The current study aimed to explore the structure and interconnectivity of a broad range of menopausal and depressive symptoms using a network analysis approach, and compare these between perimenopausal and post-menopausal women. In line with previous research, feeling tired or worn out, feeling anxious or nervous, and experiencing poor memory were the most prevalent menopausal symptoms across the entire sample (33). In terms of severity of symptoms, feeling tired or worn out, experiencing a lack of energy, and feeling anxious or nervous were the most severely rated symptoms across the entire sample. Post-menopausal women tended to report higher rates of hot flushes, dry skin, weight gain, and sexual domain symptoms, whilst a higher proportion of perimenopausal women reported feeling anxious or nervous, as well as feeling impatient with other people.
In regards to depressive symptoms as measured by the PHQ-8, fatigue, feeling slowed down or feeling restless, and sleep problems were the most severely rated symptoms, with women in the post-menopause rating the latter as significantly more severe relative to perimenopausal women. Of note, a similar proportion of peri- and post-menopausal women met criteria for major depressive disorder (MDD) according to the PHQ-8. Considering that a notable portion of women may experience either depressive symptoms or MDD throughout the menopause, healthcare professionals should endeavor to deliver depression screening. Unfortunately, mental health screening is not consistently integrated into menopausal healthcare (23), indicating a missed opportunity to identify and address potential mental health concerns that may arise.
When exploring the symptom structure of menopausal and depressive symptoms, network analysis revealed a strong association between decrease in stamina and decrease in strength, with these symptoms being identified as the most central symptoms of the network structure. Previous evidence indicates that reduced stamina can be a symptom causing confusion among menopausal individuals, leading them to question whether it should be attributed to aging or the menopause transition (34). Given the apparent importance of decreased stamina in the network structure revealed in the current study, it is crucial to ensure that public health messaging concerning menopausal symptoms is comprehensive. In the current study, decrease in stamina was also associated with psychosocial symptoms, as revealed by the clique percolation analysis. Namely, fatigue (as measured by the PHQ-8) and feelings of accomplishing less were related to a reduction in stamina. In this regard, decrease in stamina may be a key candidate for targeted therapeutic intervention. Indeed, exercise, which would increase physical stamina and strength, has been demonstrated to have a positive impact on both physical and psychosocial menopause symptoms (35–37).
In line with previous research (18), feeling depressed was also identified as a key central symptom, and was associated with both life dissatisfaction as measured by the MENQOL and the PHQ-8 item of low mood, with the latter also being implicated as highly central. In this regard, hormone replacement therapy (HRT) has been seen to be effective for managing first-onset depressive symptoms in the perimenopause and early post-menopause (38), with fluoxetine and HRT in combination being seemingly effective for depressive symptoms that meet the diagnostic threshold for MDD (39). Additionally, psychological therapies have shown moderate benefits in improving depressive symptoms in menopausal women, which are maintained long term (40). Therefore, incorporating pharmacological and/or psychological treatments alongside HRT for depressive symptoms may deliver wider benefits in women in the menopause who are experiencing such symptoms.
A change in sexual desire was also identified as a highly central symptom, demonstrating a strong association between change in sexual desire and avoiding intimacy. Sexual symptoms are frequently reported core symptoms of the menopause (18, 41), with depression being associated with decreased sexual desire in perimenopausal women (42). In line with previous research (43), a change in sexual desire was further correlated to vaginal dryness during intercourse, highlighting the importance of vaginal dryness in decreased sexual desire in the menopause, with these symptoms being inversely associated. Interestingly, vaginal dryness was the least frequently reported symptom in the current sample, contrasting previous studies demonstrating that vaginal dryness is frequent, particularly in post-menopausal women (44). Additionally, vaginal dryness was identified as the symptom with the lowest centrality (i.e., the symptom observed to have the lowest expected influence on other symptoms), despite its strong association with other highly central sexual symptoms. A reason for this finding may be the diverse range of factors and individual variability influencing menopausal symptoms, which were not assessed in the current study. One such example, is the use of vaginal lubricants, which are effective at reducing vaginal dryness in post-menopausal women (45).
