{"paper_id":"dc0f7402-ded4-49cc-941e-1b5eaee26a16","body_text":"Bilateral oophorectomy amplifies \ndepression risk following \nhysterectomy NHANES 2006–2017\nChenghui Xu1, Guangchun Zhao2, Wenlei Yao3 & Yanhua Zhang3\nThis study aims to evaluate the association between hysterectomy with bilateral salpingo-\noophorectomy (HBSO) and depressive symptoms, exploring the impact of different surgical \napproaches on the severity of depression. Data from the 2006–2017 National Health and Nutrition \nExamination Survey (NHANES) were used to analyze the relationship between surgical methods and \ndepressive symptoms.This study analyzed data from 10,780 women aged 20–80 years, with a diverse \nracial composition: 44.2% non-Hispanic White, 20.4% non-Hispanic Black, 14.7% Mexican American, \n11.0% Other Hispanic, and 9.7% Other Race.The Patient Health Questionnaire-9 (PHQ-9), a validated \ndepression screening tool, was utilized to assess depressive symptoms. Multivariable linear regression \nand binary logistic regression analyses were conducted to evaluate the association between surgical \napproaches and depressive symptoms, with results presented as odds ratios (OR) and their 95% \nconfidence intervals (CI). Subgroup analyses employed stratified regression models to investigate \ninteractions between baseline characteristics and surgical methods. Demographic analysis showed \ndifferences in age, marital status, education, income, smoking, BMI, and chronic disease prevalence \nbetween the depressive and non-depressive groups. HBSO was significantly associated with higher \nPHQ-9 scores and a higher likelihood of significant depressive symptoms (PHQ-9 ≥  10). Hysterectomy \nwas also associated with depressive symptoms, but to a lesser extent. Further analysis revealed that \nhysterectomy was significantly associated with higher depressive scores, particularly in the PHQ-9 ≥  20 \ngroup. Subgroup analysis indicated significant interaction effects between surgical types and factors \nsuch as BMI, Income-to-Poverty Ratio (IPR), smoking, and alcohol consumption. The findings suggest \na significant association between hysterectomy, particularly HBSO, and the severity of depressive \nsymptoms. Lifestyle and behavioral factors, such as BMI, smoking, and alcohol consumption, \nsignificantly influence the occurrence of postoperative depression. Thorough evaluation of patients’ \npsychological health and related factors is essential when considering gynecological surgery.\nKeywords Hysterectomy, Bilateral oophorectomy, PHQ-9, Depression, NHANES\nDepression is a prevalent and multifaceted mental health disorder with etiological factors spanning biological, \npsychological, and social dimensions. As one of the leading contributors to the global disease burden, depression \nprofoundly impacts patients’ emotional and cognitive functions and markedly diminishes their quality of life1,2. \nData from the Global Burden of Disease study in 2019 reveal that mental disorders remain among the top ten \nglobal disease burdens, with no evidence of a reduction in this burden since 19903. Clinically, depression poses \nsignificant challenges for detection, diagnosis, and management due to its diverse manifestations, unpredictable \ncourse and prognosis, and variable response to treatment. Moreover, the prevalence of depression is notably \nhigher in women compared to men, with rates approximately twice as high in females4,5. Epidemiological studies \nsuggest that the elevated incidence of depression in women may be associated with a range of factors, including \nphysiological changes, social role stressors, and psychological influences6.\nHysterectomy and bilateral oophorectomy are commonly performed procedures in the field of gynecology, \ntypically used to address a range of benign and malignant conditions such as uterine fibroids, abnormal uterine \nbleeding, cervical cancer, endometrial cancer, benign ovarian tumors, and ovarian cancer 7,8. Hysterectomy \nis the most common gynecological surgical procedure in the United States, with over 600,000 procedures \nconducted each year9. It is estimated that around 20 million women have undergone a hysterectomy for various \nobstetric and gynecological reasons10.Traditionally, it has been believed that the ovaries are crucial for hormone \n1Department of Rehabilitation Medicine, Binhai County People’s Hospital, Yancheng, China. 2Department \nof Laboratory Medicine, Binhai County People’s Hospital, Yancheng, China. 3Department of Obstetrics and \nGynecology, Binhai County People’s Hospital, Yancheng 224000, Jiangsu, China. email: 18066132759@163.com\nOPEN\nScientific Reports |        (2024) 14:31995 1| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports\n\n\nsecretion, and thus, oophorectomy might lead to a sudden drop in hormone levels due to surgical menopause. \nThis hormonal shift could result in symptoms associated with menopause, such as anxiety, insomnia, and \nemotional disturbances, potentially progressing to depression 11–14. This risk is particularly pronounced in \nyounger, premenopausal women, where the abrupt hormonal changes triggered by surgery can significantly \nincrease the likelihood of depression. Consequently, bilateral oophorectomy has been widely considered to be \nmore likely to lead to depression 15. However, recent studies have indicated that the probability of depressive \nsymptoms also increases significantly in women following hysterectomy, especially when both hysterectomy and \nbilateral oophorectomy are performed together. In contrast, bilateral oophorectomy alone does not appear to \nsignificantly elevate the incidence of depression16–18. This emerging evidence has prompted further investigation \ninto the relationship between different surgical procedures and the risk of developing depression.