• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.499-510
• /
• 2016

It is very important to select relevant variables in regression models for survival analysis. In this paper, we introduce a penalized variable-selection procedure in multi-level frailty models based on the "frailtyHL" R package (Ha et al., 2012). Here, the estimation procedure of models is based on the penalized hierarchical likelihood, and three penalty functions (LASSO, SCAD and HL) are considered. The proposed methods are illustrated with multi-country/multi-center bladder cancer survival data from the EORTC in Belgium. We compare the results of three variable-selection methods and discuss their advantages and disadvantages. In particular, the results of data analysis showed that the SCAD and HL methods select well important variables than in the LASSO method.

• 한국노년학
• /
• v.38 no.4
• /
• pp.963-979
• /
• 2018

The purpose of this study was to identify frailty profiles based on physical, psychological, and social domains of functioning and to examine the associated factors showing the differences among frailty profiles. Respondents were 70 years and older(n=403) and latent class analysis was applied to determine the optimal subgroups based on Tilberg Frailty Indicators which comprised of three domains(the physical, psychological, and social domain). Also, we performed multinominal logistic regression analysis to find out factors making differences among frailty profiles. Latent class analysis(LCA) identified three distinct types: multi-frail type(27.0%), psychologically frail type(26.8%), inadequate support type(46.2%). All three types had common difficulties in dealing with daily life problems and did not receive enough help with theses difficulties. Based on the results of the LCA three-class models, people in multi-frail type accumulated problems in physical and psychological domains and had partially social domain. On the other hands, psychologically frail type showed a relatively high anxiety disorder and depression. Lastly, people in inadequate support type reported the lack of helps, but they were relatively healthy. Comparing these groups with inadequate support type, people with multi-frail had lower educational level, poor nutritional management status and were less likely to participate in labor market. People in psychologically frail type were more likely to be male, to live in big cities rather than middle and small cities, and less likely to smoke. Based on these results, our results showed the multifaceted concept of frailty among Korean elderly people and we suggested several implications for preventing frail process.