DOI QR코드

DOI QR Code

이중구간중도절단된 생존자료의 생존함수 비교를 위한 검정: 한국인 암 예방연구 중 당뇨병에의 응용

Comparing Survival Functions with Doubly Interval-Censored Data: An Application to Diabetes Surveyed by Korean Cancer Prevention Study

  • 지선하 (연세대학교 보건대학원 국민건강증진연구소) ;
  • 남정모 (연세대학교 의과대학 예방의학교실) ;
  • 김진흠 (수원대학교 통계정보학과)
  • Jee, Sun-Ha (Institute for Health Promotion, Graduate School of Public Health Yonsei University) ;
  • Nam, Chung-Mo (Department of Preventive Medicine, Yonsei University College of Medicine) ;
  • Kim, Jin-Heum (Department of Applied Statistics, University of Suwon)
  • 발행 : 2009.06.30

초록

이중구간중도절단된 자료의 생존함수를 비교하기 위한 두 검정법을 소개하고 한국인 암 예방연구 (Jee 등, 2005) 자료에 적용하여 당뇨병 잠복시간의 분포를 성별과 연령에 따라 비교하였다. Kim 등 (2006)의 검정법을 이중구간중도절단된 자료로 확장한 검정법은 위험집합의 크기에만 의존하는 가중값을 사용하기 때문에 대용량 자료의 분석에서 Sun (2006)의 검정법보다 계산 시간을 대폭 줄일 수 있으며, 이산형 생존자료뿐만 아니라 연속형 생존자료에도 적용가능한 장점이 있다. 당뇨병의 잠복시간이 성별에 따라 매우 유의하게 달랐으며 여자의 잠복시간이 남자보다 긴 것으로 나타났다. 4개 연령그룹 간 당뇨병의 잠복시간도 성별에 관계 없이 매우 유의하게 달랐으며 여자의 경우가 남자의 경우보다 그 차이가 훨씬 더 유의했다. 한편, 소표본 모의실험을 통해 제안한 검정법과 Sun (2006)의 검정법의 검정력을 비교하였으며 제안한 검정법의 검정력이 Sun (2006)의 검정법보다 더 좋은 것으로 나타났다.

Two tests were introduced for comparing several survival functions with doubly interval-censored data and illustrated with data surveyed by Korean Cancer Prevention Study (Jee et al., 2005). The test which extended Kim et al. (2006)'s test to the doubly interval-censored data has an advantage over Sun (2006)'s test in terms of saving computation time because the proposed test only depends on the size of risk set, and also the proposed test is applicable to continuous failure time data as well as discrete failure time data unlike Sun's test. Comparing male with female groups on the incubation time of diabetes was highly different and the survival of female group was longer than that of male one. Regardless of gender, the difference in survival functions of four age groups was highly significant with p-value of less than 0.001. This trend was more remarkable for female group than for male one. Simulation results showed that the significance level of both tests was well controlled and the proposed test was better than Sun's test in terms of power.

키워드

참고문헌

  1. 김진화, 한미아, 박철진, 박일구, 신지혜, 김상용, 류소연, 배학연 (2008). 나주지역 중년인구에서 공복혈당을 통한 당뇨병 진단의 적절성 평가, <대한당뇨병학회>, 32, 328-337 https://doi.org/10.4093/kdj.2008.32.4.328
  2. 박이병 (2006). Priediabetes-risk of type 2 diabetes and CVD, <제19차 대한당뇨병학회 춘계학술대회>
  3. 통계청 (2007). <2006년 사망원인 통계연보>, 통계청
  4. De Gruttola, V. G. and Lagakos, S. W. (1989). Analysis of doubly-censored survival data, with application to AIDS, Biometrics, 45, 1-11 https://doi.org/10.2307/2532030
  5. Fay, M. P. (1996). Rank invariant tests for interval censored data under the grouped continuous model, Biometrics, 52, 811-822 https://doi.org/10.2307/2533044
  6. Fonseca, V. A. (2007). Identification and treatment of prediabetes to prevent progression to type 2 diabetes, Clinical Cornerstone, 8, 10-18 https://doi.org/10.1016/S1098-3597(09)60004-1
  7. Gomez, G. and Lagakos, S. W. (1994). Estimation of the infection time and latency distribution of AIDS with doubly censored data, Biometrics, 50, 204-212 https://doi.org/10.2307/2533210
  8. Jee, S. H., Ohrr, H., Sull, J. W., Yun, J. E., Ji, M. and Samet, J. M. (2005). Fasting serum glucose level and cancer risk in Korean men and women, Journal of the American Medical Association, 293, 194-202 https://doi.org/10.1001/jama.293.2.194
  9. Kim, J., Kang, D. R. and Nam, C. M. (2006). Logrank-type tests for comparing survival curves with intervalcensored data, Computational Statistics and Data Analysis, 50, 3165-3178 https://doi.org/10.1016/j.csda.2005.06.014
  10. Klein, J. P. and Moeschberger, M. L. (2003). Survival Analysis: Techniques for Censored and Truncated Data, Springer, New York
  11. Korea Centers for Disease Control and Prevention (2007). <국민건강영양조사 제3기 조사결과 심층분석 연구보고서: 검진부문>, 질병관리본부
  12. Little, R. J. A. and Rubin, D. B. (2002). Statistical Analysis with Missing Data, John Wiley & Sons, New York
  13. Pan, W. (2000). A two-sample test with interval censored data via multiple imputation, Statistics in Medicine, 19, 1-11 https://doi.org/10.1002/(SICI)1097-0258(20000115)19:1<1::AID-SIM296>3.0.CO;2-Q
  14. Pan, W. (2001). A multiple imputation approach to regression analysis for doubly censored data with application to AIDS studies, Biometrics, 57, 1245-1250 https://doi.org/10.1111/j.0006-341X.2001.01245.x
  15. Self, S. G. and Grossman, E. A. (1986). Linear rank tests for interval-censored data with application to PCB levels in adipose tissue of transformer repair workers, Biometrics, 42, 521-530 https://doi.org/10.2307/2531202
  16. Sun, J. (1996). A nonparametric test for interval-censored failure time data with application to AIDS studies, Statistics in Medicine, 15, 1387-1395 https://doi.org/10.1002/(SICI)1097-0258(19960715)15:13<1387::AID-SIM268>3.0.CO;2-R
  17. Sun, J. (1997). Self-consistency estimation of distributions based on truncated and doubly censored data with applications to AIDS cohort studies, Lifetime Data Analysis, 3, 305-313 https://doi.org/10.1023/A:1009609227969
  18. Sun, J. (2001). Nonparametric test for doubly interval-censored failure time data, Lifetime Data Analysis, 7, 363-375 https://doi.org/10.1023/A:1012596731505
  19. Sun, J. (2006). The Statistical Analysis of Interval-censored Failure Time Data, Springer, New York
  20. Zhao, Q. and Sun, J. (2004). Generalized log-rank test for mixed interval-censored failure time data, Statistics in Medicine, 23, 1621-1629 https://doi.org/10.1002/sim.1746