{smcl}
{* 07 Sept 2005}{...}
{hline}
help for {hi:binscrn1} {right: (version 1.1.2) }
{hline}

{title:Summary screening measures for a test with binary outcome}

{p 8 12}{cmd:binscrn1}
{it:test_var}
{it:disease_var}
{ifin}
[, {opt l:evel(#)}
{opt f:ormat(%fmt)}]


{title:Description}

{pstd}
{cmd:binscrn1} calculates true and false positive rates, positive and negative
predictive values, and diagnostic likelihood ratios for binary test outcome
data.

{pstd}
Positive and negative predictive values assume cohort data; i.e. that
the proportion of disease positive observations in the dataset represents the
population disease prevalence.

{pstd}
{it:test_var} is a 0/1 test outcome variable and
{it:disease_var} is the  0/1 (disease) group indicator variable.

{pstd}All screening measures are available as returned results after running the
program:

{col 8}   TPF: r(tp)         FPF: r(fp)
{col 8}   PPV: r(ppv)        NPV: r(npv)
{col 8}  DLR+: r(dlrpos)    DLR-: r(dlrneg)
{col 8}       disease prevalence: r(disprev)

{pstd}along with confidence limits for TPF, FPF, PPV, NPV:{p_end}
{col 6}eg. for TPF,

{col 8}Clopper-Pearson:  [r(tp_lb_ex),r(tp_ub_ex)]
{col 8}         Wilson:  [r(tp_lb_wi),r(tp_ub_wi)]
{col 8}          logit:  [r(tp_lb_lg),r(tp_ub_lg)]
{pstd}and,

{col 8} s.e.(ln(DLR+)):  r(selndlrp)
{col 8} s.e.(ln(DLR-)):  r(selndlrn)

{pstd} and the following joint confidence regions:

{col 6} (TPF, FPF): r(tpj_lb),r(tpj_ub) x r(fpj_lb),r(fpj_ub)
{col 6} (PPV, NPV): r(ppvj_lb),r(ppvj_ub) x r(npvj_lb),r(npvj_ub)
{col 6}(DLR+,DLR-): r(dlrpj_lb),r(dlrpj_ub) x r(dlrnj_lb),r(dlrnj_ub)


{title:Options}

{phang}
{opt level(#)}; see {help estimation options##level():estimation options}.

{phang}
{opt format(%fmt)} specifies the display {help format} for estimates and CI's.  The
    default is %4.2g


{title:Remarks}

{pstd} originally written for Margaret Pepe's "Statistical Methods for Evaluating Medical
Diagnostic Tests" course (Biostat 578)

{pstd} Rectangular joint confidence regions (TP, FP) and (PPV, NPV) are based on
on sqrt(1-alpha) level logit-based confidence limits for the component measures
TP, FP, PPV, NPV.

{pstd}Joint confidence regions for (DLR+, DLR-) are derived from the sqrt(1-alpha) level
logit-based limits for TP and FP.

{pstd}see {help binscrn2} for comparison of 2 screening tests (binary outcome) with
unpaired data (independent samples) and {help binscrn3} for comparison of tests
with paired data.


{title:References}

{pmore}Pepe MS.  {browse "http://www.fhcrc.org/science/labs/pepe/book":The Statistical Evaluation of Medical Tests for Classification and Prediction.}
Oxford University Press.  Chapter 2.  2003.
{break}

{pstd}calculation of confidence intervals for DLR+ and DLR-:

{pmore}Simel DL, Samsa GP, and Matchar DB.  Likelihood ratios with confidence: sample
size estimation for diagnostic test studies.  J. of Clin. Epidem. 44(8):763-70,
1991.

{pstd}Wilson confidence limits for binomial proportions:

{pmore}Brown LD, Cai TT, DasGupta A.  Interval estimation for a binomial proportion.
Statistical Science 16:101-133. 2001


{title:Examples}

{phang} .use http://www.fhcrc.org/science/labs/pepe/book/data/est1, clear {p_end}
{phang} .binscrn1 y1 d {p_end}
{phang} .binscrn1 y2 d, level(90) format(%6.2g){p_end}


{title:Author}

{pstd}Gary Longton, Fred Hutchinson Cancer Research Center, Seattle, WA.{break}
glongton@fhcrc.org