Abstract
To explore the link between depressive symptoms and fracture in midlife women, this study employed both cross-sectional and retrospective cohort analyses, a crucial step in identifying those at risk of fractures. The cross-section study data (n = 1940) were retrieved from the baseline assessment of the Study of Women’s Health Across the Nation. In the retrospective cohort study (n = 1884), women’s Center for Epidemiologic Studies Depression scores were assessed at the start, and their fracture status was recorded during a 5-year follow-up. Logistic regression models were applied to explore the connections between depressive symptoms and fractures in both cross-sectional and retrospective cohort studies. Among 379 women with fracture and 1561 without, depression could not predict fracture based on both Center for Epidemiologic Studies Depression scores (P = .216) and depressive symptoms (P = .883) after adjusting for age, body mass index, smoking, menopausal status, spine and hip bone mineral density, race/ethnicity, total family income, education, physical activity, financial strain, medication use, and history of diseases. However, among 1884 women in the retrospective cohort study, depressive symptoms was associated with a 53.6% increase in fracture risk after adjustment (odds ratio 1.536, 95% confidence interval 1.051–2.245, P = .027). Moreover, a high proportion of women with depressive symptoms was found in the fracture group, and women with depressive symptoms at the start showed a statistically significant correlation with the number of fractures. Depressive symptoms can serve as a predictor for fractures in midlife women 5 years later, independent of health and physiological factors, yet no clear link was found in the cross-sectional analysis.
Keywords: depressive symptoms, fracture, link, midlife, osteoporosis, women
1. Introduction
Bone fractures represent a significant global public health concern. In 2019, the worldwide incidence of new fractures reached 178 million cases in 2019, reflecting a 33.4% increase since 1990. This rise is partially attributed to the growing and aging global population.[1] Fractures are associated with significant adverse outcomes, including absenteeism from work, diminished occupational performance, disabilities, decreased quality of life, health problems, and considerable healthcare costs. These injuries impose substantial socioeconomic burdens on individuals, families, communities, and healthcare systems.[2,3] Therefore, it is vital to look into effective prevention strategies and modifiable risk factors for fractures.
Depression, as a leading cause of burden in the Global Burden of Disease, has received increasing attention.[4–6] The prevalence of depression exceeds 11.1% within the general population of developed countries.[7] Emerging research increasingly indicates that mental disorders and the medications used to treat them can alter bone metabolism.[8,9] Clinical depression, as well as associated symptoms such as anxiety, stress, and diminished self-perceived well-being, has been linked to reduced bone mineral density (BMD), particularly among women.[10,11] However, the association between depression and fracture remains controversial. While individuals with depression often exhibit decreased BMD and an increased susceptibility to fractures,[12] findings from a population-controlled study suggests that a direct link between depression and BMD may not exist.[13] Furthermore, a study focusing on the elderly Korean population found that males with depression were more likely to experience decreased BMD, a trend not observed in females.[14] As a result, it remains imperative to further investigate the relationship between depression and fracture risk, especially in women, who are generally considered to be at a higher risk of both conditions.[15,16]
In modern society, women, particularly those in middle-age, encounter numerous pressures. This demographic often faces significant challenges related to family, career and social roles, which can lead to the development of mental health issues, notably increasing the likehood of depression. Depression is a complex condition that not only affects women’s mood, but also has implications for bone health by influencing lifestyle and healthy behaviors, such as exercise and diet.[17,18] Exercise enhances bone health by modulating interleukin-6 released from skeletal muscles, which regulates bone cell activities and promotes fatty acid oxidation. Additionally, it applies mechanical loads, enhances myokines and muscle strain, improves balance to prevent falls, regulates remodeling hormones, and reduces age-related bone loss through telomere and epigenetic effects.[19,20] Concurrently, nutrition is a critical modifiable factor affecting bone health, with appropriate dietary patterns improving bone mineral status and reducing the risk of osteoporosis and fracture.[21] However, numerous studies examining the relationship between depression and fractures have predominantly focused on women aged 50 and above,[22,23] those in postmenopause,[24] or individuals across a broad age spectrum.[25,26] Midlife women who face numerous societal, economic, and daily life pressures have been overlooked.
