Skip to main content
. 2024 Nov 7;28:359. doi: 10.1186/s13054-024-05128-2

Table 1.

Summary of results of TSA

Effect size Incidence, or variance I2 (%) D2 (%) RIS % of RIS attained Pooled effect (TSA adjusted 95% CI) Z-curve passed the conventional boundaries? Z-curve passed the TSA boundaries? Z-curve passed the futility boundaries?
Overall mortality (12 studies, n = 802)
RRR: 10.0% 17.4% 0 0 19,155 4.19 NA Yes No No
RRR: 20.0%* 17.4% 0 0 4584 17.5 0.66 (0.17, 2.55) Yes No No
RRR: 30.0%* 17.4% 0 0 1944 41.3 0.66 (0.38, 1.14) Yes No No
Intensive care unit length of stay (6 studies, n = 380)
MIREDIF 1 day 51.1 39 69.9 7119 18.7 − 3.62 (− 14.30, 7.06) Yes No No
MIREDIF 2 days* 51.1 39 69.9 1781 21.3 − 3.62 (− 9.80, 2.56) Yes No No
MIREDIF 3 days* 51.1 39 69.9 792 48.0 − 3.62 (− 7.73, 0.49) Yes No No
Hospital length of stay (3 studies, n = 200)
MIREDIF 1 day 312.2 0 0 13,122 1.5 NA Yes No No
MIREDIF 2 days* 312.2 0 0 3281 3.1 − 7.51 (− 27.50, 12.48) Yes No No
MIREDIF 3 days* 312.2 0 0 1459 13.7 − 7.51 (− 27.50, 12.48) Yes No No
Diarrhea incidence (11 studies, n = 765)

RRR:

15.0%*

37.4 51 60.5 7644 10.0 0.70 (0.20, 2.49) Yes No No
RRR: 25.0% 37.4 51 60.5 2678 28.6 0.70 (0.38, 1.29) Yes No No

RRR:

35.0%*

37.4 51 60.5 1325 57.7 0.70 (0.45, 1.09) Yes No No

TSA trial sequential analysis, I2 Between-trial heterogeneity, D2 diversity-estimate, RRR relative risk reduction, MIREDIF minimally relevant difference, RIS required information size, NA not applicable, CI confidence interval

*Post-hoc sensitivity analyses