\nThis study analyzes data from the 2006–2017 National Health and Nutrition Examination Survey (NHANES) \nto categorize four surgical procedures: (1) No surgery; (2) Hysterectomy; (3) Bilateral Oophorectomy; and (4) \nHysterectomy with Bilateral Salpingo-Oophorectomy (HBSO). The objective is to assess the association between \nthese procedures and depression, using Patient Health Questionnaire-9 (PHQ-9) scores to measure incidence \nand severity. Key questions include whether hysterectomy or oophorectomy significantly affects depression risk \nand how surgical type correlates with depression severity. The study aims to provide evidence for improving \nmental health management in women post-gynecological surgery.\nMaterials and methods\nStudy population\nThis study utilizes a cross-sectional design, drawing data from the National Health and Nutrition Examination \nSurvey (NHANES) conducted between 2006 and 2017. NHANES, organized by the Centers for Disease Control \nand Prevention (CDC), is an ongoing survey to evaluate the health and nutritional status of the adult population \nin the United States. The survey employs a stratified, multistage, probability sampling design, ensuring national \nrepresentativeness.Our study population reflects the diverse demographic composition of the United States, with \n69.8% non-Hispanic white, 11.1% non-Hispanic black, 12.5% Hispanic (including 7.2% Mexican American and \n5.3% Other Hispanic), and 6.6% other racial/ethnic groups.For this research, NHANES data from 2006 to 2017 \nwere screened and processed. Figure 1 illustrates the data selection and inclusion/exclusion criteria, detailing the \nselection process of study participants. As shown in Fig. 1, the initial dataset included all individuals participating \nFig. 1. Flow diagram of study selection.\n \nScientific Reports |        (2024) 14:31995 2| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\nin NHANES .After applying multiple exclusion criteria—such as missing records on HBSO, incomplete PHQ-\n9 depression assessments, and missing general information like BMI, Income-to-Poverty Ratio (IPR), marital \nstatus, smoking, drinking, hypertension, and diabetes—the final sample comprised 10,780 women who met the \nstudy criteria.\nDefinition of depression\nDepressive symptoms were assessed using the PHQ-9, a widely utilized tool for screening depression. The PHQ-\n9 has been validated across diverse populations and demonstrates strong reliability and validity in previous \nNHANES studies19.The PHQ-9 consists of 9 items, each scored from 0 (not at all) to 3 (nearly every day), with \na total possible score ranging from 0 to 27. Depression was analyzed using three complementary approaches to \nprovide a comprehensive understanding of the relationship between surgical procedures and depression:\n 1.  Continuous Variable: The total PHQ-9 score was treated as a continuous variable to evaluate the severity of \ndepressive symptoms, allowing for detection of subtle variations in symptom severity.\n 2.  Binary Variable: Depression was defined based on a PHQ-9 score ≥ 10, indicating clinically significant de -\npressive symptoms20,21, which provides a clinically relevant cutoff point for identifying significant depres -\nsion.\n 3.   Ordered Categorical Variable: Depression was categorized based on PHQ-9 scores to capture the full spec-\ntrum of depression severity as follows to < 5: No depression;≥5 to < 10: Mild depression;≥10 to < 15: Mod-\nerate depression;≥15 to < 20: Moderately severe depression;≥20: Severe depressions22,23.\nDefinition of surgical procedures\nData were obtained from the reproductive health section of the NHANES questionnaire during the Mobile \nExamination Center (MEC) interview. Each participant’s hysterectomy status was determined by her response \nto the question, “Have you ever had a hysterectomy, which is the removal of the uterus?” (coded as RHD280). \nResponses to questions about bilateral oophorectomy (coded as RHQ305) and HBSO (hysterectomy plus \nbilateral salpingo-oophorectomy) were also recorded. Participants who answered “yes” to these questions were \nclassified accordingly. The primary independent variable, surgical type, was categorized into four groups based \non the interview data: None (no hysterectomy or oophorectomy), Hysterectomy, Bilateral Oophorectomy, and \nHBSO (Hysterectomy with Bilateral Salpingo-Oophorectomy).\nDetermination of covariates\nCovariates that might confound the results were selected based on literature and widely accepted academic \nstandards. Age was categorized into two groups (< 50 and ≥ 50 years) to account for the physiological transition \nof menopause, which significantly alters hormonal dynamics and potentially modifies the impact of surgical \ninterventions.These covariates included race/ethnicity (non-Hispanic white, non-Hispanic black, other Hispanic, \nand Mexican American/other), body mass index (BMI, kg/m², categorized into < 25.0, 25.0–29.9, and ≥ 30.0 kg/\nm²), marital status (widowed/divorced/separated/never married/living with a partner), education level ( less \nthan 9th Grade, 9-11th Grade, High School Grad/GED or Equivalent, Some College or AA degree, College \nGraduate or above), and the income-poverty ratio (IPR, ≤ 1.3,1.3< IPR ≤ 3.5, > 3.5 ) and systemic inflammation \nindex. Additionally, smoking status (ever smoked 100 cigarettes in a lifetime), alcohol consumption (less than \n12 drinks per year vs. 12 or more drinks per year), and self-reported chronic conditions such as hypertension, \ndiabetes, stroke, coronary heart disease, and cancer (yes/no), as well as hormone use (yes/no), were included. \nThese covariates were used as adjustment factors to minimize their potential influence on the relationship \nbetween the primary independent and dependent variables.