DOI QR코드

DOI QR Code

Power Estimation and Follow-Up Period Evaluation in Korea Radiation Effect and Epidemiology Cohort Study

원전 코호트 연구의 적정 대상규모와 검정력 추정

  • Cho, In-Seong (Department of Preventive Medicine, College of Medicine, Seoul National University) ;
  • Song, Min-Kyo (Department of Preventive Medicine, College of Medicine, Seoul National University) ;
  • Choi, Yun-Hee (Medical Research Collaborating Center, Seoul National University Hospital/Seoul National University) ;
  • Li, Zhong-Min (Institute of Radiation Effect & Epidemiology, Seoul National University Medical Research Centre) ;
  • Ahn, Yoon-Ok (Department of Preventive Medicine, College of Medicine, Seoul National University)
  • 조인성 (서울대학교 의과대학 예방의학교실) ;
  • 송민교 (서울대학교 의과대학 예방의학교실) ;
  • 최윤희 (서울대학교 의학연구협력센터) ;
  • 이충민 (서울대학교 의학연구원 원자력영향역학연구소) ;
  • 안윤옥 (서울대학교 의과대학 예방의학교실)
  • Received : 2010.06.09
  • Accepted : 2010.10.12
  • Published : 2010.11.30

Abstract

Objectives: The objective of this study was to calculate sample size and power in an ongoing cohort, Korea radiation effect and epidemiology cohort (KREEC). Method: Sample size calculation was performed using PASS 2002 based on Cox regression and Poisson regression models. Person-year was calculated by using data from '1993-1997 Total cancer incidence by sex and age, Seoul' and Korean statistical informative service. Results: With the assumption of relative risk=1.3, exposure:non-exposure=1:2 and power=0.8, sample size calculation was 405 events based on a Cox regression model. When the relative risk was assumed to be 1.5 then number of events was 170. Based on a Poisson regression model, relative risk=1.3, exposure:non-exposure=1:2 and power=0.8 rendered 385 events. Relative risk of 1.5 resulted in a total of 157 events. We calculated person-years (PY) with event numbers and cancer incidence rate in the nonexposure group. Based on a Cox regression model, with relative risk=1.3, exposure:non-exposure=1:2 and power=0.8, 136 245PY was needed to secure the power. In a Poisson regression model, with relative risk=1.3, exposure:non-exposure=1:2 and power=0.8, person-year needed was 129517PY. A total of 1939 cases were identified in KREEC until December 2007. Conclusions: A retrospective power calculation in an ongoing study might be biased by the data. Prospective power calculation should be carried out based on various assumptions prior to the study.

Keywords

References

  1. Ahn YO, Yoo KY, Park BJ, Kim DH, Bae JM, Kim DH, et al. Epidemiology: The Principles and Applications. Seoul: Seoul National University Press; 2005 p. 205-222. (Korean)
  2. Ahn YO, Yoo KY, Park BJ. Manual for Medical Statistics. Seoul: Seoul National University Press; 2005 p. 183-242. (Korean)
  3. Armitage P, McPherson CK, Rowe BC. Repeated significance tests on accumulating data. J Roy Stat Soc Ser A 1969; 132(2): 235-244. https://doi.org/10.2307/2343787
  4. Willett W. Nutritional Epidemiology. 2nd ed. New York: Oxford University Press; 1998. p. 1-18.
  5. Bender R. Introduction to the use of regression models in epidemiology. Methods Mol Biol 2009; 471: 179-195. https://doi.org/10.1007/978-1-59745-416-2_9
  6. Breslow NE, Day NE. Statistical methods in cancer research. Volume II--The design and analysis of cohort studies. IARC Sci Publ 1987;(82): 1-406.
  7. Singh H, Mahmud SM. Different study designs in the epidemiology of cancer: case-control vs. cohort studies. Methods Mol Biol 2009; 471: 217-225. https://doi.org/10.1007/978-1-59745-416-2_11
  8. Xue X, Hoover DR. Statistical methods in cancer epidemiological studies. Methods Mol Biol 2009; 471: 239-272. https://doi.org/10.1007/978-1-59745-416-2_13
  9. Brown CC, Green SB. Additional power computations for designing comparative Poisson trials. Am J Epidemiol 1982; 115(5): 752-758. https://doi.org/10.1093/oxfordjournals.aje.a113357
  10. Blettner M, Ashby D. Power calculation for cohort studies with improved estimation of expected numbers of deaths. Soz Praventivmed 1992; 37(1): 13-21. https://doi.org/10.1007/BF01369097
  11. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: L Erlbaum Associates; 1988. p. 1-18, p. 531-542.
  12. Goodman SN. Confidence limits vs power, calculations. Epidemiology 1994; 5(2): 266-268. https://doi.org/10.1097/00001648-199403000-00025
  13. Goodman SN, Berlin JA. The use of predicted confidence intervals when planning experiments and the misuse of power when interpreting results. Ann Intern Med 1994; 121(3): 200-206. https://doi.org/10.7326/0003-4819-121-3-199408010-00008
  14. Gordon I. Sample size estimation in occupational mortality studies with use of confidence interval theory. Am J Epidemiol 1987; 125(1): 158-162. https://doi.org/10.1093/oxfordjournals.aje.a114499
  15. Hoenig JM, Heisey DM. The abuse of power: the pervasive fallacy of power calculations for data analysis. Am Stat 2001; 55(1): 19-24. https://doi.org/10.1198/000313001300339897
  16. Howe GR, Chiarelli AM. Methodological issues in cohort studies. II: Power calculations. Int J Epidemiol 1988; 17(2): 464-468. https://doi.org/10.1093/ije/17.2.464
  17. Lubin JH, Gail MH. On power and sample size for studying features of the relative odds of disease. Am J Epidemiol 1990; 131(3): 552-566. https://doi.org/10.1093/oxfordjournals.aje.a115530
  18. Maxwell SE, Kelley K, Rausch JR. Sample size planning for statistical power and accuracy in parameter estimation. Annu Rev Psychol 2008; 59: 537-563. https://doi.org/10.1146/annurev.psych.59.103006.093735