Perceived stress as a mediator between negative life events and psychosocial quality of life during the menopausal transition: A path analysis Authors: Ming Jun Kuck, Nicola J. Paine, Eef Hogervorst Published: 2026 ABSTRACT BackgroundThe menopausal transition, accompanied by numerous life challenges, is characterised by significant hormonal fluctuations and various symptoms. This study investigates the intricate relationships among perceived stress, resilience, self-efficacy, and social support vs. negative life events and their combined impact on the psychosocial quality of life (QoL) in women across the menopausal transition.MethodsCross-sectional data were gathered via an anonymous online survey of women aged 40–60 at various stages of the menopausal transition (premenopausal, n = 36; perimenopausal, n = 120; postmenopausal, n = 113), classified using STRAW+10 criteria, resulting in 269 valid responses. The survey included measures for negative life events, perceived stress, social support, resilience, self-efficacy, and menopausal-specific QoL. The study's path model regarded negative life events as an exogenous variable and perceived stress, social support, resilience, self-efficacy, and psychosocial quality of life as endogenous variables, whilst controlling for age and menopausal stage.ResultsNegative life events were significantly associated with lower social support (β = −.186) and resilience (β = −.182), higher perceived stress (β = .338), and poorer psychosocial QoL (β = .164). Perceived stress emerged as the strongest predictor of psychosocial QoL (β = .578). Resilience, social support, and self-efficacy were indirectly linked to better psychosocial QoL through their effects on stress reduction. The model explained 47% of the variance in psychosocial QoL and 41% in perceived stress.ConclusionPerceived stress emerged as the central mediator linking negative life events to psychosocial QoL, with resilience, self-efficacy, and social support as indirect protective factors. Enhancing these modifiable factors through targeted interventions may improve well-being in women across the menopausal transition. Future longitudinal studies are needed to validate these findings. Introduction The menopause, defined by the end of menstruation confirmed after 12 consecutive months of amenorrhoea, is preceded by the perimenopausal transition — a period of irregular menstrual cycles and significant hormonal fluctuations — and followed by the postmenopausal period [1]. These stages are collectively referred to as the menopausal transition and are often accompanied by symptoms such as hot flushes, night sweats, insomnia, brain fog, depression, and reduced sex drive, which can persist long after menstruation ceases [1]. According to a survey carried out by the Fawcett Society in 2022, an overwhelming majority of menopausal women — 77% — described at least one symptom of menopause as ‘very difficult’ [2]. Approximately 75% of women included in other surveys reported problems with memory and concentration, while 69% experienced anxiety or depression [3], highlighting the considerable mental health burden of the menopausal transition. According to the Governmental report Shattering the Silence about Menopause: 12-Month Progress Report, approximately one-third of the entire UK female population, around 13 million, were estimated to be undergoing menopause, with the estimated annual cost of menopause-related workplace absences reaching approximately £1.8 billion [4]. Menopausal symptoms can significantly impair women's quality of life (QoL) [5,6]. Menopausal symptoms are generally classified into four main domains: vasomotor symptoms (VMS) such as hot flushes and night sweats; physical symptoms such as joint pain and fatigue; urogenital symptoms such as vaginal dryness and urinary issues; and psychological or psychosocial symptoms such as depression, anxiety, and poor memory [7]. The present study focuses specifically on the psychosocial domain, assessed using the Menopause-Specific Quality of Life Questionnaire (MenQoL), which evaluates VMS and the physical, sexual, and psychosocial aspects affected by menopause and is preferred over generic QoL tools for its clinical relevance [7,8]. QoL is defined by the World Health Organisation (WHO) as an individual's perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns. Research has shown that more severe VMS can negatively affect psychological and emotional wellbeing [[5], [6], [7], [8], [9]], with women experiencing more severe VMS reporting higher levels of anxiety, depression, stress, and difficulties in emotion regulation [6]. However, the relationship between menopausal symptoms and QoL is not determined solely by symptom severity or menopausal stage, and psychological factors may play an important role in shaping this relationship. Not all symptoms correlate with menopausal stage [10]. Perimenopausal women typically experience more severe symptoms [[10], [11], [12], [13], [14]], yet Kuck and Hogervorst found that self-reported memory issues did not differ across stages, and that age was a stronger predictor of perceived stress and anxiety, possibly due to improved coping mechanisms in older women [10]. Similarly, Falconi et al. found no significant association between neuroendocrine changes during perimenopause and perceived stress levels [15]. These findings suggest that the link between perceived stress, menopausal symptoms and poorer QoL is more influenced by psychological factors—such as self-efficacy, resilience, and perceptions of control—than by menopausal stage alone [10,15]. Building on this, the perception of negative life events plays a further role: Pimenta et al. found that women who viewed recent life events negatively reported more severe psychological symptoms, including depressive mood, anxiety, cognitive