• Journal of Preventive Medicine and Public Health
• /
• v.32 no.4
• /
• pp.452-458
• /
• 1999

Objectives: The authors conducted the study to evaluate bias when potentially diseased subjects were included in cohort members while analyzing risk factors of chronic liver diseases. Methods: Total of 14,529 subjects were followed up for the incidence of liver diseases from January 1993 to June 1997. We have used databases of insurance company with medical records, cancer registry, and death certificate data to identify 102 incident cases. The cohort members were classified into potentially diseased group(n=2,217) when they were HBsAg positive, serum GPT levels higher than 40 units, or had or has liver diseases in baseline surveys. Cox's model were used for potentially diseased group, other members, and total subjects, respectively. Results: The risk factors profiles were similar for total and potentially diseased subjects: HBsAg positivity, history of acute liver disease, and recent quittance of smoking or drinking increased the risk. while intake of pork and coffee decreased it. For the potentially diseased, obesity showed marginally significant protective effect. Analysis of subjects excluding the potentially diseased showed distinct profiles: obesity increased the risk, while quitting smoking or drinking had no association. For these intake of raw liver or processed fish or soybean paste stew increased risk; HBsAg positivity, higher levels of liver enzymes and history of acute liver diseases increased the risk. Conclusions: The results suggested the potential bias in risk ratio estimates when potentially diseased subjects were included in cohort study on chronic liver diseases, especially for lifestyles possibly modified after disease onset. The analytic strategy excluding potentially diseased subjects was considered appropriate for identifying risk factors for chronic liver diseases.

• 대한예방의학회:학술대회논문집
• /
• 1994.04a
• /
• pp.283-303
• /
• 1994

• 대한예방의학회:학술대회논문집
• /
• 2003.04a
• /
• pp.2-7
• /
• 2003

• 대한예방의학회:학술대회논문집
• /
• 2003.04a
• /
• pp.62-67
• /
• 2003

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.347-373
• /
• 2021

Over the past 50 years, explorative research on the nation's mortality decline patterns has showed a decrease in age-specific mortality rates in all age groups, but there were different improvement patterns in specific mortality rates depending on ages and periods. Greater distinct improvement was observed in mortality rates among women than men, and there was a noticeable improvement in mortality rates in certain groups especially in the more recent decades, revealing a structural change in the overall trends regarding death periods. In this paper, we compare various stochastic mortality models considering cohort effects for mortality projection using Korean female mortality data and further explore the uncertainty related to projection. It also created age-specific mortality rates and life expectancy for women until 2067 based on the results of the analysis, and compared them with future age-specific mortality rates and life expectancy provided by the national statistical office (KOISIS). The best optimal model could vary depending on data usage periods. however, considering the overall fit and predictability, the PLAT model would be regarded to have appropriate predictability in terms of the mortality rates of women in South Korea.

• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.335-342
• /
• 2015

Joint hierarchical generalized linear models proposed by Molas et al. (2013) extend the simple longitudinal model into multiple models fitted jointly. It can easily handle the correlation of multivariate longitudinal data. In this paper, we apply this method to analyze KoGES cohort dataset. Fixed unknown parameters, random effects and variance components are estimated based on a standard framework of h-likelihood theory. Furthermore, based on the conditional Akaike information criterion the correlated covariance structure of random-effect model is selected rather than an independent structure.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2003.06a
• /
• pp.161-179
• /
• 2003