Accordingly, in our study, analyses were performed on a subsample from the Study of Women’s Health Across the Nation (SWAN) to explore how depressive symptoms relate to fracture in midlife women, using both cross-sectional and retrospective cohort studies.
2. Methods
2.1. Study population and design
All data utilized in present study are from SWAN, a longitudinal community-based cohort study focusing on midlife women. Participants were enrolled during their early perimenopause and premenopause between 1995 and 1997 from 7 clinical sites across the United States. The cohort recruitment involved both face to face and telephone interviews with a response rate of 46.6%, and 16,142 women aged from 40 to 55 years old completed the survey. Of these, a total of 3302 women, comprising 1550 Caucasian, 935 African American, 286 Hispanic, 250 Chinese and 281 Japanese, were deemed eligible and subsequently enrolled into the SWAN cohort for the longitudinal survey, completing the baseline study from January 1996 to December 1997. Eligibility criteria required participants to be between 42 and 52 years old, possess a uterus and at least 1 intact ovary, and have not used hormonal medications in the past 3 months. SWAN data collection includes physical assessments, fasting morning blood draws, and both interviewer‐administered and self‐administered questionnaires. All methods were performed in accordance with the relevant guidelines and regulations. The study was approved by the Institutional Review Boards at each study site (University of Pittsburgh IRB0709006, IRB0402168; Partners Healthcare 1999P006353/MGH; University of California at Davis, 260339-17; University of California at Los Angeles, 11-002274-AM-00009; University of Michigan, 00000245; Rush University Medical Center, 13021201‑IRB01‑AM04; Albert Einstein American College of Medicine, 2005-012), and informed consent was received from all participants at each study visit.[27,28] Women lacking data on depression symptoms, fracture or covariates were excluded (Fig. 1). A total of 1940 women participated in this cross-sectional analysis.
Figure 1.
Flowchart of screening process that lead to the final included samples.
In a 5-year prospective study, the broken bones since last visit were recorded on annual follow-up visit. Out of 1940 women, 1884 were ultimately included in the analysis after excluding those with incomplete fracture data. Participants were classified as having experienced fracture if they reported at least 1 instance of a broken bone. The association between baseline depressive symptoms and the risk of future fracture was further examined.
To minimize selection bias, analyses were restricted to participants with complete baseline data on depression, bone health, and key confounding variables. We also explored the demographic characteristics of included versus excluded participants, confirming no systematic differences in critical variables.
2.2. Depression assessment
Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D). This scale measures the frequency of being bothered by depressive symptoms in the past week on a scale from 0 (rarely) to 3 (most or all of the time).[29] Responses to the 20 items are summed to yield a total score ranging from 0 to 60. A CES-D score of 16 or higher is indicative a high level of depressive symptoms.[30]
2.3. Fracture assessment
At baseline and at each annual follow-up visit, women completed a questionnaire that included a symptom checklist. The symptom checklist about fracture was worded as follows: Since you were age 20 years, has a doctor ever told you that you had a broken bone? Response choices were no (1), yes (2), missing (−9), do not know (−8) and null. As a result, we classified women with fracture based on self-reported questionnaires.
2.4. Health-related variables
The demographic characteristic considered in this study included age, body mass index (BMI), ethnicity (Black/African American, Chinese/Chinese American, Japanese/Japanese American, Caucasian/White, and Hispanic), education (less than high school, high school graduate, some college/technical school, college graduate and postgraduate), total family income (from <$19,999–$100,000 or more). Age was determined based on the self-reported date of birth, while race/ethnicity was identified through self-reported identification. BMI was calculated as weight in kilograms divided by height in meters squared. Measurement of height and weight were obtained using calibrated electronic or balance beams scales and stadiometers. According to existing literatures, aging is associated with accelerates bone loss due to disruptions in the bone remodeling process. A low BMI may elevate the risk of bone loss as a result of inadequate mechanical loading and nutritional deficiencies, including insufficient calcium and vitamin D. Differences in bone density among different racial groups may result in varying fracture risks. Furthermore, socioeconomic status affects bone health by influencing nutrition and access to healthcare, with lower socioeconomic status being linked to increased bone loss. Engaging in weight-bearing exercise facilitates bone formation, whereas immobility promotes bone resorption. Additionally, smoking and certain medications can adversely affect bone metabolism, and conditions such as diabetes and arthritis may compromise bone health through hormonal imbalances or inflammation.[31–34] Consequently, variables including age, BMI, race, socioeconomic status, physical activity, smoking, menopausal status, medication use, and the presence of diseases were included as covariates in this study due to their documented association with bone health.
