{"paper_id":"22d17f5d-7d0b-4017-9b93-5f8fdc600134","body_text":"Premature ovarian insufficiency (POI), also known as premature menopause or in former days as premature ovarian failure, is a condition in which the ovaries cease to function normally before the age of 40.1 POI is a life-changing diagnosis with profound physical, psychological, and social consequences.2 Women with POI not only experience the symptoms associated with estrogen deficiency but also experience the loss of reproductive function. A combination of these factors may lead to various psychological problems that significantly impact their quality of life (QoL).\nPOI is linked to an elevated lifetime risk for depression and anxiety. A recent meta-analysis revealed an odds ratio (OR) of 3.3 for depression and 4.9 for anxiety in women with POI compared with those without the condition.3 While the precise mechanisms underlying anxiety and depression in POI are yet to be fully elucidated, potential contributing factors may involve both neuroendocrine dysregulation and psychosocial stressors. Both a large increase in FSH levels4 and low levels of estradiol (E2)5 are associated with an increased risk of anxiety and depression. On the psychosocial front, the experience of infertility and the additional burdens imposed by estrogen deficiency, such as vasomotor symptoms, reduced short term memory function, reduced bone mineral density, and an increased risk of cardiovascular disease,6–8 may contribute to the heightened risk of anxiety and depression in women with POI.\nPrevious research has shown a vulnerability for depressive symptoms during the menopause transition.9 The incidence of depressive symptoms is 2-3 times higher in perimenopausal women compared with premenopausal women. Therefore, there seems to be a relation between depressive symptoms and menopausal status. This study suggested that hot flushes and sleep disruption caused by hot flushes are especially related to depressive symptoms. Also, women who received hormone therapy (HT) seem to have a lower risk of developing depressive symptoms compared with those who did not use HT.9\nIt would be helpful to identify factors that are associated with developing depression in women with POI. A recent study by Handing et al10 employed machine learning techniques to identify the most significant factors associated with depression risk in middle-aged European women. The findings revealed that social isolation, poor self-rated health, and family burden emerged as the primary determinants of depression susceptibility. We hypothesized that similar factors are associated with depression in women with POI. To further our understanding of the factors that contribute to depression in women with POI, we propose a large-scale cross-sectional study to identify variables that are associated with depression in this population.\nMETHODS\nWe performed a cross-sectional study to identify variables that are associated with depressive symptoms in women with POI using clinician and patient-reported outcome measures (PROM). The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Erasmus Medical Centre (The Netherlands) (protocol code MEC-2023-0094; May 10, 2023). Informed consent was obtained from all women involved in the study.\nParticipants\nAll women with POI who attended our multidisciplinary POI care unit between April 2020 and December 2023 for the first time were included. Women who did not meet the criteria for POI or did not complete the questionnaires were excluded. POI was diagnosed according to the ESHRE 2016 guideline, women should have oligo/amenorrhea for at least 4 months, and an elevated FSH level (>25 IU/L) on 2 occasions at least 4 weeks apart.1\nPatient-reported outcome measures\nThe method of collecting PROMs at our multidisciplinary POI care unit has been described before.11 In short, before the first visit, women were asked to complete multiple PROM questionnaires: Beck Depression Inventory (BDI-II), Greene Climacteric scale (GCS), and in case of grief related to infertility, the Fertility Quality of Life tool (FertiQoL). In addition, a