impairment, and loss of control, while physiological symptoms such as VMS remained directly linked to menopausal stage [16]. A systematic review identified stressful life events as major risk factors for menopausal depression and anxiety, while social support emerged as a key protective factor [[17], [18], [19]], suggesting that both internal psychological resources and external social factors shape the menopausal experience. The psychological impact of stressful life events is not limited to the period in which they occur but can persist and continue to shape menopausal symptom experience over time [20]. Women who remained distressed by a stressful life event that occurred in the past year reported 21% more VMS than those who experienced no recent stressors, highlighting the sustained influence of psychological stress on menopausal symptom burden [20]. However, social support did not mitigate the impact of stress on VMS in that context, possibly due to the severity of the stressor [21]. This distinction is important as it suggests that VMS may be more hormonally driven and less responsive to psychosocial factors, whereas psychological symptoms such as anxiety, depressed mood, and cognitive complaints are more closely tied to cognitive appraisal and therefore more amenable to psychosocial intervention [21]. Therefore, the present study focuses specifically on psychosocial QoL and its interrelationships with stress and protective psychological resources, rather than vasomotor or physical outcomes. To understand how these interactions operate, the present study draws on the Transactional Model of Stress and Coping by Lazarus and Folkman (see Fig. 1), a well-established theoretical framework which proposes that the stress response is influenced by how a person appraises a situation and their available coping responses [22,23]. Coping and adaptability to challenges and problems were seen as a consequence of transactions between a person and their environment [22]. Within this framework, internal factors such as resilience and self-efficacy enable individuals to cope with stressors, while external factors such as social support and negative life events interact with these internal resources to shape psychological outcomes. Importantly, resilience is not conceptualised here as a fixed trait, but rather as a dynamic capacity that can be shaped by exposure to negative life events [24], making it a malleable and therefore viable target for intervention.Fig. 1Theoretical framework based on Lazarus and Folkman's Transactional Theory of Stress and Coping. Theoretical framework based on Lazarus and Folkman's Transactional Theory of Stress and Coping. Increasing social support may help boost psychological resilience and alleviate menopausal symptoms, thereby improving psychosocial QoL [25,26]. Social support from family, friends, and significant others may mitigate psychological symptoms through enhanced health behaviours, prosocial behaviour, and increased self-esteem [18,19], positioning it as both a direct protective factor and an indirect pathway through which the impact of stress on psychosocial QoL may be reduced during the menopausal transition. Resilience plays an equally important role, enabling women to adapt to the challenges posed by negative life events and menopausal stressors. Higher resilience has been associated with lower perceived stress and fewer psychological complaints such as depressive mood and anxiety, ultimately contributing to better psychosocial QoL [10,27]. Self-efficacy — one's belief in their capacity to manage challenges and achieve desired outcomes — further shapes how menopause symptoms are appraised and experienced [28]. Higher self-efficacy and resilience may indirectly reduce the perceived severity of menopausal psychosocial symptoms by enhancing coping capacity and reducing the subjective threat posed by stressors [10,27,28]. Recent studies suggest that increasing self-efficacy and reducing stress may be effective treatment options for perimenopausal women, contributing to improved psychological well-being [29,30]. Together, social support, resilience, and self-efficacy represent the key modifiable protective factors examined in the present study. Building on this theoretical framework, we developed a path model (Fig. 2) to explore the interrelationships between negative life events, perceived stress, social support, resilience, self-efficacy, and psychosocial QoL in women across the menopausal transition. This study aims to understand how these factors interact and influence each other, generate hypotheses for future longitudinal research, and identify potential intervention targets to enhance the quality of life and emotional well-being of menopausal women. We hypothesised that negative life events would be associated with poorer psychosocial QoL, and that social support, resilience, self-efficacy, and perceived stress would mediate this relationship among women going through the menopausal transition, whilst controlling for age and menopausal stage.Fig. 2Proposed theoretical model. Proposed theoretical model. Theoretical framework and proposed model Table 1Sample characteristics (n = 269).Table 1Sample characteristicsN (%)Race/EthnicityWhite British/Other White Background243 (90.3%)Black/African/Caribbean/Black British3 (1.1%)Asian/Asian British16 (5.9%)Mixed/Multiple Ethnic groups4 (1.5%)Other3 (1.1%)Level of educationLess than high school9 (3.3%)High school/College/A-Levels86 (32%)University degree98 (36.4%)Postgraduate degree76 (28.3%)Employment StatusEmployed full-time150 (55.8%)Employed part-time69 (25.7%)Retired8 (3.0%)Other41 (15.2%)Marital StatusMarried/In a relationship203 (75.5%)Divorced/Separated/Single62 (23.0%)Widowed3 (1.1%)Number of ChildrenNone61 (22.7%)143 (16.0%)2-4160 (59.5%)More than 45 (1.9%)MHT useNone105 (39.0%)Past22 (8.2%)Current142 (52.8%)Menopausal StagePremenopausal36 (13.4%)Perimenopausal120 (44.6%)Postmenopausal113 (42.0%)Negative life eventsMajor change in eating habits98(36.4)Changed work situation59(21.9%)Serious illness of family member68 (25.3%)Trouble with employer31 (11.5%)Major change in financial status63 (23.4%)Major change in closeness of family member56 (20.8%)Major change in arguments with spouse48 (17.8%)Son or daughter leaving home37 (13.8%)Major personal illness or injury35 (13.0%)Retirement from work13 (4.8%)Number of negative life events experienced in the past year071 (26.4%)169 (25.7%)249 (18.2%)335 (13.0%)4 or more events45 (16.7%)Psychosocial symptomsBeing dissatisfied with my personal life145(53.9%)Feeling anxious or nervous197(73.2%)Experiencing poor memory201(74.7%)Accomplishing less than I used to178(66.2%)Feeling depressed, down or blue169 (62.8%)Being impatient with other people176 (65.4%)Feelings of wanting to be alone163 (60.6%)Number of psychosocial symptoms experienced010 (3.7%)115 (5.6%)227 (10%)326 (9.7%)433 (12.3%)545 (16.7%)654 (20.1%)759 (21.9%) Table 1Sample characteristics (n = 269).Table 1Sample characteristicsN (%)Race/EthnicityWhite British/Other White Background243 (90.3%)Black/African/Caribbean/Black British3 (1.1%)Asian/Asian British16 (5.9%)Mixed/Multiple Ethnic groups4 (1.5%)Other3 (1.1%)Level of educationLess than high school9 (3.3%)High school/College/A-Levels86 (32%)University degree98 (36.4%)Postgraduate degree76 (28.3%)Employment StatusEmployed full-time150 (55.8%)Employed part-time69 (25.7%)Retired8 (3.0%)Other41 (15.2%)Marital StatusMarried/In a relationship203 (75.5%)Divorced/Separated/Single62 (23.0%)Widowed3 (1.1%)Number of ChildrenNone61 (22.7%)143 (16.0%)2-4160 (59.5%)More than 45 (1.9%)MHT useNone105 (39.0%)Past22 (8.2%)Current142 (52.8%)Menopausal StagePremenopausal36 (13.4%)Perimenopausal120 (44.6%)Postmenopausal113 (42.0%)Negative life eventsMajor change in eating habits98(36.4)Changed work situation59(21.9%)Serious illness of family member68 (25.3%)Trouble with employer31 (11.5%)Major change in financial status63 (23.4%)Major change in closeness of family member56 (20.8%)Major change in arguments with spouse48 (17.8%)Son or daughter leaving home37 (13.8%)Major personal illness or injury35 (13.0%)Retirement from work13 (4.8%)Number of negative life events experienced in the past year071 (26.4%)169 (25.7%)249 (18.2%)335 (13.0%)4 or more events45 (16.7%)Psychosocial symptomsBeing dissatisfied with my personal life145(53.9%)Feeling anxious or nervous197(73.2%)Experiencing poor memory201(74.7%)Accomplishing less than I used to178(66.2%)Feeling depressed, down or blue169 (62.8%)Being impatient with other people176 (65.4%)Feelings of wanting to be alone163 (60.6%)Number of psychosocial symptoms experienced010 (3.7%)115 (5.6%)227 (10%)326 (9.7%)433 (12.3%)545 (16.7%)654 (20.1%)759 (21.9%) Sample characteristics (n = 269). Methods Study design This study aimed to validate a theoretical path model (Fig. 2) through a secondary analysis using data from a previously published study [10]. Negative life events was used as an exogenous variable in the model, and the endogenous variables are as follows: social support, resilience, self-efficacy, perceived stress, and psychosocial QoL. The path model explores the impact of negative life events on the psychosocial QoL of menopausal women, as well as the mediating role of perceived stress, internal factors (self-efficacy, resilience) and external factors (social support) in the relationship between negative life events and psychosocial QoL, using path analysis. Path analysis is an extension of multiple regression that helps to better understand complex relationships in intricate models [31]. Participants and data collection Women (sex assigned at birth) aged 40–60 who were going through the menopausal transition were recruited through convenience sampling. The menopausal stage was self-reported. Participants were presented with four descriptions corresponding to STRAW+10 stages and selected the one that best matched their current status: premenopausal (menstruated within the previous three months with no recent increase in cycle irregularity), early perimenopausal (menstruated within the previous three months but with increasing cycle-length irregularity over the past year), late perimenopausal (menstruated within the previous 12 months but not the previous three), and postmenopausal (no menstruation for at least 12 months). For the present analyses, the early and late perimenopausal responses were combined into a single perimenopausal group (premenopausal n = 36; perimenopausal n = 120; postmenopausal n = 113) [10]. Stage-based comparisons are reported in that prior study, supporting the validity of treating the sample as a menopausal transition continuum [10]. The online survey was accessible through a QR code on a poster shared through social media and menopause-related web-based groups. No incentives were provided, and all data collected were anonymous. The study was approved by the Loughborough University Ethics Review Sub-Committee (Project ID: 13264). Prior to participating, participants were given an information sheet to review and gave their consent to participate. Participants were invited to complete an anonymous online survey on the Bristol Online Surveys platform. This study employed a cross-sectional design. Participants completed the survey on a single occasion, and no follow-up data were collected. They were free to withdraw from the study at any point during the survey, and their data were not retained. Path analysis generally requires a sample size of at least 150 and approximately 10 to 20 times the number of observed variables [31]. Given six observed variables in our model, the recommended minimum of 10-20 observations per variable yields a