Skip to main content

Table 1 Assessment of data reconstruction quality

From: Which test for crossing survival curves? A user’s guideline

Publication

MS G1

MS G2

HR [CI]

Bang et al. (2020) [37]

5.80 (5.88)

4.30 (4.44)

0.83 [0.53, 1.31] (0.82 [0.52, 1.29])

Becker et al. (2020) [39]

not defined

6.00 (6.21)

5.50 (5.51)

Bellmunt et al. (2017) [45]

3.30 (3.24)

2.10 (2.08)

0.98 [0.81, 1.19] (0.93 [0.77, 1.13])

Cortes et al. (2019) [46]

4.90 (4.94)

4.70 (4.72)

0.63 (0.62)

Ferris et al. (2016) [41]

2.00 (2.02)

2.30 (2.29)

0.89 [0.70,1.13] (0.89 [0.70,1.14])

Fradet et al. (2019) [47]

3.30 (3.35)

2.10 (2.16)

0.96 [0.79, 1.16] (0.92 [0.77, 1.11])

Godfrey et al. (2018) [36]

1.40 [0.54, 3.61] (1.40 [0.53, 3.69])

Golan et al. (2019) [38]

18.90 (18.90)

18.10 (18.10)

0.91 [0.56, 1.46] (0.88 [0.55, 1.42])

Hammel et al. (2019) [35]

21.20 (21.36)

6.00 (5.93)

0.72 [0.41, 1.27] (0.72 [0.42, 1.24])

Jones et al. (2020) [34]

26.00 (26.00)

20.0 (18.80)

0.59 [0.34, 1.05] (0.58 [0.33, 1.02])

Jones et al. (2018) [33]

15.10 (15.08)

8.10 (8.02)

0.72 [0.45, 1.17] (0.71 [0.44, 1.15])

Kotani et al. (2019) [48]

8.60 (8.62)

8.00 (8.02)

0.74 [0.48, 1.14] (0.72 [0.47, 1.11])

Kreuzer et al. (2020) [50]

19.40 (19.40)

20.90 (21.30)

1.22 [0.60, 2.47] (1.26 [0.62, 2.56])

Lu et al. (2018) [40]

4.63 (4.68)

4.23 (4.33)

0.78 [0.60, 1.00] (0.74 [0.55, 1.01])

Malone et al. (2020) [49]

0.66 [0.41, 1.07] (0.68 [0.42, 1.10])

Motzer et al. (2015) [42]

4.60 (4.46)

4.40 (4.07)

0.88 [0.75, 1.03] (0.87 [0.98, 1.34])

Mukai et al. (2019) [44]

27.90 (27.90)

16.60 (16.60)

0.55 [0.23, 1.29] (0.55 [0.23, 1.29])

Toxopeus et al. (2018) [43]

1.02 [0.75, 1.39] (1.01 [0.75, 1.39])

  1. Quality of data reconstruction regarding the published median survival (MS) in group 1 and 2 (G1 and G2), the hazard ratio (HR) with 95% confidence intervals (CI). For each study the published statistics are given with the corresponding statistics of the reconstructed data in parentheses. Three studies did not report MS (−) and two did not provide confidence intervals