questionnaire about general health, medical history and lifestyle, and a questionnaire about emotional support (PROMIS) were conducted. Depressive symptoms were classified by a total BDI-II score above 19.12\nClinician-reported outcome measures\nDuring the first visit to our multidisciplinary POI care unit, a comprehensive patient assessment and personalized health care plan was formulated. This assessment typically encompasses the measurement of vital signs, including blood pressure, weight, and height, to calculate body mass index (BMI). A comprehensive blood panel was conducted to evaluate hormonal levels, lipid profile, and thyroid function (assessed by thyroid stimulating hormone [TSH]). To investigate neuroendocrine dysregulation, luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol, and progesterone were included in the analysis. In addition, anti-Müllerian hormone was analyzed as a proxy for ovarian reserve, and vitamin D levels were included due to their association with depression.\nStatistical analysis\nBased on a literature review and predefined hypotheses, potential variables associated with depressive symptoms were selected. To account for uncertainty in variable inclusion, all available variables were initially examined using univariate logistic regression models. Variables with a P-value <0.20 were then included in a multivariate model. All variables were standardized for interpretability. Categorical variables included a missing category to accommodate missing data in clinical practice. We used country of birth as a proxy for ethnicity, with the Netherlands serving as the reference group. Similarly, being single was used as the reference category for relationship status. In a multivariate backwards elimination procedure, variables that did not (significantly, P<0.05) contribute to the dependent measure were removed from the model one by one. Four sub-analyses were conducted: (1) among women who expressed a desire to conceive, (2) young women aged under 30, (3) among women with idiopathic POI, and (4) to identify specific domains and questions in the GCS, FertiQol, and PROMIS for emotional support that were significantly associated with depressive symptoms. Given the exploratory nature of this study, we have not applied formal corrections for multiple testing. The statistical analyses were performed using the Statistical Package of Social Sciences version 24.0 for Windows (SPSS Inc., Chicago, IL).\nRESULTS\nBetween April 2020 and December 2023, 345 women with POI attended our multidisciplinary POI care unit. Baseline characteristics of these women are shown in Table 1. Of the 345 women, 103 (29.9%) reported depressive symptoms, while 242 (70.1%) did not. In this cohort, a considerable number of women (42.3%) were already using estrogen plus progestogen therapy (EPT) at baseline. However, there was no difference in depressive symptoms between women using EPT and those not using EPT (P=0.89).\nTABLE 1 -\nBaseline characteristics of women with POI\n|\nWomen with POI (N = 345) |\nWomen with depressive symptoms (n = 103) |\nWomen without depressive symptoms (n = 242) |\nP\n|\n| Age, mean (SD) (y) |\n35.0 (7.7) |\n33.9 (7.5) |\n35.5 (7.7) |\n0.07 |\n| Age at diagnosis, median (IQR) (y) |\n32.0 (24.0-36.0) |\n29.0 (23.0-35.0) |\n33.0 (25.0-37.0) |\n0.02 |\n| Years since diagnosis, median (IQR) |\n2.5 (1.0-7.7) |\n2.5 (1.1-8.2) |\n2.5 (1.0-7.4) |\n0.85 |\n| BMI, mean (SD) (kg/m2) |\n24.6 (4.8) |\n25.2 (5.5) |\n24.4 (4.4) |\n0.18 |\n| POI diagnoses cause |\n|\n|\n|\n0.05 |\n| Idiopathic, n (%) |\n242 (70.1) |\n78 (75.7) |\n164 (67.8) |\n|\n| Iatrogenic, n (%) |\n75 (21.7) |\n23 (22.3) |\n52 (21.5) |\n|\n| Genetic, n (%) |\n24 (7.0) |\n2 (1.9) |\n22 (9.1) |\n|\n| Autoimmune, n (%) |\n4 (1.2) |\n0 |\n4 (1.7) |\n|\n| Country of birth |\n|\n|\n|\n0.16 |\n| The Netherlands, n (%) |\n282 (82.5) |\n80 (77.7) |\n202 (84.5) |\n|\n| Other, n (%) |\n60 (17.5) |\n23 (22.3) |\n37 (15.5) |\n|\n| Missing, n (%) |\n3 (0.9) |\n0 |\n3 (1.2) |\n|\n| Relationship status |\n|\n|\n|\n0.19 |\n| Partner, n (%) |\n224 (64.9) |\n63 (61.2) |\n161 (66.5) |\n|\n| No partner, n (%) |\n70 (20.3) |\n27 (26.2) |\n43 (17.8) |\n|\n| Missing, n (%) |\n51 (14.8) |\n13 (12.6) |\n38 (15.7) |\n|\n| Wish to conceive |\n|\n|\n|\n0.27 |\n| Yes, n (%) |\n120 (38.3) |\n40 (38.8) |\n80 (33.1) |\n|\n| No, n (%) |\n193 (55.9) |\n57 (55.3) |\n136 (56.2) |\n|\n| Missing, n (%) |\n32 (9.3) |\n6 (5.8) |\n26 (10.7) |\n|\n| Parity |\n|\n|\n|\n0.08 |\n| Nulliparous, n (%) |\n198 (59.3) |\n63 (61.2) |\n135 (55.8) |\n|\n| ≥1, n (%) |\n136 (39.4) |\n40 (38.8) |\n96 (39.7) |\n|\n| Missing, n (%) |\n11 (3.2) |\n0 |\n11 (4.5) |\n|\n| History of depression, n (%)a |\n9 (2.6) |\n7 (6.8) |\n2 (0.8) |\n0.001 |\n| Use of antidepressants, n (%) |\n31 (9.0) |\n19 (18.4) |\n12 (5.0) |\n<0.001 |\n| EPT use during intake, n (%) |\n146 (42.3) |\n43 (41.7) |\n103 (42.6) |\n0.89 |\n| Use of oral HT, n (%) |\n103 (29.9) |\n27 (26.2) |\n76 (31.4) |\n0.33 |\n| Use of dermal HT, n (%) |\n43 (12.5) |\n16 (15.5) |\n27 (11.2) |\n0.26 |\n| Number of days EPT use before intake, median (IQR) |\n42.5 (0-785.8) |\n71 (0-672) |\n18 (0-974) |\n0.88 |\n| OCP use during intake, n (%) |\n44 (12.8) |\n12 (11.7) |\n32 (13.2) |\n0.69 |\n| Hormone levels |\n| TSH mU/L, median (IQR) |\n1.63 (1.17-2.26) |\n1.50 (1.10-2.01) |\n1.65 (1.23-2.42) |\n0.04 |\n| LH IU/L, median (IQR) |\n26.5 (9.8-40.5) |\n27.8 (11.8-46.0) |\n25.8 (9.4-38.3) |\n0.16 |\n| FSH IU/L, median (IQR) |\n50.2 (18.6-79.9) |\n53.1 (24.7-84.4) |\n47.9 (15.6-78.6) |\n0.28 |\n| Estradiol, pmol/L, median (IQR) |\n76.0 (<55-238.5) |\n70 (<55-200) |\n79 (<55-252) |\n0.33 |\n| Progesterone, nmol/L, median (IQR) |\n<0.3 (<0.3-<0.3) |\n<0.3 (<0.3-<0.3) |\n<0.3 (<0.3-<0.3) |\n0.60 |\n| AMH µg/L, median (IQR) |\n<0.01 (<0.01-0.03) |\n<0.01 (<0.01-0.02) |\n<0.01 (<0.01-0.03) |\n0.46 |\n| Vitamin D, nmol/L, mean (SD) |\n66.5 (26.6) |\n66.5 (51.0-78.0) |\n66.0 (51.0-83.0) |\n0.46 |\n| Total GCS score, mean (SD) |\n20.7 (11.0) |\n29.6 (9.6) |\n16.9 (9.2) |\n<0.001 |\n| Total FertiQoL score, mean (SD) |\n64.0 (15.5) |\n53.8 (16.4) |\n69.4 (11.8) |\n<0.001 |\n| Total PROMIS score for emotional support, median (IQR) |\n18.0 (15.5-20.0) |\n16.0 (13.0-19.0) |\n19.0 (16.0-20.0) |\n<0.001 |\n| Total BDI-II score, mean (SD) |\n15.4 (10.0) |\n27.6 (7.2) |\n10.1 (5.3) |\n<0.001 |\nValues of baseline characteristics are displayed as mean with SD, median with interquartile range (IQR) or numbers with percentages (%), depending on the distribution.\nAMH, anti-Müllerian hormone; BDI-II, Beck Depression Inventory-II; BMI, body mass index; EPT, estrogen plus progestogen therapy; FertiQoL, Fertility Quality of Life; FSH, follicle-stimulating hormone; GCS, Greene climacteric scale; HT, hormone therapy; LH, luteinizing hormone; OCP, oral contraceptive pill; POI, premature ovarian insufficiency; PROMIS, Patient-Reported Outcomes Measurement Information System; TSH, thyroid-stimulating hormone; y, years.\naFive women who reported a history of depression were still using antidepressants (all in the group of women with depressive symptoms).\nWomen with depressive symptoms were younger at the time of diagnosis (29.0 vs. 33.0, P=0.02), the distribution of POI diagnosis cause was borderline significantly different (P=0.05), and women with depressive symptoms had a significantly lower TSH (1.50 vs. 1.65 mU/L, P=0.04). Furthermore, women with depressive symptoms had significantly higher GCS scores, lower fertiQoL scores and lower total PROMIS score for emotional support (P<0.001). Furthermore, age, BMI, and relationship status were not significantly different between the 2 groups. In total, 35 women had a score indicative of severe depression (BDI >30), of these women 4 reported havinga history of depression and 8 were using antidepressants.