2.5. Statistical analysis
The normality of all continuous variables was assessed using the Kolmogorov–Smirnov test. Data that followed a normal distribution were summarized as mean ± standard deviation, while non-normally distributed data were described using the median and interquartile range. To compare the means of age, BMI, CES-D score, and BMD between the non-fracture and facture groups, Student t test was employed for normally distributed data, and the Mann–Whitney U test was used for non-normally distributed data. Chi-square tests were conducted to compare the percentages of categorical variables between women with and without fractures, as well as the incidence of fractures among women with low (<16) versus high (≥16) CES-D scores. Logistic regression analysis was utilized to investigate the associations between CES-D scores or depressive symptoms and the occurrence of fractures. All above analyses were conducted using R software (version 4.2.1, R Foundation for Statistical Computing, Vienna, Austria) and SPSS (version 20.0; IBM Corp., Armonk).
3. Results
3.1. Cross-sectional study
Table 1 presents the characteristics of the study sample stratified by fracture status. The total sample comprised 1940 women with an approximately 46 years, predominantly identifying as Caucasian/White Non-Hispanic. The median CES-D score for the entire sample was 16, and approximately 19.5% of the participants had experienced fractures. The median (or mean) age was 46 for both the 379 women with fractures and the 1561 without. Significant differences were observed between women with and without fractures in terms of BMI (P = .001), race/ethnicity (P < .001), education (P = .013), income (P < .001), physical activity (P < .001), financial strain (P = .002), history of diseases (P = .010), and tendency towards depression (P = .039). The baseline CES-D scores among women with fractures were significantly higher than those in women without fractures (P = .001). A prior history of depressive symptoms was documented in 225 out of the 379 fracture cases and 835 out of 1561 non-fracture cases (P = .039). In addition, individuals exhibiting depressive symptoms tended to have lower BMD, as indicated by P = .007 for total spine bone mineral density (SPBMDT) and P < .001 for total hip bone mineral density (HPBMDT), which indirectly suggested an elevated risk of fractures.
Table 1.
Baseline characteristics in the nested cross-sectional study.
| Total (n = 1940) | Non-fracture group (n = 1561) | Fracture group (n = 379) | P | |
|---|---|---|---|---|
| Age (medians [IQR]) | 46 (44–48) | 46 (44–48) | 46 (44–48) | .717 |
| BMI (kg/m2) | 25.95 (22.42–31.68) | 25.47 (22.31–31.41) | 27.28 (23.10–33.07) | .001 |
| Race/ethnicity (n [%]) | <.001 | |||
| Black/African American | 514 (26.5%) | 421 (27.0%) | 93 (24.5%) | |
| Chinese/Chinese American | 208 (10.7%) | 192 (12.3%) | 16 (4.2%) | |
| Japanese/Japanese American | 206 (10.6%) | 183 (11.7%) | 23 (6.1%) | |
| Caucasian/White Non-Hispanic | 1012 (52.2%) | 765 (49.0%) | 247 (65.2%) | |
