원전 코호트 연구의 적정 대상규모와 검정력 추정
Power Estimation and Follow-Up Period Evaluation in Korea Radiation Effect and Epidemiology Cohort Study
1
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
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 405events 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, 136245PY 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.
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 405events 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, 136245PY 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.
참고문헌 (18)
- Ahn YO, "Epidemiology: The Principles and Applications", pp.205~222 (2005), Seoul National University Press
- Ahn YO, "Manual for Medical Statistics", pp.183~242 (2005), Seoul National University Press
- Armitage P, "Repeated significance tests on accumulating data", J Roy Stat Soc Ser A, 132(2), pp.235~244 (1969)
- Willett W, "Nutritional Epidemiology. 2nd ed", pp.1~18 (1998), Oxford University Press
- Bender R, "Introduction to the use of regression models in epidemiology", Methods Mol Biol, 471, pp.179~195 (2009)
- Breslow NE, "Statistical methods in cancer research. Volume II--The design and analysis of cohort studies", IARC Sci Publ, 1~406 (1987)
- Singh H, "Different study designs in the epidemiology of cancer: case-control vs. cohort studies", Methods Mol Biol, 471, pp.217~225 (2009)
- Xue X, "Statistical methods in cancer epidemiological studies", Methods Mol Biol, 471, pp.239~272 (2009)
- Brown CC, "Additional power computations for designing comparative Poisson trials", Am J Epidemiol, 115(5), pp.752~758 (1982)
- Blettner M, "Power calculation for cohort studies with improved estimation of expected numbers of deaths", Soz Praventivmed, 37(1), pp.13~21 (1992)
- Cohen J, "Statistical Power Analysis for the Behavioral Sciences. 2nd ed", pp.1~18 (1988), L Erlbaum Associates
- Goodman SN, "Confidence limits vs power, calculations", Epidemiology, 5(2), pp.266~268 (1994)
- Goodman SN, "The use of predicted confidence intervals when planning experiments and the misuse of power when interpreting results", Ann Intern Med, 121(3), pp.200~206 (1994)
- Gordon I, "Sample size estimation in occupational mortality studies with use of confidence interval theory", Am J Epidemiol, 125(1), pp.158~162 (1987)
- Hoenig JM, "The abuse of power: the pervasive fallacy of power calculations for data analysis", Am Stat, 55(1), pp.19~24 (2001)
- Howe GR, "Methodological issues in cohort studies. II: Power calculations", Int J Epidemiol, 17(2), pp.464~468 (1988)
- Lubin JH, "On power and sample size for studying features of the relative odds of disease", Am J Epidemiol, 131(3), pp.552~566 (1990)
- Maxwell SE, "Sample size planning for statistical power and accuracy in parameter estimation", Annu Rev Psychol, 59, pp.537~563 (2008)