Table II.
The efficacy of the propensity score matching.
| Group | Sample (n) | Ps R2 | LR χ2 | P>χ2 | MeanBias | MedBias | B | R | %Var |
|---|---|---|---|---|---|---|---|---|---|
| Non-surgery vs. surgery | Unmatched (1,413) | 0.309 | 484.81 | <0.01 | 58.20 | 50.50 | 157.8a | 1.15 | 71.00 |
| Matched (294) | 0.015 | 12.59 | 0.13 | 9.30 | 8.50 | 29.4a | 1.22 | 0.00 | |
| Sublobectomy vs. lobe/s | Unmatched (200) | 0.030 | 10.19 | 0.25 | 11.50 | 11.20 | 41.5a | 1.09 | 0.00 |
| Matched (84) | 0.011 | 2.53 | 0.96 | 6.00 | 5.10 | 24.50 | 1.40 | 0.00 | |
| Lobe/s vs. pneumonectomy | Unmatched (200) | 0.161 | 12.98 | 0.11 | 42.10 | 31.70 | 111.8a | 1.78 | 0.00 |
| Matched (10) | 0.353 | 9.78 | 0.28 | 20.10 | 19.40 | 109.9a | 7.50* | 0.00 |
a
B>25%, R outside (0.5; 2); B, absolute standardized difference of the means of the linear index of the propensity score in the unmatched and matched groups; R, ratio of unmatched to matched variances of the propensity score index; %Var, the percentage of continuous variables that have variance ratios that exceed the 2.5th and 97.5th percentiles of the F-distribution; Lobe/s, lobectomy or bilobectomy.