Skip to main content
BMC Medicine logoLink to BMC Medicine
. 2025 Jul 15;23:424. doi: 10.1186/s12916-025-04248-y

A population-based observational study using statistical modeling to assess the association between depressive symptom severity and sleep disorders in postmenopausal women

Ying Cui 1, Huimin Du 2,✉
PMCID: PMC12261605  PMID: 40660163

Abstract

Background

This study aimed to investigate the association between depressive symptom severity and sleep disorders in postmenopausal women.

Methods

This observational study included data from 4808 postmenopausal women derived from a nationally representative sample in the USA. Depressive symptom severity was assessed using the Patient Health Questionnaire-9, while sleep disorders were identified based on self-reported physician diagnoses. Weighted multivariable logistic regression models were used to analyze the association between depressive symptom severity and sleep disorders, adjusting for potential confounders. Restricted cubic splines were applied to evaluate possible nonlinear relationships, and subgroup analyses were conducted across key sociodemographic, health, and behavioral factors. Additionally, Lasso regression and logistic regression were used to identify the most influential predictors of sleep disorders. Supplementary and sensitivity analyses were performed using alternative sleep outcomes and modified depressive symptom scales to test robustness and item-level overlap.

Results

Depressive symptom severity was positively associated with sleep disorders, demonstrating a dose–response relationship (P for trend < 0.001). Each unit increase in depressive symptom score was associated with a 10% higher risk of sleep disorders (OR = 1.10, 95% CI: 1.07–1.13). RCS analysis confirmed a linear association (P for nonlinear = 0.4696). Subgroup analyses identified CVD as a significant effect modifier (P for interaction = 0.019), with a stronger association in individuals with CVD (OR = 1.11, 95% CI: 1.09–1.13) compared to those without (OR = 1.07, 95% CI: 1.03–1.11). Lasso and logistic regression analyses consistently ranked depressive symptoms as the strongest predictor of sleep disorders. The association remained robust and specific across both supplementary outcomes and sensitivity models using modified depressive symptom scales.

Conclusions

This study demonstrated a linear dose–response association between depressive symptom severity and sleep disorders in postmenopausal women, which was further amplified among individuals with CVD. Depressive symptoms were identified as the most critical predictor, underscoring the importance of mental health in managing sleep health. These findings highlight the need for integrated interventions combining mental health screening, lifestyle modifications, and community-based care approaches to mitigate the dual burden of depressive symptoms and sleep disorders in this vulnerable population.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12916-025-04248-y.

Keywords: Depressive symptoms, Sleep disorders, Postmenopausal women, Cardiovascular disease, Dose–response relationship

Background

Sleep disorders represent a significant health concern among postmenopausal women, often becoming more severe and persistent than those experienced during premenopausal years [1, 2]. While sleep disturbances can occur at any life stage, women undergoing menopause are particularly vulnerable due to the physiological and hormonal changes associated with this transition. Prior to menopause, estrogen and progesterone help regulate sleep–wake cycles, promote sleep stability, and reduce nighttime awakenings [3]. However, the decline in these hormones after menopause often leads to disruptions in circadian rhythms, increased sleep fragmentation, and diminished overall sleep quality [4, 5]. Consequently, many postmenopausal women experience difficulty initiating and maintaining sleep, as well as nonrestorative sleep, adversely affecting daily functioning and overall well-being [6].

The underlying factors contributing to sleep disorders in postmenopausal women are complex and multifactorial, involving hormonal fluctuations [7], autonomic nervous system dysregulation [8], and metabolic alterations [6, 9]. Additionally, lifestyle and health-related factors, including hot flashes [10], obesity [11], sedentary behavior [12], and comorbidities such as hypertension [13] and diabetes [14], further elevate the risk of sleep disturbances. Societal stigma and misconceptions surrounding menopause can also negatively impact emotional health [15], potentially intensifying psychological distress and worsening sleep outcomes.

Depressive symptoms represent another major health concern in this population [16]. Declining estrogen and progesterone levels during menopause contribute to increased susceptibility to depressive symptoms through disruptions in key neurotransmitter systems [17], particularly serotonin [18] and gamma-aminobutyric acid (GABA) [19]. Moreover, menopause-related psychosocial stressors, such as negative societal attitudes, insufficient social support, and life transitions, can exacerbate emotional distress [20]. Given the frequent coexistence of depressive symptoms and sleep disorders, and their substantial combined impact on quality of life, understanding the relationship between depressive symptom severity and sleep disturbances among postmenopausal women may yield critical insights for clinical management and intervention strategies, ultimately improving health outcomes in this population.

This study aimed to investigate the association between depressive symptom severity and physician-diagnosed sleep disorders using data from a nationally representative US population. Additionally, subgroup analyses were conducted to explore whether this association varied by demographic and clinical characteristics, providing deeper insights into potential differences across subpopulations. Identifying these subgroup variations may inform targeted interventions and improve clinical management strategies tailored specifically for postmenopausal women at elevated risk for depressive symptoms and sleep disturbances.

Methods

Study design and participants

The National Health and Nutrition Examination Survey (NHANES) is a program implemented by the National Center for Health Statistics (NCHS) to evaluate the health and nutritional status of the US population. It combines interviews, physical examinations, laboratory tests, and detailed dietary assessments to obtain comprehensive health data across all age groups. Annually, approximately 5000 individuals are selected through a randomized scientific sampling process, ensuring that the sample accurately reflects the diverse US population. The research protocol was sanctioned by the NCHS Research Ethics Review Board, and every participant provided written informed consent [21].

This study utilized data from five consecutive NHANES cycles (2005–2006, 2007–2008, 2009–2010, 2011–2012, and 2013–2014), selected because these cycles provided complete information on doctor-diagnosed sleep disorders [22]. The initial dataset comprised 50,965 participants. To focus on postmenopausal women, males (N = 25,260) and premenopausal females (N = 19,511) were excluded. Further exclusions were made for participants with missing sleep disorder data (N = 8), missing PHQ-9 score data (N = 112), and incomplete covariate data (N = 1266), yielding a final analytical sample of 4808 participants. A detailed flowchart illustrating the participant selection process is presented in Fig. 1.

Fig. 1.

Fig. 1

Flowchart of the study participants

Identification of postmenopausal women

In this study, postmenopausal status was defined using a two-step approach based on items from the NHANES reproductive health questionnaire. First, women who answered “No” (code 2) to the item “Had regular periods in the past 12 months” (RHQ031) were selected. Since this question reflects the regularity of menstruation rather than complete cessation, we applied an additional criterion to improve specificity. Specifically, participants were further assessed based on their response to the follow-up question (RHD043: “What is the reason that you have not had regular periods?”). This item included several options such as pregnancy, breastfeeding, hysterectomy, menopause/change of life, medical conditions or treatments, uncertainty (“don’t know”), and other causes. To ensure a precise classification, we included only women who selected either “Hysterectomy” or “Menopause/Change of life” as the reason for their amenorrhea. Women who reported temporary, pathological, or uncertain reasons—such as pregnancy, breastfeeding, medical treatments, or other unspecified causes—were excluded. This approach minimized potential misclassification and allowed for a more accurate identification of postmenopausal women.

Assessment of depressive symptoms

Depressive symptoms were assessed using the Patient Health Questionnaire-9 (PHQ-9), a widely validated screening tool that is included in the NHANES Mental Health—Depression Screener module. This instrument, which aligns with the nine DSM-IV criteria for depression, captures the frequency of symptoms experienced over the preceding 2 weeks [23]. Each item is scored from 0 (“Not at all”) to 3 (“Nearly every day”), generating a composite score (variable prefix DPQ) that ranges from 0 to 27. The overall scores are categorized into minimal (0–4), mild (5–9), moderate (10–14), moderately severe (15–19), and severe (20–27) levels of depressive symptoms [24]. A threshold score of 10 or above is commonly used in epidemiological studies to indicate elevated depressive symptom levels, with reported sensitivity and specificity of approximately 88% [25]. In this study, PHQ-9 was used solely to quantify depressive symptom severity and was not applied to establish a clinical diagnosis.

Assessment of sleep disorders

Sleep disorders were identified using data from the NHANES Sleep Disorders module. The assessment is based on item SLQ060, which inquires if a doctor or other healthcare professional has ever informed the participant of having a sleep disorder. The precise wording of the question is “Have you ever been told by a doctor or other health professional that you have/he has a sleep disorder?” This item, offering a simple “Yes” or “No” response, is applicable to both male and female subjects aged 16 to 150 years. This item does not differentiate among specific types of sleep disorders (e.g., insomnia, obstructive sleep apnea, narcolepsy) and thus reflects a broad spectrum of clinically diagnosed sleep-related conditions.

