Ecologic studies are widely used in all fields of public health on account of accesibility of data. However, two problems related to these studies have been brought up. The first is ecological fallacy occurred in the course of interpreting the ecologic level of exposure-disease associations into individual level. The second is exposure isclassification which leads to serious bias. Nevertheless there is few methodologic study dealing joint effects of the two problems in ecologic study. This study was conducted to suggest an ecologic model not having an ecologic fallacy due to model linkage failure and a methodology for correcting the misclassification bias due to exposure-confounder misclassification. Finally, we suggest a criteria for the ecologic model selection. Main results are as follows: 1. A linear ecologic regression model has a serious ecological fallacy due to model linkage failure and the misclassification bias due to the exposure-confounder misclassification. 2. An interaction ecologic regression model has no ecological fallacy due to model linkage failure, but it is affected seriously by the exposure misclassification. However misclassification bias could be removed mathematically if the information related to the misclassification was known. 3. A log-linear ecologic regression model has an ecological fallacy due to model linkage failure. It is seriously biased as the individual risk ratio are increased, but relatively less affected by the exposure misclassification than interaction ecologic regression model. 4. One of the two ecologic regression model-interaction ecologic regression model and log-linear ecologic regression model- would be selected according to the information of individual risk ratio and exposure misclassification. But using a linear ecologic regression model should be avoided in any circumstance. The above results are only valid in case that there is no other source of ecological fallacy except model linkage failure. Also exposure and confounder are independent each other, measured binary, and having nondifferential misclassification. Since the above assumptions are somewhat strong in considering the real situations of ecologic studies, it is necessary to extend the scope of this study.
Bias (Epidemiology) ; Joints ; Odds Ratio ; Public Health

No abstract available.
Moyamoya Disease*

PURPOSE: According to the staging system for breast cancer by the 2003 revised American Joint Committee on Cancer (AJCC), the patients with 10 or more positive axillary nodes are classified as N3 and also as a new stage, i.e., stage IIIC. The aim of this study was to investigate the prognosis of patients with 10 or more positive nodes. METHODS: The database of 125 patients with 10 or more positive axillary nodes who underwent surgery at Korea Cancer Center Hospital between 1997 and 2001 were reviewed. The age of the patients, the T stage, the number and site of the positive nodes, the hormone receptor status, the HER-2 over-expression, and the treatment modalities were examined in reference to the disease-free survival (DFS). RESULTS: At the median follow-up time of 40 months, 16 cases (13%) of locoregonal recurrence and 57 cases (46%) of systemic relapse had developed in 67 patients (54%). The DFS and overall survival rates at 5 years were 46% and 55%, respectively. On univariate analysis, the T stage (p<0.001), hormone receptor status (p=0.001), and neoadjuvant chemotherapy (p=0.014) were predictive factors of recurrence. On multivariate analysis, the T stage (p=0.002) and hormone receptor status (p=0.02) were independent predictors of recurrence. The patients with hormone receptor positive tumor had a 58%, 5-year DFS rate. On the contrary, in 9 of 10 patients with T4 tumor, recurrence developed within 2 years after the initial treatment. CONCLUSIONS: This study showed that stage IIIC according to the revised AJCC staging system was not a prognostically homogeneous group. Some notably high survival rates were observed in a subgroup of patients, and especially for those patients with hormone receptor positive tumor. In contrast, the prognosis of patients with T4 tumor was significantly worse than that of the patients with the other stage IIIC disease. Thus, we suggest that the stage IIIC group in the new AJCC staging system needs to be refined to provide more reliable prognostic information for the patients with advanced breast cancer.
Breast Neoplasms* ; Breast* ; Disease-Free Survival ; Drug Therapy ; Follow-Up Studies ; Humans ; Joints ; Korea ; Lymph Nodes* ; Multivariate Analysis ; Neoplasm Staging ; Prognosis* ; Recurrence ; Survival Rate

BACKGROUND: The purpose of this study was to investigate the relationship between 7 health habits and health-related quality of life using the SF-36 questionnaires. METHODS: The subjects for this study were 550 workers in Seongnam City, Kyonggi Province, Korea. Data were collected by using constructed questionnaires from 11 September 2004 to 25 September 2004. Statistical analyses were conducted using SPSS. RESULTS: There were positive relationships of favorable health habits as indicated by Alameda 7 to health-related quality of life. After controlling for the effects of confounding cofactors that included age, sex, diseases, and other health habits, multiple regression analysis indicated as follows. 1) The item of 'eating breakfast everyday' were significantly related to the grade of social function (SF) and bodily pain (BP). 2) The item of 'sleeping for 7~8 hours a day' were significantly related to all domains of the SF-36 except physical function (PF). 3) The item of 'normal Body Mass Index' were significantly related to the grade of Role limitations-emotional (RE). 4. The item of 'regular physical exercise' were significantly related to the grade of general health (GH), physical function (PF), Vitality (VT), and mental health (MH). Other health habits like not snacking between meals, never smoking, and low alcohol consumption were not associated with any of the SF-36 profiles. CONCLUSION: There were positive relationships of favorable health habits to health-related quality of life. It is important to have favorable health habits for good health-related quality of life.
Alcohol Drinking ; Breakfast ; Gyeonggi-do ; Korea ; Meals ; Mental Health ; Quality of Life* ; Smoke ; Smoking ; Snacks