| Hispanic | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | |
| Education (n [%]) | .013 | |||
| Less than high school | 75 (3.9%) | 65 (4.2%) | 10 (2.6%) | |
| High school graduate | 330 (17.0%) | 285 (18.3%) | 45 (11.9%) | |
| Some college/technical school | 656 (33.8%) | 515 (33.0%) | 141 (37.2%) | |
| College graduate | 421 (21.7%) | 339 (21.7%) | 82 (21.6%) | |
| Postgraduate | 458 (23.6%) | 357 (22.9%) | 101 (26.6%) | |
| Income (n [%]) | <.001 | |||
| Less than $19,999 | 220 (11.3%) | 159 (10.2%) | 61 (16.1%) | |
| $20,000 to $49,999 | 675 (34.8%) | 553 (35.4%) | 122 (32.2%) | |
| $50,000 to $99,999 | 741 (38.2%) | 603 (38.6%) | 138 (36.4%) | |
| $100,000 or more | 304 (15.7%) | 246 (15.8%) | 58 (15.3%) | |
| Physical activity [n (%)] | <.001 | |||
| Never | 402 (20.7%) | 322 (20.6%) | 80 (21.1%) | |
| Less than once a month | 280 (14.4%) | 233 (14.9%) | 47 (12.4%) | |
| Once a month | 109 (5.6%) | 87 (5.6%) | 22 (5.8%) | |
| 2–3 times a month | 218 (11.2%) | 179 (11.5%) | 39 (10.3%) | |
| Once a week | 231 (11.9%) | 193 (12.4%) | 38 (10.0%) | |
| More than once a week | 700 (36.1%) | 547 (35.0%) | 153 (40.4%) | |
| Financial strain (n [%]) | .002 | |||
| Very hard | 150 (7.7%) | 107 (6.9%) | 43 (11.3%) | |
| Somewhat hard | 510 (26.3%) | 429 (27.5%) | 81 (21.4%) | |
| Not very hard at all | 1280 (66.0%) | 1025 (65.7%) | 255 (67.3%) | |
| Smoking currently (n [%]) | .990 | |||
| No | 518 (26.7%) | 398 (25.5%) | 120 (31.7%) | |
| Yes | 294 (15.2%) | 226 (14.5%) | 68 (17.9%) | |
| Menopausal status (n [%]) | .078 | |||
| Early perimenopausal | 889 (45.8%) | 700 (44.8%) | 189 (49.9%) | |
| Premenopausal | 1051 (54.2%) | 861 (55.2%) | 190 (50.1%) | |
| Medication use (current) (n [%]) | .073 | |||
| Thyroid drugs | 109 (5.6%) | 87 (5.6%) | 22 (5.8%) | |
| Insulin drugs | 57 (2.9%) | 40 (2.6%) | 17 (4.5%) | |
| Psychotropic drugs | 203 (10.5%) | 146 (9.4%) | 57 (15.0%) | |
| Steroid pills | 32 (1.6%) | 21 (1.3%) | 11 (2.9%) | |
| Vitamins or minerals | 1054 (54.3%) | 827 (53.0%) | 227 (59.9%) | |
| History of diseases (n [%]) | .010 | |||
| High blood pressure | 348 (17.9%) | 276 (17.7%) | 72 (19.0%) | |
| Diabetes | 96 (4.9%) | 76 (4.9%) | 20 (5.3%) | |
| Heart attack or angina | 35 (1.8%) | 20 (1.3%) | 15 (4.0%) | |
| Arthritis | 316 (16.3%) | 227 (14.5%) | 89 (23.5%) | |
| Osteoporosis | 26 (1.3%) | 17 (1.1%) | 9 (2.4%) | |
| Tendency towards depression (n [%]) | .039 | |||
| CES-D score <16 | 880 (45.4%) | 726 (46.5%) | 154 (40.6%) | |
| CES-D score ≥ 16 | 1060 (54.6%) | 835 (53.5%) | 225 (59.4%) | |
| CES-D score (Medians [IQR]) | 16 (13–20) | 16 (13–20) | 17 (14–22) | .001 |
| Bone mineral density (g/cm2) | ||||
| Total spine BMD | 1.07 (0.98–1.16) | 1.07 (0.98–1.17) | 1.07 ± 0.14 | .453 |
| Total hip BMD | 0.95 (0.86–1.05) | 0.95 (0.86–1.05) | 0.96 ± 0.13 | .350 |
Bold values indicate statistical significance.
BMD = bone mineral density, BMI = body mass index, CES-D = Center for Epidemiologic Studies Depression.