Covariates

In accordance with the Andersen Behavioral Model—which is widely recognized for its ability to integrate diverse determinants of health [26]—the covariates were classified into three distinct categories: predisposing factors, enabling factors, and need factors. Predisposing factors included demographic and social characteristics such as age, race, education level, and marital status. Enabling factors comprised socioeconomic and living condition indicators, specifically the poverty income ratio (PIR) and housing characteristics. Need factors encompassed clinical and biological parameters, including body mass index (kg/m2), waist circumference (cm), weight (kg), total cholesterol (mg/dL), HDL-C (mg/dL), uric acid (mg/dL), as well as health behaviors and conditions like smoking, hypertension, diabetes, cardiovascular disease (CVD), cancer, and prescription medication use.

Health behaviors and medical conditions were defined using self-reported data from the NHANES health questionnaires. Smoking status was determined using item SMQ020; participants were classified as smokers if they reported having smoked at least 100 cigarettes in their lifetime. This 100-cigarette cutoff is a widely accepted threshold established by the US CDC to operationalize smoking history in epidemiological studies. All medical conditions were defined based on self-reported physician diagnoses. Specifically, hypertension was defined as a physician-diagnosed history of high blood pressure (BPQ020 = “Yes”), and diabetes as a physician-diagnosed condition, excluding gestational diabetes (DIQ010 = “Yes”). CVD was defined as a composite indicator, based on a self-reported physician diagnosis of any of the following: congestive heart failure (MCQ160B), coronary heart disease (MCQ160C), angina pectoris (MCQ160D), myocardial infarction (MCQ160E), or stroke (MCQ160F). Although hypertension is a well-established risk factor for cardiovascular disease, it was treated as an independent covariate to enable the separate evaluation of its association with the outcome. The composite CVD variable did not include hypertension. This modeling strategy allows for a more precise estimation of the individual contributions of hypertension and CVD, given their potentially differential associations with both depressive symptoms and sleep disorders. Cancer status was defined as a self-reported physician diagnosis of any malignancy or cancer (MCQ220 = “Yes”).

Prescription medication use was included as a binary covariate to control for potential pharmacological confounding. Participants were first identified using the RXDUSE variable, which indicates whether they had taken any prescription medications in the past month. Among those who responded “yes,” all reported medications were cross-referenced with the Multum Lexicon Therapeutic Classification Scheme. Individuals were classified as “prescription medication users” if they had taken at least one drug belonging to any of the following four therapeutic classes: sedatives and hypnotics (e.g., benzodiazepines, barbiturates), antidepressants (e.g., SSRIs, tricyclics), hormonal agents (e.g., estrogens, corticosteroids), and cardiovascular agents (e.g., beta-blockers, ACE inhibitors, calcium channel blockers). These drug classes corresponded to Multum first-level category codes 57 (central nervous system agents, with sleep-related medications defined under second-level code 67), 242 (psychotherapeutic agents), 97 (hormones/hormone modifiers), and 40 (cardiovascular agents). Participants who used any of these medications were coded as “Yes” for this covariate, while all others—including those who reported no prescription medication use—were coded as “No.” This composite variable was created to minimize model complexity, avoid potential multicollinearity due to overlapping medication use across classes, and control for the general pharmacological exposure that may affect both depressive symptoms and sleep outcomes.

Statistical analysis

All data were weighted according to NHANES sampling guidelines to ensure national representativeness, with adjustments made for the complex survey design to improve precision and generalizability. Initially, weighted Rao–Scott second-order corrected chi-square tests were used to compare categorical characteristics between participants with and without sleep disorders, while weighted Mann–Whitney tests were applied to analyze continuous variables.

Depressive symptom severity was modeled using two complementary approaches to enhance statistical robustness and clinical interpretability. First, the total PHQ-9 score (range: 0–27) was treated as a continuous variable to assess trends across the full score distribution. Second, the score was categorized based on established clinical cutoff points into five levels: minimal (0–4), mild (5–9), moderate (10–14), moderately severe (15–19), and severe (20–27). Both forms were analyzed in parallel to evaluate the consistency and clinical relevance of the associations observed.

Subsequently, multivariable logistic regression models were employed to examine the association between depressive symptom severity and sleep disorders, with results expressed as odds ratios (ORs) and 95% confidence intervals (CIs). Three models were developed to adjust for potential confounders: Model 1 was unadjusted; Model 2 controlled for predisposing and enabling factors (age, race, education level, marital status, poverty income ratio, and housing characteristics); and Model 3 further adjusted for need factors (BMI, waist circumference, weight, total cholesterol, HDL-C, uric acid, smoking status, hypertension, diabetes, CVD, cancer, and prescription medication use). Collinearity was assessed using variance inflation factors (VIFs), all of which were below 5, indicating no significant collinearity issues.

Restricted cubic splines (RCS) were applied in Model 3 to assess the potential nonlinearity in the relationship between depressive symptom severity and the risk of sleep disorders, with P for nonlinear used as an indicator. To evaluate the robustness of the findings, subgroup analyses were conducted based on age, race, education level, marital status, housing characteristics, BMI, smoking status, hypertension, diabetes, cardiovascular disease, cancer, and prescription medication use, with interaction P values reported. Additionally, variable importance was assessed using all baseline variables through a combined approach of least absolute shrinkage and selection operator (LASSO) and logistic regression, with the top five factors highlighted in the results.

To further examine the robustness of the primary findings, we conducted supplementary analyses using two alternative sleep-related outcomes available in NHANES: (1) self-reported history of sleep difficulties (SLQ050), treated as a binary outcome, and (2) self-reported sleep duration (SLD010H), treated as a continuous outcome. These supplementary outcomes were analyzed using the same multivariable modeling framework and covariate adjustments as the primary analysis.

Finally, sensitivity analyses were conducted to address concerns regarding the potential inflation of depressive symptom scores in menopausal women due to symptom overlap with physiological changes. Specifically, we constructed two modified depressive symptom scales: (1) PHQ-8, which excluded the sleep-related item (item 3), and (2) PHQ-7, which excluded both the sleep-related item (item 3) and the fatigue item (item 4)—two symptoms that may reflect menopausal transitions rather than depressive pathology. Each modified score was analyzed using the same multivariable framework as the primary analysis (Models 1–3). To further examine whether these symptoms exert independent effects, we also constructed models that included PHQ-8 or PHQ-7 alongside the excluded item(s) to test for their additive associations with sleep disorder diagnosis.

All statistical analyses were conducted using R (version 2024.12.0 + 467; Posit Software, PBC), and a two-sided P value of < 0.05 was considered statistically significant.

Results

Sample characteristics of the participants

Table 1 outlines the unweighted demographic profile of the cohort. In total, 4808 individuals were analyzed, of whom 452 (9.40%) reported having a physician-diagnosed sleep disorder. Among these participants, 317 (70.13%) had a PHQ-9 score below 10, indicating lower levels of depressive symptoms, while 135 (29.87%) had scores of 10 or higher, reflecting elevated symptom severity. In contrast, among the 4356 individuals without reported sleep disorder symptoms, 3960 (90.91%) recorded PHQ scores under 10, and 396 (9.09%) scored 10 or above. These unweighted statistics offer a straightforward depiction of the sample’s characteristics.

Table 1.

