Linear inference for mixed treatment comparison meta-analysis: a two-stage approach

Authors

  • Guobing Lu,

    Corresponding author
    • School of Social and Community Medicine, University of Bristol, Bristol, U.K.
    Search for more papers by this author
  • Nicky J. Welton,

  • Julian P. T. Higgins,

  • Ian R. White,

  • A. E. Ades

Errata

This article corrects:

  1. Linear inference for mixed treatment comparison meta-analysis: A two-stage approach Volume 2, Issue 1, 43–60, Article first published online: March 2011

Correspondence to: Guobing Lu, School of Social and Community Medicine, University of Bristol, Canynge Hall, 39 Whatley Road, Bristol BS8 2PS, U.K.

E-mail: Guobing.Lu@bristol.ac.uk

Research Synthesis Methods 2(1): 43–60

Several errors have been found in the paper and can be corrected as follows:

  1. In page 46, in the line following Equation (6), N should be defined as N = ∑ g(Tg − 1).
  2. In page 48, in the line above Equation (13) and the line above Equation (14), (1-α)% should be 100(1-α)%.
  3. In Equation (14), W−1 should be W.
  4. In Equation (17), “log” is redundant and should be removed.
  5. In the second line below Equation (18), inline image should be inline image, respectively.
  6. In the line above Equation (22), “multiple” should be “multi-arm”.
  7. In Table 1, in the headings for the two First-stage direct estimate columns, the subscripts should be g instead of i. And in the Effect parameter column, dAD should be right below dAC, and dBD should be right below dBC, so that they are on the same line with the entries in the subsequent columns. In addition, the total numbers of participants in groups g = 1,2,3 should be 2864, 12 566 and 40, respectively.

We apologize to the readers for the many errors that appeared in this paper. Our special thanks are due to Prof. David C. Hoaglin, who pointed to us these errors and the many other tiny typos (not listed here), for his enthusiasm to our paper and his unusually keen sense to find errors in mathematical symbols.

Ancillary