range of 60-120; however, the absolute minimum of n ≥ 150 was used as the threshold. Our achieved sample size (n = 269), after removing three responses due to having over 10% missing data, comfortably exceeds this. Measures Negative life experiences Ten selected items from The Life Experiences Survey were included based on their relevance to the aim of the study [32]. These ten items were 1) major change in eating habits, 2) changed work situation, 3) serious illness of a close family member, 4) trouble with employer, 5) major change in financial status, 6) major change in closeness of family members, 7) major change in number of arguments with spouse, 8) son or daughter leaving home, 9) major personal illness or injury, 10) retirement from work. Participants rated how the specific events had affected their life in the past year based on a 7-point Likert scale ranging from −3 ‘extremely negative’ to +3 ‘extremely positive’. The score for negative life events was calculated; women who evaluated their experience of life events as more negative received higher scores. The internal consistency of the scale was acceptable (Cronbach α = .68). Perceived stress The Perceived Stress Scale-10 (PSS-10) assessed one's experience of stress-related feelings and thoughts within the last month on a 5-point Likert scale [33]. The PSS-10 consists of four positive items and six negative items, with each item rated from 0 (never) to 4 (very often). The total score was calculated after reversing the four positive items' scores, with a higher score indicating a higher level of perceived stress. Cronbach's α for this scale was α = .87 in the present study. Social support The Multidimensional Scale of Perceived Social Support measures an individual's perceived level of support from three sources: family, friends, and significant others. Each of the three subscales consists of 4 items, adding up to a total of 12 items [34]. The items were measured using a 7-point Likert scale, and participants ranked each item from ‘very strongly disagree’ (1) to ‘very strongly agree’ (7). The mean scale score was then used to interpret support as low (mean score between 1.00 and 2.99), moderate (score of 3.00-5.00) and high (score from 5.10 to 7.00). The internal consistency of the scale was excellent (Cronbach's α = .94). Resilience The Brief Resilience Scale (BRS) comprises 6 Likert items, with half of the items reversely coded to ensure high internal validity of the scale [35]. The scale measures one's subjective perception of their ability to bounce back or cope with stress. Participants ranked each item on a 5-point scale from ‘strongly disagree’ to ‘strongly agree’. The resilience score was calculated by dividing the sum by the total number of questions answered, with a higher score indicating a higher level of resilience. The internal consistency of the scale in this study was good, with α = .82. Self-efficacy The General Self-Efficacy scale (GSE) consists of 10 items on a 4-point scale to assess one's sense of perceived self-efficacy (control) [36]. It had a high internal consistency of α = .90 in this study. The GSE assesses an individual's belief in their abilities and competencies. A total score ranging from 10 to 40 was calculated by summing up the responses to all ten items, with no recoding required. A higher total score represents a higher level of perceived self-efficacy; a lower total score represents a lower level of self-efficacy. Psychosocial quality of life Psychosocial QoL was measured through the MenQoL scale, which consists of four domains with 29 items: vasomotor, psychosocial, physical, and sexual [8]. Participants selected ‘Yes’ or ‘No’ to indicate whether they had experienced the problem in the past month, followed by how bothered they were, from not at all bothered to extremely bothered. The conversion score ranged from 1 to 8; participants received a conversion score of ‘1’ when they did not experience the symptom, ‘2’ when they experienced the symptom but were not at all bothered by it, and ‘8’ when they were extremely bothered by the menopausal symptom. Each domain was scored separately, and the overall questionnaire score was the average of the domain means. Higher scores indicate poorer quality of life. In the present study, psychosocial QoL refers specifically to the psychosocial subscale of the MenQoL. The Cronbach's alpha coefficient for the psychosocial domain was α = .83. Data analysis To assess the validity of the model, we first tested the bivariate associations between all six observed variables in the model using Pearson's r correlation. To estimate the relationships among the variables and the validity of the proposed theoretical model in Fig. 2, a path analysis was performed. Missing data (N:44, less than 5% missing values for each variable) were replaced with median values for ordinal data [37]. The normality of the data was examined with skewness and kurtosis, and all variables had an absolute skewness of ≤3 and an absolute kurtosis of <10, confirming normality. A sensitivity power analysis indicated that our sample (n = 269) provided ≥80% power to detect standardised path coefficients of β ≥ .17 (f2 ≥ .03) at α = .05, which encompasses all direct paths in the model [38]. The weakest direct path (NLE → psychosocial QoL, β = .164) achieved 77% power, marginally below the conventional 80% threshold. For the primary mediation path (NLE → perceived stress → psychosocial QoL, indirect β = .195), power exceeded 99%. Smaller serial mediation paths (indirect β = .032–.039) were likely underpowered at the current sample size, which is acknowledged as a limitation for the interpretation of the secondary indirect effects. The overall model fit was comprehensively assessed using several fit statistics, including Chi-square (χ2), degrees of freedom (df), comparative