Unweighted characteristics of the study population

Characteristic Total
N = 4808
Without sleep disorders
N = 4356
With sleep disorders
N = 452
PHQ < 10
N = 3960
PHQ ≥ 10
N = 396
PHQ < 10
N = 317
PHQ ≥ 10
N = 135
Age
    Mean (SD) 63.5 (10.8) 64.2 (10.7) 59.8 (11.0) 62.3 (10.4) 57.4 (9.72)
    Median [Min, Max] 43.00 [20.0, 85.0] 64.0 [20.0, 85.0] 60.0 [27.0, 85.0] 62.0 [30.0, 85.0] 58.0 [34.0, 80.0]
Race
    Mexican American 611 (12.7%) 495 (12.5%) 66 (16.7%) 28 (8.8%) 22 (16.3%)
    Non-Hispanic Black 410 (8.5%) 316 (8.0%) 45 (11.4%) 29 (9.1%) 21 (15.6%)
    Non-Hispanic White 2511 (52.2%) 2089 (52.8%) 184 (46.5%) 180 (56.8%) 58 (43.0%)
    Other Hispanic 985 (20.5%) 809 (20.4%) 85 (21.5%) 63 (19.9%) 28 (20.7%)
    Other race 291 (6.1%) 252 (6.4%) 16 (4.0%) 17 (5.4%) 6 (4.4%)
Education levels
    High school or below 2278 (52.3%) 2021 (51.0%) 257 (64.9%) 161 (50.8%) 73 (54.1%)
    Some college or higher 2078 (47.7%) 1939 (49.0%) 139 (35.1%) 156 (49.2%) 62 (45.9%)
Marital status
    Married/living with partner 2228 (51.1%) 2075 (52.4%) 153 (38.6%) 152 (47.9%) 56 (41.5%)
    Widowed/divorced/separated 1854 (42.6%) 1640 (41.4%) 214 (54.0%) 139 (43.8%) 69 (51.1%)
    Naver married 247 (6.3%) 245 (6.2%) 29 (7.3%) 26 (8.2%) 10 (7.4%)
PIR
    Mean (SD) 2.60 (1.59) 2.69 (1.59) 1.74 (1.36) 2.50 (1.59) 1.61 (1.28)
    Median [Min, Max] 2.22 [0, 5.00] 2.31 [0, 5.00] 1.26 [0, 5.00] 2.04 [0, 5.00] 1.16 [0, 5.00]
Housing characteristics
    Owned or being bought 3271 (75.1%) 3025 (76.4%) 246 (62.1%) 233 (73.3%) 68 (50.4%)
    Rented 988 (22.7%) 847 (21.4%) 141 (35.6%) 76 (24.0%) 61 (45.2%)
    Other arrangement 97 (2.2%) 88 (2.2%) 9 (2.3%) 8 (2.5%) 6 (4.4%)
BMI (kg/m2)
    Mean (SD) 29.8 (6.68) 29.2 (6.57) 31.1 (7.55) 33.2 (8.32) 33.6 (8.32)
    Median [Min, Max] 28.4 [13.2, 67.0] 28.3 [13.2, 67.0] 30.0 [15.2, 64.0] 32.1 [17.2, 82.1] 32.6 [17.1, 61.7]
Waist circumference (cm)
    Mean (SD) 98.6 (14.7) 98.2 (14.5) 102 (15.8) 107 (17.0) 108 (16.2)
    Median [Min, Max] 97.7 [61.5, 157] 97.4 [61.5, 157] 102 [64.0, 149] 107 [66.9, 162] 107 [71.8, 154]
    Weight (kg)
    Mean (SD) 75.1 (18.4) 74.7 (18.2) 78.9 (19.7) 85.8 (22.2) 86.1 (22.3)
    Median [Min, Max] 72.1 [33.2, 174] 71.8 [33.2, 174] 76.5 [38.2, 146] 84.7 [39.8, 188] 82.0 [46.5, 163]
Total cholesterol (mg/dL)
    Mean (SD) 208 (41.9) 208 (41.4) 207 (45.9) 201 (41.7) 206 (46.9)
    Median [Min, Max] 206 [89.0, 463] 206 [89.0, 463] 205 [93.0, 383] 201 [116, 460] 202 [124, 440]
    HDL-C (mg/dL)
    Mean (SD) 58.8 (16.5) 59.3 (16.6) 54.4 (15.1) 55.6 (15.4) 53.4 (14.8)
    Median [Min, Max] 57.0 [17.0, 188] 57.0 [17.0, 188] 53.0 [20.0, 119] 54.0 [11.0, 109] 51.0 [23.0, 108]
Uric acid (mg/dL)
    Mean (SD) 5.22 (1.36) 5.22 (1.35) 5.26 (1.47) 5.53 (1.45) 5.02 (1.47)
    Median [Min, Max] 5.10 [1.10, 11.5] 5.10 [1.10, 11.5] 5.00 [1.90, 10.3] 5.40 [2.40, 10.9] 4.90 [1.50, 11.2]
Smoking
    Yes 1784 (41.0%) 1567 (39.6%) 217 (54.8%) 136 (42.9%) 84 (62.2%)
    No 2572 (59.0%) 2393 (60.4%) 179 (45.2%) 181 (57.1%) 51 (37.8%)
Hypertension
    Yes 2373 (54.5%) 2121 (53.6%) 252 (63.6%) 213 (67.2%) 89 (65.9%)
    No 1983 (45.5%) 1839 (46.4%) 144 (36.4%) 104 (32.8%) 46 (34.1%)
Diabetes
    Yes 806 (18.5%) 694 (17.5%) 112 (28.3%) 77 (24.3%) 48 (35.6%)
    No 3550 (81.5%) 3266 (82.5%) 284 (71.7%) 240 (75.7%) 87 (64.4%)
CVD
    Yes 604 (13.9%) 498 (12.6%) 107 (26.8%) 58 (18.3%) 35 (25.9%)
    No 3752 (86.1%) 3462 (87.4%) 290 (73.2%) 259 (81.7%) 100 (74.1%)
Cancer
    Yes 693 (15.9%) 619 (15.6%) 74 (18.7%) 55 (17.4%) 22 (16.3%)
    No 3663 (84.1%) 3341 (84.4%) 322 (81.3%) 262 (82.6%) 113 (83.7%)
Prescription medication use
    Yes 2966 (68.1%) 2667 (69.3%) 229 (75.5%) 273 (86.1%) 123 (91.1%)
    No 1390 (31.9%) 1293 (32.7%) 97 (24.5%) 44 (13.9%) 12 (8.9%)
Depressive Symptom Score (PHQ-9)
    Mean (SD) 3.34 (4.26) 2.27 (2.46) 14.1 (3.57) 3.40 (2.73) 14.8 (4.11)
    Median [Min, Max] 2.00 [0.00, 27.0] 1.00 [0, 9.00] 13.0 [10.0, 27.0] 3.00 [0, 9.00] 14.0 [10.0, 27.0]

PIR poverty-to-income ratio, BMI body mass index, WC waist circumference, HDL-C high-density lipoprotein cholesterol, CVD cardiovascular disease

Table 2 presents the weighted characteristics of the study population. The average age of participants was 62.00 ± 10.98 years, with a mean BMI of 29.38 ± 6.98 kg/m2. Individuals with sleep disorders exhibited significantly higher depressive symptom scores compared to those without physician-diagnosed sleep disorders (6.16 ± 5.62 vs. 3.10 ± 3.97, P < 0.001). Furthermore, depressive symptom severity showed a distinct distribution between the two groups (P < 0.001). Specifically, the proportion of participants with minimal depressive symptoms (PHQ-9 score < 5) was lower among those with sleep disorders (52.05%) than those without (73.27%). In contrast, the prevalence of moderate (15.26% vs. 4.95%), moderately severe (7.16% vs. 2.01%), and severe depressive symptoms (3.05% vs. 0.56%) was notably higher among individuals with physician-diagnosed sleep disorders. In addition, significant differences were observed between the two groups in terms of age, PIR, BMI, waist circumference, weight, HDL-C, uric acid levels, smoking status, hypertension, diabetes, CVD, and prescription medication use (Pall < 0.05).

Table 2.

Sample characteristics and comparison between participants with and without physician-diagnosed sleep disorders (weighted)

