Table 4.
Certainty of evidences quality.
| References | Outcomes | Simple | Limitations | Inconsistency | Indirectness | Imprecision |
Publication bias |
Quality |
|---|---|---|---|---|---|---|---|---|
| Tong et al. (19) | HAMD | MD −1.27 (−2.55, 0.01) | -1 | -1 | 0 | 0 | 0 | Low |
| EPD | SMD −0.49 (−1.01, 0.02) | -1 | -1 | 0 | -1 | -1 | Very low | |
| Estradiol level | MD 63.99 (13.47, 114.51) | -1 | -1 | 0 | -1 | -1 | Very low | |
| Effect rate | RR 1.20 (1.09, 1.33) | -1 | 0 | 0 | 0 | 0 | Moderate | |
| Li et al. (20) | HAMD | SMD −1.08 (−2.11, −0.05) | -1 | -1 | 0 | 0 | 0 | Low |
| Estradiol levels | SMD 1.96 (−0.01, 3.93) | -1 | 0 | 0 | -1 | -1 | Very low | |
| Effect rate | RR 1.00 (0.89, 1.12) | -1 | -1 | 0 | 0 | 0 | Low | |
| Li et al. (21) | HAMD | MD −1.38 (−3.40, 0.64) | -1 | -1 | 0 | 0 | 0 | Low |
| EPDS | MD 1.08 (1.09, 3.26) | -1 | -1 | 0 | -1 | -1 | Very low | |
| Effective rate | RR 1.15 (1.06, 1.24) | -1 | 0 | 0 | -1 | 0 | Low | |
| Estradiol levels | MD 36.92 (23.14, 50.71) | -1 | -1 | 0 | 0 | 0 | Low | |
| Cao et al. (22) | HAMD | MD 0.45 (−0.52,1.41) | -1 | -1 | 0 | 0 | 0 | Low |
| EPDS | MD 0.55 (0.18, 0.92) | -1 | 0 | 0 | -1 | -1 | Very low | |
| Effective rate | RR 0.93 (0.70, 1.23) | -1 | -1 | 0 | 0 | 0 | Low | |
| Estradiol levels | MD 0.20 (−0.19, 0.58) | -1 | 0 | 0 | 0 | 0 | Moderate | |
| Wang et al. (23) | HAMD | MD −1.27 (−2.55,0.01) | -1 | -1 | 0 | 0 | 0 | Low |
| EPDS | MD −0.47 (−0.92, −0.03) | -1 | 0 | 0 | -1 | 0 | Low | |
| Estradiol levels | WMD 63.99 (13.39, 114.60) | -1 | -1 | 0 | 0 | 0 | Low | |
| Effective rate | OR 3.15 (2.19, 4.55) | -1 | 0 | 0 | -1 | 0 | Low | |
| Pang and Shi (24) | HAMD | MD −1.03 (−2.58,0.52) | -1 | -1 | 0 | -1 | -1 | Very low |
| Effective rate | RR 0.98 (0.84, 1.14) | -1 | 0 | 0 | -1 | -1 | Very low |
RR, Risk Ratio; OR, odds ratio; SMD, SMD, standardized mean difference; WMD, Weighted Mean Difference; AT, acupuncture therapy; ST, sham acupuncture; CM, conventional medication; HAMD, Hamilton Rating Scale for Depression; EPDS, Edinburgh Postnatal Depression Scale. , The design of the experiment with a large bias in random, distributive hiding or blind; , The confidence interval overlaps less, the heterogeneity test P is very small, and the I2 is larger; , Confidence interval is not narrow enough; , Funnel graph asymmetry; , Fewer studies are included and there may be greater publication bias.