fit index (CFI), Tucker–Lewis index (TLI), standardised root mean square residual (SRMR), and root mean square error of approximation (RMSEA). The standardisation coefficient and 95% confidence interval (CI) were calculated with 5000 nonparametric bootstrapping for the direct and indirect paths. All data were processed with the IBM Statistical Package for the Social Sciences (SPSS) version 29, and path analysis was conducted with Amos (version 29). Results Participants’ general characteristics and relationship among variables Table 1 displays the general characteristics of the participants. Almost half of the women reported they were perimenopausal (N = 120, 45%). The mean age of women was 51.3 (SD = 4.8) years old. Over half were currently using menopausal hormone therapy (MHT) (52.8%). Most were White British and White other (90.3%), married or in a relationship (75.5%), employed full time (55.8%), and had 2-4 children (59.5%). Only 9 (3.3%) women had received less than a high school level of education. No significant differences were found between the participants’ psychosocial QoL based on their general characteristics. Women in this study reported substantially higher perceived stress than general population norms (d = 1.02), marginally lower resilience (d = −.41), and slightly lower social support (d = −.38). Self-efficacy scores were consistent with general population norms (d = .05). Mean psychosocial QoL was x‾ = 4.07 (range 1-8, higher scores indicating poorer QoL), as shown in Table 2.Table 2Mean and standard deviation values of the study variables.Table 2M (SD)Normative M (SD)Cohen's dSkewnessKurtosisResilience3.22 (.87)3.53(.60)−.41−.43−.22Self-efficacy29.64 (4.87)29.39(5.08).05−.41.82Perceived Stress20.38 (7.53)13.02(6.89)1.02−.06−.61Social support5.14 (1.39)5.59(.94)−.38−1.01.82Psychosocial QoL4.07 (1.82)--.11−.93Vasomotor QoL2.80 (2.00)--.99−.06Physical QoL3.65 (1.48)--.40−.53Sexual QoL3.33 (2.19)--.54−.91Negative life events4.00 (4.50)--1.8694.969Note: Potential maximum and minimum scores for each scale are as follows: Resilience (1-5); Self-efficacy (10-40); Perceived Stress (0-40); Psychosocial QoL (1-8); Negative life events (0-30); Social support (1-7). All four MenQoL domains are reported descriptively for completeness; only psychosocial QoL was modelled, consistent with the study's theoretical focus. Table 2Mean and standard deviation values of the study variables.Table 2M (SD)Normative M (SD)Cohen's dSkewnessKurtosisResilience3.22 (.87)3.53(.60)−.41−.43−.22Self-efficacy29.64 (4.87)29.39(5.08).05−.41.82Perceived Stress20.38 (7.53)13.02(6.89)1.02−.06−.61Social support5.14 (1.39)5.59(.94)−.38−1.01.82Psychosocial QoL4.07 (1.82)--.11−.93Vasomotor QoL2.80 (2.00)--.99−.06Physical QoL3.65 (1.48)--.40−.53Sexual QoL3.33 (2.19)--.54−.91Negative life events4.00 (4.50)--1.8694.969Note: Potential maximum and minimum scores for each scale are as follows: Resilience (1-5); Self-efficacy (10-40); Perceived Stress (0-40); Psychosocial QoL (1-8); Negative life events (0-30); Social support (1-7). All four MenQoL domains are reported descriptively for completeness; only psychosocial QoL was modelled, consistent with the study's theoretical focus. Mean and standard deviation values of the study variables. Note: Potential maximum and minimum scores for each scale are as follows: Resilience (1-5); Self-efficacy (10-40); Perceived Stress (0-40); Psychosocial QoL (1-8); Negative life events (0-30); Social support (1-7). All four MenQoL domains are reported descriptively for completeness; only psychosocial QoL was modelled, consistent with the study's theoretical focus. Relationship between variables Prior to model testing, the correlations among the variables were analysed (Table 3). Psychosocial QoL was significantly correlated with all other variables, but the strongest correlation was seen between perceived stress and psychosocial QoL. Age was only significantly correlated with increased resilience, lower perceived stress, and increased psychosocial QoL (as a lower score indicates better QoL).Table 3Correlation analysis between study's variables with Spearman's rank. (n = 269).Table 3Variables12345671. Age--2. Resilience.139**--3. Self-efficacy.093.595**--4. Perceived Stress−.211*−.457**−.439**--5. Negative life events.037−.208**−.204**.444**--6. Social support.085.217**.198**−.332**−.186**--7. Psychosocial QoL8. Menopausal stage−.207**.609**−.353**.055−.275**.067.673**−.115.417**−.009−.212**.114--−.090**. Correlation is significant at p < .001 (2-tailed). Table 3Correlation analysis between study's variables with Spearman's rank. (n = 269).Table 3Variables12345671. Age--2. Resilience.139**--3. Self-efficacy.093.595**--4. Perceived Stress−.211*−.457**−.439**--5. Negative life events.037−.208**−.204**.444**--6. Social support.085.217**.198**−.332**−.186**--7. Psychosocial QoL8. Menopausal stage−.207**.609**−.353**.055−.275**.067.673**−.115.417**−.009−.212**.114--−.090**. Correlation is significant at p < .001 (2-tailed). Correlation analysis between study's variables with Spearman's rank. (n = 269). **. Correlation is significant at p < .001 (2-tailed). Goodness of fit of the hypothetical model The overall model fit was evaluated using various fit statistics: likelihood-ratio χ2 with p > .05, a comparative fit index (CFI) of ≥.95, Tucker-Lewis Index (TLI) of >.90, standardised root mean square residual (SRMR) of ≤.08, and a root mean square error of approximation (RMSEA) of ≤.06. Menopausal stage did not significantly predict any variable in the present model and was therefore excluded, consistent with our prior study using the same cohort in which no significant differences were found across menopausal stages for any key variable [10]. The final model, as illustrated in Fig. 3, demonstrated a good fit: p = .222, χ2/df = 1.350, CFI: .995, TLI: .984, SRMR: .031, and RMSEA: .036.Fig. 3Standardised regression weights of the final path analysis model, adjusting for age. Standardised regression weights of the final path analysis model, adjusting for age. Path analysis Direct The direct effects of the study's variables associated with the psychosocial QoL of menopausal women are presented below in Table 4. Negative life events, the exogenous variable, were associated with lower perceived social support (β = −.186; p = .004) and lower resilience (β = −.182; p = .009). A significant positive direct association was seen between negative life events and perceived stress (β = .338; p < .001) and poorer psychosocial QoL (β = .164; p = .002, as a higher score means worse QoL). Higher perceived stress level was associated with poorer psychosocial QoL (β = .578; p < .001). Resilience had a positive association with higher self-efficacy (β = .595; p < .001) and higher social support was also associated with higher resilience (β = .172; p = .006). Within the model, perceived stress functions in a dual role: as an endogenous outcome variable with respect to social support, resilience, and self-efficacy, and simultaneously as the primary mediator between these variables and psychosocial QoL. Social support (β = −.174; p = .002), resilience (β = −.214; p < .001), and self-efficacy (β = −.198; p < .001) each showed significant negative associations with perceived stress, indicating that higher levels of these protective factors were associated with lower perceived stress. However, social support, resilience, and self-efficacy did not directly and significantly affect psychosocial QoL in this model. Thus, these direct paths were removed from the final path model alongside the control, menopausal stage, as it did not significantly predict any variables. This is consistent with Kuck & Hogervorst [10], where menopausal stage ceased to predict stress and anxiety once age was covaried.Table 4Direct and Indirect paths (standardised).Table 4PathwayβSEtp95% CILowerUpperDirectNLE → SS−.186.019−3.103.004−.308−.062NLE → RS−.182.011−3.058.009−.304−.042NLE → PS.338.0816.942<.001.239.431NLE → QoL.164.0203.290.002.068.257RS → SE.595.27512.109<.001.503.673RS → PS−.214.516−3.548<.001−.338−.086RS → QoL−.094.121−1.613.147−.217.032SE → PS−.198.089−3.379<.001−.318−.078SE → QoL.071.0211.260.245−.046.189PS → QoL*.578.01211.412<.001*.491.656SS → RS.172.0372.908.002.044.306SS → PS−.174.259−3.575<.001−.288−.063SS → QoL.022.061.462.659−.072.118IndirectNLE→SS→PS.033.0252.120.003.016.117NLE→RS→PS.039.0341.882.005.015.149NLE→RS→SE→PS.021.0181.944.005.009.083NLE→SS→RS→PS.007.0071.571.004.002.034NLE→SS→RS→SE→PS.004.0041.500.003.001.019NLE→PS (total indirect effect).103.0493.490<.001.083.275NLE→PS→QoL.195.0136.077<.001.056.106NLE→SS→PS→QoL.019.0042.000.003.002.016NLE→RS→PS→QoL.023.0051.800.005.002.022NLE→RS→SE→PS→QoL.012.0031.667.005.001.012NLE→SS→RS→SE→PS→QoL.002.0011.000.003.000.003NLE→SS→RS→PS→QoL.004.0012.000.004.000.005NLE→QoL (total indirect).255.0147.286<.001.078.132Note: NLE: negative life events, SS: social support, PS: perceived stress, RS: resilience, SE: self-efficacy, QoL: psychosocial quality of life. * Indicates the strongest significant direct path. Table 4Direct and Indirect paths (standardised).Table 4PathwayβSEtp95% CILowerUpperDirectNLE → SS−.186.019−3.103.004−.308−.062NLE → RS−.182.011−3.058.009−.304−.042NLE → PS.338.0816.942<.001.239.431NLE → QoL.164.0203.290.002.068.257RS → SE.595.27512.109<.001.503.673RS → PS−.214.516−3.548<.001−.338−.086RS → QoL−.094.121−1.613.147−.217.032SE → PS−.198.089−3.379<.001−.318−.078SE → QoL.071.0211.260.245−.046.189PS → QoL*.578.01211.412<.001*.491.656SS → RS.172.0372.908.002.044.306SS → PS−.174.259−3.575<.001−.288−.063SS → QoL.022.061.462.659−.072.118IndirectNLE→SS→PS.033.0252.120.003.016.117NLE→RS→PS.039.0341.882.005.015.149NLE→RS→SE→PS.021.0181.944.005.009.083NLE→SS→RS→PS.007.0071.571.004.002.034NLE→SS→RS→SE→PS.004.0041.500.003.001.019NLE→PS (total indirect effect).103.0493.490<.001.083.275NLE→PS→QoL.195.0136.077<.001.056.106NLE→SS→PS→QoL.019.0042.000.003.002.016NLE→RS→PS→QoL.023.0051.800.005.002.022NLE→RS→SE→PS→QoL.012.0031.667.005.001.012NLE→SS→RS→SE→PS→QoL.002.0011.000.003.000.003NLE→SS→RS→PS→QoL.004.0012.000.004.000.005NLE→QoL (total indirect).255.0147.286<.001.078.132Note: NLE: negative life events, SS: social support, PS: perceived stress, RS: resilience, SE: self-efficacy, QoL: psychosocial quality of life. * Indicates the strongest significant direct path. Direct and Indirect paths (standardised). Note: NLE: negative life events, SS: social support, PS: perceived stress, RS: resilience, SE: self-efficacy, QoL: psychosocial quality of life. * Indicates the strongest significant direct path. Indirect The total indirect effect of all variables on psychosocial QoL was β = .255 (p < .001). Perceived stress was the strongest significant mediator between having experienced more negative life events and poorer psychosocial QoL (β = .195; p < .001). As there were no significant direct relationship between social support, self-efficacy, resilience, and psychosocial QoL, we investigated these factors as potential mediators between negative life events and perceived stress, which was the primary driver of psychosocial QoL. The indirect association of experiencing more negative life events with higher perceived stress (β = .103, p < .001), was mediated by both resilience (β = .039; p = .005) and social support (β = .033; p = .003). Serial mediation analysis was applied to test whether social support, resilience, self-efficacy, and perceived stress level mediated the association between negative life events and psychosocial QoL. All five serial mediation pathways were tested simultaneously with AMOS and were all significant, as presented in Table 4. Collectively, the final path model (Fig. 3) explained 47% of the variance