Table 2 illustrates the association between CES-D scores and fracture incidence using multivariable logistic regression analysis. The crude regression coefficient with a 95% confidence interval of 1.029 (1.011–1.047), indicated a positive associate between CES-D scores and fracture risk. After adjusting for age, BMI, smoking, menopausal status, SPBMDT, and HPBMDT in Model 1, the results were attenuated compared to the crude model. Although the association was further reduced after controlling for race/ethnicity, total family income, education, physical activity, financial strain, medication use, and disease history in Model 2, a positive correlation between depression scores and fracture risk persisted. Furthermore, analogous findings have been documented in research examining depressive symptoms and fracture incidence after adjustment (P = .883) as presented in Table S1, Supplemental Digital Content, https://links.lww.com/MD/Q14.
Table 2.
Multivariable logistic regression models presenting the associations of CES-D scores with fracture.
| Crude | Model 1 | Model 2 | |
|---|---|---|---|
| OR | 1.029 (1.011–1.047) | 1.021 (1.003–1.040) | 1.013 (0.993–1.033) |
| P | .001 | .022 | .216 |
Crude: no adjustment.
Model 1: adjusted for age, BMI, smoking, menopausal status, SPBMDT, and HPBMDT.
Model 2: adjusted for age, BMI, smoking, menopausal status, SPBMDT and HPBMDT, race/ethnicity, total family income, education, physical activity, financial strain, medication use and history of diseases.
BMI = body mass index, HPBMDT = total hip bone mineral density, OR = odds ratio, SPBMDT = total spine bone mineral density.
A stratified analysis was conducted based on menopausal status, racial/ethnic subgroups and the use of psychotropic drugs. In models with no or minimal covariates adjustment, a weak correlation between CES-D scores and an elevated fracture risk was identified in certain subgroups, specifically among early perimenopausal women, Black/African Americans, and both users and nonusers of psychotropic drugs. However, this correlation did not reach statistical significance in any subgroup when additional covariates were controlled for in Model 2 (Table S2, Supplemental Digital Content, https://links.lww.com/MD/Q14). Meanwhile, comparable outcomes were noted in studies addressing depressive symptoms and fractures risk post-adjustment (Table S3, Supplemental Digital Content, https://links.lww.com/MD/Q14). These findings imply that the potential relationship between depression and fracture risk may be affected by confounding variables, and generally lacks robustness or statistical significance across most subgroups within a cross-sectional study.
3.2. Retrospective cohort study
Of the initial 1940 participants, follow-up assessments were conducted over 5 visits. After excluding individuals with missing self-reported fracture data, a total of 1884 participants were included in the final analysis. Table S4, Supplemental Digital Content, https://links.lww.com/MD/Q14 provides an overview of the study sample’s characteristics based on fracture status. The women in the overall sample had a median (or mean) age of approximately 46 years, and the median median CES-D was 16. Significant differences were observed in BMI, race, financial strain, and medication use. Table 3 illustrates the association between depressive symptoms and fracture indicence using multivariable logistic regression analysis. The unadjusted regression coefficient, with a 95% confidence interval of 1.523 (1.101–2.105), indicated a positive association between depressive symptoms and fracture risk (P = .011). After adjusting for age, BMI, smoking, menopausal status, SPBMDT, and HPBMDT in Model 1, the results remained consistent with those of the unadjusted model (P = .005). Although the association was slightly attenuated after further adjustment for race/ethnicity, total family income, education, physical activity, financial strain, medication use, and disease history in Model 2, the association remained statistically significant (P = .027). Additionally, a prior history of depressive symptoms was documented in 110 out of the 173 fracture cases and 914 out of 1711 non-fracture cases (P = .011), revealing a higher prevalence of depressive symptoms in the fracture group (63.58%) compared to the non-fracture (53.42%). Furthermore, our findings indicated that women with depressive symptoms experienced a greater number of fractures (P = .010).
Table 3.
