新藥又稱作(原廠藥)的研發平均需花10~12年的時間和八百億美元，因此新藥的研發是非常耗時又需花費巨大金額，為反映成本和賺取利潤，通常新藥的價格非常昂貴，於是政府為了降低藥物價格，同意原研發藥廠等到專利期滿，一般藥廠就可仿照原廠藥製造出具相同療效的藥又稱作(學名藥)，在1984年美國食品與藥物管理局(FDA)准許學名藥上市，只須証明學名藥與原廠藥是否具有平均生體相等性( bioequivalence)即可，因為學名藥不須長期的臨床試驗，故學名藥廠可以節省大量的金錢和研發時間。
目前，針對於評估平均生體相等性，FDA建議採用一兩序列兩期重複的交叉試驗設計，而評估平均生體相等性是一種區間假設檢定，有許多學者提出不同的方法來評定，此種區間假設檢定包括Schuirmann提出的合併單邊檢定方法(Schuirmann’s Two One-Sided Tests)，及Anderson和Hauck提出的方法(Anderson and Hauck’s Test)以及許多貝氏估計方法。基於Anderson 和 Hauck方法下，本文透過拔靴重複抽樣法(Bootstrap Resampling Methods)提出一個模擬研究，改善原來方法中對於非中心參數估計的盲點，也比較許多不同方法和非常態分配假設下，在型I誤差的機率和檢定力的優劣。

Only after the patent of a brand-name innovative drug product is expired, its generic copies are allowed to market. However, regulatory approval requires evidence of bioequivalence based on the pharmacokinetic responses derived from the time-plasma concentration curve of the active ingredients. Currently, about the evaluation of interval hypotheses testing of average bioequivalence, Schuirmann’s two one-sided tests and Anderson and Hauck’s test would be introduced.
However, there exists some problem within the Anderson and Hauck’s test. It need to calculate the noncentrality parameter in test statistics. The Bootstrap resampling method is introduced to perform the empirical distribution. The real test statistics distribution is performed by a simulation study on the various combinations of parameters and sample size under 2 2 crossover design. The simulation study was conducted to empirically examine and compare the size and power of the original test and our proposed four methods. In the simulation study, we also recommend exponential distribution, uniform distribution and Cauchy distribution to replace with the normal distribution.

1. Anderson, S. and Hauck, W.W. (1983). A new procedure for testing equivalence in comparative bioavailability and other clinical trials. Communications in Statistics-Theory and Methods, 12, 2663-2692.
2. Babu, G.J., and Rao, C. R. (1993). Bootstrap methodology. In Handbook of Statistics 9, 627-659.
3. Cheng, R. C. H., and Jones, O. D. (2000). Analysis of simulation factorial experiments by EDF resample statistics. In Proceeding of the 2000 Winter Simulation Conference, ed. J. A. Joines, R. R. Barton, L. Kang, and P. A. Fishwick, 697-703. Piscataway, New Jersey: Institute of Electrical and Electronics Engineers.
4. Chernick, M.R. (1999). Bootstrap Methods A Practitioner’s Guide. New York, Wiley.
5. Chinchilli V.J. and Esinhar (1996) Design and analysis of intra-subject variability in crossover experiments. Statistics in Medicine, 15:1619-1634.
6. Chow, S.C. and Shao, J. (1990). An alternative approach for the assessment of bioequivalence between two formulations of a drug. Biometrical Journal, 32, No. 8, 969-976.
7. Chow, S.C., Cheng, J. and Shao, J. (1990). A nonparametric bootstrap approach for assessing bioequivalence. Unpublished manuscript.
8. Chow, S.C. and Liu, J.P. (2000) Design and Analysis of Bioavailability and Bioequivalence Studies. Marcel Dekker, New York.
9. Cornell, R.G. (1980). Evaluation of Bioavailability Data Using Nonparametric Statistics in Drug Absorption and Disposition: Statistical Considerations. ed. by K.S. Albert, American Pharmaceutical Association, Academy of Pharmaceutical Sciences, 51-57. Washington, D.C.
10. Dunnett, C.W. and Gent, M. (1977). Significance testing to establish equivalence between treatments, with special reference to data in the form of 2 X 2 tables. Biometrics, 33, 593-602.
11. Efron, B. (1982). The Jackknife, the Bootstrap and Other Resampling Plans, SIAM, Philadelphia.
12. Efron, B. and Tibshirani R.J. (1986). Bootstrap methods for standard errors, confidence intervals and other measures of statistical accuracy, Statistical Science 1, 54-77.
