Randomized controlled trials (RCT) were included in the systematic review if they met the following criteria: 1) including perimenopausal, and postmenopausal women; 2) studying parallel or crossover groups; and 3) including participants with depression or anxiety.

Consort Checklist was used to assess the quality of included RCTs. This checklist include items such as blinding, drop-out rate, randomization technique, and intention to treat report.

The data extracted by two separate authors included sample size, type of intervention and control, duration of treatment, dose, and results. Disagreements were resolved by consensus-based discussion.

Balk et al.9 compared the effect of soy flour containing 100 mg isoflavones and wheat cereal as placebo on 19 menopausal women. There was no statistical difference between groups regarding depression symptoms (P = 0.58) at the end of a six-month period. Also, the two groups did not show any improvement compared to the baseline. Kotsopoulos et al.10 examined the effectiveness of soy on menopausal women. They randomized menopausal women in two groups of Soy protein dietary supplements (n = 44) and Casein placebo (n = 50). The score of depression dropped by 25% in soy group but it remained unchanged in placebo group. The score of anxiety dropped by 17% in the soy group and 16% in the placebo group. Statistical analysis by multivariate analysis of covariance (MANOCA) showed no significant differences between groups for psychological symptoms (irritability, depression, anxious, tiredness and sleeplessness). Hanachi and Golkho11 designed a randomized three-group study comparing the efficacy: 1) soymilk supplementation; 2) soymilk supplementation and exercise; and 3) placebo on depression and anxiety. Both groups were not different from placebo group regarding anxiety and depression. The effect on soymilk supplementation and exercise on depression and nervousness was less significant than Soymilk supplementation. To sum up, soy improved slightly anxiety and depression.

A study by Aghamiri et al.12 compared the efficiency of Hops with placebo. Intervention group received 500 mg hop once daily for 12 weeks. The Hops group showed a statistically significant decrease in anxiety compared to the placebo group in three intervals at 2 (P < 0.001), 4 (P < 0.001), 8 (P < 0.001) and 12 (P < 0.001) weeks. The corresponding values for depression were 2 (P < 0.001), 4 (P < 0.001), 8 (P < 0.001) and 12 (P < 0.001) respectively.

In Cagnacci et al. study,4 all patient received 1 g/day of calcium and then were randomized into three groups of control group (no treatment), 100 mg Kava and 200 mg Kava. Mean score of depression dropped from 37.0 ± 1.8 to 34.9 ± 1.7 in the calcium d, from 38.2 ± 2.4 to 33.7 ± 1.4 in Kava-Kava (100 mg) and from 38.7 ± 2 to 33.5 ± 1.9 in Kava (200 mg) groups. The reduction between the baseline and 3-month follow-up was only significant in the group of Kava 200 mg. Comparison of the three groups using ANOVA test did not show any statistically significant difference between groups in terms of depression score. The score anxiety remained unchanged in the calcium group. In 100 mg and 200 mg groups, anxiety score dropped significantly from 47.3 ± 2.2 to 22.5 and from 46.6 ± 1.6 to 41.3 ± 1.6 respectively.

Lipovac et al.13 conducted a double-blind, randomized, crossover study (with a wash-out period of one week) on two groups (red clover and placebo) for 12 weeks. After administering red clover, Hospital Anxiety subscale decreased from (9.98 ± 4.65 to 2.40 ± 2.53) in red clover group and from 9.98 ± 4.65 to 8.05 ± 4.76 (P < 0.001) in placebo group (P < 0.001). Hospital depression subscale decreased from 6.91 ± 4.02 to 1.50 ± 2.06 (P < 0.001) in red clover group and from 6.91 ± 4.02 to 5.23 ± 3.65 (P < 0.001) in placebo group. Zung Self-Rating Depression Scale revealed a significant decrease (P < 0.001 compared to baseline and placebo). Tice et al.14 compared three groups of Promensil (82 mg red clover-derived isoflavones), Rimostil (82 mg red clover-derived isoflavones) and placebo. They did not find any significant improvement in subscale of ‘depression’ either in Promensil (P = 0.23) or Rimostil (0.79) groups compared to the placebo group. The corresponding values for subscale of anxiety were (P = 0.33) and (P = 0.80) respectively. Hidalgo et al.15 conducted a double-blind, randomized, crossover study (with a wash-out period of one week) on two groups of red clover (80 mg) and placebo for 90 days. According to Kupperman index, the subscales of ‘depression’ and ‘nervousness’ revealed a significant improvement in the red clover group compared to the placebo group (both, P = 0.05). Geller et al.16 randomized menopausal women into four groups of black cohosh, red clover, placebo and 0.625 mg conjugated equine estrogens (CEE) plus 2.5 mg medroxyprogesterone acetate (MPA) (CEE/MPA; n = 89). The comparison of two groups in terms of Greene Anxiety Score did not show any significant difference after 12 weeks (P = 0.12); however, at 12 months, significant levels (P = 0.04) were observed. The comparison between Placebo and Black cohosh was non-significant. However, there was a significant improvement in CEE/MPA group compared to placebo at 3 months and 12 months.

In their study, Bu and Lephart17 investigated the levels of brain Bcl-2-associated death protein (a proapoptotic member of Bcl-2 protein family) and neuron-specific β-III tubulin (an early marker of neuronal differentiation/survival) in male rats, which were fed with either a standard chow rich of soy isoflavones (Phyto-600) or isoflavones free of soy (Phyto-free) in their entire life (since conception until tissue collection). In the group of rats fed with the Phyto-600 diet, a significant decline in the levels of β-III tubulin was observed in amygdala, frontal cortex, hippocampus and medial basal hypothalamus (MBH) compared to Phyto-free group. In short, the results of this study provided evidences for the neuroprotective power of soy isoflavones in the amygdala, frontal cortex, hippocampus and MBH. This suggests that soy isoflavones can potentially improve learning and memory, behaviors related to anxiety, and recovery from trauma.

Koppula and Kumar18 conducted a study on anti-stress and memory-enhancing features of fennel extract in lab rats. Fennel is a phytoestrogen. The extract was administered orally with an orogastric tube at doses of 50, 100, and 200 mg/kg body weight. The anti-stress activity of rats was assessed based on urinary levels of vanillylmandelic acid (VMA) and ascorbic acid. Daily administration of Foeniculum vulgare (F. vulgare) extract 1 h before the induction of stress had a significant effect (P < 0.05) of alleviating stress-induced urinary biochemical levels of VMA. It should be noted that these changes were in a dose-dependent fashion. As such, F. vulgare could be useful in the management of stress and stress-related disorders.

Future studies can compare the effect of herbal medicines and current treatments such as Fluoxetine. Given the weak methodology of this study, it is recommended that future studies adopt a design based on Consolidated Standards of Reporting Trials (CONSORT) guidelines.

There were several limitations in our study including poor methodology (such as inadequate treatment allocation, small sample size and unclear blinding method), few clinical trials and limited review of relevant studies. Only one study 13 measured depression and anxiety level by specific questionnaire. The rest of studies measured the anxiety and depression symptoms base on menopausal symptom questionnaire. The random sequence generation, Allocation concealment and blinding of outcome assessment was either not performed or not reported in sufficient detail in most of studies. The overall methodological quality was moderate to high.