in psychosocial QoL and 41% of the variance in perceived stress. Discussion This study examined the impact of negative life events on psychosocial QoL in women undergoing the menopausal transition using a cross-sectional path analysis based on the Transactional Model of Stress and Coping [22]. Greater exposure to negative life events was associated with lower social support and lower resilience, both of which were in turn associated with higher perceived stress and poorer psychosocial QoL. Resilience, but not social support, was directly linked to self-efficacy in the model. An important consideration is whether the associations identified in this study are specific to menopausal women or reflect broader psychosocial processes in adulthood or midlife. Our prior study using the same cohort [10] found that resilience and self-efficacy were associated with psychological complaints across the full menopausal transition, regardless of menopausal stage, suggesting that the mediation pathways identified in the present model likely reflect transdiagnostic psychosocial processes rather than menopause-specific mechanisms. Similar associations have been reported in non-menopausal adult populations, consistent with broader stress and coping frameworks [22]. What may be specific to the menopausal context is the particular content of psychosocial QoL impairment captured by the MenQoL, where symptoms such as poor memory, depressed mood, and anxiety are experienced and appraised within the context of the menopausal transition, potentially intensifying their perceived impact on QoL. The hormonal changes of the menopausal transition, in which declining oestrogen may heighten HPA axis reactivity and stress sensitivity, may further shape how these pathways operate in this specific population. Future research comparing these pathways in menopausal and non-menopausal samples, including men of similar age, would be needed to test this in more detail. The strongest direct path in the model was the association between resilience and self-efficacy (β = .595), representing the most clinically actionable finding of the present study. Increased resilience and self-efficacy were associated with lower perceived stress, leading to fewer psychological complaints such as depressive mood and anxiety, ultimately enhancing psychosocial QoL in menopausal women [10]. This is consistent with Bandura's self-efficacy and social cognitive theories, which emphasise effort, mastery, and perseverance as key components of self-efficacy development [39,40]. Similarly, Lee and Lee [29] found that strengthening self-efficacy while minimising stress significantly improved the psychological well-being of middle-aged women. Crucially, both resilience and self-efficacy are modifiable internal factors, meaning their association with perceived stress and psychosocial QoL represents a direct target for psychological intervention. Evidence-based approaches such as cognitive behavioural therapy, mindfulness-based stress reduction, and resilience training programmes may therefore represent effective non-pharmacological strategies for improving psychosocial QoL across the menopausal transition, independent of menopausal stage [10]. Among all the items listed in the psychosocial QoL domain, poor memory was the most frequently reported (74.7%), suggesting a robust association between perceived stress and subjective memory complaints in midlife women. This finding is consistent with Bell et al. [41], who found that higher perceived stress was associated with increased memory complaints in older adults, and with Potter et al. [42], who demonstrated that perceived stress was independently associated with memory complaints regardless of depression, anxiety, and activity levels. Our prior study using the same cohort found a similar association between poor memory and perceived stress, with no significant links to resilience, self-efficacy, or menopausal stage, even when accounting for age [10], suggesting this relationship is driven primarily by stress appraisal rather than hormonal or stage-specific factors. These findings collectively highlight the central role of perceived stress in shaping subjective cognitive appraisal, particularly memory [10,41,42]. This has broader clinical relevance, given evidence that elevated perceived stress and exposure to two or more stressful life events heighten the risk of dementia and mild cognitive impairment in later life [43,44]. Women face a higher dementia risk than men, partly attributed to the decline in neuroprotective sex steroids during the menopausal transition [45], though these observations are drawn from the wider literature and cannot be directly tested in the present study, which did not include hormonal assays or cognitive assessments. They are noted here as relevant background context for future research incorporating biological and cognitive measures. Over 73% of participants reported experiencing at least one negative life event in the past year, with 31.2% reporting two to three events and 16.7% reporting four or more. The most common stressors, including changes in eating habits, family illness, financial instability, work transitions, and shifts in family closeness, were reported by over 20% of participants, and were comparable to the life events observed in the longitudinal Seattle Midlife Women's Health Study [46], where approximately 86% of women experienced sustained stress with only a slight decline over a decade [46]. This underscores the persistent nature of stress exposure during midlife. Consistent with this high prevalence of negative life events, perceived stress levels in the present sample were substantially elevated relative to general population norms (d = 1.02). This considerably higher stress burden observed in this sample may in part explain why