The non-central X2 distribution can be used to calculate power for tests detecting departure from a null hypothesis. Required sample size can also be calculated because it is proportional to the non-centrality parameter for the distribution. We demonstrate how these calculations can be carried out in Stata using the example of calculating power and sample size for case–control studies of gene–gene and gene–environment interactions. Do-files are available for these calculations.
AlbertP. S., RatnasingheD., TangreaJ., and WacholderS.2001. Limitations of the case-only design for identifying gene-environment interactions. American Journal of Epidemiology154: 687–693.
2.
AndrieuN., and GoldsteinA.M.2000. A case-combined design using both population based- and related-controls: a potential alternative for increasing power in gene–environment interaction detection. Genetic Epidemiology19: 235–236.
3.
AndrieuN., GoldsteinA. M., ThomasD. C., and LangholzB.2001. Counter-matching in studies of gene–environment interaction: efficiency and feasibility. American Journal of Epidemiology153: 265–274.
4.
BeggC. B., and BerwickM.1997. A note on the estimation of relative risks of rare genetic susceptibility markers. Cancer Epidemiology Biomarkers Prevention6: 99–103.
5.
BrennanP.2002. Gene-environment interaction and aetiology of cancer: what does it mean and how can we measure it?Carcinogenesis23: 381–387.
6.
BrownB. W., LovatoJ., and RussellK.1999. Asymptotic power calculations: description, examples, computer code. Statistics in Medicine18: 3137–3151.
7.
GaudermanW. J.2002. Sample size requirements for association studies of gene–gene interaction. American Journal of Epidemiology155: 478–484.
8.
LongmateJ. A.2001. Complexity and power in case–control association studies. American Journal of Human Genetics68: 1229–1237.
9.
PiegorschW. W., WeinbergC. R., and TaylorJ.A.1994. Non-hierarchical logistic models and case-only designs for assessing susceptibility in population-based case–control studies. Statistics in Medicine13: 153–162.
10.
SelfS. G., MauritsenR. H., and OharaJ.1992. Power calculations for likelihood ratio tests in generalized linear models. Biometrics48: 31–39.
11.
SiegmundK. D., and LangholzB.2001. Stratified case sampling and the use of family controls. Genetic Epidemiology20: 316–327.
12.
SturmerT., and BrennerH.2002. Flexible matching strategies to increase power and efficiency to detect and estimate gene–environment interactions in case–control studies. American Journal of Epidemiology155: 593–602.
13.
WeinbergC. R., and UmbachD.M.2000. Choosing a retrospective design to assess joint genetic and environmental contributions to risk. American Journal of Epidemiology152: 197–203.
14.
WilksS. S.1938. The large sample distribution of the likelihood ratio for testing composite hypotheses. Annals of Mathematical Statistics9: 60–62.
15.
WitteJ. S., GaudermanW. J., and ThomasD.C.1999. Asymptotic bias and efficiency in case–control studies of candidate genes and gene–environment interactions: basic family designs. American Journal of Epidemiology149: 693–705.
