The R Journal: accepted article

This article will be copy edited and may be changed before publication.

cchs: An R Package for Stratified CaseCohort Studies PDF download
Edmund Jones

Abstract The cchs package contains a function, also called cchs, for analyzing data from a stratified case–cohort study, as used in epidemiology. For data from a stratified case–cohort study, cchs calculates Estimator III of Borgan et al. (2000), which is a score-unbiased estimator for the regression coefficients in the Cox proportional hazards model. From the user’s point of view the function is similar to coxph and other widely used model-fitting functions. Convenient software was not previously available for Estimator III because it is complicated to calculate. SAS and S-Plus code-fragments for the calculation have been published (Langholz and Jiao, 2007a; Cologne et al., 2012), but cchs is much easier to use on new datasets, and works differently so that it is far more efficient in terms of time and memory and can cope with much bigger datasets. It also avoids several minor approximations and simplifications.

Received: 2017-12-28; online 2018-05-16, supplementary material, (1.3 Kb)
CRAN packages: cchs, cchs, survival, survey, NestedCohort
CRAN Task Views implied by cited CRAN packages: Survival, SocialSciences, ClinicalTrials, Econometrics, OfficialStatistics

CC BY 4.0
This article and supplementary materials are licensed under a Creative Commons Attribution 4.0 International license.

  author = {Edmund Jones},
  title = {{cchs: An R Package for Stratified CaseCohort Studies}},
  year = {2018},
  journal = {{The R Journal}},
  url = {}