help sampsi_fleming-------------------------------------------------------------------------------

Title

Exact sample size calculation for single-stage designs

Syntax

sampsi_fleming[,options]

optionsDescription ------------------------------------------------------------------------- Mainp0(#) specifies the null proportion.p1(#) specifies the alternative proportion.alpha(#) specifies the type I error, the default is 0.05.power(#) specifies the power, the default is 0.9.start(#) specifies the start sample size in the initial sample size search, the default is 1.end(#) specifies the largest sample size in the initial sample size search, the default is 40. -------------------------------------------------------------------------

DescriptionA single stage design requires all

npatients to receive and experimental treatment. Within a specified time period they will be observed to have responded (success) or not (failure). The probability of a success in the trial ispand this is compared to a fixed valuep0, which is the probability of a success under the null hypothesis, this could represent no treatment or standard treatment. The trial therefore tests the null hypothesisH0:p=p0against the alternative hypothesisH1:p>p0. A particular valuep1(p1>p0) is considered to be the desired probability of success.The number of successes,

S, will follow a Binomial distribution with parametersnandp. A sample size is found by using a search such thatP(S>=u|p0)<=alphaandP(S>=u|p1)>=power. The design is therefore characterised by the sample sizenand the critical valueu, the null hypothesis is rejected if there areuor more successes in a trial of sizen.

Latest VersionThe latest version is always kept on the SSC website. To install the latest version click on the following link

ssc install sampsi_fleming, replace.

Options+------+ ----+ Main +-------------------------------------------------------------

p0(#)specifies the null proportion.

p1(#)specifies the alternative proportion.

alpha(#)specifies the type I error, the default is 0.05.

power(#)specifies the power, the default is 0.9.

start(#)specifies the start sample size in the initialgrid search, the default is 1.

end(#)specifies the largest sample size in the initial grid search, the default is 40.

Examples

The default sample size calculation

sampsi_fleming

The default design has a sample size of 47 and a critical value of 15, the type I error is 0.0366 and power is 0.9012.

sampsi_fleming, p0(0.45) p1(0.5) s(800)

A sample size of 861 and a critical value of 412 gives a type I error of 0.0499 and a power of 0.9024.

sampsi_fleming, a(0.01) p(0.8)

A sample size of 52 and a critical value of 18 gives a type I error of 0.0099 and a power of 0.8245.

sampsi_fleming, p0(0.1) p1(0.35) a(0.01) p(0.8)

A sample size of 25 and a critical value of 7 gives a type I error of 0.0095 and a power of 0.8266.

AuthorAdrian Mander, MRC Biostatistics Unit, Cambridge, UK. Email adrian.mander@mrc-bsu.cam.ac.uk

ReferencesR.P.A'Hern (2001) Sample size tables for exact single-stage phase II designs.

Statistics in Medicine20:859-866.J.Whitehead (2008) Bayesian sample size for exploratory clinical trials incorporating historical data.

Statistics in Medicine27:2307-2327.T.R. Fleming (1982) One-sample multiple testing procedure for phase II clinical trials.

Biometrics38:143-151.

Also seeRelated commands

HELP FILES SSC installation links Description

samplesize (if installed) (ssc install samplesize) Sample Size graphics sampsi_reg (if installed) (ssc install sampsi_reg) Sample Size for linear regression sampsi_mcc (if installed) (ssc install sampsi_mcc) Sample Size for matched case/control studies sampsi_rho (if installed) (ssc install sampsi_rho) Sample Size for Pearson correlation