Combined therapies rank best for perimenopausal mood

Anxiety, low mood and panic symptoms are common around the menopausal transition, and the options offered vary widely: antidepressants and other medication, hormone-based approaches, acupoint stimulation such as acupuncture, and talking therapies. Trials have compared these in isolation, but women and clinicians are usually choosing between them, or between a single treatment and a combination. That comparative question has stayed unresolved, with claims of superiority on all sides. This review set out to rank the strategies against one another across mood, menopausal symptom, sleep and hormonal outcomes.

The authors searched PubMed, Embase, the Cochrane Library, Web of Science, CNKI, Wanfang, VIP and SinoMed from inception to 14 June 2026 for randomised controlled trials in perimenopausal women with emotional disorders. They included 131 trials covering 11,457 women and grouped interventions into pharmacotherapy, acupoint stimulation and psychotherapy, alone and in combination. A Bayesian network meta-analysis pooled direct and indirect comparisons, and Surface Under the Cumulative Ranking Curve values were calculated to rank each option. Outcomes included the Hamilton Depression and Anxiety scales, the Kupperman Index, self-rating depression and anxiety scales, the Pittsburgh Sleep Quality Index, serum hormones and adverse events.

Medication combined with psychotherapy achieved the highest ranking probabilities for Hamilton depression scores (SUCRA 92.4%), the Kupperman Index (97.9%), self-rated depression (94.5%), sleep quality (98.1%) and follicle-stimulating hormone reduction (96.1%). Acupoint stimulation with psychotherapy ranked highest for Hamilton anxiety reduction (93.7%), and psychotherapy alone ranked highest for self-rated anxiety (98.9%). Medication plus acupoint stimulation ranked top for luteinising hormone reduction. Within the drug subgroups, antidepressants combined with traditional Chinese medicine ranked highest for depression and for safety. The control condition ranked best on safety outcomes overall, and no single strategy led across every measure.

The results are consistent with the general clinical impression that combining a biological and a psychological treatment tends to outperform either alone, and they extend it to menopausal symptom burden and sleep as well as mood. But network meta-analyses rank options largely through indirect comparison, so a high SUCRA value is a probability of being best, not a demonstrated win. The included trials span very different treatments, populations and scales, and the safety ranking simply reflects that doing nothing carries no treatment side effects. Which specific combination suits which symptom profile remains an open question the authors frame as individualised choice.

For anyone weighing up options for mood symptoms during the menopausal transition, this is a reason to think of combinations rather than a single winner, and to notice that talking therapy ranked strongly for anxiety in its own right. It also shows that different outcomes favour different approaches, so what matters is which symptoms are most troubling. Because these are indirect rankings from trials of varying quality, they cannot tell an individual what will work for her. Treatment decisions of this kind belong with a clinician.