• Korea journal of population studies
• /
• v.29 no.2
• /
• pp.53-88
• /
• 2006

The major aim of this paper is to develop a hypothetical set of age-specific fertility rates which are logically derived and reasonably accurate in the projection of future population. The first procedure is to select a generalized log-gamma distribution model, which includes Coale-McNeil nuptiality model, in order to estimate and project a set of age-specific fertility rates by birth cohort and birth order. The second is to apply the log-gamma model with an empirical adjustment to the actual data to estimate and project the future fertility rates for relatively young birth cohorts who did not complete their reproductive career. This study reconstructs or translates a set of cohort age-specific fertility rates into a set of period age-specific fertility rates which must be hypothesized in order to establish the broader framework of future population projection. For example, the fertility at age 20 in the year of 2020 is the fertility at age 20 for the cohort born in 1990, while the fertility at age 21 in the year of 2020 is the fertility at 21 for the cohort born in 1989. In turn, once a set of age-specific fertility rates for the cohorts who were born up to the year of 2010, it is possible for one to establish an hypothetical set of period age-specific fertility rates which will be needed to project the future population until the year of 2055. The difference in the hypothetical system of age-specific fertility rates between this study and the 2005 special population projection comes from the fact that the fertility estimation/projection model used in this study was skillfully exploited to reflect better actual trend of fertility decline caused by rise in marriage age and increasing proportion of those who remain single until their end of reproduction. In this regard, this paper argues that the set of age-specific fertility rates derived from this study is more logical and reasonably accurate than the set of those used for the 2005 special projection. In the population projection, however, the fundamental issue of the hypothetical setting of age-specific fertility rates in relation to the fertility estimation/projection model is about how skillfully one can handle the period effects. It is not easy for one to completely cope with the problem of period effects except for the a minor period adjustment based on recent actual data, along with the given framework of a cohort-based fertility estimation/projection model.

• Korea journal of population studies
• /
• v.30 no.1
• /
• pp.97-124
• /
• 2007

국내 생화학 분야 학자들의 연구성과에 영향을 주는 요인은 무엇인가? 특히 젠더는 연구성과에 어떤 영향을 주는가? 이 연구는 다음과 같은 항목에 초점을 두었다. 첫째 연구성과의 특징에 따른 차이를 보기 위해, 연구성과를 학술지 유형에 따라 SCI논문과 비SCI논문으로 구분해서 비교분석했다. 둘째, 교육기간부터 경력기간별로 연구성과에 미치는 요인이 변화할 가능성을 분석하기 위해, 연구성과를 종속변수로 하는 모델을 대학원 기간과 졸업 이후 경력 기간별로 구성했다. 셋째, 연구성과에 대한 이론적 요인으로서 사회인구학적 요인, 조직적 요인, 그리고 이전 연구업적의 누적효과를 음이항모델을 사용해서 분석했다. 연구 결과는 다음과 같다. 첫째, SCI논문성과의 경우 젠더 효과는 대학원 기간 동안에는 관찰되지 않았지만, 경력 초기부터 다음 기간 연구성과에 영향을 주는 것으로 나타났다. SCI학술지 연구성과의 경우 조직적 요인 중에서 외국 학위가 중요한 것으로 파악되었다. 이전 연구성과 역시 누적적으로 그 후 연구성과에 영향을 주었다. 비SCI논문 수를 분석한 결과, 대학원 기간과 졸업 후 3년간 기간에는 젠더 차이가 관찰되지 않았고, 졸업 후 3-6년 기간에 젠더 효과가 발견되었다. 또한 대학원 기간의 비SCI논문의 누적효과가 지속적으로 나타났다.

• Korea journal of population studies
• /
• v.26 no.1
• /
• pp.143-170
• /
• 2003

During the last few decades dramatic expansion of education occurred for women as well as men in Korea. Taking into account such a rapid expansion of education, this study examines trends in the effects of education on first occupational attainment among Korean women. Using the data from "the 4th Survey on Women's Employment," conducted by Korean Women's Development Institute in 2001, this study investigates the trends across three cohorts classified on the basis of the year of labor force entry after schooling: before 1980, 19801989, and 1990 or later. First, log-linear models are applied to the data to detect the temporal change in the overall association between education and first occupational attainment controlling for marginal distribution. The log-linear analysis shows that the strength of association between education and first occupation has declined over time. An additional analysis of OLS regression is conducted to see how the effects of each level of educational attainment on occupational prestige have changed across the three cohorts. The results of OLS regression suggest that the differences in prestige scores between the lowest and each of other educational levels are narrower in recent cohorts.t cohorts.