screening test of our county.

Methods : TSH levels were measured in 4,708 subjects that participated in periodic health examination at department of family medicine, Hanil hospital. The subjects with TSH abnormality were carried out test of T3 and T4.

Results : There were 4,649 men and 59 women in study population. According to area, subjects were distributed to 1,935 in rural area, 1,651 in small and medium cities and 1,122 in large cities. The number of subjects with TSH abnormality was 123 (2.6%) and the number of subjects with TFT abnomality among male was 32(0.7%). There was no statistically difference by age and sex, there was revealed as statistically significant higher rate in small and medium cities by area. The TSH was revealed as left skewed distribution. The normal ragne of TSH was 0.34~3.28mIU/L after log transformation and positive predictive value was 26%. After supposing sensitivity was 100%, specificity was etimated as 98.1%.

Conclusion : The prevalence of T3, T4 abnormality in Korea as similar as other countries was 0.7% and positive predictive value of T3, T4 abnormality by EIA methods was 26% that was low. Therefore it seems that there are restrictions in cost effectiveness as community screening test, although sensitivity and specificity were high.

Background
: In considerable portion of patients complaining upper gastrointestinal symptoms, the causes of the symptoms are not certain. This study was purposed to identify the factors related to the upper gastrointestinal symptoms.

Methods : the subjects were selected among the people that visited outpatient department of family medicine for the purpose of health screening and were examined by gastrofiberoscopy or upper gastrointestine series. All subjects were surveyed to determine the existence of 8 upper gastrointestinal symptoms and 15 factors composed of social environment, life style, diet patterns.

Results : The number of complaining symptoms was correlated significantly with sex, diet pattern, sleep pattern, exercise, and bowel habit. The bowel habit was correlated with 7 symptoms. Sleep pattern was correlated with 6 symptoms and diet pattern was correlated with 5 symptoms. The pepticulcer was correlated significantly with male, diet pattern, alcoho drinking, cigarette smoking. There was no significant correlation between peptic ulcer and number of complained sympotoms. Also there was no significant correlation between each symptom and peptic ulcer.

Conclusion : Upper gastrointestinal symptoms were affected by sex, diet pattern, exercise, sleep pattern, and bowel habit rather than organic causes. and these factors should be considered in management of patients presenting with upper gastrointestinal symptoms.