perceived stress emerged as the strongest predictor of psychosocial QoL impairment in the path model. Increased exposure to negative life events was directly associated with lower resilience, and both were independently associated with higher perceived stress. This is consistent with García-León et al. [47], where resilience levels were unrelated to chronic objective stress (measured via hair cortisol), but were associated with the number and intensity of stressful life events and perceived stress. Their study also found that individuals with lower resilience exhibited higher levels of psychopathological symptoms, including phobic anxiety, obsessions and compulsions, and interpersonal sensitivity [47]. Social support represents a key modifiable external factor in the management of psychosocial QoL during the menopausal transition. Although it was a weaker mediator in our analyses, social support still played a significant role as both a parallel and serial mediator in the relationship between negative life events, perceived stress, and psychosocial QoL. This is consistent with a recent systematic review, which found social support to be a common element in effective psychosocial interventions targeting non-physiological menopausal symptoms, including depression, anxiety, cognition, and QoL [48]. The present findings further suggest that social support may operate not only as a direct protective factor but also as an indirect pathway through which resilience and self-efficacy are enhanced, thereby reducing perceived stress and improving psychosocial QoL. This aligns with the broader literature suggesting that support from family, friends, and significant others may mitigate psychological symptoms through enhanced health behaviours, prosocial behaviour, and increased self-esteem [18,19], and is consistent with studies from Iran and Vietnam, where higher perceived social support was also linked to better overall QoL and psychosocial QoL in postmenopausal women [49,50]. Together, these findings underscore the value of social support as both a standalone intervention target and a facilitator of other protective psychological resources. Methodological considerations Our findings should be viewed in light of the following considerations. First, the proposed analyses for this study were not pre-registered, although the hypotheses and analytical model were theoretically grounded in the Transactional Model of Stress and Coping prior to data collection. Second, smaller serial indirect paths (indirect β = .032–.039) were likely underpowered, representing a limitation for these secondary findings. Third, menopausal stage was assessed via self-report using STRAW+10 criteria; reliance on self-report staging is an acknowledged limitation. Fourth, MHT use was assessed via a single yes/no item; given that 52.8% of participants reported MHT use, heterogeneity of regimens (type, route, dose) may represent a confounding factor. Fifth, analyses focused on the psychosocial MenQoL subscale; excluding the vasomotor, physical, and sexual domains limits generalisability to the full menopausal QoL spectrum. Sixth, the negative life events measure did not account for recurrence or cumulative impact. Seventh, the cross-sectional design prevents causal inferences; longitudinal studies incorporating repeated measures are needed to examine how these relationships evolve and whether interventions targeting resilience and social support have sustained benefits. The sample comprised predominantly white, educated, UK-based individuals, limiting generalisability. Given that menopausal symptoms may vary across ethnic and socioeconomic groups [51], future studies should include more diverse populations. Conclusion Perceived stress emerged as the central mediator linking negative life events to psychosocial QoL in women across the menopausal transition, with resilience, self-efficacy, and social support as key modifiable protective factors. The substantially elevated perceived stress in this sample relative to population norms (d = 1.02) highlights the psychological burden of this life stage and the importance of targeted intervention. The strong resilience–self-efficacy association (β = .595) represents the most clinically actionable finding, suggesting that evidence-based interventions targeting these internal resources — alongside social support — may offer effective non-pharmacological strategies for improving psychosocial QoL, independent of menopausal stage. Future longitudinal research is needed to establish causal directionality and assess the long-term effectiveness of such approaches. Declarations Please note that an abstract of this paper was presented as a poster at the Menopause Society 2024 Annual Meeting and the 19th IMS World Congress on Menopause. Ethics approval and consent to participate The study was approved by the Loughborough University Ethics Review Sub-Committee (Project ID: 13264). Prior to participating, participants were given an information sheet to review and gave their consent to participate. Consent for publication n/a. Data availability statement The data that support the findings of this study are available on reasonable request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Funding No funding was received for the development of this paper. CRediT authorship contribution statement Ming Jun Kuck: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing – original draft, Writing – review & editing. Nicola J. Paine: Supervision, Validation, Writing – review & editing. Eef Hogervorst: Project administration, Supervision, Validation, Writing – review & editing. Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Eef Hogervorst acted as expert on dementia and MHT for NICE (UK Guidelines) and ESHRE (EU Guidelines). None of the other authors have any conflict of interest.