\nStatistical analysis\nDescriptive statistics were presented as count (percentage) for categorical variables and mean ± standard \ndeviation (SD) for continuous variables.Between-group comparisons were conducted using Chi-square tests or \nFisher’s exact tests as appropriate.Multiple linear and binary logistic regression analyses were performed to assess \nthe association between surgical procedures and depressive symptoms. The following models were constructed: \nModel 1: Unadjusted crude model. Model 2: Adjusted for age and race/ethnicity to control these potential \nconfounding factors.Model 3: Further adjusted for multiple social and health factors, including education \nlevel, marital status, BMI, smoking and alcohol consumption status, chronic diseases, and hormone use, to \ncomprehensively evaluate the independent association between surgical procedures and depressive symptoms. \nFor the analysis of depression severity, multinomial logistic regression models were utilized to examine the risk \nof depression across different PHQ-9 score ranges for various surgical procedures. Results were presented as \nodds ratios (ORs) with corresponding 95% confidence intervals (CIs) and p-values.\nSubgroup analyses were conducted using stratified regression models to explore potential interactions between \ndifferent baseline characteristics (such as BMI, income-to-poverty ratio, smoking and alcohol consumption \nstatus, hypertension, diabetes, etc.) and the relationship between surgical procedures and depressive symptoms.\nAll statistical analyses were weighted according to NHANES guidelines using the MEC weights, specifically \n“WTMEC2YR/5, ” to ensure nationally representative estimates from the five NHANES cycles (2007–2016). Data \nanalysis and processing were performed using R software (version 4.2.2) and EmpowerStats software. A two-\ntailed p-value of < 0.05 was considered statistically significant.\nScientific Reports |        (2024) 14:31995 3| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\nResults\nComparative analysis of surgical methods, depressive symptoms, and baseline \ncharacteristics\nIn the study sample, Table 1 compares the depressive symptom group (PHQ-9 score ≥ 10) and the non-depressive \nsymptom group. The results indicate that the depressive symptom group had a significantly higher proportion of \nindividuals younger than 50 years (57.4% vs. 53.4%, P = 0.0353) and a lower proportion of married individuals \n(36.6% vs. 54.8%, P < 0.0001). Lower educational attainment and income were more prevalent in the depressive \nsymptom group, with significant differences observed ( P < 0.0001). Additionally, the depressive symptom \ngroup exhibited a higher smoking rate (59.1% vs. 37.2%, P < 0.0001), a greater proportion of individuals with \na BMI > 30, and a significantly increased prevalence of various chronic diseases (all P < 0.0001). Laboratory \ndata also revealed higher inflammatory markers in the depressive symptom group ( P < 0.0001). These findings \nsuggest significant differences in multiple characteristics between the depressive and non-depressive groups.\nTable 2 compares the baseline characteristics of participants across different surgical methods, including no \nsurgery, hysterectomy, bilateral oophorectomy, and HBSO. The results show that the proportion of individuals \nyounger than 50 years was highest in the no-surgery group (63.1%), whereas the proportion of individuals \naged ≥ 50 years significantly increased in the HBSO group (86.2%). The highest percentage of non-Hispanic \nwhites was observed in the HBSO group (81.0%). Significant differences in marital status, educational \nattainment, BMI, smoking rates, and the prevalence of chronic diseases were observed across the groups, with \nthe HBSO group exhibiting the highest prevalence of hypertension, diabetes, stroke, and coronary artery disease. \nThe PHQ-9 scores indicated the highest occurrence of depressive symptoms in the HBSO group (14.4%). These \nfindings highlight the potential associations between surgical methods and health and psychological status.\nAssociation between surgical methods and depression\nThe results from the linear regression and binary logistic regression analyses are presented in Table  3.\nHysterectomy was significantly associated with depressive symptoms across all three models. In the unadjusted \nModel 1, the β coefficient for the hysterectomy group was 0.79 (p = 0.001). This association remained significant \nin the age- and race-adjusted Model 2 (β = 1.00, p < 0.001) and the fully adjusted Model 3 (β = 0.73, p = 0.002).In \nthe binary logistic regression for PHQ-9 score ≥ 10, the hysterectomy group also showed significantly elevated \nodds ratios of 1.39 (p = 0.019), 1.57 (p = 0.003), and 1.35 (p = 0.047) in Models 1, 2, and 3, respectively.Compared \nto the non-surgical group, the HBSO group demonstrated a stronger association with higher PHQ-9 scores, \nwith β coefficients of 0.91, 1.26, and 0.98 across the three models (all p < 0.001). The HBSO group also had \nsignificantly higher odds of severe depressive symptoms, with ORs of 1.66, 2.04, and 1.77 in Models 1, 2, and 3 \n(all p < 0.001).In contrast, bilateral oophorectomy alone did not show a significant association with depressive \noutcomes in any of the models (p > 0.05). This may be due to the small sample size, leading to unstable results.\n Relationship between surgical methods and depression severity\nIn multiple logistic regression analyses (Table  4), the distribution of different severity levels of depression \nacross surgical methods was examined. The results showed that hysterectomy was significantly associated with \nhigher depressive scores, particularly in the PHQ-9 ≥ 20 group, with an OR of 2.10 ( P = 0.002), indicating a \nsignificant association between hysterectomy and severe depression. HBSO surgery also demonstrated a similar \npositive association, especially in the PHQ-9 score groups of 10 ≤ PHQ-9 < 15 and PHQ-9 ≥ 20, with ORs of \n1.38 (P = 0.004) and 1.78 (P = 0.017), respectively, further supporting the increased risk of depression following \nsurgery. The association between bilateral oophorectomy and depression severity was unclear. The small sample \nsize for bilateral oophorectomy ( n = 33) made it difficult to draw reliable conclusions about the relationship \nbetween this surgical method and the severity of depressive symptoms. Overall, the data support the association \nbetween hysterectomy, HBSO surgery, and the severity of depressive symptoms.