Yousung Park

Body Mass Index and Mortality in the General Population and in Subjects with Chronic Disease in Korea: A Nationwide Cohort Study (2002-2010)

Background The association between body mass index (BMI) and mortality is not conclusive, especially in East Asian populations. Furthermore, the association has been neither supported by recent data, nor assessed after controlling for weight changes. Methods We evaluated the relationship between BMI and all-cause or cause-specific mortality, using prospective cohort data by the National Health Insurance Service in Korea, which consisted of more than one million subjects. A total of 153,484 Korean adults over 30 years of age without pre-existing cardiovascular disease or cancer at baseline were followed-up until 2010 (mean follow-up period = 7.91 ± 0.59 years). Study subjects repeatedly measured body weight 3.99 times, on average. Results During follow-up, 3,937 total deaths occurred; 557 deaths from cardiovascular disease, and 1,224 from cancer. In multiple-adjusted analyses, U-shaped associations were found between BMI and mortality from any cause, cardiovascular disease, and cancer after adjustment for age, sex, smoking status, alcohol consumption, physical activity, socioeconomic status, and weight change. Subjects with a BMI < 23 kg/m2 and ≥ 30 kg/m2 had higher risks of all-cause and cause-specific mortality compared with the reference group (BMI 23–24.9 kg/m2). The lowest risk of all-cause mortality was observed in subjects with a BMI of 25–26.4 kg/m2 (adjusted hazard ratio [HR] 0.86; 95% CI 0.77 to 0.97). In subgroup analyses, including the elderly and those with chronic diseases (diabetes mellitus, hypertension, and chronic kidney disease), subjects with a BMI of 25–29.9 kg/m2 (moderate obesity) had a lower risk of mortality compared with the reference. However, this association has been attenuated in younger individuals, in those with higher socioeconomic status, and those without chronic diseases. Conclusion Moderate obesity was associated more strongly with a lower risk of mortality than with normal, underweight, and overweight groups in the general population of South Korea. This obesity paradox was prominent in not only the elderly but also individuals with chronic disease.

Chitosan microparticle preparation for controlled drug release by response surface methodology

The objectives were to investigate the effects of formulation variables on the release of drug and to optimize the formulation of chitosan microparticles loaded with drug for controlled release using response surface methodology. Chitosan microparticles were prepared by dropping a chitosan solution into sodium tripolyphosphate (TPP) through ionic cross-linking. The release behaviour of felodipine as a model drug was affected by preparation variables. A central composite design was used to evaluate and optimize the effect of preparation variables, chitosan concentration (X1), the pH of the TPP solution (X2) and cross-linking time (X3) on the cumulative per cent drug release (Y) in 24 h. Chitosan concentration and cross-linking time affected negatively the release of felodipine, while the pH of the TPP did so positively and was the highest influential factor. The optimum rate of drug release, 100% in 24 h, was achieved at 1.8% chitosan concentration, a pH 8.7 for the TPP solution and 9.7 min cross-linking time.

Some asymptotic properties in INAR(1) processes with Poisson marginals

The first-order integer-valued autoregressive (INAR(1)) process with Poisson marginal distributions is considered. It is shown that the sample autocovariance function of the model is asymptotically normally distributed. We derive asymptotic distribution of Yule-Walker type estimators of parameters. It turns out that our Yule-Walker type estimators are better than the conditional least squares estimators proposed by Klimko and Nelson (1978) and Al-Osh and Alzaid (1987). also, we study the relationship between the model andM/M/∞ queueing system.

Bayesian analysis for incomplete multi-way contingency tables with nonignorable nonresponse

We propose Bayesian methods with five types of priors to estimate cell probabilities in an incomplete multi-way contingency table under nonignorable nonresponse. In this situation, the maximum likelihood (ML) estimates often fall in the boundary solution, causing the ML estimates to become unstable. To deal with such a multi-way table, we present an EM algorithm which generalizes the previous algorithm used for incomplete one-way tables. Three of the five types of priors were previously introduced while the other two are newly proposed to reflect different response patterns between respondents and nonrespondents. Data analysis and simulation studies show that Bayesian estimates based on the old three priors can be worse than the ML regardless of occurrence of boundary solution, contrary to previous studies. The Bayesian estimates from the two new priors are most preferable when a boundary solution occurs. We provide an illustrating example using data for a study of the relationship between a mothers smoking and her newborns weight.