Characteristic Na Overall
N = 38,423,426b
Sleep disorders P valuec
Without
N = 34,775,227b
With
N = 3,648,198b
Age 4808 62.00 ± (10.98) 62.22 ± (11.03) 59.90 ± (10.33)  < 0.001
Race 4808 0.317
    Mexican American 611 (4.46%) 561 (4.51%) 50 (4.03%)
    Other Hispanic 410 (3.34%) 360 (3.23%) 50 (4.33%)
    Non-Hispanic White 2511 (77.98%) 2273 (77.96%) 238 (78.22%)
    Non-Hispanic Black 985 (9.76%) 894 (9.72%) 91 (10.15%)
    Other race 291 (4.46%) 268 (4.59%) 23 (3.26%)
Education level 4808 0.983
    High school or below 2512 (43.67%) 2278 (43.67%) 234 (43.60%)
    Some college or higher 2296 (56.33%) 2078 (56.33%) 218 (56.40%)
Marital status 4808 0.748
    Married/living with partner 2436 (58.86%) 2228 (59.04%) 208 (57.20%)
    Widowed/divorced/separated 2062 (35.98%) 1854 (35.87%) 208 (36.98%)
    Naver married 310 (5.16%) 274 (5.09%) 36 (5.81%)
PIR 4808 3.07 ± (1.60) 3.10 ± (1.59) 2.73 ± (1.63) 0.002
Housing characteristics 4808 0.098
    Owned or being bought 3572 (81.61%) 3271 (82.03%) 301 (77.65%)
    Rented 1125 (16.55%) 988 (16.19%) 137 (19.92%)
    Other arrangement 111 (1.84%) 97 (1.78%) 14 (2.43%)
BMI (kg/m2) 4808 29.38 ± (6.98) 29.02 ± (6.77) 32.74 ± (7.99)  < 0.001
Waist circumference (cm) 4808 98.66 ± (15.41) 97.83 ± (14.99) 106.61 ± (17.03)  < 0.001
Weight (kg) 4808 76.32 ± (19.10) 75.36 ± (18.52) 85.52 ± (21.93)  < 0.001
Total cholesterol (mg/dL) 4808 209.47 ± (41.81) 209.95 ± (41.46) 204.89 ± (44.77) 0.093
HDL-C (mg/dL) 4808 59.41 ± (16.77) 59.78 ± (16.80) 55.94 ± (16.09)  < 0.001
Uric acid (mg/dL) 4808 5.15 ± (1.35) 5.12 ± (1.33) 5.36 ± (1.49) 0.006
Smoking 4808 0.002
    Yes 2004 (44.49%) 1784 (43.65%) 220 (52.41%)
    No 2804 (55.51%) 2572 (56.35%) 232 (47.59%)
Hypertension 4808  < 0.001
    Yes 2675 (51.26%) 2373 (50.02%) 302 (63.09%)
    No 2133 (48.74%) 1983 (49.98%) 150 (36.91%)
Diabetes 4808  < 0.001
    Yes 931 (14.61%) 806 (13.77%) 125 (22.54%)
    No 3877 (85.39%) 3550 (86.23%) 327 (77.46%)
CVD 4808  < 0.001
    Yes 697 (12.29%) 604 (11.63%) 93 (18.56%)
    No 4111 (87.71%) 3752 (88.37%) 359 (81.44%)
Cancer 4808 0.225
    Yes 770 (17.80%) 693 (17.48%) 77 (20.84%)
    No 4038 (82.20%) 3663 (82.52%) 375 (79.16%)
Prescription medication use 4808  < 0.001
    Yes 3362 (69.66%) 2966 (67.83%) 396 (87.04%)
    No 1446 (30.34%) 1390 (32.17%) 56 (12.96%)
Depressive Symptom Score (PHQ-9) [0–27] 4808 3.39 ± (4.25) 3.10 ± (3.97) 6.16 ± (5.62)  < 0.001
Level of depressive symptoms 4808  < 0.001
    Minimal [0–5] 3420 (73.27%) 3205 (75.49%) 215 (52.05%)
    Mild [5–10] 857 (17.50%) 755 (16.98%) 102 (22.48%)
    Moderate [10–15] 321 (5.93%) 249 (4.95%) 72 (15.26%)
    Moderately severe [15–20] 153 (2.50%) 111 (2.01%) 42 (7.16%)
    Severe [20–27] 57 (0.80%) 36 (0.56%) 21 (3.05%)

PIR poverty-to-income ratio, BMI body mass index, WC waist circumference, HDL-C high-density lipoprotein cholesterol, CVD cardiovascular disease

aN not missing (unweighted)

bMean ± (SD); n (unweighted) (%)

cAdjustment chi-squared test with Rao and Scott's second-order correction; Mann–Whitney test for complex survey samples

Association between depressive symptom severity and physician-diagnosed sleep disorders in postmenopausal women

Table 3 summarizes the association between depressive symptom severity and physician-diagnosed sleep disorders. When depressive symptoms were treated as a continuous variable, the results from weighted multivariable logistic regression models demonstrated a significant positive correlation. In the fully adjusted model (Model 3), each one-unit increase in depressive symptom score was associated with a 10% higher risk of sleep disorders (OR = 1.10, 95% CI: 1.07–1.13, P < 0.001), with the association remaining robust across all models.

Table 3.

Association between depressive symptom severity and physician-diagnosed sleep disorders (weighted)

Exposure Model 1 Model 2 Model 3
OR (95% CI) P value OR (95% CI) P value OR (95% CI) P value
Depressive symptom score (PHQ-9) [0–27] 1.13 (1.10,1.16)  < 0.001 1.12 (1.09,1.15)  < 0.001 1.10 (1.07,1.13)  < 0.001
Level of depressive symptoms
Minimal [0–5] 1 (Ref.) 1 (Ref.) 1 (Ref.)
Mild [5–10] 1.92 (1.34,2.75)  < 0.001 1.81 (1.23,2.66) 0.003 1.51 (1.02,2.23) 0.040
Moderate [10–15] 4.47 (2.94,6.80)  < 0.001 3.99 (2.42,6.57)  < 0.001 3.01 (1.78,5.09)  < 0.001
Moderately severe [15–20] 5.16 (3.24,8.22)  < 0.001 4.43 (2.72,7.21)  < 0.001 3.42 (2.08,5.62)  < 0.001
Severe [20–27] 7.86 (4.05,15.3)  < 0.001 6.56 (3.22,13.4)  < 0.001 4.93 (2.30,10.6)  < 0.001
P for trend  < 0.001  < 0.001  < 0.001

Model 1: Not adjusted

Model 2: Adjusted for predisposing and enabling factors (age, race, education level, marital status, PIR, and housing characteristics)

Model 3: Additionally adjusted for need factors (BMI, WC, weight, TC, HDL-C, uric acid, smoking, hypertension, diabetes, CVD, cancer, and prescription medication use)

PIR poverty-to-income ratio, BMI body mass index, WC waist circumference, TC total cholesterol, HDL-C high-density lipoprotein cholesterol, CVD cardiovascular disease, OR odds ratio, CI confidence interval

When depressive symptoms were categorized into five severity levels, a significant dose–response relationship was observed, with increasing depressive symptom severity significantly correlating with a higher likelihood of sleep disorders (P for trend < 0.001). In Model 3, compared to individuals with minimal depressive symptoms, those with mild symptoms had a 51% increased risk of sleep disorders (OR = 1.51, 95% CI: 1.02–2.23, P = 0.040), while those with moderate symptoms exhibited more than a threefold increase in risk (OR = 3.01, 95% CI: 1.78–5.09, P < 0.001). The risk was even higher in the moderately severe (OR = 3.42, 95% CI: 2.08–5.62, P < 0.001) and severe depressive symptom groups (OR = 4.93, 95% CI: 2.30–10.6, P < 0.001). This trend remained consistent across all models, supporting the observed association between depressive symptom severity and physician-diagnosed sleep disorder risk in postmenopausal women.

RCS analysis

To further explore the nature of this relationship, RCS was applied in Model 3. As illustrated in Fig. 2a, the results indicated a linear association between depressive symptom severity and the risk of sleep disorders, with no significant evidence of nonlinearity (P for nonlinear = 0.4696).

Fig. 2.

Fig. 2

a Restricted cubic spline curve for the association between depressive symptom severity and physician-diagnosed sleep disorders. b Restricted cubic spline curve for the association between depressive symptom severity and physician-diagnosed sleep disorders by CVD status. Lines represent odds ratios, and areas represent 95 % confidence interval intervals

Subgroup analyses

Subgroup analyses were conducted based on age, race, education level, marital status, housing characteristics, BMI, smoking status, hypertension, diabetes, CVD, cancer, and prescription medication use. As illustrated in Fig. 3, the interaction test showed that CVD status was the only significant effect modifier (P for interaction = 0.019), while no significant interactions were observed for other subgroups (P all > 0.05). Among participants with CVD, the association between depressive symptom severity and sleep disorder risk was stronger (OR = 1.11, 95% CI: 1.09–1.13), compared to those without CVD, where the effect size was slightly attenuated (OR = 1.07, 95% CI: 1.03–1.11). The significant interaction suggests that the relationship between depressive symptom severity and sleep disorders differs by CVD status, with a more pronounced association observed in individuals with CVD. This pattern was further supported by the RCS curves stratified by CVD status, as shown in Fig. 2b.

Fig. 3.

Fig. 3

Subgroup analyses of the association between depressive symptom severity and physician-diagnosed sleep disorders

Importance of variables

Additionally, LASSO regression and logistic regression were applied to baseline data to identify the key variables most strongly associated with sleep disorders in postmenopausal women. As shown in Fig. 4a, the LASSO regression results ranked depressive symptoms as the most influential factor, followed by prescription medication use, waist circumference, age, and weight. Similarly, Fig. 4b presents the logistic regression rankings, where depressive symptoms remained the strongest predictor, followed by prescription medication use, age, waist circumference, and education level. These findings consistently underscore depressive symptoms as the primary determinant of sleep disorders in postmenopausal women.

Fig. 4.

