The psychological toll of perimenopause

Perimenopause is the time leading up to a women’s final menstrual period and the 12  months that follow. It typically begins in the mid-40s, although symptoms can start earlier. 

Research has shown that perimenopause is associated with an increased risk of depression and a range of mood symptoms, including irritability, anxiety, and depressive mood, and that these can affect quality of life and work.^1,2,3 But the degree to which psychological symptoms during perimenopause differ from premenopause, and how this varies across countries, had not been studied on a global scale until now.

Writing in Menopause, the flagship journal of The Menopause Society, our science team reports findings from a large international survey of Flo app members, carried out in collaboration with researchers from the Center for Women's Health, Mayo Clinic.

Who took part and how we measured symptoms

In total, 7,975 Flo app members aged 35 or older took part in our survey, which we conducted in English, French, Spanish, and Portuguese. Of these, 2,353 identified as being in perimenopause, and 5,622 said they were not yet in perimenopause or postmenopause.

To measure psychological symptoms, we used the psychological domain of the Menopause Rating Scale, which is a validated clinical questionnaire used widely in menopause research. This domain covers four symptoms: 

• Depressive mood (feeling down, sad, on the verge of tears, lack of drive, mood swings)
• Irritability (feeling nervous, inner tension, feeling aggressive)
• Anxiety (inner restlessness, feeling panicky)
• Physical and mental exhaustion (general decrease in performance, impaired memory, decrease in concentration, forgetfulness)

Each symptom is scored from 0 to 4, giving a total psychological domain score of 0 to 16. Higher scores mean more severe symptoms.

Participants also shared information about their age, education, income, ethnicity, body mass index (BMI), smoking and alcohol habits, and whether they had an existing diagnosis of anxiety or depression. This allowed us to examine whether differences between groups held up after accounting for these factors.

How perimenopause takes its toll on mental health

The difference between the two groups was clear and consistent. Women in perimenopause had an average psychological domain score of 7.3, compared with 4.8 in the premenopause group. Every individual symptom — depressive mood, irritability, anxiety, and exhaustion — was significantly higher in the perimenopause group.

The biggest gap was in exhaustion. Women in perimenopause scored an average of 2.0 for exhaustion, compared with 1.3 in the premenopause group — the largest difference of any individual symptom.

Importantly, these differences held up after we adjusted for a wide range of other factors, including age, education, income, ethnicity, existing diagnoses of anxiety or depression, BMI, smoking, and alcohol consumption. Perimenopause was still independently associated with higher psychological symptom scores across the board.

“We know that anxiety and depression get a lot of attention when it comes to perimenopause, and for good reason — they matter. But what stood out to us in this study was the scale of the difference in exhaustion and irritability between the perimenopause and premenopause groups,” said Adam Cunningham, PhD, lead research scientist at Flo and co-first author of the study. “These are symptoms that can have a real impact on daily life and work, and they deserve to be part of the conversation.”

Note that because this study surveyed women at a single point in time, we can describe these associations but cannot say with certainty what drives them. Women taking hormonal contraception or hormone therapy were not included, so we cannot say how treatment might affect these findings.

How symptoms vary around the world

More than 2,000 women in the perimenopause group took part from 20 countries across six continents, giving us a first look at how psychological symptom burden varies internationally.

We found significant differences between countries across all four symptoms and in the overall domain score. Nigeria had the lowest mean psychological domain score (3.5), followed by Venezuela (4.8), Chile (5.1), Germany (5.3), and Colombia (5.4). The highest scores were seen in Brazil (8.8), the United Kingdom (8.4), Australia (8.1), Canada (8.0), and the United States (7.9).

These patterns are exploratory and should be interpreted carefully. Because our study used an app-based sample, participation rates and demographics will have differed by country, and we did not adjust the international analysis for social determinants of health or other key factors. The findings are intended as hypothesis-generating rather than definitive. Still, the variation across countries suggests that cultural attitudes, healthcare access, and local awareness of perimenopause may all play a role in how these symptoms are experienced and reported.

“One of the strengths of this study is its international reach,” said Liudmila Zhaunova, PhD, director of science at Flo and senior study author. “Including women from 20 countries allowed us to begin exploring how psychological symptoms during perimenopause may vary across populations. While these findings are exploratory, they reinforce the need for more globally representative research to better understand women’s experiences and inform future education, research, and care.”

References: 

1. Cunningham, Adam C., et al. “Perimenopause Symptoms, Severity, and Healthcare Seeking in Women in the US. npj Womens Health, vol. 3, 2025, https://doi.org/10.1038/s44294-025-00061-3.
2. D’Angelo, Stefania, et al. “Impact of Menopausal Symptoms on Work: Findings from Women in the Health and Employment after Fifty (HEAF) Study.” International Journal of Environmental Research and Public Health, vol. 20, no. 1, 24 Dec. 2022, https://doi.org/10.3390/ijerph20010295. 
3. Liu, Xiangrong, et al. “Investigation of the Quality of Life and Influencing Factors among Perimenopausal Women.” Archives of Gynecology and Obstetrics, vol. 312, 2025, pp. 1253–65, https://doi.org/10.1007/s00404-025-08116-1.