\nSupplementary Tables 1 and Fig. 2 illustrate the predicted probabilities of depressive symptoms (as measured \nby PHQ-9 scores) across different surgical methods. Among patients who did not undergo surgery (None \ngroup), most exhibited mild depressive symptoms, with the highest probability of a PHQ-9 score < 5 (71.5%) \nand the lowest probability of severe depressive symptoms (PHQ-9 ≥ 20) at 1.1%. In contrast, patients undergoing \nhysterectomy (Hysterectomy group) and HBSO surgery were more likely to experience moderate to severe \ndepressive symptoms, with probabilities of PHQ-9 scores ≥ 20 at 2.0% and 1.8%, respectively. The relationship \nfor bilateral oophorectomy remained unclear.\nThe results of the ordinary logistic regression for depressive symptoms by different surgeries are presented \nin Supplementary Table 2.The analysis shows that hysterectomy (OR = 1.40, 95% CI: 1.23 to 1.59, adjusted \nP < 0.017) is significantly associated with increased odds of depressive symptoms. Similarly, the combined \nprocedure of hysterectomy and bilateral salpingo-oophorectomy (H-BSO) (OR = 1.31, 95% CI: 1.16 to 1.48, \nadjusted P < 0.017) also shows a significant association. However, bilateral oophorectomy alone (OR = 0.88, 95% \nCI: 0.42 to 1.87, adjusted P = 0.747) does not show a significant association with depressive symptoms, likely due \nto the limited sample size for this specific procedure.\nSubgroup analysis\nThe results of the subgroup analysis on the relationship between surgical methods and depression are presented \nin the form of a forest plot (Fig.  3). The analysis indicated significant differences in the association between \ndifferent surgical types (particularly HBSO) and depressive symptoms across various subgroups, a trend validated \nin the overall population. Specifically, hysterectomy was significantly associated with depressive symptoms, with \nthe addition of oophorectomy further enhancing this association, whereas bilateral oophorectomy alone was not \nsignificantly associated with depressive symptoms. Moreover, significant interactions with BMI, INR, smoking, \nand alcohol consumption were observed (all interactions P < 0.0001), suggesting that these factors played a \nScientific Reports |        (2024) 14:31995 4| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\nCharacteristic Total mean(95%CI)/%(95%CI) Depressive Group(PHQ-9 score < 10) Non-Depressive Group(PHQ-9 score ≥  10) P-value\nAge(years),% 0.256\n<50 5526 (51.3%) 4870 (51.1%) 656 (52.8%)\n≥50 5254 (48.7%) 4667 (48.9%) 587 (47.2%)\nRace,% < 0.001\nMexican American 1585 (14.7%) 1399 (14.7%) 186 (15.0%)\nOther Hispanic 1190 (11.0%) 1008 (10.6%) 182 (14.6%)\nNon-Hispanic White 4766 (44.2%) 4232 (44.4%) 534 (43.0%)\nNon-Hispanic Black 2194 (20.4%) 1928 (20.2%) 266 (21.4%)\nOther Race 1045 (9.7%) 970 (10.2%) 75 (6.0%)\nMarital status,% < 0.001\nMarried 5064 (47.0%) 4651 (48.8%) 413 (33.2%)\nWidowed 1173 (10.9%) 1041 (10.9%) 132 (10.6%)\nDivorced 1439 (13.3%) 1198 (12.6%) 241 (19.4%)\nSeparated 415 (3.8%) 316 (3.3%) 99 (8.0%)\nNever married 1908 (17.7%) 1666 (17.5%) 242 (19.5%)\nLiving with partner 781 (7.2%) 665 (7.0%) 116 (9.3%)\nEducation level,% < 0.001\nLess Than 9th Grade 966 (9.0%) 799 (8.4%) 167 (13.4%)\n9-11th Grade 1473 (13.7%) 1208 (12.7%) 265 (21.3%)\nHigh School Grad/GED or Equivalent 2308 (21.4%) 2037 (21.4%) 271 (21.8%)\nSome College or AA degree 3484 (32.3%) 3086 (32.4%) 398 (32.0%)\nCollege Graduate or above 2549 (23.6%) 2407 (25.2%) 142 (11.4%)\nIncome-to-poverty ratio,% < 0.001\n≤1.3 3664 (34.0%) 2975 (31.2%) 689 (55.4%)\n>1.3,≤3.5 3994 (37.1%) 3609 (37.8%) 385 (31.0%)\n>3.5 3122 (29.0%) 2953 (31.0%) 169 (13.6%)\nBMI(kg/m2),% < 0.001\n≤25 3248 (30.1%) 2974 (31.2%) 274 (22.0%)\n>25, ≤ 30 3049 (28.3%) 2761 (29.0%) 288 (23.2%)\n>30 4483 (41.6%) 3802 (39.9%) 681 (54.8%)\nSmoking,% < 0.001\n≥ 100 cigarettes in life 3965 (36.8%) 3272 (34.3%) 693 (55.8%)\n< 100 cigarettes in life 6815 (63.2%) 6265 (65.7%) 550 (44.2%)\nAlcohol, % < 0.001\n≥ 12 drinks/year 6651 (61.7%) 5824 (61.1%) 827 (66.5%)\n< 12 drinks/year 4129 (38.3%) 3713 (38.9%) 416 (33.5%)\nHypertension,% < 0.001\nyes 3971 (36.8%) 3383 (35.5%) 588 (47.3%)\nno 6809 (63.2%) 6154 (64.5%) 655 (52.7%)\nDiabetes,% < 0.001\nyes 1318 (12.2%) 1067 (11.2%) 251 (20.2%)\nno 9218 (85.5%) 8258 (86.6%) 960 (77.2%)\nborderline 244 (2.3%) 212 (2.2%) 32 (2.6%)\nStroke, % < 0.001\nyes 382 (3.5%) 295 (3.1%) 87 (7.0%)\nno 10,398 (96.5%) 9242 (96.9%) 1156 (93.0%)\nCoronary heart disease,% < 0.001\nyes 265 (2.5%) 208 (2.2%) 57 (4.6%)\nno 10,515 (97.5%) 9329 (97.8%) 1186 (95.4%)\nCancer, % 0.027\nyes 1067 (9.9%) 922 (9.7%) 145 (11.7%)\nno 9713 (90.1%) 8615 (90.3%) 1098 (88.3%)\nSurgery,% < 0.001\nNone 8348 (77.4%) 7453 (78.1%) 895 (72.0%)\nHysterectomy 1146 (10.6%) 987 (10.3%) 159 (12.8%)\nContinued\nScientific Reports |        (2024) 14:31995 5| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\nmoderating role in the relationship between surgery and depressive symptoms. In contrast, interactions with \nage, hypertension, and diabetes were insignificant (interaction P values: 0.572, 0.154, and 0.752, respectively).\nDiscussion\nThis study conducted an in-depth analysis of the association between various surgical procedures and depressive \nsymptoms. The results indicate that hysterectomy is associated with a higher prevalence of depressive symptoms, \neven after adjusting for multiple social and health factors. When hysterectomy is accompanied by bilateral \nsalpingo-oophorectomy (HBSO), the prevalence of depressive symptoms is further increased (OR = 1.77, \n95%CI 1.34–2.34, P = 0.0002). These findings underscore the psychological impact of hysterectomy on women, \nparticularly in cases where both ovaries are removed, leading to a higher prevalence of depressive symptoms.The \nresults of this study suggest the following key considerations for patients undergoing inevitable hysterectomy \nor hysterectomy with bilateral salpingo-oophorectomy (HBSO): Comprehensive preoperative mental health \nscreening, active psychological support during the perioperative period, referral to mental health professionals \nfor patients with prior risk factors, individualized psychological support plans, hormone replacement therapy to \nalleviate the psychological impact of bilateral salpingo-oophorectomy.