Fig. 4

Variable importance according to a Lasso regression and b logistics regression

Supplementary analyses

To test the robustness of our findings, we conducted supplementary analyses using two alternative sleep-related outcomes available in the NHANES dataset. First, we analyzed SLQ050 as a binary outcome, based on participants’ responses to whether they had ever been told by a doctor or health professional that they had trouble sleeping. Second, we examined SLD010H, which directly captures participants’ responses to the question, “How much sleep do you usually get at night on weekdays or workdays?” and treated it as a continuous outcome. Both outcomes were analyzed using the same multivariable models and covariate adjustments as the main analysis, including prescription medication use. The results were consistent with those of the primary analysis. Specifically, greater depressive symptom severity was significantly associated with increased odds of self-reported sleep difficulties and with shorter sleep duration.

In the fully adjusted model (Model 3), each one-point increase in PHQ-9 score was associated with a 15% increase in the odds of sleep difficulties (OR = 1.15, 95% CI: 1.12–1.17, P < 0.001). When depressive symptom severity was modeled categorically, the odds of reporting sleep difficulties progressively increased across symptom levels: mild (OR = 2.18, 95% CI: 1.72–2.75), moderate (OR = 3.47, 95% CI: 2.41–5.00), moderately severe (OR = 4.76, 95% CI: 3.14–7.20), and severe (OR = 7.30, 95% CI: 3.21–16.6), with a significant linear trend (P for trend < 0.001) (Additional file 1: Table S1).

Similarly, depressive symptom severity was negatively associated with self-reported sleep duration in the linear regression model. Each one-point increase in PHQ-9 score was associated with a 0.06-h reduction in sleep duration (β =  − 0.06, 95% CI: − 0.07 to − 0.04, P < 0.001). Compared to participants with minimal depressive symptoms, those with higher severity levels reported significantly shorter sleep durations: mild (β =  − 0.30, 95% CI: − 0.45 to − 0.15), moderate (β =  − 0.63, 95% CI: − 0.86 to − 0.38), moderately severe (β =  − 0.98, 95% CI: − 1.36 to − 0.60), and severe (β =  − 0.88, 95% CI: − 1.58 to − 0.19), with a significant linear trend (P for trend = 0.002) (Additional file 1: Table S2). These supplementary analyses reinforce the robustness and consistency of our findings, demonstrating that the association between depressive symptom severity and impaired sleep persists across multiple operational definitions of sleep disturbance.

Sensitivity analyses

To assess whether overlapping item content between the PHQ-9 and the sleep disorder outcome might bias the results—particularly due to shared symptom domains such as sleep disturbance and fatigue—two sets of sensitivity analyses were conducted using the same multivariable modeling framework as the primary analysis.

First, to evaluate the potential impact of sleep-related measurement redundancy, we constructed a modified depressive symptom score excluding item 3 of the PHQ-9 (“trouble falling or staying asleep, or sleeping too much”), referred to as PHQ-8. The PHQ-8 score remained significantly associated with physician-diagnosed sleep disorders in the fully adjusted model (Model 3; OR = 1.11, 95% CI: 1.07–1.14, P < 0.001). Moreover, when both PHQ-8 and item 3 were included simultaneously in the model, each retained statistical significance, indicating independent associations with the outcome (Additional file 1: Table S3).

Second, to further address concerns that menopause-related somatic symptoms may inflate depressive symptom scores, we created a PHQ-7 score by excluding both item 3 (sleep disturbance) and item 4 (fatigue or low energy). In the fully adjusted model (Model 3), the PHQ-7 score remained significantly associated with sleep disorders (OR = 1.11, 95% CI: 1.08–1.15, P < 0.001). When PHQ-7, item 3, and item 4 were jointly entered into the model, all three variables remained independently and significantly associated with the outcome (Additional file 1: Table S4). These findings indicate that although certain PHQ-9 items—such as sleep disturbance and fatigue—may conceptually overlap with physiological symptoms experienced during menopause, their association with physician-diagnosed sleep disorders persists even after these items are removed. This suggests that the observed relationship reflects independent and clinically meaningful effects, rather than false positives resulting from menopausal somatic symptoms.

Discussion

This study, utilizing data from five NHANES cycles, is the first to comprehensively examine the relationship between depressive symptom severity and sleep disorders in postmenopausal women. The findings revealed a significant linear dose–response association, where increasing depressive symptom severity was consistently associated with a higher prevalence of sleep disorders. Notably, the relationship remained robust across multiple analytical approaches, including logistic regression, RCS, and variable importance ranking methods. Moreover, subgroup analyses identified CVD as a significant effect modifier, with a stronger association observed in individuals with CVD compared to those without. This finding indicates that the relationship betweendepressive symptom severity and sleep disorders may be more pronounced in postmenopausal women with CVD. Additionally, variable importance rankings derived from LASSO and logistic regression consistently identified depressive symptoms as the most influential predictor of sleep disorders. These findings emphasize the pivotal role of mental health and metabolic–cardiovascular factors in the development of sleep disorders among postmenopausal women, underscoring the necessity for integrated interventions that address both psychological and physiological health determinants.

In addition to the primary findings, supplementary and sensitivity analyses were performed to strengthen the robustness and interpretability of the results. Supplementary analyses using two alternative sleep-related outcomes—self-reported sleep difficulties and sleep duration—demonstrated consistent associations with depressive symptom severity. The dose–response trend remained evident across both binary and continuous models, reinforcing the reliability of the observed relationship regardless of how sleep disturbances were defined. Furthermore, sensitivity analyses were conducted to address potential measurement overlap between depressive symptom items and the sleep disorder outcome, particularly in the context of menopausal symptomatology. Using modified depressive symptom scores (PHQ-8 and PHQ-7), we showed that the core association persisted even after removing sleep- and fatigue-related items. When the excluded items were reintroduced alongside the modified scores, each retained an independent association with the outcome. These findings suggest that the observed relationship is not merely an artifact of overlapping symptom definitions or menopausal physiological changes but reflects distinct and additive associations between depressive symptomatology and sleep disorders.

The observed dose–response relationship between depressive symptom severity and sleep disorders aligns with previous research indicating that greater depressive symptoms are associated with an increased likelihood of sleep disturbances [27–29]. Prior studies have reported similar positive correlations, yet many have focused on broader populations [27, 28, 30, 31] or specific clinical groups [29, 32, 33] rather than postmenopausal women, who may experience distinct physiological and psychological changes that exacerbate this relationship. Unlike some earlier research suggesting a potential threshold effect, where sleep disorder risk sharply rises beyond a certain depressive symptom severity [34–36], our findings demonstrated a consistent linear trend among postmenopausal women, reinforcing the notion that even mild depressive symptoms can contribute significantly to sleep disturbances in this specific population. This indicates that interventions targeting depressive symptom management in postmenopausal women may be beneficial even at lower levels of symptom severity, potentially preventing further sleep disruption and improving overall health outcomes.

In postmenopausal women, the association between depressive symptom severity and sleep disorders may be driven by two key biological pathways: hormonal dysregulation [37] and neurotransmitter imbalance [38], both of which play a crucial role in mood regulation and sleep homeostasis. Postmenopause is characterized by a substantial decline in estrogen [37] and progesterone levels [39], which can adversely impact both emotional stability and sleep regulation [40]. Estrogen plays a key role in modulating serotonin and GABA pathways [41, 42], both of which are essential for maintaining mood balance and sleep architecture [43]. The reduction in estrogen levels has been linked to increased depressive symptoms and impaired sleep quality, manifesting as shortened sleep duration and frequent nocturnal awakenings [44–46]. Additionally, progesterone, known for its sedative effects, declines sharply after menopause, contributing to prolonged sleep latency and difficulty maintaining sleep [39]. These hormonal fluctuations may exacerbate depressive symptoms [47], which in turn further disrupt sleep patterns, creating a bidirectional cycle between depressive symptoms and sleep disorders. Beyond hormonal changes, neurotransmitter imbalances serve as another critical mechanism underlying this relationship [38]. Depressive symptoms are strongly associated with dysfunctions in serotonin, dopamine, and norepinephrine systems [48], all of which play fundamental roles in both mood regulation and sleep–wake control [49, 50]. Serotonin deficiency, a well-established feature associated with depressive symptoms [51], is particularly relevant in postmenopausal women as it directly influences melatonin production, the key hormone regulating the circadian sleep–wake cycle [52]. A decline in serotonin availability may not only intensify depressive symptoms but also lead to disruptions in sleep initiation and maintenance [53, 54]. Additionally, dopaminergic and noradrenergic dysregulation may contribute to hyperarousal states [55], further impairing sleep continuity and increasing the risk of chronic insomnia. Given that postmenopausal women already experience age-related declines in neurotransmitter function [38], these neurochemical alterations may compound the risk of bothdepressive symptoms and sleep disturbances in this population.