\nDeterminants of depressive symptoms (BDI-II >19)\nTo explore variables associated with depressive symptoms among women with POI, we first conducted an univariate analysis. In the univariate models (Table 2), age (OR=0.97, P=0.07), age at diagnosis (OR=0.97, P=0.04), BMI (OR=1.04, P=0.14), POI diagnosis cause genetic (OR=0.19, P=0.03), country of birth as a proxy for ethnicity (OR=1.57, P=0.13), relationship status (OR=0.62, P=0.10), TSH (OR=0.80, P=0.09), LH (OR=1.01, P=0.14), anti-Müllerian hormone (OR=0.12, P=0.16), total GCS score (OR=1.14, P<0.001), total FertiQoL score (OR=0.92, P<0.001), and total PROMIS score for emotional support (OR=0.82, P<0.001) had a P-value <0.20 and were therefore included in the multivariable model. The multivariable logistic regression model showed that more severe menopausal symptoms (GCS score, OR=1.13, P<0.001), lower emotional support (PROMIS score, OR=0.86, P<0.001) and younger age at diagnosis (OR=0.95, P=0.01) were significantly associated with higher odds of reporting depressive symptoms. POI caused by known genetic variants (OR=0.10, P=0.04) was associated with a lower odds of reporting depressive symptoms. This multivariable logistic regression model explained almost half of the variance in depressive symptoms (R2=0.45).\nTABLE 2 -\nUnivariate model: determinants of depressive symptoms (BDI-II >19)\n| Determinants |\nOR (95% CI) |\nP\n|\n| Age |\n0.97 (0.94-1.00) |\n0.07d |\n| Age at diagnosis |\n0.97 (0.94-1.00) |\n0.04d |\n| Years since diagnosis |\n1.01 (0.97-1.05) |\n0.76 |\n| BMI |\n1.04 (0.99-1.09) |\n0.14d |\n| POI diagnoses cause—iatrogenic |\n0.93 (0.53-1.63) |\n0.80 |\n| POI diagnoses cause—genetic |\n0.19 (0.04-0.83) |\n0.03d |\n| POI diagnoses cause—autoimmune |\n0.00 |\n0.99 |\n| Country of birtha |\n1.57 (0.88-2.81) |\n0.13d |\n| Relationship statusb |\n0.62 (0.36-1.09) |\n0.10d |\n| Wish to conceivec |\n1.19 (0.73-1.95) |\n0.48 |\n| Parity |\n1.12 (0.70-1.80) |\n0.64 |\n| EPT use during intake |\n0.97 (0.61-1.54) |\n0.89 |\n| Use of oral HT |\n0.78 (0.46-1.30) |\n0.34 |\n| Use of dermal HT |\n1.46 (0.75-2.85) |\n0.26 |\n| Number of days EPT use before intake |\n1.00 (1.00-1.00) |\n0.35 |\n| OCP use during intake |\n0.87 (0.43-1.76) |\n0.69 |\n| TSH |\n0.80 (0.62-1.04) |\n0.09d |\n| LH |\n1.01 (1.00-1.02) |\n0.14d |\n| FSH |\n1.00 (1.00-1.01) |\n0.28 |\n| Estradiol |\n1.00 (1.00-1.00) |\n0.22 |\n| Progesterone |\n1.01 (0.97-1.05) |\n0.71 |\n| AMH |\n0.12 (0.01-2.31) |\n0.16d |\n| Vitamin D |\n1.00 (0.99-1.01) |\n0.46 |\n| Total GCS score |\n1.14 (1.10-1.17) |\n<0.001d |\n| Total FertiQoL score |\n0.92 (0.90-0.95) |\n<0.001d |\n| Total PROMIS score for emotional support |\n0.82 (0.76-0.88) |\n<0.001d |\nValues are displayed in odds ratios (OR) with 95% confidence interval (CI).\nAMH, anti-Müllerian Hormone; BDI-II, Beck Depression Inventory-II; BMI, body mass index; EPT, estrogen plus progestogen therapy; FertiQoL, Fertility Quality of Life; FSH, follicle-stimulating hormone; GCS, Greene climacteric scale; HT, hormone therapy; LH, luteinizing hormone; OCP, oral contraceptive pill; POI, premature ovarian insufficiency; PROMIS, Patient-Reported Outcomes Measurement Information System; TSH, thyroid-stimulating hormone.\naReference is “The Netherlands.”\nbReference is “No partner.”\ncReference is “No wish to conceive.”\ndP-value <0.20 and therefore included in the multivariate model.\nSub-analysis\nIn a subgroup of women who filled in the FertiQoL (this questionnaire is only filled in when women feel grief related to infertility), the univariate model identified total GCS score, total PROMIS score for social support, total FertiQoL score, POI diagnose cause, LH, FSH, estradiol, TSH, partner status, years since diagnosis, age at diagnosis, BMI and use of dermal HT were significantly associated with depressive symptoms. The multivariable logistic regression model showed that a worse GCS score (OR=1.09, P<0.001), a worse FertiQoL score (OR=0.93, P<0.001) and the use of dermal HT (OR=3.00, P=0.05) were significantly associated with higher odds of reporting depressive symptoms (Table 3). This multivariable logistic regression model explained 46% of the variance in depressive symptoms (R2=0.46).