\nThis outcome aligns with previous research. For example, one study reported that hysterectomy alone is \nalready associated with a higher prevalence of depression, but when combined with oophorectomy, the association \nwith depressive symptoms becomes even stronger16. Another study emphasized that women who undergo both \nhysterectomy and oophorectomy, especially younger women, have a significantly higher prevalence of developing \ndepression18. Additionally, some research suggests that bilateral oophorectomy may be associated with a reduced \nprevalence of postoperative depressive symptoms in women without baseline depressive symptoms24. However, \nthe small sample size in our bilateral oophorectomy group (n = 33) limits our ability to draw conclusive findings, \nand further investigation is needed to confirm these relationships.\nTo understand the reasons behind the heightened depressive symptoms in patients undergoing hysterectomy \nor HBSO, it is crucial first to consider the profound psychological impact of hysterectomy itself. The uterus, as \na vital reproductive organ, represents not just a physiological entity but also a significant psychological one. \nMany women may experience psychological stress related to the loss of reproductive ability or the symbolic \nrepresentation of femininity following a hysterectomy, which could directly cause or exacerbate depressive \nsymptoms25,26. Even in cases where the ovaries are not removed, this psychological stress alone may lead to the \nonset of depressive symptoms27.\nWhen an oophorectomy is performed in addition to a hysterectomy, the prevalence of depressive symptoms \nsignificantly increases. This may be attributed to the sharp decline in estrogen and progesterone levels following \noophorectomy, which further diminishes the neuroprotective effects within the female body28,29. The deficiency \nof estrogen may lead to neurotransmitter imbalances, such as those involving serotonin and dopamine, thereby \nbeing associated with an increased prevalence of depressive symptoms30. Furthermore, oophorectomy may affect \nthe stability of the HPA axis, exacerbating psychological health issues 29,30. Thus, the worsening of depressive \nsymptoms following HBSO may stem from a combination of psychological trauma from hysterectomy and \nhormonal imbalances after oophorectomy29.However, our study did not conduct a mediation analysis to confirm \nthese mechanistic pathways, and further research is needed to explore these potential mechanisms.\nAdditionally, the trauma of surgery and issues related to pain management during postoperative recovery may \ncontribute significantly to the increase in depressive symptoms. Patients who undergo hysterectomy may require \nlong-term management of pelvic pain, which has a bidirectional relationship with depressive symptoms 31,32. \nInadequate pain management or persistent chronic pain post-surgery could be a critical factor in the onset \nof depressive symptoms 33,34. Moreover, systemic inflammatory responses triggered by surgery could lead to \nelevated levels of inflammatory markers such as interleukin-6 (IL-6) and C-reactive protein (CRP), which have \nCharacteristic Total mean(95%CI)/%(95%CI) Depressive Group(PHQ-9 score < 10) Non-Depressive Group(PHQ-9 score ≥  10) P-value\nBilateral Oophorectomy 33 (0.3%) 31 (0.3%) 2 (0.2%)\nHBSO 1253 (11.6%) 1066 (11.2%) 187 (15.0%)\nFemale hormones,% 0.335\nyes 2043 (19.0%) 1803 (18.9%) 240 (19.3%)\nno 8699 (80.7%) 7697 (80.8%) 1002 (80.6%)\nunknown 31 (0.3%) 30 (0.3%) 1 (0.1%)\nLaboratory data, (109/L), mean\nPlatelet count 258.6 ± 67.0 257.7 ± 66.2 265.9 ± 72.3 < 0.001\nNeutrophils number 4.3 ± 1.7 4.2 ± 1.7 4.6 ± 2.0 < 0.001\nLymphocyte number 2.2 ± 0.9 2.2 ± 0.9 2.3 ± 0.9 < 0.001\nNLR 2.1 ± 1.1 2.1 ± 1.1 2.2 ± 1.2 0.008\nSII 547.4 ± 337.5 541.4 ± 324.5 593.7 ± 421.5 < 0.001\nTable 1. Baseline characteristics of participants with and without depressive symptoms. Data in the Table: \nMean ± SD / N (%), P-values: For continuous variables, P-values are derived using the Kruskal-Wallis rank sum \ntest. For count variables with theoretical cell counts < 10, P-values are derived using Fisher’s exact test. None: \nNo Hysterectomy or Oophorectomy, HBSO hysterectomy with bilateral. Salpingo-Oophorectomy.\n \nScientific Reports |        (2024) 14:31995 6| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\nCharacteristic\nTota l Surgery\nP-valueMean(95%CI)/%(95%CI) None Hysterectomy Bilateral Oophorectomy H-BSO\n8348 1146 33 1253\nAge(years),% < 0.001\n<50 5526 (51.3%) 5102 (61.1%) 276 (24.1%) 6 (18.2%) 142 (11.3%)\n≥50 5254 (48.7%) 3246 (38.9%) 870 (75.9%) 27 (81.8%) 1111 (88.7%)\nRace,% < 0.001\nMexican American 1585 (14.7%) 1320 (15.8%) 136 (11.9%) 4 (12.1%) 125 (10.0%)\nOther Hispanic 1190 (11.0%) 960 (11.5%) 129 (11.3%) 3 (9.1%) 98 (7.8%)\nNon-Hispanic White 4766 (44.2%) 3524 (42.2%) 521 (45.5%) 15 (45.5%) 706 (56.3%)\nNon-Hispanic Black 2194 (20.4%) 1631 (19.5%) 303 (26.4%) 5 (15.2%) 255 (20.4%)\nOther Race 1045 (9.7%) 913 (10.9%) 57 (5.0%) 6 (18.2%) 69 (5.5%)\nMarital status,% < 0.001\nMarried 5064 (47.0%) 3870 (46.4%) 557 (48.6%) 13 (39.4%) 624 (49.8%)\nWidowed 1173 (10.9%) 676 (8.1%) 209 (18.2%) 9 (27.3%) 279 (22.3%)\nDivorced 1439 (13.3%) 990 (11.9%) 231 (20.2%) 2 (6.1%) 216 (17.2%)\nSeparated 415 (3.8%) 335 (4.0%) 40 (3.5%) 2 (6.1%) 38 (3.0%)\nNever married 1908 (17.7%) 1775 (21.3%) 71 (6.2%) 3 (9.1%) 59 (4.7%)\nLiving with partner 781 (7.2%) 702 (8.4%) 38 (3.3%) 4 (12.1%) 37 (3.0%)\nEducation level,% < 0.001\nLess Than 9th Grade 966 (9.0%) 729 (8.7%) 112 (9.8%) 2 (6.1%) 123 (9.8%)\n9-11th Grade 1473 (13.7%) 1100 (13.2%) 199 (17.4%) 6 (18.2%) 168 (13.4%)\nHigh School Grad/GED or Equivalent 2308 (21.4%) 1692 (20.3%) 279 (24.3%) 8 (24.2%) 329 (26.3%)\nSome College or AA degree 3484 (32.3%) 2684 (32.2%) 378 (33.0%) 11 (33.3%) 411 (32.8%)\nCollege Graduate or above 2549 (23.6%) 2143 (25.7%) 178 (15.5%) 6 (18.2%) 222 (17.7%)\nIncome-to-poverty ratio,% < 0.001\n≤1.3 3664 (34.0%) 2916 (34.9%) 367 (32.0%) 15 (45.5%) 366 (29.2%)\n>1.3,≤3.5 3994 (37.1%) 2996 (35.9%) 466 (40.7%) 11 (33.3%) 521 (41.6%)\n>3.5 3122 (29.0%) 2436 (29.2%) 313 (27.3%) 7 (21.2%) 366 (29.2%)\nBMI(kg/m2),% < 0.001\n≤25 3248 (30.1%) 2721 (32.6%) 240 (20.9%) 12 (36.4%) 275 (21.9%)\n>25, ≤ 30 3049 (28.3%) 2305 (27.6%) 341 (29.8%) 7 (21.2%) 396 (31.6%)\n>30 4483 (41.6%) 3322 (39.8%) 565 (49.3%) 14 (42.4%) 582 (46.4%)\nSmoking, % < 0.001\n≥ 100 cigarettes in life 3965 (36.8%) 2907 (34.8%) 487 (42.5%) 17 (51.5%) 554 (44.2%)\n< 100 cigarettes in life 6815 (63.2%) 5441 (65.2%) 659 (57.5%) 16 (48.5%) 699 (55.8%)\nAlcohol, % < 0.001\n≥ 12 drinks/year 6651 (61.7%) 5272 (63.2%) 658 (57.4%) 21 (63.6%) 700 (55.9%)\n< 12 drinks/year 4129 (38.3%) 3076 (36.8%) 488 (42.6%) 12 (36.4%) 553 (44.1%)\nHypertension,% < 0.001\nyes 3971 (36.8%) 2482 (29.7%) 659 (57.5%) 16 (48.5%) 814 (65.0%)\nno 6809 (63.2%) 5866 (70.3%) 487 (42.5%) 17 (51.5%) 439 (35.0%)\nDiabetes,% < 0.001\nyes 1318 (12.2%) 805 (9.6%) 238 (20.8%) 5 (15.2%) 270 (21.5%)\nno 9218 (85.5%) 7392 (88.5%) 871 (76.0%) 28 (84.8%) 927 (74.0%)\nborderline 244 (2.3%) 151 (1.8%) 37 (3.2%) 0 (0.0%) 56 (4.5%)\nStroke, % < 0.001\nyes 382 (3.5%) 200 (2.4%) 72 (6.3%) 2 (6.1%) 108 (8.6%)\nno 10,398 (96.5%) 8148 (97.6%) 1074 (93.7%) 31 (93.9%) 1145 (91.4%)\nCoronary heart disease,% < 0.001\nyes 265 (2.5%) 132 (1.6%) 59 (5.1%) 3 (9.1%) 71 (5.7%)\nno 10,515 (97.5%) 8216 (98.4%) 1087 (94.9%) 30 (90.9%) 1182 (94.3%)\nCancer, % < 0.001\nyes 1067 (9.9%) 556 (6.7%) 209 (18.2%) 5 (15.2%) 297 (23.7%)\nno 9713 (90.1%) 7792 (93.3%) 937 (81.8%) 28 (84.8%) 956 (76.3%)\nFemale hormones,% < 0.001\nContinued\nScientific Reports |        (2024) 14:31995 7| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\nbeen shown to be significantly associated with depression 35–38.This study could not adjust for inflammatory \nmarkers such as interleukin-6 (IL-6) due to its absence in the nhance database. We used the SII as a surrogate \nmarker for systemic inflammation, but future research should further investigate the role of inflammation in \nthis process.\nThe modulatory effects of baseline characteristics such as high BMI, smoking, and alcohol consumption \nfurther support our conclusions. These factors are closely associated with the increase in depressive symptoms \nfollowing hysterectomy, indicating that metabolic, lifestyle, and behavioral factors play a critical role in the \nonset of postoperative depression. For example, high BMI may negatively impact mental health through social \nModel 1 Model 2 Model 3\nOutcome: PHQ-9 score, continuous N β (95%CI) P-value β (95%CI) P-value β (95%CI) P-value\nSurgery\nNone 8348 Ref. Ref. Ref.\nHysterectomy 1146 0.79 (0.36, 1.21) 0.001 1.00 (0.56, 1.44) < 0.001 0.73 (0.29, 1.17) 0.002\nBilateral Oophorectomy 33 0.20 (-1.73, 2.12) 0.842 0.53 (-1.36, 2.43) 0.583 0.12 (-1.52, 1.76) 0.885\nHBSO 1253 0.91 (0.54, 1.28) < 0.001 1.26 (0.87, 1.64) < 0.001 0.98 (0.62, 1.34) < 0.001\nOutcome: PHQ-9 score ≥  10, \nBinary OR (95%CI) P-value OR (95%CI) P-value OR (95%CI) P-value\nSurgery\nNone 8348 Ref. Ref. Ref.\nHysterectomy 1146 1.39 (1.06, 1.81) 0.019 1.57 (1.18, 2.09) 0.003 1.35 (1.01, 1.81) 0.047\nBilateral Oophorectomy 33 1.14 (0.17, 7.58) 0.896 1.36 (0.20, 9.45) 0.754 0.99 (0.13, 7.57) 0.992\nH-BSO 1253 1.66 (1.32, 2.09) < 0.001 2.04 (1.56, 2.67) < 0.001 1.77 (1.34, 2.34) < 0.001\nTable 3. Association between surgical methods and depressive symptoms. None: No Hysterectomy or \nOophorectomy, H-BSO hysterectomy with bilateral salpingo-oophorectomy. For PHQ-9 score as continuous \nestimated results were expressed as β (95% CI), and for PHQ-9 score ≥ 10, Binary estimated results were \nexpressed as OR (95% CI); β, Partial regression coefficient; OR, Odds Ratio; CI, Confidence Interval. Model 1: \nCovariates were not adjusted at all. Model 2: Adjusted for age and race. Model 3: Adjusted for age, race, marital \nstatus, education level, ratio of family income to poverty, BMI, smoking, alcohol consumption, hypertension, \ndiabetes, and systemic inflammation index (SII).\n \nCharacteristic\nTota l Surgery\nP-valueMean(95%CI)/%(95%CI) None Hysterectomy Bilateral Oophorectomy H-BSO\nyes 2043 (19.0%) 842 (10.1%) 372 (32.5%) 12 (36.4%) 817 (65.3%)\nno 8699 (80.7%) 7480 (89.7%) 771 (67.3%) 20 (60.6%) 428 (34.2%)\nunknown 31 (0.3%) 21 (0.3%) 2 (0.2%) 1 (3.0%) 7 (0.6%)\nLaboratory data, (109/L), mean\nPlatelet count 258.6 ± 67.0 260.5 ± 66.3 254.6 ± 67.0 269.5 ± 123.5 249.7 ± 68.2 < 0.001\nNeutrophils number 4.3 ± 1.7 4.3 ± 1.7 4.2 ± 1.7 4.4 ± 1.6 4.3 ± 1.8 0.118\nLymphocyte number 2.2 ± 0.9 2.2 ± 0.9 2.2 ± 1.2 2.3 ± 0.8 2.2 ± 0.9 0.377\nNLR 2.1 ± 1.1 2.1 ± 1.1 2.1 ± 1.0 2.1 ± 0.9 2.2 ± 1.3 0.392\nSII 547.4 ± 337.5 550.3 ± 333.0 530.2 ± 315.4 560.8 ± 333.8 543.1 ± 383.7 0.001\nPHQ-9 score, mean 3.8 ± 4.6 3.6 ± 4.5 4.4 ± 5.0 3.5 ± 3.1 4.4 ± 5.1 < 0.001\nPHQ-9 score Binary classification,% < 0.001\n< 10 9537 (88.5%) 7453 (89.3%) 987 (86.1%) 31 (93.9%) 1066 (85.1%)\n≥ 10 1243 (11.5%) 895 (10.7%) 159 (13.9%) 2 (6.1%) 187 (14.9%)\nPHQ-9 score Ordered multicategory classification ,% < 0.001\n< 5 7559 (70.1%) 5970 (71.5%) 734 (64.0%) 24 (72.7%) 831 (66.3%)\n≥ 5, < 10 1978 (18.3%) 1483 (17.8%) 253 (22.1%) 7 (21.2%) 235 (18.8%)\n≥ 10, < 15 759 (7.0%) 556 (6.7%) 94 (8.2%) 2 (6.1%) 107 (8.5%)\n≥ 15, < 20 350 (3.2%) 250 (3.0%) 42 (3.7%) 0 (0.0%) 58 (4.6%)\n≥ 20 134 (1.2%) 89 (1.1%) 23 (2.0%) 0 (0.0%) 22 (1.8%)\nTable 2. Baseline characteristics of participants with different surgery. Data in the Table: Mean ± SD / N \n(%), P-values: For continuous variables, P-values are derived using the Kruskal-Wallis rank sum test. For \ncount variables with theoretical cell counts < 10, P-values are derived using Fisher’s exact test. None: No \nHysterectomy or Oophorectomy, H-BSO hysterectomy with bilateral salpingo-oophorectomy.\n \nScientific Reports |        (2024) 14:31995 8| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\npressure and reduced self-esteem39,40. Patients who smoke or drink alcohol tend to have poorer postoperative \nrecovery, with these unhealthy behaviors further exacerbating depressive symptoms41.