We found a strong correlation between vasomotor symptoms, namely, hot flushes and sweating, with hot flushes being further associated with night sweats. Vasomotor symptoms were observed to cluster within one community revealed in the clique percolation analysis, with night sweats also being implicated in the ‘sleep problems’ community. This corroborates previous findings establishing the impact of vasomotor symptoms on sleep quantity or quality (46), with night sweats specifically being associated with sleep disturbances (47). In turn, sleep problems as measured using the PHQ-8 belonged to both the "sleep problems" community and also the "depression-related physical symptoms" community, which was further linked to the "cognitive-depression symptoms" and the "depressive mood and anxiety symptoms" communities. This finding is in line with the proposed “domino” theory of the menopause, which posits that sleep disturbances potentially mediate the relationship between vasomotor and mood symptoms (48), such that vasomotor symptoms disturb sleep, and sleep disruption causes low mood (49). In this regard, it has been posited that HRT may alleviate low mood and sleep problems by decreasing vasomotor symptoms (50).
Notably, despite the apparent interconnectivity of vasomotor and other menopausal and depressive symptoms in the current study, vasomotor symptoms appeared to have a comparatively low influence in the network. Critically, there is some discourse regarding the validity of classing non-vasomotor symptoms as symptoms of the menopause, as they may be more directly associated with sociodemographic characteristics, primarily age (20). Despite this, given the frequency and burdensomeness of non-vasomotor symptoms associated with the menopause as demonstrated both in the current study and in previous research (18), including such symptoms in the definition of the menopause whilst recognising that their aetiology may be dually attributed to aging and the menopause is practical for both research and clinical care. Unfortunately, the perception of vasomotor symptoms as being the core symptoms of the menopause can have a detrimental impact on the delivery of care for the menopause and the menopause transition, with accounts from menopausal women of healthcare professionals (HCPs) withholding support due to a lack of vasomotor symptoms (23).
Interestingly the network structure did not differ between the perimenopause and post-menopause groups, despite group differences in the frequency and severity of vasomotor, psychological, physical, and sexual symptoms, as well as symptom duration. This indicates that irrespective of frequency, severity, and duration of symptoms, the symptom structure and interactions may be stable across the span of the menopause transition and post-menopause. In this regard, early intervention is likely to be beneficial in relieving both menopausal and depressive symptoms. Critically, women in the perimenopause may be less likely to receive treatment or support compared to women in the post-menopause period (23). This may be due to a lack of awareness and understanding among HCPs regarding the potential impact of perimenopausal symptoms and the benefits of early intervention, with research indicating that the perimenopause stage may hold a particular significance when it comes to the success of hormone-based treatments for menopausal symptoms (51, 52). In turn, intervening early has been demonstrated to prevent or reduce the severity of other health issues throughout this challenging phase of life (51–56).
Limitations
It is worth noting that the sociodemographic characteristics may have contributed to the network structure identified in the current study. As the majority of the sample was white, the revealed network structure may not be widely generalisable to Black, Asian, and Minority Ethnic (BAME) groups, given the differences in menopause symptom presentation between ethnic groups (57). Therefore, we strongly advocate for further research applying these analyses techniques to menopause and depressive symptom data from diverse sociodemographic groups. Additionally, the social media recruitment strategy emphasised the study's focus on the menopause and mental health, therefore it is likely that individuals with more severe menopausal and associated mental health symptoms would have been more inclined to participate. This will have shaped the network, and the connections within it, and so the findings of the current study may not be representative of menopausal women with milder symptom severity. In addition, as the MENQOL measures psychological symptoms, some of which overlap with the items within the PHQ-8, the edge weights may have been inadvertently exaggerated.