Background
: This study was intended to look into the characteristics and problems when FACES III are used as tools to evaluate family function of adolescents.

Methods : From May to June1996, self-reported questionnaires for FACES III were conducted on 2,430 middle school and high school students(males:1,190, female:1,240) in Seoul and Yangpyung, Kyunggi-Do. There were eight schools(four middle school, four high school) in Seoul and two schools(one middle school, one high school) in Yangpyung, Kyunggi-Do in this study. Also, education level of parents, occupational status of parents, family types(extended/nuclear), number of family member and numbers of siblings were assessed.

Results : The mean values of adaptability were significantly higher in Seoul students, in females, and in nuclear families. The mean values were higher in single sibling than two or more and in parents with higher education. The mean values of cohesion were significantly higher in Seoul students, in middle school students and in family members of 5 and under. The mean values were higher in those without a sibiling and in those with above college education. In stepwise regression analysis, educational level of parents, gender, area, occupational status of mother and grade were associated significantly with the mean values of adaptability and educational level of mother and grade were associated significantly with the mean values of adaptability and educational level of parents, grade and numbers of siblings were significantly associated with the mean values of cohesion.

Conclusion : We consider that there are problems in deciding the cut-off point and in the questionnaires of survey when the FACES III is used as a tool for evaluating family function of adolescents. Trials of combining other tools of family function or consideration of lower concepts are needed in other to be applicable to the individual conversation and clinical surroundings.

BACKGROUND
It is very important for family physicians to practice with comprehensive approach to patients with chronic illness. A study was done to assess psychosocial aspects of patients with chronic illness and disability and to help physicians manage patients effectively.
METHODS
SCL-MPD questionnaires were done in 93 patients with chronic illness and disability who were admitted between April and May in 1997. The results were statistically analyzed by chisquare test and Mann-Whitney U test.
RESULTS
1. The mean score of SCL-MPD were 17.5 for `depression' and 13.78 for `somatization' with statistical significance(P<0.005). 2. Comparison of dimensions of patients with back pain syndrome had higher scoresn for `depression (25.2)', `somatization (22.1)', `hypochondriasis (14.7)', `phobic anxiety (11.6)', `neurasthenic (8.1)' and `interpersonal sensitivity(8.0)' than other disease groups with statistical significance (P<0.05). 3. In comparison of dimensions between males and females, the female group gradually had higher scores than males and, especially, 'somatization' dimension was significantly higher in score (P<0.05). 4. Symptoms ranked top in male and female patients were `54-over coscern about your health' in males and `16-over concern with the fear of having a serious illness' in females.
CONCLUSION
Family physicians have to consider not only physical impairment but also psychological aspect of patients with chronic illness for comprehensive management in medical practice.

Background
: This study was conducted to investigate the reason why H.pylori infected people develop different patterns of gastric diseases and in some infected patients have peptic ulcers, but the others have not.

Methods : We analyzed 177 cases of H. pylori positive patients who were confirmed by histologic evaluation of gastric endoscopic biopsy specimens from January 1998 to July 1998 in Health Promotion Center of one medical college hospital. Among these 177 cases, 63 cases(35.6%) were peptic ulcer patients and 114 cases(64.4%) were non ulcer patients. We anlayzed the difference of sociodemographic finding, alcohol and smoking habit, past history, family history, drug history, emotional factors, body mass index and serum biochemical profiles between ulcer and non ulcer patients.

Results : 1. There were no differences of age, sex and occupation between peptic ulcer and non ulcer patients. 2.Heavy smoking(p<0.001), heavy drink(p=0.011), past history of upper gastrointestinal disease(p<0.001), fatigue(p<0.001) and depressive mood(p=0.001) were significantly more frequent in H.pylori positive peptic ulcer patients than non ulcer patients. 3.We found several related factors that might affect peptic ulcer development in H. pylori infected patients through logistic regression analysis of variables. These factors were heavy smoking(odds ratio 5.91, 95% CI 2.54-13.78), past history (odds ratio=4.24, 95% CI 1.72-10.42), depressed mood(odds ratio 2.76, 95% CI 1.15-6.63), and fatigue(odds ratio 2.42, 95% CI 1.03-5.69).

Conclusion : We conclude that past history, heavy smoking, depressed mood, and fatigue are related factors of peptic ulcer disease in H.pylori positive patients.

BACKGROUND
Psychiatric disorders have been known to occur by the interaction of biologic, psychologic and social factors. A family can affect the occurrence of psychiatric disorders as a primary social factor. On the other hand, psychiatric disorders can affect family function. There fore, we studied to compare family function between the families of psychiatric inpatients and those of normal persons, using Family APGAR and FACES III.
METHODS
Between July 1, 1999 and August 31, 1999, we examined the subject group as 40 relatives to visit psychiatric inpatient and the control group as 40 persons whose family members did not have chronic diseases or psychiatric disorders, using questionnaire of socio-demographic factors and family function.
RESULTS
Family APGAR scores in subject group were lower than those of control group, but not statistically different. But the degrees of 'Partnership' of the Family APGAR in subject group were significantly lower than those of control group.(P<0.05) Adaptability scores of FACES III in two groups were not different. Cohesion scores of FACES III in subject group were lower than those of control group but not statistically different. The Family APGAR scores in the families of the schizophrenics were significantly lower than those of control group. The duration of psychiatric disorders did not affect family function.
CONCLUSION
The family function of subject group was worsened but not statistically different. But in the families of the schizophrenics, the Family APGAR scores were significantly lower than those of control group. Therefore, we should consider serious family dysfunction in the families of schizophrenics.

