This page contains information about software modules developed by members of the Centre for Biostatistics.


clsampsi calculates the power for randomised trials and the number of clusters and cluster sizes required for the difference of means or proportions in the presence of differential clustering effects between study arms. A ‘rough’ approximation to the optimum allocation ratio for such a trial, given a desired power, is also available (under the optimal option).

Authors: Eva Batistatou and Chris Roberts
Further information: clsampsi

dr (double robust)

dr computes a double-robust effect estimate for the effect of a treatment on an outcome given a set of confounders. The confounding effect is adjusted for in two ways: by modeling the effect of both treatment and confounders on the outcome (outcome model), and by weighting observations according to the inverse of their probability of receiving the treatment they actually received (propensity model). If either the outcome model or the propensity model is correct, the doubly robust estimate is unbiased. 

Author: Richard Emsley and Mark Lunt
Reference:Emsley R, Lunt M, Pickles A and Dunn G. The Stata Journal, 2008, Vol 8(3), pp.334-353.
Further information: the command can be installed within Stata by typing 'search dr' at the command prompt, or from the following link:Download the dr command

gllamm (generalised linear latent and mixed models)

gllamm fits generalized linear latent and mixed models. These models include Multilevel generalized linear regression models (extensions of the simple random intercept models that may be fitted in Stata using xtreg, xtlogit, xtpois to include multilevel and random coefficient models), Multilevel factor models and Multilevel structural equation models. The latent variables (or random effects) can be assumed to have a multivariate normal distribution or to be discrete allowing nonparametric maximum likelihood estimation. The common links and families of generalized linear models are available and responses can be of mixed type including continuous, censored, discrete, dichotomous, ordered categorical and unordered categorical.

Author: Andrew Pickles
Further information:the GLLAMM website and within Stata type 'ssc describe gllamm'

metaan (module for performing fixed- or random-effects meta-analyses)

The metaan command performs a meta-analysis on a set of studies and calculates the overall effect and a confidence interval for the effect. The command also displays various heterogeneity measures: Cochrane's Q, I-squared, H-squared and the between-study variance estimate. Cochrane's Q is the same across all methods, but the between-study variance estimate (and hence I-squared and H-squared) can vary between the dl and ml methods.  Only one method option must be selected.

Authors: Evan Kontopantelis and David Reeves
Reference: Kontopantelis E, Reeves D. (2010). The Stata Journal, Vol. 10, Issue 3, pp395-407.
Further information: within Stata type ‘ssc describe metaan’


metaeasy implements meta-analysis methodology in an Microsoft (Excel) add-in which is freely available and incorporates more meta-analysis models (including the iterative maximum likelihood and profile likelihood) than are usually available, while paying particular attention to the user-friendliness of the package.

Authors: Evan Kontopantelis and David Reeves
Reference: Kontopantelis E, Reeves D. (2009). Journal of Statistics Software, Vol. 30, Issue 7.
Further information:

metaeff (meta-analysis module for effect sizes calculations)

The metaeff command provides a way to calculate the effect sizes (and the respective standard errors) of research studies, for use with meta-analysis methods. The methods used for the calculations have been derived from the Cochrane Collaboration handbook.

Authors: Evan Kontopantelis and David Reeves
Further information: within Stata type ‘ssc describe metaeff’

paramed (causal mediation analysis using parametric regression models)

paramed performs causal mediation analysis using parametric regression models. Two models are estimated: a model for the mediator conditional on treatment (exposure) and covariates (if specified), and a model for the outcome conditional on treatment(exposure), the mediator and covariates (if specified). It extends statistical mediation analysis (widely known as Baron and Kenny procedure) to allow for the presence of treatment(exposure)-mediator interactions in the outcome regression model using counterfactual definitions of direct and indirect effects. paramed allows continuous, binary or count outcomes, and continuous or binary mediators, and requires the user to specify an appropriate form for the regression models. paramed provides estimates of the controlled direct effect, the natural direct effect, the natural indirect effect and the total effect with standard errors and confidence intervals derived using the delta method by default, with a bootstrap option also available.

Authors: Richard Emsley and Hanhua Liu

Further information: within Stata type 'ssc describe paramed'

skbim (skewed bimodal data generator)

The program generates random numbers from a bimodal distribution. The two unimodal distributions that make up the bimodal can be normal or skewed-normal (see sknor for more details). Different arguments can be inputted to the function, as specified by 'option'.

Author: Evan Kontopantelis
Further information: within Stata type ‘ssc describe skbim’

sknor (skewed normal data generator)

The program generates random numbers from a skewed normal distribution (right-skew being the default).

Author: Evan Kontopantelis
Further information: within Stata type ‘ssc describe sknor’