Notably, this study also identified a significant interaction effect between CVD status and the association between depressive symptom severity and sleep disorders in postmenopausal women. The findings suggest that the relationship between depressive symptoms and sleep disorders is more pronounced among individuals with CVD, as indicated by the stronger effect size observed in this subgroup. This may be attributed to shared pathophysiological mechanisms, including chronic inflammation, autonomic nervous system dysfunction, and vascular dysregulation [56, 57], which are known to be involved in both depressive symptoms and sleep disturbances. Additionally, individuals with CVD often experience higher levels of psychological stress, increased medication use, and poorer overall health status [58, 59], all of which may further exacerbate the impact of depressive symptoms on sleep disorders. Moreover, due to advancing age and estrogen deficiency, postmenopausal women are particularly prone to developing central obesity and cardiometabolic disorders [60], conditions closely linked to both sleep disturbances and depressive symptoms, potentially amplifying this observed relationship.

These findings highlight the necessity of comprehensive mind–body health interventions to alleviate the combined burden of depressive symptoms and sleep disorders in postmenopausal women. Beyond conventional medical treatments, integrating nursing, behavioral, and lifestyle interventions is essential to addressing the biological, psychological, and social aspects of these conditions. However, menopause-related stigma remains a barrier to seeking care, often leading to the underrecognition of mental health and sleep disturbances [15, 61]. Raising awareness through education, open dialogue, and community-based initiatives can encourage timely intervention and support for affected individuals. At the community level, incorporating menopause education, mental health screenings, and sleep assessments into primary programs can facilitate early detection and timely intervention. Peer-led support groups and public health campaigns may also normalize discussions around menopause, reducing misconceptions and social stigma. Clinically, evidence-based interventions such as cognitive-behavioral therapy (CBT) for depressive symptoms [62] and insomnia (CBT-I) [63], psychoeducational programs, and structured relaxation training can enhance emotional stability and sleep regulation. For postmenopausal women with CVD, tailored lifestyle interventions—including structured physical activity [64], CBT-I, and dietary modifications [60]—may simultaneously address mental health concerns and cardiovascular risk factors. Integrating multidisciplinary care models that combine mental health, sleep management, and cardiovascular health strategies could significantly improve the well-being of postmenopausal women at heightened risk for these conditions.

This study has several strengths. First, it utilizes nationally representative NHANES data across five cycles, ensuring high generalizability to postmenopausal women in the USA. Second, the study employs multiple analytical approaches, including multivariable logistic regression, RCS, and variable importance ranking, to robustly assess the association between depressive symptom severity and sleep disorders. Third, the identification of a linear dose–response relationship and the significant interaction with CVD status provide valuable insights into high-risk subgroups, highlighting the need for targeted interventions.

Despite the strengths of this study, several limitations should be acknowledged. First, due to the cross-sectional nature of NHANES, depressive symptoms (assessed over the past two weeks via PHQ-9) and sleep disorders (based on lifetime physician diagnosis) were measured at different time points, limiting causal interpretation. Second, sleep disorders were self-reported and not subclassified, potentially introducing recall bias and limiting the ability to differentiate between insomnia, sleep apnea, and other conditions. Third, although key covariates were adjusted for, residual confounding from unmeasured factors—such as time since menopause, psychosocial stressors, and specific medications—may still exist. Notably, NHANES lacks information on time since menopause, which precluded an evaluation of how symptom patterns vary across the menopausal transition; prior studies suggest symptom severity may peak during early transition and decline over time [65]. Fourth, menopausal status was identified through self-reported menstrual patterns and reasons for amenorrhea. While this two-step method aligns with previous NHANES-based research [66], misclassification remains possible due to the absence of hormonal markers. Lastly, depressive symptoms were assessed using PHQ-9 for severity estimation rather than clinical diagnosis, and structured psychiatric interviews were not available.

Conclusions

This study provides strong evidence of a positive linear association between depressive symptom severity and sleep disorders in postmenopausal women, confirming a dose–response relationship. The findings also highlight CVD as a significant effect modifier, with a stronger association observed in individuals with CVD. Additionally, depressive symptoms emerged as the strongest predictor of sleep disorders, underscoring their critical role in sleep health within this population.

These findings emphasize the need for integrated, multidisciplinary interventions that incorporate mental health screenings, lifestyle-based strategies, and community-centered care models into routine CVD and primary care management. Future research should further explore causal mechanisms, longitudinal trends, and the effectiveness of targeted interventions to improve mental well-being and sleep health in postmenopausal women.

Supplementary Information

Acknowledgements

Not applicable.

Clinical trial number

Not applicable.

Abbreviations

CBT

Cognitive-behavioral therapy

CBT-I

Cognitive-behavioral therapy for insomnia

CI

Confidence interval

CVD

Cardiovascular disease

GABA

Gamma-aminobutyric acid

LASSO

Least absolute shrinkage and selection operator

NCHS

National Center for Health Statistics

NHANES

National Health and Nutrition Examination Survey

OR

Odds ratio

PHQ-9

Patient Health Questionnaire-9

PIR

Poverty income ratio

RCS

Restricted cubic splines

VIF

Variance inflation factor

Authors' contributions

CY was responsible for the conceptualization of the study, methodology design, statistical analyses, interpretation of the results, drafting of the manuscript, and critical revision for intellectual content. DHM provided critical revisions to the manuscript. All authors have read and approved the final manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data underlying this study are freely accessible to all interested researchers through the National Health and Nutrition Examination Survey (NHANES) database. The data can be obtained by visiting the following website: https://www.cdc.gov/nchs/nhanes/?CDC_AAref_Val=. https://www.cdc.gov/nchs/nhanes/index.htm..