Background
: The purpose of this study was to identify health practices that would be helpful in order to reduce the prevalence of correctable cardiovascular risk factors (CVR), i.e. hypertension, hypercholesterolemia, and hyperglycemia in male workers by examining the association between health practices and the prevalence of the three risk factors.

Methods : For 1,815 male workers at a car manufacturer who participated in the basic health examination, survey on socioeconomic status and seven healthy practices such as not currenly smoking, moderate alcohol drinking, proper weight control, regular exercise, 7-8 hour sleep, eating breakfast regularly, and not eating between meals was done and the prevalence of hypertension, hypercholesterolemia, and hyperglycemia as obtained. χ²test, Fisher's exact test, and multiple logistic regression analyses were done for workers under 40 years old and workers 40 and over in two, separate groups by adjusting socioeconomic variables.

Results : The prevalence of hypertension and hypercholesterolemia was 1.84 times and 5.22 times, respectively, showing, higher in overweight younger workers than in those of normal weight. The prevalence of hypercholesterolemia was 2.02 times higher in younger smoking than in younger non-smokers. In contrast, the prevalence of hyperglycemia was 4.99 times higher in older smokers than in older non-smokers. The prevalence of hyperglycemia was 16.5 times higher in older workers with regular snacking habit than in those without such a habit. The prevalence of hypercholesterolemia was 7.47 times higher only in younger workers practicing healthy behaviors below 3 than in those practicing at least 6 healthy behaviors.

Conclusion : Our study suggests that overweight and smoking habit among younger workers, and smoking habit among older workers sould be controlled in order to decrease the prevalence of hypertension, hypercholesterolemia, and hyperglycemia.

BACKGROUND
The purpose of this study was to evaluate the association between social support and health status for the elderly in a rural area where social network has been decreased.
METHODS
One hundred and 58 elderly patients who visited free medical services participated in the survey that contained items about social support, heath status, socioeconomic status, health behaviors, and depression. Subjects were divided into three groups according to their social support index scores and two groups according to their health status. The health behaviors, socioeconomic status, and depression were considered as confounders. The associations between social support and confounders were analyzed by oneway analysis of variance, Cochran Mantel-Haenszel test, x2-test, and t-test. Then, multiple logistic regression analyses were done to find the adjusted association between social support and health status.
RESULTS
Social support was related to education (P=0.012), economic level (P=0.007), smoking status (P=0.028), depression score (P<0.001), and health status (P<0.001). Health status was related to age (P=0.02), economic level (P<0.001), and depression score (P<0.001). The unadjusted prevalence odds ratios (POR) for poor health status were 7.3 (95% CI: 2.8-19.2) in the poorly supported group and 1.3 (95% CI: 0.5-3.5) in moderately supported group compared with the highly supported group. The corresponding adjusted POR increased to 14.3 (95% CI: 2.8-73.3) in poorly supported group and 6.2 (95% CI: 1.2-31.9) in moderately supported group.
CONCLUSION
The poor social support was associated with poor health status and the relationship increased more with adjustment of confounders. Several methods to improve social support is necessary to maintain and enhance health status in the rural elderly.

Background
: Tuberculosis is still one of the major public health problems in Korea. The patient's compliance with anti-tuberculosis therapy is the most important matter in managing tuberculosis. Research, particularly on this issue, is needed. Therefore, we analyzed the factors associated with compliance of pulmonary tuberculosis patients using the Health Belief Model.

Methods : One hundred and 47 patients who visited the outpatient Department of Family Medicine in Asan Medical Center from January 1995 to December 1998, and who were diagnosed as having pulmonary tuberculosis and treated with anti-tuberculosis drug were chosen through chart reviews. Questions were given by telephone with questionnaire based on Health Belief Model.

Results : One hundred and two patients out of the total 147 participated in the study. The group was divided into two: 79 patients as a compliance group, and 23 as a non-compliance group. Recognition of their diagnosis, recognition of the duration of taking anti-tuberculosis drugs, disturbance of daily living with taking the drugs, efficacy expectations were statistically significant factors between the two groups.

Conclusion : Higher the knowledge about their diagnosis as having tuberculosis, higher their knowledge about the durations of having to take the anti-tuberculosis drugs, less disturbance they had experieced on their daily living, and higher their expectation for the efficacies of taking the drugs resulted in higher compliance with anti-tuberculosis medication.