GeMTC R package updated

09 Aug 2013 — Gert van Valkenhoef

Version 0.4 of the GeMTC R package has several new features:

  • Fixed effect models
  • Poisson/log models for survival data
  • Allow continuous data to be entered with standard error instead of only with standard deviation and sample size

In addition, an important bug has been fixed: in version 0.3, some models would not run using WinBUGS.

For a full changelog, see the NEWS file on the GeMTC CRAN page.

MCDA web app released

26 Jul 2013 — Gert van Valkenhoef

We have just released the first version (0.1) of our web based multiple criteria decision analysis (MCDA) preference elicitation tool. This component will be used in ADDIS 2.0 to support benefit-risk decisions. It is currently available online at

You will need a recent browser; we recommend Mozilla Firefox or Google Chrome. Several examples are provided. You can also load your own benefit-risk assessments by exporting them from ADDIS 1.16.4 (released last week). We would be glad to hear your feedback.

R packages updated

18 Jul 2013 — Gert van Valkenhoef

The following R packages have been updated this week:

  • smaa version 0.1: this is the first release of our SMAA R package.
  • gemtc version 0.3: this major update adds the following new features:
    • support for relative effect data
    • support for mixed arm-based and relative effect data
    • binom/cloglog likelihood/link for rate data
    • unrelated mean effects (UME) model
    • unrelated study effects (USE) model
    • analysis of heterogeneity with heterogeneity plot (EXPERIMENTAL)
    • full access to generated code, data structures, and parameters
    • guard against "impossible" initial values
    And also includes a number of bugfixes and documentation updates.
  • hitandrun version 0.2-1: bugfix release.

ADDIS status and future

19 Mar 2013 — Gert van Valkenhoef

The ADDIS 1.x software has been very successful as a prototype, and our decision to develop it as a desktop application has allowed us to focus on essential problems in evidence synthesis and decision analysis. However, the ADDIS vision can not be fully realized in a single user environment. The ADDIS 1.x retrospective provides an overview of design decisions made for ADDIS 1.x and their positive and negative consequences. It also identifies key changes that need to be made going forward.

The ADDIS 2.x requirements and architecture document summarizes the long-term vision and current status of ADDIS, identifies key requirements and proposes an architecture for ADDIS 2.x. Note that this document is focused on architecturally relevant non-functional requirements rather than on specific functional requirements (such as the types of analyses that should be supported). We're currently hiring new developers (updated 2013-04-26) to start implementing the new architecture.