\nTABLE 3 -\nMultivariate model: determinants of depressive symptoms (BDI-II >19)\n|\nOR (95% CI) |\nP\n|\n| Determinants |\n|\n|\n|\nTotal GCS score |\n1.13 (1.09-1.17) |\n<0.001 |\n|\nTotal PROMIS score for emotional support |\n0.86 (0.79-0.94) |\n<0.001 |\n|\nPOI diagnoses cause—iatrogenic |\n0.98 (0.47-2.04) |\n0.95 |\n|\nPOI diagnoses cause—genetic |\n0.10 (0.01-0.95) |\n0.04 |\n|\nPOI diagnoses cause—autoimmune |\n0.00 |\n0.99 |\n|\nAge at diagnosis |\n0.95 (0.91-0.99) |\n0.01 |\n| Sub-analysis FertiQoL filled in (n = 212) |\n|\nDeterminants |\n|\n|\n|\nTotal GCS score |\n1.09 (1.04-1.13) |\n<0.001 |\n|\nTotal FertiQoL score |\n0.93 (0.90-0.96) |\n<0.001 |\n|\nUse of dermal HT |\n3.00 (1.01-8.96) |\n0.05 |\n| Sub-analysis in women with idiopathic POI (n = 242) |\n|\nTotal GCS score |\n1.13 (1.09-1.17) |\n<0.001 |\n|\nTotal PROMIS score for emotional support |\n0.85 (0.77-0.94) |\n0.002 |\n| Sub-analysis in women aged <30 y (n = 86) |\n|\nTotal GCS score |\n1.18 (1.08-1.28) |\n<0.001 |\n|\nTotal PROMIS score for emotional support |\n0.81 (0.67-0.98) |\n0.03 |\n| Sub-analysis in women aged >30 y (n = 239) |\n|\nTotal GCS score |\n1.12 (1.08-1.16) |\n<0.001 |\n|\nTotal PROMIS score for emotional support |\n0.86 (0.77-0.95) |\n0.003 |\n|\nAge at diagnosis |\n0.93 (0.87-0.98) |\n0.01 |\n|\nTSH |\n0.71 (0.47-1.07) |\n0.10 |\nValues are displayed as odds ratio (OR) with 95% CI.\nBDI-II, Beck Depression Inventory-II; FertiQoL, Fertility Quality of Life; GCS, Greene Climacteric Scale; HT, hormone therapy; POI, premature ovarian insufficiency; PROMIS, Patient-Reported Outcomes Measurement Information System; TSH, thyroid-stimulating hormone; y, years.\nIn women with idiopathic POI, similar variables were identified in the univariate model. In the multivariable logistic regression model, more severe menopausal symptoms (GCS score; OR=1.13, P<0.001) and lower emotional support (PROMIS; OR=0.85, P=0.002) were significantly associated with higher odds of reporting depressive symptoms (Table 3). This multivariable logistic regression model explained 42% of the variance in depressive symptoms (R2= 0.42).\nAmong women younger than 30 years, more severe menopausal symptoms (GCS scores) and lower emotional support (PROMIS scores) were significantly associated with higher odds of reporting depressive symptoms. In women older than 30 years, a younger age at diagnosis and lower TSH levels were also associated with depressive symptoms (Table 3). The full model explained 54% of the variance in depressive symptoms in women under 30 and 41% in women over 30.\nFinally, we investigated which domains and specific questions in the GCS, FertiQol, and PROMIS for emotional support were significantly associated with depressive symptoms (Supplemental Tables 1–3, Supplemental Digital Content 1, https://links.lww.com/MENO/B399). Of the GCS, higher scores in the anxiety and depression domains were associated with a significantly higher odds of reporting depressive symptoms. The questions regarding panic attacks and/or anxiety, fatigue and listlessness, feeling depressed or not happy, tingling or numbness in the skin and libido loss were most significant in the multivariable logistic regression model associated with depressive symptoms. We also performed a sensitivity analysis by excluding the depression domain and the psychological domain (depression and anxiety) of the GCS and found for both scores the same significant association with depressive symptoms (P<0.001). In the FertiQoL, worse scores in the domains of mind and body, relation and social were associated with a higher odds of reporting depressive symptoms. The questions regarding quality of life and talking about infertility with the partner were most significant in the multivariable logistic regression model associated with depressive symptoms. In the PROMIS for emotional support, a lower score on the statement “I have someone who makes me feel appreciated” was associated with depressive symptoms.\nDISCUSSION\nIn this study, we have investigated the key variables associated with depressive symptoms in women with POI. The findings reveal several significant associations between depressive symptoms and various clinical and patient-reported factors. One-third of the women with POI suffered from depressive symptoms. This highlights the significant impact of depressive symptoms in women with POI, underscoring the need for health care providers to prioritize this aspect of care. An increase in the severity of menopausal symptoms, a lack of emotional support, and younger age at the time of POI diagnosis were independently related to depressive symptoms. Interestingly, a genetic cause for POI was associated with lower depressive symptoms. Subgroup analysis showed that among women experiencing fertility-related grief, lower fertility–related quality of life and limited emotional support were related to depressive symptoms.