Skip to content
National Cancer Institute National Cancer Institute U.S. National Institutes of Health
Division of Cancer Prevention logo
Home Site Map Contact DCP
Programs & Resources

Biometry Research Group

Statistical Software

Periodic Screening Evaluation

(Written by Stuart G. Baker)

New Approach (Simplified Approximation): See Baker SG. Evaluating periodic cancer screening without a randomized control group: a simplified design and analysis. In: Duffy SW, Hill C, Esteve J (eds). Methods of Evaluation of Cancer Screening. To appear.

Original Approach: Periodic Screening Evaluation is a method for estimating the reduction in mortality from starting periodic cancer screening at various ages based on either data from a special nonrandomized or randomized design. The methodology is described in Baker SG. Evaluating the age to begin periodic breast cancer screening using data from a few regularly scheduled screens. Biometrics 1998;54:1569-1578. It simplifies and extends Baker SG, Chu KC. Evaluating screening for the early detection and treatment of cancer without using a randomized control group. Journal of the American Statistical Association 1990;85:321-327.

It runs in Mathematica 2.2 or 3.0 and requires the files listed below.


Download All (ZIP, 20KB)

File name / sizeDescription
pse.m (M File, 13KB)main program (which loads others)
psederiv.m (M File, 5KB)derivatives used for evaluation
psedata.m (M File, 5KB)source of data
psefit.mfitting models to the data
psepreg.m (M File, 3KB)Poisson regression
pseplot.m (M File, 6KB)plotting program
statplot.m (M File, 3KB)plotting confidence intervals
matrixlx.m (M File, 31KB) matrix programming language
psefalse.m (M File, 802KB) computation for false positives

To reproduce calculations in the manuscript, load pse.m and use the function PSEHIP[dataHIP, a, z].

Data are available in psedata.m: dataOBS is for nonrandomized study; dataRAN is for a randomized study. These should be modified with new data.

To run PSE for ages, say 40 to 50, with mortality endpoint use:

  • PSEOBS[dataOBS,a,z,{40,50},"MORT"] for a nonrandomized study
  • PSERAN[dataRAN,a,z,{40,50},"MORT"] for a randomized study.

Back to top