Declarations

Ethics approval and consent to participate

The NHANES protocols were approved by the Ethics Review Board of the National Center for Health Statistics, Centers for Disease Control and Prevention. Specifically, Protocol #2005–06 covered the 2005–2006 cycle and was continued through the 2007–2010 cycles, while Protocol #2011–17 covered the 2011–2012 cycle and was continued through the 2013–2014 cycle. (see: https://www.cdc.gov/nchs/nhanes/about/erb.html). All participants provided written informed consent prior to the home interview and physical examination procedures.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Zolfaghari S, Yao C, Thompson C, Gosselin N, Desautels A, Dang-Vu TT, Postuma RB, Carrier J. Effects of menopause on sleep quality and sleep disorders: Canadian longitudinal study on aging. Menopause-J N Am Menopause Soc. 2020;27(3):295–304. 10.1097/GME.0000000000001462. [DOI] [PubMed] [Google Scholar]
  • 2.Baral G, Sharma R. Sleep and menopause. In: Garg, R. (eds) Management of Menopause. Springer, Singapore. 2024. 10.1007/978-981-96-0239-1_4
  • 3.Baker FC. Optimizing sleep across the menopausal transition. Climacteric. 2023;26(3):198–205. 10.1080/13697137.2023.2173569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Ballot O, Ivers H, Ji X, Morin C. Sleep disturbances during the menopausal transition: the role of sleep reactivity and arousal predisposition. Behav Sleep Med. 2022;20(4):500–12. 10.1080/15402002.2021.1937171. [DOI] [PubMed] [Google Scholar]
  • 5.Schaedel Z, Holloway D, Bruce D, Rymer J. Management of sleep disorders in the menopausal transition. Post Reprod Health. 2021;27(4):209–14. 10.1177/20533691211039151. [DOI] [PubMed] [Google Scholar]
  • 6.Ahmady F, Niknami M. Quality of sleep in women with menopause and its related factors. Sleep Sci. 2022;15(S 01):209–14. 10.5935/1984-0063.20220021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Motlani V, Motlani G, Pamnani S, Sahu A, Acharya N. Endocrine changes in postmenopausal women: a comprehensive view. Cureus. 2023, 15(12). 10.7759/cureus.51287. [DOI] [PMC free article] [PubMed]
  • 8.Salin S, Savukoski S, Tulppo M, Pesonen P, Auvinen J, Suvanto E, Puukka K, Niinimäki M. Does climacteric status impact regulation of the autonomic nervous system at the age of 46 years? Climacteric. 2022;25(6):586–94. 10.1080/13697137.2022.2052842. [DOI] [PubMed] [Google Scholar]
  • 9.Lankila H, Kekäläinen T, Hietavala EM, et al. A mediating role of visceral adipose tissue on the association of health behaviours and metabolic inflammation in menopause: a population-based cross-sectional study. Sci Rep. 2025;15:1999. 10.1038/s41598-025-85134-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Pei M, Gibson CJ, Schembri M, Raghunathan H, Grady D, Ganz P, Huang AJ. Hot flashes and sleep disruption in a randomized trial in menopausal women. Am J Obstet Gynecol. 2025;232(1):102 e101-102. e109. 10.1016/j.ajog.2024.07.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Wang Y, Liu H, Zhou B, Yue W, Wang M, Hu K. Menopause and obstructive sleep apnea: revealing an independent mediating role of visceral fat beyond body mass index. BMC Endocr Disord. 2025;25(1):21. 10.1186/s12902-025-01850-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Leng R, Guo A, Qian G, Mao S. Influence of sedentary behavior on sleep quality in postmenopausal women in high-altitude regions of China: a cross-sectional study. Front Neurol. 2025;15: 1476010. 10.3389/fneur.2024.1476010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Haghayegh S, Strohmaier S, Hamaya R, Eliassen AH, Willett WC, Rimm EB, Schernhammer ES. Sleeping difficulties, sleep duration, and risk of hypertension in women. Hypertension. 2023;80(11):2407–14. 10.1161/HYPERTENSIONAHA.123.21350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.LeBlanc ES, Zhang S, Hedlin H, Clarke G, Smith N, Garcia L, Hale L, Hery CB, Liu S, Ochs-Balcom H. Sleep characteristics are associated with risk of treated diabetes among postmenopausal women. Am J Med. 2024;137(4):331–40. 10.1016/j.amjmed.2023.12.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Li Q, Gu J, Huang J, Zhao P, Luo C. “They see me as mentally ill”: the stigmatization experiences of Chinese menopausal women in the family. BMC Womens Health. 2023;23(1):185. 10.1186/s12905-023-02350-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Li J, Liu F, Liu Z, Li M, Wang Y, Shang Y, Li Y. Prevalence and associated factors of depression in postmenopausal women: a systematic review and meta-analysis. BMC Psychiatry. 2024;24(1):431. 10.1186/s12888-024-05875-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Turek J, Gąsior Ł. Estrogen fluctuations during the menopausal transition are a risk factor for depressive disorders. Pharmacol Rep. 2023;75(1):32–43. 10.1007/s43440-022-00444-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.McLaren S, Seidler K, Neil J. Investigating the role of 17β-estradiol on the serotonergic system, targeting soy isoflavones as a strategy to reduce menopausal depression: a mechanistic review. J Am Nutr Assoc. 2024;43(3):221–35. 10.1080/27697061.2023.2255237. [DOI] [PubMed] [Google Scholar]
  • 19.Tran KH, Luki J, Hanstock S, Hanstock CC, Seres P, Aitchison K, Shandro T, Le Melledo J-M. Decreased GABA+ levels in the medial prefrontal cortex of perimenopausal women: a 3T 1H-MRS study. Int J Neuropsychopharmacol. 2023;26(1):32–41. 10.1093/ijnp/pyac066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Cortés YI, Marginean V, Berry D. Physiologic and psychosocial changes of the menopause transition in US Latinas: a narrative review. Climacteric. 2021;24(3):214–28. 10.1080/13697137.2020.1834529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Zipf G, Chiappa M, Porter KS, Ostchega Y, Lewis BG, Dostal J. Health and nutrition examination survey plan and operations, 1999-2010. Vital Health Stat. 2013;56:1–37 PMID: 25078429. [PubMed] [Google Scholar]
  • 22.Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data, 2005–2014. Hyattsville, MD. U.S. Department of health and human services, centers for disease control and prevention; 2025. Available from: https://www.cdc.gov/nchs/nhanes/.
  • 23.Costantini L, Pasquarella C, Odone A, Colucci ME, Costanza A, Serafini G, Aguglia A, Murri MB, Brakoulias V, Amore M. Screening for depression in primary care with Patient Health Questionnaire-9 (PHQ-9): a systematic review. J Affect Disord. 2021;279:473–83. 10.1016/j.jad.2020.09.131. [DOI] [PubMed] [Google Scholar]
  • 24.Harry ML, Sanchez K, Ahmedani BK, Beck AL, Coleman KJ, Coley RY, Daida YG, Lynch FL, Rossom RC, Waring SC. Assessing the differential item functioning of PHQ-9 items for diverse racial and ethnic adults with mental health and/or substance use disorder diagnoses: a retrospective cohort study. J Affect Disord. 2023;338:402–13. 10.1016/j.jad.2023.04.091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Chen T, Qian Y, Deng X. Relationship between atherosclerotic burden and depressive symptoms in hypertensive patients: a cross-sectional study based on the NHANES database. J Affect Disord. 2024;361:612–9. 10.1016/j.jad.2024.06.087. [DOI] [PubMed] [Google Scholar]
  • 26.Alkhawaldeh A, Albashtawy M, Rayan A, Abdalrahim A, Musa A, Eshah N, Khait AA, Qaddumi J, Khraisat O. Application and use of Andersen’s behavioral model as theoretical framework: a systematic literature review from 2012–2021. Iran J Public Health. 2023;52(7):1346. 10.18502/ijph.v52i7.13236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Chunnan L, Shaomei S, Wannian L. The association between sleep and depressive symptoms in US adults: data from the NHANES (2007–2014). Epidemiol Psychiatr Sci. 2022;31: e63. 10.1017/S2045796022000452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Milaneschi Y, Kappelmann N, Ye Z, Lamers F, Moser S, Jones PB, Burgess S, Penninx BW, Khandaker GM. Association of inflammation with depression and anxiety: evidence for symptom-specificity and potential causality from UK Biobank and NESDA cohorts. Mol Psychiatry. 2021;26(12):7393–402. 10.1038/s41380-021-01188-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Hyun S, Hahm HC, Wong GTF, Zhang E, Liu CH. Psychological correlates of poor sleep quality among US young adults during the COVID-19 pandemic. Sleep Med. 2021;78:51–6. 10.1016/j.sleep.2020.12.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Fang H, Tu S, Sheng J, Shao A. Depression in sleep disturbance: a review on a bidirectional relationship, mechanisms and treatment. J Cell Mol Med. 2019;23(4):2324–32. 10.1111/jcmm.14170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Palagini L, Hertenstein E, Riemann D, Nissen C. Sleep, insomnia and mental health. J Sleep Res. 2022;31(4): e13628. 10.1111/jsr.13628. [DOI] [PubMed] [Google Scholar]
  • 32.Chorney DB, Detweiler MF, Morris TL, Kuhn BR. The interplay of sleep disturbance, anxiety, and depression in children. J Pediatr Psychol. 2008;33(4):339–48. 10.1093/jpepsy/jsm105. [DOI] [PubMed] [Google Scholar]
  • 33.Cheng H, Hu W, Luo S, Feng X, Chen Z, Yu X, Li H, Feng X, Chen W, Guo VY. Pathways linking loneliness and depressive symptoms among Chinese adolescents: the mediating role of sleep disturbance. J Affect Disord. 2025;370:235–40. 10.1016/j.jad.2024.11.006. [DOI] [PubMed] [Google Scholar]