\nThis study is the first to investigate variables that are associated with depressive symptoms in a large cohort of 345 women with POI. While previous research on POI has explored associations between women with and without depressive symptoms, our findings contribute novel insights. Consistent with our results, women with a history of depression have been shown to experience a more severe burden of menopausal symptoms, particularly in the psychosocial and sexual domains.13 Our findings and those of previous studies highlight the complex interplay between menopausal symptoms, sleep disturbances, anxiety, and depression in women experiencing menopausal complaints. Research in breast cancer survivors has shown that vasomotor symptoms can disrupt sleep, leading to increased anxiety, which in turn, is a primary predictor of depression.14 While sleep disturbances were not directly associated with depression in this population, their strong correlation with anxiety and depression in women with POI15 suggests a potential indirect effect.\nUnexpectedly, although a higher burden of menopausal symptoms was independently associated with depressive symptoms, vasomotor symptoms specifically (night sweats and flushes) were not, suggesting they may not be the key contributors to this association. This is consistent with the finding that hormone therapy did not significantly impact depressive symptoms. But this contradicts previous research in women with a normal menopausal age, which has demonstrated a positive association between vasomotor symptoms and depression.16 One potential explanation for this discrepancy lies in the distinct characteristics of POI. Compared with natural menopause, POI occurs at a younger age and often involves a more pronounced psychological impact, possibly due to the concept of “biographical disruption.” Biographical disruption refers to the significant disruption of an individual’s life trajectory and self-identity.17 In the context of POI, this disruption could encompass altered life goals, loss of sense of control, social stigma, and disrupted social roles. Notably, a recent study suggests that the impact of POI on daily life can be characterized by this concept.17\nIt is plausible that other domains of the menopausal experience, such as psychosocial challenges or reproductive loss, may play a more prominent role than vasomotor complaints in the association with depressive symptoms in women with POI. While the psychological impact of infertility is well-established,18 there is limited research on how infertility affects depressive symptoms in women with POI. A study in young breast cancer survivors demonstrated that concerns about premature menopause, menopausal symptoms, and infertility are associated with depressive symptoms,19 highlighting the potential importance of this loss of fertility.\nOur findings underline the need for comprehensive assessment and management of menopausal symptoms, including psychosocial and sexual well-being, in women with POI. Addressing the psychological impact of POI, including the loss of fertility, is probably crucial in preventing or mitigating depressive symptoms. In addition, lower levels of social support were independently associated with depressive symptoms, which can be helpful for early detection and preventive strategies. Finally, mental health professionals should explore the specific role of fertility loss in the relationship with depression in women with POI.