A computer program for the analysis of matched case-control studies

https://doi.org/10.1016/0010-4809(81)90031-8Get rights and content

Abstract

The use of regression methods for matched epidemiologic studies is discussed. A computer program is given for use in analysis of 1-1 and 1-R matched case-control data as well as the more general stratified analysis, where m cases are matched with n controls in each of K strata. The program permits variable matching ratios. An illustration is given.

References (24)

  • D.G. Seigel et al.

    Validity in estimating relative risk in case-control studies

    J. Chron. Dis.

    (1973)
  • N.E. Breslow et al.

    Estimation of multiple relative risk functions in matched case-control studies

    Amer. J. Epidemiol.

    (1978)
  • N.E. Breslow et al.

    Are there two logistic regressions for retrospective studies?

    Biometrics

    (1978)
  • J.H. Lubin

    The Development and Examination of Some Regression Methodologies in Case-Control Studies

  • R.L. Prentice

    Use of the logistic model in retrospective studies

    Biometrics

    (1976)
  • J.A. Anderson

    Separate sample logistic discriminination

    Biometrika

    (1972)
  • R.L. Prentice et al.

    Logistic disease incidence models and case-control data

    Biometrika

    (1979)
  • N. Mantel

    Synthetic retrospective studies and related topics

    Biometrics

    (1973)
  • B. MacMahon et al.
  • O.S. Miettinen

    Matching and design efficiency in retrospective studies

    Amer. J. Epidemiol.

    (1970)
  • E.C. Hammand

    Smoking and death rates: A riddle in cause and effect

    Amer. Sci.

    (1958)
  • G.R. Chase

    On the efficiency of matched paris in Bernoulli trials

    Biometrika

    (1968)
  • Cited by (115)

    • Risk factors associated with secondary sarcomas in childhood cancer survivors: A report from the childhood cancer survivor study

      2012, International Journal of Radiation Oncology Biology Physics
      Citation Excerpt :

      In 15 case patients and 2 control subjects, the quality of the records available did not permit a radiation dose assignment (because in some cases it is not confirmed if the survivor received radiation at all). STATA 11.0 and the Epicure module PECAN were used to estimate odds ratios (ORs), model dose–response relations, perform likelihood ratio tests, and calculate 95% likelihood-based confidence intervals (CI) (7). Statistical tests were two-sided and based on an α-level of 0.05.

    • Role of radiation dose in the risk of secondary leukemia after a solid tumor in childhood treated between 1980 and 1999

      2010, International Journal of Radiation Oncology Biology Physics
      Citation Excerpt :

      The function that is widely accepted is a product of an increasing linear-quadratic term and a decreasing exponential term, which represents the competing effects of cell killing. From the computer program written in FORTRAN (6), which is a modified version of Lubin's PECAN program (17), we adapted this program to statistical software R to improve its performance and calculation capacities, using the nlm (nonlinear minimization) function. This function uses Newton-Rapshon methods.

    • Sociodemographic indicators and risk of brain tumours

      2003, International Journal of Epidemiology
    • Cellular-telephone use and brain tumors

      2001, New England Journal of Medicine
    View all citing articles on Scopus
    View full text