  • 34.Dong L, Xie Y, Zou X. Association between sleep duration and depression in US adults: a cross-sectional study. J Affect Disord. 2022;296:183–8. 10.1016/j.jad.2021.09.075. [DOI] [PubMed] [Google Scholar]
  • 35.Yin J, Wang H, Li S, Zhao L, You Y, Yang J, Liu Y. Nonlinear relationship between sleep midpoint and depression symptoms: a cross-sectional study of US adults. BMC psychiatry. 2023;23(1):671. 10.1186/s12888-023-05130-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Seizer L, Martínez-Albert E, Löchner J. Nonlinear and symptom specific associations between chronotype and depression. Sci Rep. 2024;14(1):28696. 10.1038/s41598-024-79868-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Davis SR, Pinkerton J, Santoro N, Simoncini T. Menopause—biology, consequences, supportive care, and therapeutic options. Cell. 2023;186(19):4038–58. 10.1016/j.cell.2023.08.016. [DOI] [PubMed] [Google Scholar]
  • 38.Liang G, Kow ASF, Yusof R, Tham CL, Ho Y-C, Lee MT. Menopause-associated depression: impact of oxidative stress and neuroinflammation on the central nervous system—a review. Biomedicines. 2024;12(1): 184. 10.3390/biomedicines12010184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Caufriez A, Leproult R, L’Hermite-Balériaux M, Kerkhofs M, Copinschi G. Progesterone prevents sleep disturbances and modulates GH, TSH, and melatonin secretion in postmenopausal women. J Clin Endocrinol Metab. 2011;96(4):E614–23. 10.1210/jc.2010-2558. [DOI] [PubMed] [Google Scholar]
  • 40.Thurston RC, Thomas HN, Castle AJ, et al. Menopause as a biological and psychological transition. Nat Rev Psychol. 2025. 10.1038/s44159-025-00463-9. [Google Scholar]
  • 41.Sun Q, Li G, Zhao F, Dong M, Xie W, Liu Q, Yang W, Cui R. Role of estrogen in treatment of female depression. Aging (Albany NY). 2024;16(3):3021. 10.18632/aging.205507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Sikes-Keilp C, Rubinow DR. GABA-ergic modulators: new therapeutic approaches to premenstrual dysphoric disorder. CNS Drugs. 2023;37(8):679–93. 10.1007/s40263-023-01030-7. [DOI] [PubMed] [Google Scholar]
  • 43.Hantsoo L, Payne JL. Towards understanding the biology of premenstrual dysphoric disorder: from genes to GABA. Neurosci Biobehav Rev. 2023;149: 105168. 10.1016/j.neubiorev.2023.105168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Brown AM, Gervais NJ. Role of ovarian hormones in the modulation of sleep in females across the adult lifespan. Endocrinology. 2020;161(9):bqaa128. 10.1210/endocr/bqaa128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Lord C, Sekerovic Z, Carrier J. Sleep regulation and sex hormones exposure in men and women across adulthood. Pathol Biol (Paris). 2014;62(5):302–10. 10.1016/j.patbio.2014.07.005. [DOI] [PubMed] [Google Scholar]
  • 46.Morssinkhof MWL, van Wylick DW, Priester-Vink S, van der Werf YD, den Heijer M, van den Heuvel OA, Broekman BFP. Associations between sex hormones, sleep problems and depression: a systematic review. Neurosci Biobehav Rev. 2020N;118:669–80. 10.1016/j.neubiorev.2020.08.006. [DOI] [PubMed] [Google Scholar]
  • 47.Stefaniak M, Dmoch-Gajzlerska E, Jankowska K, Rogowski A, Kajdy A, Maksym RB. Progesterone and its metabolites play a beneficial role in affect regulation in the female brain. Pharmaceuticals (Basel). 2023;16(4): 520. 10.3390/ph16040520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Haenisch B, Bönisch H. Depression and antidepressants: insights from knockout of dopamine, serotonin or noradrenaline re-uptake transporters. Pharmacol Ther. 2011;129(3):352–68. 10.1016/j.pharmthera.2010.12.002. [DOI] [PubMed] [Google Scholar]
  • 49.Belujon P, Grace AA. Dopamine system dysregulation in major depressive disorders. Int J Neuropsychopharmacol. 2017;20(12):1036–46. 10.1093/ijnp/pyx056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Ruhé HG, Mason NS, Schene AH. Mood is indirectly related to serotonin, norepinephrine and dopamine levels in humans: a meta-analysis of monoamine depletion studies. Mol Psychiatry. 2007;12(4):331–59. 10.1038/sj.mp.4001949. [DOI] [PubMed] [Google Scholar]
  • 51.Erritzoe D, Godlewska BR, Rizzo G, Searle GE, Agnorelli C, Lewis Y, Ashok AH, Colasanti A, Boura I, Farrell C. Brain serotonin release is reduced in patients with depression: a [11C] Cimbi-36 positron emission tomography study with a d-amphetamine challenge. Biol Psychiatry. 2023;93(12):1089–98. 10.1016/j.biopsych.2022.10.012. [DOI] [PubMed] [Google Scholar]
  • 52.Ahmad SB, Ali A, Bilal M, Rashid SM, Wani AB, Bhat RR, Rehman MU. Melatonin and health: insights of melatonin action, biological functions, and associated disorders. Cell Mol Neurobiol. 2023;43(6):2437–58. 10.1007/s10571-023-01324-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Drakatos P, O’Regan D, Liao Y, Panayiotou C, Higgins S, Kabiljo R, Benson J, Pool N, Tahmasian M, Romigi A. Profile of sleep disturbances in patients with recurrent depressive disorder or bipolar affective disorder in a tertiary sleep disorders service. Sci Rep. 2023;13(1):8785. 10.1038/s41598-023-36083-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Embang JEG, Tan YHV, Ng YX, Loyola GJP, Wong LW, Guo Y, Dong Y. Role of sleep and neurochemical biomarkers in synaptic plasticity related to neurological and psychiatric disorders: a scoping review. J Neurochem. 2025;169(1): e16270. 10.1111/jnc.16270. [DOI] [PubMed] [Google Scholar]
  • 55.Jain R, Chepke C, Davis LL, McIntyre RS, Raskind MA. Dysregulation of noradrenergic activity: its role in conceptualizing and treating major depressive disorder, schizophrenia, agitation in Alzheimer’s disease, and posttraumatic stress disorder. J Clin Psychiatry. 2024;85(4):57634. 10.4088/JCP.plunaro2417ah. [DOI] [PubMed] [Google Scholar]
  • 56.Schwarz KG, Vicencio SC, Inestrosa NC, Villaseca P, Del Rio R. Autonomic nervous system dysfunction throughout menopausal transition: a potential mechanism underpinning cardiovascular and cognitive alterations during female ageing. J Physiol. 2024;602(2):263–80. 10.1113/JP285126. [DOI] [PubMed] [Google Scholar]
  • 57.Lee E, Anselmo M, Tahsin CT, Vanden Noven M, Stokes W, Carter JR, Keller-Ross ML. Vasomotor symptoms of menopause, autonomic dysfunction, and cardiovascular disease. Am J Physiol Heart Circ Physiol. 2022;323(6):H1270–80. 10.1152/ajpheart.00477.2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Kwapong YA, Boakye E, Khan SS, Honigberg MC, Martin SS, Oyeka CP, Hays AG, Natarajan P, Mamas MA, Blumenthal RS. Association of depression and poor mental health with cardiovascular disease and suboptimal cardiovascular health among young adults in the United States. J Am Heart Assoc. 2023;12(3): e028332. 10.1161/JAHA.122.028332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Levine GN. Psychological stress and heart disease: fact or folklore? Am J Med. 2022;135(6):688–96. 10.1016/j.amjmed.2022.01.053. [DOI] [PubMed] [Google Scholar]
  • 60.Tan A, Dunseath G, Thomas RL, Prior SL, Bracken RM, Churm R. Effect of home‐based exercise with or without a Mediterranean‐style diet on adiposity markers in postmenopausal women: a randomized‐control trial. Physiol Rep. 2025, 13(3):e70239. 10.14814/phy2.70239 [DOI] [PMC free article] [PubMed]
  • 61.Dahlgren MK, Kosereisoglu D, Smith RT, Sagar KA, Lambros AM, El-Abboud C, Gruber SA. Identifying variables associated with menopause-related shame and stigma: results from a national survey study. J Womens Health (Larchmt). 2023;32(11):1182–91. 10.1089/jwh.2023.0185. [DOI] [PubMed] [Google Scholar]
  • 62.Gautam M, Tripathi A, Deshmukh D, Gaur M. Cognitive behavioral therapy for depression. Indian J Psychiatry. 2020;62(Suppl 2):S223–9. 10.4103/psychiatry.IndianJPsychiatry_772_19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Morin CM. Cognitive-behavioral therapy of insomnia. Sleep Med Clin. 2006;1(3):375–86. 10.1016/j.jsmc.2006.06.008. [Google Scholar]
  • 64.O’Donnell E, Craig J. Habitual aerobic exercise in healthy postmenopausal women does not augment basal cardiac autonomic activity yet modulates autonomic-metabolic interactions. Menopause. 2022;29(6):714–22. 10.1097/GME.0000000000001963. [DOI] [PubMed] [Google Scholar]
  • 65.Baker FC, De Zambotti M, Colrain IM, Bei B. Sleep problems during the menopausal transition: prevalence, impact, and management challenges. Nat Sci Sleep. 2018:73–95. 10.2147/NSS.S125807. [DOI] [PMC free article] [PubMed]
  • 66.Wang K, Cui Y, Wu T. Association between tinnitus and suicidal ideation in postmenopausal women: an observational study. Am J Geriatr Psychiatry. 2025;33(4):405–12. 10.1016/j.jagp.2024.10.012. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data Availability Statement

The data underlying this study are freely accessible to all interested researchers through the National Health and Nutrition Examination Survey (NHANES) database. The data can be obtained by visiting the following website: https://www.cdc.gov/nchs/nhanes/?CDC_AAref_Val=. https://www.cdc.gov/nchs/nhanes/index.htm..


Articles from BMC Medicine are provided here courtesy of BMC

RESOURCES