51 Trials on Hormone Therapy, Mood, Anxiety, and Sleep

Hormone therapy improved depression, anxiety, mood, and sleep quality scores in the largest randomized evidence base yet assembled on menopausal psychological symptoms: a July 10, 2026 synthesis in Frontiers in Medicine pooling 51 randomized trials in 41,821 women. All four outcomes moved in the same direction.

The more useful part of the answer is in which women, and after how long.

What the pooled numbers actually measure

The included trials did not all use the same questionnaires. One might score sleep with the Pittsburgh Sleep Quality Index, another with a different instrument entirely. To combine them, the authors converted each trial's result into a standardized mean difference (SMD), which expresses the change in units of standard deviation rather than in raw points. That makes trials comparable to each other. It also means the number is not a symptom score. An SMD does not translate into hours of sleep gained or points dropped on a depression scale, and by convention a value near 0.7 is read as a moderate to large effect.

With that in mind, here is what the 51 hormone therapy trials showed when pooled:

• Sleep quality: SMD 0.74 (95% CI 0.65 to 0.83)
• Depression: SMD 0.73 (95% CI 0.66 to 0.79)
• Anxiety: SMD 0.68 (95% CI 0.56 to 0.80)
• Mood: SMD 0.55 (95% CI 0.41 to 0.69)

The authors also ran a trial sequential analysis. This method asks whether enough participants have accumulated across all the trials for a pooled result to be treated as settled, rather than as a finding that could still reverse as more data arrives. For mood, anxiety, and sleep quality, the cumulative evidence crossed the monitoring boundaries. For depression, stress, and quality of life, it did not, and the authors state those results require further validation.

That ordering is worth sitting with. Depression had one of the tightest confidence intervals of the four, and it is still the outcome the authors flag as unsettled.

Who benefited, and after how long

Sleep split sharply by menopausal stage

The widest gap in the whole analysis was in sleep, between women at different stages.

• Postmenopausal women: SMD 0.91 (95% CI 0.88 to 0.94)
• Perimenopausal women: SMD 0.25 (95% CI 0.13 to 0.38)

Same intervention, same outcome measure, and a large pooled effect in one group against a small one in the other. The synthesis reports that difference but does not establish why it exists. It is also a subgroup comparison rather than a randomized head to head, so it is best read as a signal about where the sleep benefit concentrated in these trials, not as evidence that it does nothing during perimenopause.

The mood benefit appeared with longer use

Mood showed a different pattern, this one about time rather than stage. In trials of longer-term use, the pooled effect on mood was SMD 0.77 (95% CI 0.73 to 0.82). In short-term trials it was 0.12, essentially nothing.

Sleep and vasomotor symptoms are often described as the fast-moving outcomes in this area. In this synthesis, mood was not one of them. Anyone reading a single early-weeks experience as the final verdict on mood is reading against what these trials found.

A prevention trial points the other direction on stage

The pooled analysis asked what happens to symptoms women already have. A 2018 randomized trial in JAMA Psychiatry asked something different: can hormone therapy keep depressive symptoms from developing in the first place?

It enrolled 172 women aged 45 to 60, perimenopausal or early postmenopausal, all of whom were free of clinically significant depressive symptoms at baseline. They were randomized, double blind, to 12 months of transdermal estradiol with intermittent oral micronized progesterone, or to placebo. Over those 12 months, 17.3 percent of the hormone therapy group developed clinically significant depressive symptoms on the CES-D scale, compared with 32.3 percent on placebo (OR 2.5, 95% CI 1.1 to 5.7, P = .03).

This is a single trial of 172 women, small enough that the confidence interval is wide, and it reports what happened in those participants rather than what will happen in anyone else. Two of its subgroup findings are still notable. The benefit was concentrated in women in the early menopause transition and was not evident in the late transition or after menopause, which runs opposite to the stage pattern the 51-trial synthesis found for sleep. And mood benefit was larger in women who reported more stressful life events in the preceding six months.

The regulatory backdrop, and what it does not settle

All of this is being read in a changed regulatory context. On November 10, 2025, HHS and the FDA announced they were initiating removal of the broad boxed warnings on menopausal hormone therapy products, specifically the references to cardiovascular disease, breast cancer, and probable dementia. As of that announcement the removal was initiated rather than complete, with the agency working with manufacturers on labeling. The boxed warning for endometrial cancer on systemic estrogen-alone products was not removed.

A label change is not a safety finding, and this synthesis is not one either. It measured psychological symptoms. Its authors note that adverse events were defined inconsistently across the included studies, which makes it the wrong place to look for an answer about risk.

What this means if you are weighing it

The evidence base on menopausal mood, anxiety, and sleep is now considerably larger than it was, and it points in a consistent direction. It is also averages across trial populations, not a forecast for any individual. The authors are direct about the limits: participants varied in comorbidities and in how severe their symptoms were at baseline, and they write that the pooled results cannot be directly applied to real-world clinical practice.

The practical takeaway is that stage and duration are not footnotes here. Where you are in the transition tracked with how much the sleep outcome moved, and how long treatment ran tracked with whether the mood outcome moved at all. Whether hormone therapy makes sense for you depends on those factors plus your own history and risk profile, which is a conversation for a licensed provider who can look at all of it together.