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

1. In page 46, in the line following Equation (6), N should be defined as N= P g(Tg 1). 2. In page 48, in the line above Equation (13) and the line above Equation (14), (1-a)% should be 100(1-a)%. 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), W 1 1 ; W 1 2 ; W 1 3 should be W 1 AB ; W 1 AC ; W 1 BC , 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.