\nOur finding that genetic causes for POI are associated with a lower risk of depression warrants further investigation to understand the underlying mechanisms. One potential explanation could be that women with a genetically linked POI diagnosis might have received more information and support due to a family history of the condition. This could lead to better emotional preparedness and coping mechanisms compared with women with a nongenetic cause of POI. However, the time between POI diagnosis and study intake was not significantly associated with depressive symptoms, suggesting that the length of time living with the diagnosis may not be strongly related to depressive symptoms in this cross-sectional sample.\nHowever, some limitations are important to acknowledge. The observational design of the study prevents definitive conclusions about causal relationships between the identified variables and depression. Given the large number of statistical tests conducted, the likelihood of false positive findings may be increased. Therefore, the results should be interpreted with caution. We are fully aware of the statistical implications of conducting multiple comparisons, and we emphasize that the findings should be considered hypothesis-generating rather than confirmatory. We highlight the need for future studies to test these associations using predefined hypotheses and appropriate methods to adjust for multiple comparisons. Another limitation of the study is that the assessment of depressive symptoms, menopausal symptoms, and social support was conducted at a single time point at intake in women with POI, which prevents analysis of changes over time and limits our ability to identify individuals who may be at increased risk of developing depressive symptoms. However, we aim to collect additional data in the coming years, including follow-up assessments, to enable longitudinal analyses that may help clarify trajectories and support future risk prediction. Furthermore, an age-matched premenopausal control group could help delineate the independent and interactive effects of POI diagnosis, chronic condition status, and psychosocial variables on observed depressive symptoms. In addition, the study population was recruited from a single center, which may limit the generalizability of our findings to a broader population, although women with POI from all over the country attended our center of expertise. Finally, the questionnaires used may have overlapping domains, potentially inflating the observed association between some measures. We identified significant differences in other domains and a sensitivity analysis showed similar results when we excluded the “depression” and “psychological” domains from the GCS total score. Future studies may benefit from including a wider range of variables beyond those captured by the current questionnaires.\nCONCLUSION\nThis study provides valuable insights into the complex factors associated with depressive symptoms in women with POI. Our findings underscore the high prevalence of depressive symptoms in this population and highlight the importance of comprehensive care addressing both physical and psychological aspects of menopause at an early age. By identifying key variables such as fertility-related grief, more severe menopausal symptoms, lack of emotional support, and younger age at POI diagnosis, which might enhance a deeper understanding of this issue. While the unexpected finding that vasomotor symptoms were not the primary driver challenges previous research, it emphasizes the need for further exploration of the unique psychological impact of POI. The observed association between social support and lower depressive symptoms, as well as between fertility-related grief and depressive symptoms, suggest potential areas for future research. To fully address the mental health needs of women with POI, a multidisciplinary approach is essential, incorporating psychological support, symptom management strategies, and tailored interventions to address the specific challenges faced by this population.