\nOur study population demonstrated notable racial and ethnic diversity, comprising 69.8% non-Hispanic \nwhite, 11.1% non-Hispanic black, 12.5% Hispanic (7.2% Mexican American and 5.3% Other Hispanic), and \n6.6% other racial/ethnic groups (including Asian Americans, Pacific Islanders, and Native Americans). This \ndistribution generally aligns with the U.S. demographic composition, though with some variations. Importantly, \nwe observed differences in surgery prevalence across ethnic groups, with 71.6% of non-Hispanic white women \nFig. 2. Predicted probabilities of depressive symptoms (PHQ-9 scores) across different surgical methods.\n \nPHQ-9 < 5 (Reference) 5 ≤  PHQ-9 < 10 10 ≤  PHQ-9 < 15 15 ≤  PHQ-9 < 20 PHQ-9 ≥  20\nOR (95%CI) P-value OR (95%CI) P-value OR (95%CI) P-value OR (95%CI) P-value OR (95%CI) P-value\n(Intercept) 1.0 (ref.) 0.25 (0.23, 0.26) < 0.001 0.09 (0.09, 0.10) < 0.001 0.04 (0.04, 0.05) < 0.001 0.01 (0.01, 0.02) < 0.001\nHysterectomy 1.0 (ref.) 1.39 (1.19, 1.62) < 0.001 1.38 (1.09, 1.73) 0.007 1.37 (0.98, 1.91) 0.068 2.10 (1.32, 3.35) 0.002\nBilateral Oophorectomy 1.0 (ref.) 1.17 (0.51, 2.73) 0.709 0.89 (0.21, 3.80) 0.880 0.00 (0.00, 0.00) < 0.001 0.01 (0.00, inf.) 0.959\nH-BSO 1.0 (ref.) 1.14 (0.97, 1.33) 0.102 1.38 (1.11, 1.72) 0.004 1.67 (1.24, 2.24) 0.001 1.78 (1.11, 2.85) 0.017\nTable 4. Multinomial Logistic Regression analysis of PHQ-9 scores (ordered multi-category classification). \nNone: No Hysterectomy or Oophorectomy, H-BSO hysterectomy with bilateral salpingo-oophorectomy.\n \nScientific Reports |        (2024) 14:31995 9| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\nhaving undergone hysterectomy compared to 14.1% of non-Hispanic black women and 9.4% of Hispanic women \n(5.0% Mexican American and 4.4% Other Hispanic). These variations might reflect underlying disparities in \nhealthcare access, cultural attitudes toward gynecological surgery, or differences in the prevalence of conditions \nrequiring these procedures42 .\nSeveral limitations of our study should be acknowledged. First, the cross-sectional nature of NHANES data \nprevents us from establishing causal relationships between surgical procedures and depression. Second, we lack \ninformation about the timing of surgery relative to depression onset, which could influence the interpretation \nFig. 3. Forest plot of subgroup analysis on the relationship between surgical methods and depression.\n \nScientific Reports |        (2024) 14:31995 10| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\nof our results. Third, while PHQ-9 is a validated screening tool, cultural differences in depression expression \nand reporting might affect its accuracy across different ethnic groups. Fourth, the relatively small sample size in \ncertain subgroups, particularly in the bilateral oophorectomy group (n = 33), limits our ability to draw definitive \nconclusions about some associations. Finally, we were unable to account for certain potential confounders such \nas pre-existing mental health conditions, detailed hormone replacement therapy regimens, or specific surgical \nindications, which might influence the relationship between surgery and depression.\nIn summary, this study elucidates the complex relationship between hysterectomy, HBSO, and the \ndevelopment of depressive symptoms. This relationship is influenced not only by hormonal changes but also \nby psychological trauma, chronic pain, inflammatory responses, and individual lifestyle and behavioral factors.\nFurthermore, the observed racial and ethnic differences in surgery prevalence and outcomes suggest the need \nfor culturally sensitive approaches in both research and clinical practice. Future research should continue to \nexplore the interplay of these factors and their long-term impact on women’s mental health to provide more \ncomprehensive evidence for clinical decision-making.\nConclusion\nHysterectomy is significantly associated with the onset of depressive symptoms, and the association is further \nincreased when accompanied by oophorectomy. However, the underlying mechanisms, including the roles of \npsychological trauma, chronic pain, and inflammatory responses, were not fully explored in this cross-sectional \nanalysis. These findings underscore the importance of preoperative mental health assessment and postoperative \npsychological support to reduce depression and enhance patient well-being.Future research is needed to \nelucidate the specific pathways linking these surgical procedures to mental health outcomes.\nData availability\nData availabilityThe datasets analyzed in this study are available in the National Health and Nutrition Examina-\ntion Survey (NHANES) repository and are openly accessible online (www.cdc.gov/nchs/nhanes/).\nReceived: 15 September 2024; Accepted: 16 December 2024\nReferences\n 1. Prince, M. et al. (2007). Lancet(Lancet (London, England)), 370 (9590): 859–877 .\n 2. Malhi, G. S. & Mann, J. J. Lancet(Lancet (London, England)), 392 (10161): 2299–2312. (2018).\n 3. Lancet Psychiatry(The lancet. Psychiatry), 9 (2): 137–150. (2022).\n 4. Kuehner, C. Lancet Psychiatry(the lancet. Psychiatry) 4 (2), 146–158 (2017).\n 5. Sassarini, D. J. Maturitas(Maturitas), 94: 149–154. (2016).\n 6. Perlis, R. H. et al. Depress. Anxiety(Depression Anxiety), 38 (10): 1026–1033. (2021).\n 7. Harvey, S. V ., Pfeiffer, R. M., Landy, R., Wentzensen, N. & Clarke, M. A. Am. J. Obstet. Gynecol(American J. Obstet. Gynecology), \n227 (4): 611. (2022).\n 8. Marsh, E. E., Wegienka, G. & Williams, D. R. JAMA(JAMA), 331 (17): 1492–1493. (2024).\n 9. Simms, K. T. et al. Gynecol. Oncol(Gynecologic Oncology), 158 (3): 710–718. (2020).\n 10. Broni, E. K. et al. Am. J. Obstet. Gynecol(American J. Obstet. Gynecology), 231 (4): 441–448. (2024).\n 11. Kim, H. et al. J. Affect. Disord(Journal Affect. Disorders), 285: 47–54. (2021).\n 12. Bräuner, E. V . et al. (2022). Menopause(Menopause (New Y ork, N.Y .)), 29 (3):276–283 .\n 13. Lin, K. Y . et al. J. Obstet. Gynecol(Taiwanese J. Obstet. Gynecology), 59 (6): 899–905. (2020).