Multivariable logistic regression models presenting the associations of depressive symptoms with fracture at visit 5.
| Crude | Model 1 | Model 2 | |
|---|---|---|---|
| OR | 1.523 (1.101–2.105) | 1.695 (1.174–2.447) | 1.536 (1.051–2.245) |
| P value | .011 | .005 | .027 |
Crude: no adjustment.
Model 1: adjusted for age, BMI, smoking, menopausal status, SPBMDT, and HPBMDT.
Model 2: adjusted for age, BMI, smoking, menopausal status, SPBMDT and HPBMDT, race/ethnicity, total family income, education, physical activity, financial strain, medication use, and history of diseases.
BMI = body mass index, HPBMDT = total hip bone mineral density, OR = odds ratio, SPBMDT = total spine bone mineral density.
A stratified analysis was conducted based on menopausal status, racial/ethnic subgroups and psychotropic drug use was performed (Table S5, Supplemental Digital Content, https://links.lww.com/MD/Q14). This analysis highlighted significant subgroup differences in the association between depressive symptoms and fracture risk. In early perimenopausal women, consistent and significant positive correlation between depressive symptoms and fracture risk persisted even after multiple adjustments. Among Caucasian/White Non-Hispanic individuals, a potential association approached significance following adjustment. In contrast, no significant associations were observed in premenopausal women, Black/African Americans, Chinese/Chinese Americans, or Japanese/Japanese Americans. Although no statistically significant association was found among psychotropic drug users after final adjustment, the relatively high odds ratio values indicated the need for further validation with larger samples. Overall, the impact of depressive symptoms on fracture risk might differ based on menopausal status and racial background, with early perimenopausal women constituting a high-risk subgroup warranting targeted attention.
In addition, a Kaplan–Meier survival plot, presented in Figure S1, Supplemental Digital Content, https://links.lww.com/MD/Q13, illustrates the probability of remaining fracture-free over a 5-year period for women exposed and unexposed to depressive symptoms at baseline. The findings demonstrated that the incidence of fractures is higher among individuals with depressive symptoms during this period compared to those without depressive symptoms.
4. Discussion
To explore the association between depression and fractures, many studies have already provided pertinent evidence. Depression is a prevalent mental health issue that may be significantly linked to bone health. Current research on the relationship between depression and fractures has predominantly focused on women over the age of 50, postmenopausal women, or those spanning a broad age range.[22–26] In this study, we aim to in midlife women aged 42 to 52 years who may experience heightened economic, social, life and physiological pressures due to complex role expectations. Consequently, we explored the relationship between depressive symptoms and bone fractures among women in midlife. In examining the impact of depressive symptoms on fracture incidence in women, we considered various factors as essential covariate, including race, socioeconomic status, educational attainment, physical activity levels, menstrual status, medication use, and overall health status, due to their significant associations with living conditions and health outcomes. These covariates are supported by multiple studies on bone health.[31–34] The above-discussed covariates are fully considered in our investigation of the relationship between depressive symptoms and fractures.
Many studies have demonstrated that individuals with depression tend to have lower BMD, potentially increasing their susceptibility to fractures.[12,35] A study targeting middle-aged and elderly populations revealed a significant association between depression and both fractures and decreased bone density, particularly among women.[36] Our study similarly identified lower BMD in participants with depressive symptoms (P = .007 for SPBMDT and P < .001 for HPBMDT). In addition, this cross-sectional study demonstrated that higher CES-D scores were associated with 1.029-fold increase in the odds of fracture status (P = .001) (Table 2), while depressive symptoms were related to 1.27-fold increase in frature risk with no adjustment (P = .040) (Table S1, Supplemental Digital Content, https://links.lww.com/MD/Q14). However, no significant association was found between CES-D scores (odds ratio [OR] = 1.013, P = .216) or depressive symptoms (OR = 1.019, P = .833) in our cohort, even after adjusting for age, BMI, smoking, menopausal status, SPBMDT and HPBMDT, race/ethnicity, total family income, education, physical activity, financial strain, medication use, and disease history. These data align with the findings of Williams study, which demonstrated that exposure to depression was associated with a 1.45-fold increase in odds of fracture, albeit marginally significant (P = .05), among women aged 20 to 93 years (median age 50.4 years) participating in the Geelong Osteoporosis Study, after adjusting for age. Neverthless, The association between depression and fracture risk was attenuated when further adjusted for age, education level, BMI, smoking, physical activity, BMD, history of prior fractures, falls, and medication use (OR = 1.52, P = .06).[26] It is assumed that in the cross-sectional study, the onset of depressive symptoms could not be precisely determined, potentially resulting in a random association between depressive symptoms and fractures, thereby leading to an insignificant correlation after controlling for confounding factors.