No abstract available.
Pancreaticoduodenectomy*

No abstract available.
Doxorubicin* ; Drug Therapy, Combination* ; Etoposide* ; Fluorouracil* ; Mitomycin* ; Retrospective Studies* ; Stomach Neoplasms*

Traditionally, radical hysterectomy is the main surgical method for the treatment of early cervical carcinoma and always results in the loss of fertility. But, large numbers of young women are recently being diagnosed with cervical carcinoma and fertility preservation has become a concern. So, there has been a move towards more conservative approaches for the treatment of cervical carcinoma in recent years. Radical trachelectomy, which allows preservation of uterus but removes the cervix, parametrium and upper one third of the vagina, is a conservative but locally radical procedure. We performed radical abdominal trachelectomy with pelvic lymphadenectomy for two cases of invasive cervical carcinomas, which may be the first report in korea. One patient was 37 years old single women who had stage Ib cervical cancer and was disease free for 17 months after treatment. The other patient was 19 years old student who was also single and had stage IIa cervical cancer and was disease free for 14 months after treatment. We report the first two cases and review the literature on radical trachelectomy.
Adult ; Cervix Uteri ; Female ; Fertility ; Fertility Preservation ; Humans ; Hysterectomy ; Korea ; Lymph Node Excision* ; Uterine Cervical Neoplasms ; Uterus ; Vagina ; Young Adult

BACKGROUND AND OBJECTIVES: The purpose of this study was to investigate the familial correlation and heritability of the LDL cholesterol level through a segregation analysis. SUBJECTS AND METHODS: This study was based on the data from 9,884 Korean individuals, ascertained as pedigrees, who had participated in the 1998 and 2001 Korean National Health & Nutrition Examination Survey. The subjects of the segregation analysis were 3,613 Korean with more than five family members. RESULTS: The LDL cholesterol level revealed a strong familial correlation among spouses, parental-offspring and siblings; a low correlation with spouses, but a high correlation with parental-offspring and siblings. The heritability of the LDL cholesterol level was 37.3%, when controlled for age, age2, gender, age x gender, waist circumference, smoking, alcohol drinking, exercise and education. After the segregation analysis, models of the major gene effects were rejected in the general population. However, the Mendelian dominant and co-dominant models were found in high risk families. CONCLUSION: In conclusion, the randomly ascertained Korean families of this study showed strong familial aggregation in the LDL cholesterol level. This suggested that variations in the level of LDL cholesterol may be influenced by a major gene effect.
Alcohol Drinking ; Cholesterol, LDL ; Education ; Humans ; Siblings ; Smoke ; Smoking ; Spouses ; Waist Circumference

The purpose of this study is to examine the relationship between pesticide use and mortality, and cancer mortality which are recognized as health hazards of pesticides. Data were analyzed from a cohort of 6,066 people aged fifty-five or over and who have been residing in the main island of Kangwha county. Death certificates, computerized citizenship registers searches, and household interview survey lead to get more than ninety-seven percents follow-up rate for the first five year observation period. Important findings are as follows: 1. Age specific mortality rates of pesticide users are significantly lower than those of pesticides nonusers. The SMR of male is 0.80, and 0.58 for females respectively probably due to healthy worker effects. But, age specific cancer mortality rates are significantly higher than those of pesticide nonusers (SMR=1.59) in males. this finding is not observed in females (SMR=0.85), however. 2. Logistic regression analyses showed that self-reported health status, drinking, and smoking histories in male are significantly associated with total mortality rate. The histories of pesticide use are also calculated to be highly associated with cancer mortality as in univariate analyses in males. In female, self-reported health status, age of first delivery are found to be significantly related to total mortality rate. Only drinking history is calculated to be associated with cancer mortalities in females. Data from further observation of 'Kangwha cohort' and indepth analyses of these are highly expected.
Cohort Studies* ; Death Certificates ; Drinking ; Family Characteristics ; Female ; Follow-Up Studies ; Healthy Worker Effect ; Humans ; Logistic Models ; Male ; Morinda ; Mortality* ; Pesticides ; Smoke ; Smoking

No abstract available.
Biopsy, Fine-Needle* ; Breast Neoplasms* ; Breast* ; Diagnosis*