stdca.hlp

{smcl}
{* *! version 1.1  22sep2013}{...}
{viewerdialog twoway "dialog twoway"}{...}
{vieweralsosee "[G-2] graph twoway line" "mansection G-2 graphtwowayline"}{...}
{vieweralsosee "" "--"}{...}
{vieweralsosee "[G-2] graph twoway scatter" "help scatter"}{...}
{vieweralsosee "" "--"}{...}
{vieweralsosee "[G-2] graph twoway fpfit" "help twoway_fpfit"}{...}
{vieweralsosee "[G-2] graph twoway lfit" "help twoway_lfit"}{...}
{vieweralsosee "[G-2] graph twoway mband" "help twoway_mband"}{...}
{vieweralsosee "[G-2] graph twoway mspline" "help twoway_mspline"}{...}
{vieweralsosee "[G-2] graph twoway qfit" "help twoway_qfit"}{...}
{viewerjumpto "Syntax" "line##syntax"}{...}
{viewerjumpto "Description" "line##description"}{...}
{viewerjumpto "Options" "line##options"}{...}
{viewerjumpto "Remarks" "line##remarks"}{...}
{title:Title}

{p2colset 5 32 34 2}{...}
{p2col :Decision Curve Analysis } {p_end}
{p2colreset}{...}

{marker syntax}{...}
{title:Syntax}

{p 8 24 2}
{opt stdca} {indepvars} {ifin} {cmd:, timepoint(#)} [{it:options}]

{marker description}{...}
{title:Description}

{pstd}
Diagnostic and prognostic models are typically evaluated with measures of accuracy that do not address clinical consequences. Decision-analytic techniques allow assessment of clinical outcomes but often require collection of additional information and may be cumbersome to apply to models that yield a continuous result. Decision curve analysis is a method for evaluating and comparing prediction models that incorporates clinical consequences, requires only the data set on which the models are tested,and can be applied to models that have either continuous or dichotomous results. The {cmd:stdca} function performs decision curve analysis for time to event or survival outcomes.

{pstd}
See {browse "http://www.decisioncurveanalysis.org"} for more information.

{synoptset 22}
{p2col:{it:options}}Description{p_end}
{p2line}
{synopt :{opt t:imepoint(#)}}specifies the timepoint at which the decision curve analysis is perfomred. e.g. if you're interested in evaluating 5 year survival, then timepoint(5).{p_end}
{synopt :{opt prob:ability(yes|no)}}specifies whether or not each of the independent variables are probabilities. The default is yes.{p_end}
{synopt :{opt harm(numlist)}}specifies the harm(s) associated with the independent variable(s). The default is none.{p_end}
{synopt :{opt intervention}}plot net reduction in interventions.{p_end}
{synopt :{opt interventionper(#)}}number of net reduction in interventions per interger. Default is 1.{p_end}
{synopt :{opt smooth}}smooth net benefit curve.{p_end}
{synopt :{opt smoother(method)}}smoothing method to apply. Default is 3rssh.{p_end}
{synopt :{opt nograph}}do not display graph.{p_end}
{synopt :{opt xstart(#)}}starting value for x-axis (threshold probability) between 0 and 1. default is 0.01.{p_end}
{synopt :{opt xstop(#)}}stopping value for x-axis (threshold probability) between 0 and 1. default is 0.99.{p_end}
{synopt :{opt xby(#)}}increment for threshold probability; default is 0.01.{p_end}
{synopt :{opt ymin(#)}}minimum bound for graph. default is -0.05.{p_end}
{synopt :{opt compet#(integer)}}if evaluating outcome in presence of acompeting risk(s), {cmd:compet#} is used to specify the value of the event(s) associated with failure due to the competing risk(s).  The {cmd:stcompet} function is used to calculate the risk for the event of interest.  Up to 6 competing risk events may be specified.{p_end}
{synopt :{help prefix_saving_option:{bf:{ul:sa}ving(}{it:filename}{bf:, ...)}}}save
	decision curve analysis results to {it:filename}{p_end}
INCLUDE help gr_twopt
INCLUDE help gr_axlnk
INCLUDE help gr_conopt

{p2line}
{p 4 6 2}
{it:connect_options} discusses options for one {it:y} versus one {it:x};
see {it:{help scatter##connect_options:connect_options}} in 
{helpb scatter:[G-2] graph twoway scatter} when plotting
multiple {it:y}s against one {it:x}.

{marker examples}{...}
{title:Examples}

   {hline}
{pstd}Setup{p_end}
{phang2}{cmd:. webuse stan3}{p_end}

{pstd}Decision Curve Analysis{p_end}
{phang2}{cmd:. stdca age surgery, timepoint(250) prob(no no) smooth legend(cols(1))}{p_end}
{phang2}{cmd:. stdca age, timepoint(250) prob(no) smooth intervention legend(cols(1)) xstart(0.4) xstop(0.6)}{p_end}
{phang2}{cmd:. stdca surgery, timepoint(250) prob(no no) smooth legend(cols(1)) lcolor(black gs8 red green) lpattern(solid solid dash dash) title("Decision Curve Analysis Example", size(4) color(red))}{p_end}

{pstd}Setup{p_end}
{phang2}{cmd:. webuse hypoxia}{p_end}
{phang2}{cmd:. stset dftime, failure(failtype==1)}{p_end}

{pstd}Decision Curve Analysis{p_end}
{phang2}{cmd:. stdca tumsize, timepoint(2) compet1(2) prob(no) smooth}{p_end}
    {hline}

{marker saved_results}{...}
{title:Saved results}

{pstd}
{cmd:dca} saves the following in {cmd:r()}:

{synoptset 20 tabbed}{...}
{p2col 5 20 24 2: Macros}{p_end}
{synopt:{cmd:r(N)}}number of observations{p_end}
{p2colreset}{...}