In the retrospective cohort study, depressive symptoms were associated with a 53.6% increase in fractures risk (P = .027) after adjustment for age, BMI, smoking, menopausal status, SPBMDT and HPBMDT, race/ethnicity, total family income, education, physical activity, financial strain, medication use, and disease history. Moreover, the presence of depressive symptoms at baseline was also significantly associated with fracture risk. These findings are consistent with prior cohort studies that have identified a correlation between depression and an increased risk of fracture.[37] Meta-analyses of such studies have reported that depression is associated with a 24% to 39% increase in fracture risk.[12,38] The specific mechanisms through which depression increases fracture risk are not yet fully understood. Depression may contribute to fractures through several pathways. Firstly, it may induce inflammatory and hormonal changes that affect bone remodeling, thereby indirectly reducing bone strength.[39–41] Elevated proinflammatory cytokines associated with depression, such as interleukin-6 and C-reactive protein, can promote osteoclast formation.[39] Research by Atteritano et al demonstrated that postmenopausal women with depression exhibited higher levels of parathyroid hormone (72.26 vs 39.64 pg/mL) and lower levels of vitamin D (35.08 vs 80.70 nmol/L), which adversely affect bone density.[40] Secondly, neuropathological changes in specific brain regions of individuals with depression may impair balance, judgment, and coordination, thereby increasing the likelihood of falls and subsequent fractures. A study by Whooley et al found that elderly women with depression had a higher fall rate (70%) compared to those without depression (59%). After controlling for other variables, the risk remained 1.4 times greater.[22] In Williams study, depressed patients reported higher fall rates.[26] Thirdly, depression-related unhealthy behaviors, such as smoking and reduced physical activity, have been shown to negatively impact bone metabolism.[42,43] According to Alghadir study, smokers had significantly lower BMD (1.79 ± 0.13 g/cm²) compared to nonsmokers (2.1 ± 0.14 g/cm²), while young individuals with moderate-to-vigorous physical activity (≥600 MET-min/week) demonstrated much higher BMD than those who are less active.[42] Lastly, psychotropic drugs may affect bone health by directly impacting bone tissue or by increasing fall risk through side effects such as drowsiness and dizziness. Spangler et al reported a higher risk of spinal fractures among users of selective serotonin reuptake inhibitors (HR = 1.36), potentially due to their influence on serotonin transporters and signaling pathways.[24] Forsén et al found that adjusting for psychotropic medication use reduced the association between depression and hip fractures, with the relative risk decreasing from 2.1 to 1.5, indicating that these medications may heighten fall risk.[23] Collectively, these findings underscore the importance of effective management in maintaining bone health, particularly for women in the early perimenopausal phases. It is recommended that regular depression screenings be conducted alongside monitoring of bone density and fall risk. Antidepressants with minimal impact on bone metabolism should be prioritized, and psychotherapy may serve to mitigate risks associated with falls and bone loss.
To the best of our knowledge, this study presents both strengths and limitations. A major strength is the utilization of a nationwide sample of middle-aged women experiencing multiple pressures and complex roles. The study design incorporated both cross-sectional and retrospective cohort methodologies to comprehensively investigate the relationship between fractures and depressive symptoms. Importantly, depressive symptoms emerged as a statistically significant risk factor for fractures, even after adjusting for confounding variables. However, several limitations must be acknowledged. First, fracture data and depressive symptoms were gathered using self-reported questionnaires without clinical confirmation, which might result in misclassification bias. Secondly, despite accounting for certain covariates, other unmeasured variables could still affect the outcomes of multiple comparisons. Thirdly, the lack of a standardized diagnostic tool, coupled with reliance on a depression symptom scale, might introduce recall bias.