One-level rotation design balanced on time in monthly sample and in rotation group

We introduce the two-way balanced one-level rotation design for which balancing is done on interview time in sample and in rotation group and provide the necessary and sufficient condition for the two-way balancing and an algorithm to construct such design. Using this design, we obtain generalized composite estimators (GCE) and the minimum variance linear unbiased estimator (MVLUE). We then calculate their variance and mean squared error (MSE) when there exist two types of correlations and the rotation group bias. Minimizing the weighted sum of variances (or MSEs), we derive one set of the compromise coefficients for all important estimators and characteristics to retain the consistency in total. We compared the GCE with the compromise coefficients to MVLUE and other GCEs with the previous coefficients. We show that the second-order correlations have very significant effects on the variance and MSE. We also investigate the design efficiencies for some selected two-way balanced designs.

Forecasting cause-age specific mortality using two random processes

Mortality forecasts are critical information for assessing the health of a population and are necessary for making informed decisions about how best to direct health-related resources and activities. Timeliness in making health statistics available is crucial to identify and address current health problems. Being motivated to meet these needs, we propose a method to forecast the number of cause-age specific deaths through a two random processes model. Unlike the previous methods, the new method incorporates both cross-sectional and longitudinal correlations into our model without a high-dimensional problem. A bootstrap confidence interval is presented to measure the validity of our model and to detect an unusual occurrence of deaths. Our data analysis demonstrates that our method gives promising results compared with the true final counts.

Robust Regression Using Data Partitioning and M-Estimation

We propose a new robust regression estimator using data partition technique and M estimation (DPM). The data partition technique is designed to define a small fixed number of subsets of the partitioned data set and to produce corresponding ordinary least square (OLS) fits in each subset, contrary to the resampling technique of existing robust estimators such as the least trimmed squares estimator. The proposed estimator shares a common strategy with the median ball algorithm estimator that is obtained from the OLS trial fits only on a fixed number of subsets of the data. We examine performance of the DPM estimator in the eleven challenging data sets and simulation studies. We also compare the DPM with the five commonly used robust estimators using empirical convergence rates relative to the OLS for clean data, robustness through mean squared error and bias, masking and swamping probabilities, the ability of detecting the known outliers, and the regression and affine equivariances.

A Method for Construction of Life Table in Korea

The life table is a statistical model for life expectancy and reflects mortality experiences exposed to a particular group of people. The following three issues are prerequisite for constructing the life table : a selection of how to estimate the death probability from observed death rates, a graduation method to smooth irregularity of the death probabilities, and an extension method of the death probabilities for oldest-old ages. To construct the life table that is fittest to Korean mortality experiences, we examine five estimation methods such as Chiangs and Grevilles for the death probability, three graduation techniques including Beers and Grevilles formulae, and twelve mathematical functions for the extension of death probabilities for oldest-old ages. We also propose a method to resolve the cross-over problem arising from construction the life table.

A Study on the Population Structure of Democratic People's Republic of Korea

The re-unification of the two Koreas is seen as a potential solution to the aging problem and low fertility, along with the increase in the old population structure of the Republic of Korea. Population structure is an indicator of national competitiveness or growth, but little is known officially about the populations of Democratic People’s Republic of Korea (DPRK). Understanding the population structure of the DPRK is important to prepare for re-unification and estimate the socio-economic costs of national welfare under a re-unified Korea. In this paper, we assess reliability of the two modern national censuses of DPRK and use the limited resources available to reconstruct the intercensal populations between these two censuses. Excess deaths from the 1995 famine are estimated at 489,972 to 574,306 and are close to the estimates of Goodkind et al. (2011) and the reconstructed populations in the of DPRK implies a big difference between two Koreas.

Nonignorable Nonresponse Imputation and Rotation Group Bias Estimation on the Rotation Sample Survey

We propose proper methods to impute the item nonresponse in 4-8-4 rotation sample survey. We consider nonignorable nonresponse mechanism that can happen when survey deals with sensitive question (e.g. income, labor force). We utilize modeling imputation method based on Bayesian approach to avoid a boundary solution problem. We also estimate a interview time bias using imputed data and calculate cell expectation and marginal probability on fixed time after removing estimated bias. We compare the mean squared errors and bias between maximum likelihood method and Bayesian methods using simulation studies.