\n 14. Hickey, M. et al. Gynecol. Oncol(Gynecologic Oncology), 161 (2): 527–534. (2021).\n 15. Rocca, W . A. et al. (2018). Menopause(Menopause (New Y ork, N.Y .)), 25 (11): 1275–1285 .\n 16. Y ang, Y . et al. Sci. Rep(Scientific Reports), 13 (1): 9744. (2023).\n 17. Richards, D. H. Lancet(Lancet (London, England)), 2 (7826): 430–433. (1973).\n 18. Hassan, H. et al. Am. J. Obstet. Gynecol(American J. Obstet. Gynecology), 230 (1): 44–57. (2024).\n 19. Liang, J. et al. JMIR Public. Health Surveill(JMIR Public. Health Surveillance), 9: e45776. (2023).\n 20. Ettman, C. K. et al. JAMA Netw. Open(JAMA Netw. open), 3 (9): e2019686. (2020).\n 21. Costantini, L. et al. J. Affect. Disord(Journal Affect. Disorders), 279: 473–483. (2021).\n 22. de Oliveira, R. D. et al. JAMA Netw. Open(JAMA Netw. open), 6 (11): e2345138. (2023).\n 23. Kroenke, K., Spitzer, R. L. & Williams, J. B. J. Gen. Intern. Med(Journal Gen. Intern. Medicine), 16 (9): 606–613. (2001).\n 24. Rohl, J., Kjerulff, K., Langenberg, P . & Steege, J. Am. J. Obstet. Gynecol(American J. Obstet. Gynecology), 199 (1): 21–22. (2008).\n 25. Choi, H. G., Rhim, C. C., Y oon, J. Y . & Lee, S. W . (2020). Menopause(Menopause (New Y ork, N.Y .)), 27 (5): 543–549 .\n 26. Bahri, N. et al. Iran. Red Crescent Med. J(Iranian Red Crescent Med. Journal), 18 (1): e30493. (2016).\n 27. Ielmini, M. et al. J. Clin. Med(Journal Clin. Medicine), 13 (11). (2024).\n 28. Farquhar, C. M., Harvey, S. A., Yu, Y ., Sadler, L. & Stewart, A. W . Am. J. Obstet. Gynecol(American J. Obstet. Gynecology), 194 (3): \n711–717. (2006).\n 29. Wise, P . M., Dubal, D. B., Wilson, M. E., Rau, S. W . & Böttner, M. Endocrinology(Endocrinology), 142 (3): 969–973. (2001).\n 30. Herrera-Pérez, J. J. et al. Front. Psychiatry(Frontiers Psychiatry), 15: 1408878. (2024).\n 31. Poleshuck, E. L. et al. Pain med(Pain medicine (Mass)), 14 (5): 686–691. (2013).\n 32. Lirk, P ., Thiry, J., Bonnet, M. P ., Joshi, G. P . & Bonnet, F . Reg anesth pain med(Regional anesthesia and pain medicine), 44 (4): \n425–436. (2019).\n 33. Bougie, O., Blom, J., Zhou, G., Murji, A. & Thurston, J. Best pract res clin Obstet Gynaecol(best practice & research. Clin. Obstet. \nGynaecology). 85 (Pt B), 23–34 (2022).\n 34. Rawal, N. Best Pract Res Clin Anaesthesiol(best practice & research. Clin. Anaesthesiol.  37 (2), 123–132 (2023).\n 35. Ting, E. Y ., Y ang, A. C. & Tsai, S. J. Int. J. Mol. Sci(International J. Mol. Sciences), 21 (6). (2020).\n 36. Beurel, E., Toups, M. & Nemeroff, C. B. Neuron(Neuron), 107 (2): 234–256. (2020).\n 37. Harsanyi, S., Kupcova, I., Danisovic, L. & Klein, M. Int. J. Mol. Sci(International J. Mol. Sciences), 24 (1). (2022).\n 38. Kofod, J., Elfving, B., Nielsen, E. H., Mors, O. & Köhler-Forsberg, O. Eur. Neuropsychopharmacol(European Neuropsychopharmacology: \nJ. Eur. Coll. Neuropsychopharmacology), 54 :116–125. (2022).\n 39. Karageorgiou, V . et al. BMC Med(BMC Medicine), 21 (1): 355. (2023).\n 40. Scott, K. M., Mcgee, M. A., Wells, J. E. & Oakley, B. M. J. Psychosom. Res(Journal Psychosom. Research), 64 (1): 97–105. (2008).\nScientific Reports |        (2024) 14:31995 11| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/\n\n 41. Y onek, J. C., Meacham, M. C., Shumway, M., Tolou-Shams, M. & Satre, D. D. Drug alcohol depend(Drug and alcohol dependence), \n227: 108922. (2021).\n 42. Jacoby, V . L., Fujimoto, V . Y ., Giudice, L. C., Kuppermann, M. & Washington, A. E. Am. J. Obstet. Gynecol(American J. Obstet. \nGynecology), 202 (6): 514–521. (2010).\nAcknowledgements\nWe would like to acknowledge all the participants.\nAuthor contributions\nC.X.: Writing – original draft, Visualization, Validation, Conceptualization. G. Z.:Software, Methodology, For-\nmal analysis, Visualization, Conceptualization.W . Y .: Writing – review & editing, Supervision, Methodology, \nFormal analysis.Y .Z. : Writing – review & editing, Software, Supervision, Project administration.\nFunding\nNone.\nDeclarations\nCompeting interests\nThe authors declare no competing interests.\nEthics approval and consent to participate\nThe NHANES project was approved by the National Ethical Review Board for Health Statistics Research, \nand the data are publicly available on the project website (https://wwwn.cdc.gov/nchs/nhanes). The patients’ \ninformation was anonymized, and thus, the need for informed consent was waived for this study.\nConsent for publication\nAll authors approved the final manuscript and the submission to this journal.\nAdditional information\nSupplementary Information The online version contains supplementary material available at  h t t p s : / / d o i . o r g / 1 \n0 . 1 0 3 8 / s 4 1 5 9 8 - 0 2 4 - 8 3 6 7 5 - y     .  \nCorrespondence and requests for materials should be addressed to Y .Z.\nReprints and permissions information is available at www.nature.com/reprints.\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and \ninstitutional affiliations.\nOpen Access  This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives \n4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in \nany medium or format, as long as you give appropriate credit to the original author(s) and the source, provide \na link to the Creative Commons licence, and indicate if you modified the licensed material. Y ou do not have \npermission under this licence to share adapted material derived from this article or parts of it. The images or \nother third party material in this article are included in the article’s Creative Commons licence, unless indicated \notherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence \nand your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to \nobtain permission directly from the copyright holder. To view a copy of this licence, visit  h t t p : / / c r e a t i v e c o m m o \nn s . o r g / l i c e n s e s / b y - n c - n d / 4 . 0 /     .  \n© The Author(s) 2024 \nScientific Reports |        (2024) 14:31995 12| https://doi.org/10.1038/s41598-024-83675-y\nwww.nature.com/scientificreports/","source_license":"CC0","license_restricted":false}