------------------------------------------------------------------------------- help forstepp_tail,stepp_window,stepp_plotPatrick Royston -------------------------------------------------------------------------------

Subpopulation treatment effect pattern plot (STEPP)

stepp_tailregression_cmd[yvar]zvar[adjvars] [if] [in],options

stepp_windowregression_cmd[yvar]zvar[adjvars] [if] [in],options

stepp_plotstubname[,options]

regression_cmdmay beclogit,cnreg,glm,intreg,logistic,logit,mlogit,nbreg,ologit,oprobit,poisson,probit,qreg,regress,stcox,streg, orxtgee.

optionsDescription -------------------------------------------------------------------------gen(stubname)creates five new variables containing results of STEPP analysisg(#)(stepp_tailonly) defines the number of subpopulation groupsn1(#)(stepp_windowonly) defines the number of individuals belonging only to one of two neighbouring subpopulationsn2(#)(stepp_windowonly) defines the number of individuals in a subpopulationtreatment(trt_varlist)defines the list of variables whose interactions withzvarare to be studiedregression_cmd_optionsoptions forregression_cmd

optionsforstepp_plotvn(#)variable number intreatment()plot(plot)adds other plots to the generated graphgraph_optionsoptions ofgraph twoway-------------------------------------------------------------------------

All weight types supported by

regression_cmdare allowed; see weight.

yvaris not allowed forstregandstcox. For these commands, you must firststsetyour data.

Description

ststep_tailandststep_windowcompute Bonetti & Gelber (2000, 2004)'s STEPP estimators for graphical exploration of a treatment/covariate interaction.ststep_tailprovides the tail-oriented estimator, andststep_windowthe sliding-window estimator. Plotting the results may be done by usingstepp_plot, in which casestubnameis the same as in thegen(stubname)option ofstepp_tailandstepp_window.

zvaris the continuous covariate whose interaction with treatment is to be studied, andadjvarsis a list of other covariates used to linearly adjust each model fitted to the treatment variable(s) defined bytreatment().

Options

Options forstepp_tailand:stepp_window

gen(stubname)creates five new variables calledstubnameb,stubnamese,stubnamemean,stubnamelb,stubnameub.stubnamebis the estimated regression coefficient in each subpopulation,stubnameseis its standard error,stubnamemeancontains the mean ofzvarin each subpopulation, andstubnamelbandstubnameubare pointwise 95% confidence limits forstubnameb. Iftreatment()includes more than one variable, the created variables have 2, 3, ... appended to the names, e.g.stubnameb2.

g(#)(stepp_tailonly) defines the number of subpopulation groups. The actual number of subpopulations used is 2 * # - 1.

n1(#)(stepp_windowonly) defines the number of individuals belonging only to one of two neighbouring subpopulations.

n2(#)(stepp_windowonly) defines the number of individuals in a subpopulation. The overlap between two neighbouring subpopulations isn2()minusn1()individuals.

treatment(trt_varlist)defines the list of variables whose interactions withzvarare to be studied. Typicallytrt_varlistwill comprise just one binary variable, representing the two arms of a parallel-group clinical trial.

regression_cmd_optionsare options forregression_cmd.

Options for:stepp_plot

vn(#)#is an integer defining the variable number intreatment(), when more than one variable is specified. When only one variable is specified,vn()is not required.

plot(plot)provides a way to add other plots to the generated graph; see help plot option.

graph_optionsare options ofgraph twoway, such asxtitle(),ytitle(), etc.

Examples

. stepp_tail regress y x a1 a2, g(10) gen(z) treatment(t)

. stepp_window stcox x a1 a2, n1(40) n2(50) gen(z) treatment(t)

. stepp_plot z, xtitle("Serum rhubarb") ytitle("log relative hazard")name(myplot)

AuthorPatrick Royston, MRC Clinical Trials Unit, London. pr@ctu.mrc.ac.uk

ReferencesM. Bonetti and R. D. Gelber. 2000. A graphical method to assess treatment-covariate interactions using the Cox model on subsets of the data.

Statistics in Medicine19: 2595-2609.M. Bonetti and R. D. Gelber. 2004. Patterns of treatment effects in subsets of patients in clinical trials.

Biostatistics5: 465-481.P. Royston and W. Sauerbre. 2009. Two techniques for investigating interactions between treatment and continuous covariates in clinical trials.

Stata Journal9(2): 230-251.W. Sauerbrei, P. Royston and K. Zapien. 2007. Detecting an interaction between treatment and a continuous covariate: a comparison of two approaches.

Computational Statistics and Data Analysis51: 4054-4063.

Also seeArticle:

Stata Journal, volume 9, number 2: st0164Online: mfpi