Table 3. Subgroup and meta-regression analyses between NLR and OS.
Variables | Number of outcomes | HR [95% CI] | P value | Heterogeneity | Model | P3-value of Meta-regression | ||
---|---|---|---|---|---|---|---|---|
I2 (%) | P2 value | Univariate | Multivariate | |||||
(1) Ethnicity | 0.329 | 0.762 | ||||||
Asian | 14 | 1.35 [1.22–1.51] | < 0.00001 | 61 | 0.001 | random | ||
Caucasian | 7 | 1.40 [1.16–2.21] | 0.005 | 71 | 0.002 | random | ||
(2) Cut-off value | 0.659 | 0.476 | ||||||
< 4.0 | 12 | 1.30 [1.21–1.39] | < 0.00001 | 40 | 0.08 | fixed | ||
≥ 4.0 | 9 | 1.36 [1.11–1.67] | 0.003 | 78 | < 0.0001 | random | ||
(3) Sample size | 0.017 | 0.031 | ||||||
N < 150 | 6 | 1.89 [1.50–2.38] | < 0.00001 | 0 | 0.73 | fixed | ||
150 < N ≤ 200 | 6 | 1.26 [1.16–1.36] | < 0.00001 | 30 | 0.21 | fixed | ||
N > 200 | 9 | 1.34 [1.11–1.59] | 0.002 | 77 | < 0.0001 | random | ||
(4) Clinical stage | 0.563 | 0.925 | ||||||
L | 5 | 1.47 [1.03–2.10] | 0.04 | 75 | 0.003 | random | ||
E | 3 | 1.51 [1.08–2.10] | 0.02 | 72 | 0.03 | random | ||
L + E | 13 | 1.38 [1.22–1.57] | < 0.00001 | 61 | 0.002 | random |
L:limited stage; E:extensive stage