\nREFERENCES\n1. Webber L, et alReproduction European Society for Human, Embryology Guideline Group on POI. ESHRE Guideline: management of women with premature ovarian insufficiency. Hum Reprod 2016;31:926-937. doi:\nhttps://doi.org/10.1093/humrep/dew027\n2. Maclaran K, Panay N. Current concepts in premature ovarian insufficiency. Womens Health (Lond) 2015;11:169-182. doi:\nhttps://doi.org/10.2217/whe.14.82\n3. Xi D, Chen B, Tao H, Xu Y, Chen G. The risk of depressive and anxiety symptoms in women with premature ovarian insufficiency: a systematic review and meta-analysis. Arch Womens Ment Health 2023;26:1-10. doi:\nhttps://doi.org/10.1007/s00737-022-01289-7\n4. Ryan J, Burger HG, Szoeke C, et al. A prospective study of the association between endogenous hormones and depressive symptoms in postmenopausal women. Menopause 2009;16:509-517. doi:\nhttps://doi.org/10.1097/gme.0b013e31818d635f\n5. Payne JL. The role of estrogen in mood disorders in women. Int Rev Psychiatry 2003;15:280-290. doi:\nhttps://doi.org/10.1080/0954026031000136893\n6. Atsma F, Bartelink MLEL, Grobbee DE, van der Schouw YT. Postmenopausal status and early menopause as independent risk factors for cardiovascular disease: a meta-analysis. Menopause 2006;13:265-279. doi:\nhttps://doi.org/10.1097/01.gme.0000218683.97338.ea\n7. Costa GPO, Ferreira-Filho ES, Simoes RS, Soares-Junior JM, Baracat EC, Maciel GAR. Impact of hormone therapy on the bone density of women with premature ovarian insufficiency: a systematic review. Maturitas 2023;167:105-112. doi:\nhttps://doi.org/10.1016/j.maturitas.2022.09.011\n8. Stevenson JC, Collins P, Hamoda H, et al. Cardiometabolic health in premature ovarian insufficiency. Climacteric 2021;24:474-480. doi:\nhttps://doi.org/10.1080/13697137.2021.1910232\n9. Cohen LS, Soares CN, Vitonis AF, Otto MW, Harlow BL. Risk for new onset of depression during the menopausal transition: the Harvard Study of Moods and Cycles. Arch Gen Psychiatry 2006;63:385-390. doi:10.1001/archpsyc.63.4.385\n10. Handing EP, Strobl C, Jiao Y, Feliciano L, Aichele S. Predictors of depression among middle-aged and older men and women in Europe: a machine learning approach. Lancet Reg Health Eur 2022;18:100391. doi:\nhttps://doi.org/10.1016/j.lanepe.2022.100391\n11. van Zwol-Janssens C, Jiskoot G, Schipper J, Louwers YV. Introducing a value-based healthcare approach for women with premature ovarian insufficiency (POI): Recommendations for patient-centered outcomes in clinical practice. Maturitas 2024;184:107971. doi:\nhttps://doi.org/10.1016/j.maturitas.2024.107971\n12. Beck AT, Steer RA, Brown G. Beck depression inventory–II. Psychological assessment. 1996. doi:\nhttps://doi.org/10.1037/t00742-000\n13. Allshouse AA, Semple AL, Santoro NF. Evidence for prolonged and unique amenorrhea-related symptoms in women with premature ovarian failure/primary ovarian insufficiency. Menopause 2015;22:166-174. doi:10.1097/GME.0000000000000286\n14. Vincent AJ, Ranasinha S, Sayakhot P, Mansfield D, Teede HJ. Sleep difficulty mediates effects of vasomotor symptoms on mood in younger breast cancer survivors. Climacteric 2014;17:598-604. doi:10.3109/13697137.2014.900745\n15. Ates S, Aydin S, Ozcan P, Bakar RZ, Cetin C. Sleep, depression, anxiety and fatigue in women with premature ovarian insufficiency. J Psychosom Obstet Gynaecol 2022;43:482-487. doi:10.1080/0167482X.2022.2069008\n16. Kravitz HM, Colvin AB, Avis NE, Joffe H, Chen Y, Bromberger JT. Risk of high depressive symptoms after the final menstrual period: the Study of Women’s Health Across the Nation (SWAN). Menopause 2022;29:805-815. doi:10.1097/GME.0000000000001988\n17. Johnston-Ataata K, Flore J, Kokanović R, et al. “My relationships have changed because I’ve changed”: biographical disruption, personal relationships and the formation of an early menopausal subjectivity. Sociol Health Illn 2020;42:1516-1531. doi:10.1111/1467-9566.13143\n18. Stanhiser J, Steiner AZ. Psychosocial aspects of fertility and assisted reproductive technology. Obstet Gynecol Clin North Am 2018;45:563-574. doi:10.1016/j.ogc.2018.04.006\n19. Howard-Anderson J, Ganz PA, Bower JE, Stanton AL. Quality of life, fertility concerns, and behavioral health outcomes in younger breast cancer survivors: a systematic review. J Natl Cancer Inst 2012;104:386-405. doi:10.1093/jnci/djr541","source_license":"CC0","license_restricted":false}