The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Analysis of medical panel binary data using marginalized models

주변화 모형을 이용한 의료 패널 이진 데이터 분석

  • Chaeyoung Oh (Department of Statistics, Sungkyunkwan University) ;
  • Keunbaik Lee (Department of Statistics, Sungkyunkwan University)
  • 오채영 (성균관대학교 통계학과) ;
  • 이근백 (성균관대학교 통계학과)
  • Received : 2024.01.10
  • Accepted : 2024.04.01
  • Published : 2024.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

Longitudinal data are measured repeatedly over time from the same subject, so there is a correlation from the repeated outcomes. Therefore, when analyzing this correlation, both serial correlation and between-subject variation must be considered in longitudinal data analysis. In this paper, we will focus on the marginalized models to estimate the population average effect of covariates among models for analyzing longitudinal binary data. Marginalized models for longitudinal binary data include marginalized random effects models, marginalized transition models, and marginalized transition random effect models, and in this paper, these models are first reviewed, and simulations are conducted using complete data and missing data to compare the performance of the models. When there were missing values in the data, there is a difference in performance depending on the model in which the data was generated. We analyze Korea Health Panel data using marginalized models. The Korean Medical Panel data considers subjective unhealthy responses as response variables as binary variables, compares models with several explanatory variables, and presents the most suitable model.

경시적 자료는 같은 개체를 반복 측정함으로써 시간의 흐름에 따른 반복 측정된 자료들 간의 상관관계가 존재한다. 따라서 경시적 자료분석에서는 이 상관관계를 분석할 때 개체 내 상관관계와 개체 간 변동성 모두를 고려해야 한다. 본 논문에서는 경시적 이진 자료를 분석하기 위한 모형 중 공변량의 모집단 평균 효과의 추정을 위해 주변화 모형에 집중하고자 한다. 경시적 이진 자료분석을 위한 주변화 모형으로는 주변화 임의효과, 주변화 전이, 주변화 전이 임의효과 모형이 있으며, 본 논문에서 이들 모형을 먼저 고찰하고, 그리고 모형들의 성능을 비교하기 위해 결측치가 없는 자료와 결측치가 있는 자료로 나눠서 모의실험을 진행한다. 모의실험에서 자료에 결측치가 있는 경우에 자료가 생성된 모형에 따른 성능 차이가 있음을 확인하였다. 마지막으로 주변화 모형을 이용하여 한국의료패널자료를 분석한다. 한국의료패널자료는 반응변수로 주관적 불건강 응답을 이진변수로 고려하였고, 여러 설명변수를 가진 모형을 비교하고 가장 적합한 모형을 제시한다.

Keywords

Acknowledgement

이 성과는 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 기초연구사업임 (NRF-2022R1A2C1002752), 이 논문은 오채영의 석사논문의 일부를 발췌하였음.

References

  1. Breslow NE and Clayton DG (1993). Approximate inference in generalized linear mixed models, Journal of the American Statistical Association, 88, 9-25.
  2. Daniels MJ and Hogan JW (2008). Missing Data in Longitudinal Studies: Strategies for Bayesian Modeling and Sensitivity Analysis, CRC Press, Boca Raton, FL.
  3. Heagerty PJ (1999). Marginally specified logistic-normal models for longitudinal binary data, Biometrics, 55, 688-698.
  4. Heagerty PJ (2002). Marginalized transition models and likelihood inference for longitudinal categorical data, Biometrics, 58, 342-351.
  5. Jung D-J (2014). The Effects of self-rated health on depression to disabled elderly: The moderating effects of psychosocial resources, Health and Social Welfare Review, 34, 247-275.
  6. Kang E-J (2007). Clustering of lifestyle behaviors of Korean adults using smoking, drinking, and physical activity, Health and Social Welfare Review, 27, 44-66.
  7. KHP (2023). Korea health panel survey, Available from: https://www.khp.re.kr:444/
  8. Kim GS (2021). Self-rated health, depression and anxiety in family caregivers of terminal cancer patients: The mediating effects of bonding social capital and bridging social capital, Health and Social Welfare Review, 41, 212-233.
  9. Kim H-R (2005). The relationship of socioeconomic position and health behaviors with morbidity in Seoul, Korea, Health and Social Welfare Review, 25, 3-35.
  10. Lee K (2022). Longitudinal Data Analysis: Using R, Free Academy, Paju.
  11. Liang K-Y and Zeger SL (1986). Longitudinal data analysis using generalized linear models, Biometrika, 73, 13-22.
  12. Little RJA and Rubin DB (2002). Statistical Analysis with Missing Data (2nd ed), Wiley, New York.
  13. Moon S-J, Sohn M-S, and Choi M-K (2017). The effects of changes in economic activity of the physically disabled on the self-rated health: Focusing on the mediating effect of self-esteem, Disability & Employment, 27, 217-239.
  14. Oh Y-H, Bae H-O, and Kim Y-S (2006). A study on physical and mental function affecting self-perceived health of older persons in Korea, Journal of the Korea Gerontological Society, 26, 461-476.
  15. Park E-J, Jun J, and Kim N-S (2015). The association of multiple risky health behaviors with self-reported poor health, stress, and depressive symptom, Health and Social Welfare Review, 35, 136-157.
  16. Park Y-K, Kim C-Y, and Hwang S-S (2018). Interaction effects of income and unmet healthcare needs to subjective health status: Using the Korea health panel, 2009-2014, Health and Social Science, 47, 57-83.
  17. Schildcrout JS and Heagerty PJ (2007). Marginalized models for moderate to long series of longitudinal binary response data, Biometrics, 63, 322-331.
  18. Song M-S, Song H-J, and Mok J-Y (2003). Community based cross-sectional study on the related factors with perceived health status among the elderly, Journal of the Korea Gerontological Society, 23, 127-142.
  19. Yoon B-J (2016). Differential effects on self-rated health by socioeconomic class, Journal of Health Informatics Statistics, 41, 35-42.