13. Efron, B. and Tibshirani R.J. (1993). An introduction to the Bootstrap New York and London: Chapman & Hall.
14. Frick, H. (1987). On level and power of Anderson and Hauck’s procedure for testing equivalence in comparative bioavailability. Communications in Statsitics-Theory and Methods, 16, 2771-2778.
15. Griece, A.P. (1985). A Bayesian analysis of the two-period changeover design and its use in clinical trials. Biometrics, 21, 467-480.
16. Hauschke, D. , Steinijans, V.W., and Diletti, E. (1990). A distribution-free procedure for the statistical analyses of bioequivalence studies. Int. J. Clin. Pharmacol. Ther. Toxic., 28, 72-78.
17. Kendall, M.G. and Stuart, A. (1961). The Advance Theory of Statistics. Vol. Ⅱ, Griffen, London, England.
18. Kleijnen, J. P. C., Cheng, R. C. H. and Bettonvil, B. (2000). Validation of trace-driven simulation models: more on bootstrap tests. In Proceeding of the 2000 Winter Simulation Conference, ed. J. A. Joines, R. R. Barton, L. Kang, and P. A. Fishwick, 882-892. Piscataway, New Jersey: Institute of Electrical and Electronics Engineers.
19. Lehamnn, E.L. (1959). Testing Statistical Hypotheses. John Wiley and Sons, New York, NY.
20. Liu, J.P. (1991). Bioequivalence and intrasubject variability. Journal of Biopharmaceutical Statistics, 1, No. 2 205-219.
21. Locke, C.S. (1984). An exact confidence interval for untransformed effect data for the ratio of two formulations means. J. Pharmacokin. Biopharm., 12, 649-655.
22. Mandallaz, D. and Mau, J. (1981). Comparison of different methods for decision-making in bioequivalence assessment. Biometrics, 37, 213-222.
23. Metzler, C.M. (1974). Bioavailability: A problem in bioequivalence. Biometrics, 30, 309-317.
24. Mller-Cohrs, J. (1990). The power of the Anderson-Hauck test and double t-test. Biometrical Journal, 32, 259-266.
25. Munk, A. (2000) Connections between average and individual bioequivalence. Statistics in Medicine, Vol. 19, 2843-2854.
26. Racing-Poon, A., Grieve, A.P., Fluehler, H., and Smith, A.F. (1986). Bayesian methods in practice: experiences in the pharmaceutical industry (with discussion). Applied Statistics, 35, 93-150.
27. Racing-Poon, A., Grieve, A.P., Fluehler, H., and Smith, A.F. (1987). A two-stage procedure for bioequivalence studies. Biometrics, 43, 847-856.
28. Rodda, B.E. and Davis, R.L. (1980). Determining the probability of an important difference in bioavailability. Clin. Pharmacol. Ther., 28, 247-252.
29. Schuirmann, D.J. (1981). On hypothesis testing to determine if the mean of a normal distributions is continued in a known interval. Biometrics, 37, p.617(Abstract).
30. Schuirmann, D.J. (1987). A comparison of the two one-sided teste procedure and the power approach for assessing the equivalence of average bioavailability. J. of Pharmacokin. Biopharm., 15, 657-680.
31. Steinijans, V.W. and Diletti, E. (1983). Statistical analysis of bioavailability studies: parametric and nonparametric confidence intervals. Europ. J. Clin. Pharmacol., 24, 127-136.
32. Steinijans, V.W. and Diletti, E. (1985). Generalized of distribution-free confidence intervals for bioavailability ratios. Europ. J. Clin. Pharmacol., 28, 85-88.
33. US FDA. Guidance for industry on statistical approaches to establishing bioequivalence (2001). Center for Drug Evaluation and Research, Food and Drug Administration, Rockville, Maryland.
34. Westlake, W.J. (1972). Use of confidence intervals in analysis of comparative bioavailability trials. Journal of Pharmaceutical Sciences, 61, 1340-1341.
35. Westlake, W.J. (1973). The design and analysis of comparative blood-level trials. Current Concepts in the Pharmaceutical Sciences. ed. by J. Swarbrick, Lea & Febiger, Philadelphia, PA.
36. Westlake, W.J. (1974). The use of balanced incomplete block designs in comparative bioavailability trials. Biometrics, 30, 319-327.
37. Westlake, W.J. (1976). Symmetrical confidence intervals for bioequivalence trials. Biometrics, 32, 741-744.
38. Westlake, W.J. (1981). Bioequivalence testing- A need to rethink (Reader reaction response). Biometrics, 37, 591-593.
39. Vladimir Dragalin, Valerii Fedorov, Scott Patterson, Byron Jones (2003). Kullback-Leibler divergence for evaluating bioequivalence. Statistics in Medicine, Vol. 22,913-930.