In summary, our study indicates that depressive symptoms constitute a significant risk factor for fractures in women aged 42 to 52 years. Given the potentially clinically significant comorbidity between mental health and bone health, it is imperative to prioritize screening in at-risk populations, particularly among middle-aged women. However, due to the observational design of this study, it is not possible to infer causal relationships, despite the cohort study’s findings demonstrating statistical significance.
Acknowledgments
We extend our gratitude to all individuals who contributed to the planning and execution of the SWAN project, as well as to those who facilitated the accessibility of the SWAN datasets on their online platform. Furthermore, we appreciate the efforts of the reviewers and editors who dedicated their time to evaluating our submission.
Author contributions
Data curation: Wenjie Lou, Ni Wang, Rui Xiong.
Formal analysis: Wenjie Lou.
Investigation: Wenjie Lou.
Methodology: Wenjie Lou.
Supervision: Kai Zou, Biao Che.
Validation: Wenjie Lou.
Visualization: Wenjie Lou, Kai Zou, Biao Che.
Writing – original draft: Wenjie Lou.
Writing – review & editing: Kai Zou, Biao Che.
Supplementary Material
Abbreviations:
- BMD
- bone mineral density
- BMI
- body mass index
- CES-D
- Center for Epidemiologic Studies Depression
- HPBMDT
- total hip bone mineral density
- OR
- odds ratio
- SPBMDT
- total spine bone mineral density
- SWAN
- the Study of Women’s Health Across the Nation
The authors have no funding and conflicts of interest to disclose.
The datasets generated during and/or analyzed during the current study are publicly available.
Supplemental Digital Content is available for this article.
How to cite this article: Lou W, Wang N, Xiong R, Zou K, Che B. Associations of depressive symptoms with fracture in midlife women: An observational study from the Study of Women’s Health Across the Nation. Medicine 2025;104:38(e44643).
KZ and BC contributed to this article equally.
Contributor Information
Wenjie Lou, Email: louwenjie0718@163.com.
Ni Wang, Email: wangniwhwh@163.com.
Rui Xiong, Email: rongruiwh@163.com.
Kai Zou, Email: zoukaiwh@163.com.
References
- [1].Wu A-M, Bisignano C, James SL, et al. Global, regional, and national burden of bone fractures in 204 countries and territories, 1990–2019: a systematic analysis from the Global Burden of Disease Study 2019. Lancet Healthy Longev. 2021;2:e580–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Pike C, Birnbaum HG, Schiller M, Sharma H, Burge R, Edgell ET. Direct and indirect costs of non-vertebral fracture patients with osteoporosis in the US. PharmacoEcon. 2010;28:395–409. [DOI] [PubMed] [Google Scholar]
- [3].Borgström F, Karlsson L, Ortsäter G, et al. Fragility fractures in Europe: burden, management and opportunities. Arch Osteoporos. 2020;15:59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Ferrari AJ, Charlson FJ, Norman RE, et al. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS Med. 2013;10:e1001547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Yang F, Lodder P, Huang N, Liu X, Fu M, Guo J. Thirty-year trends of depressive disorders in 204 countries and territories from 1990 to 2019: an age-period-cohort analysis. Psychiatry Res. 2023;328:115433. [DOI] [PubMed] [Google Scholar]
- [6].Rong J, Cheng P, Li D, Wang X, Zhao D. Global, regional, and national temporal trends in prevalence for depressive disorders in older adults, 1990–2019: an age-period-cohort analysis based on the global burden of disease study 2019. Ageing Res Rev. 2024;100:102443. [DOI] [PubMed] [Google Scholar]
- [7].Bromet E, Andrade LH, Hwang I, et al. Cross-national epidemiology of DSM-IV major depressive episode. BMC Med. 2011;9:90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Fernandes BS, Hodge JM, Pasco JA, Berk M, Williams LJ. Effects of Depression and serotonergic antidepressants on bone: mechanisms and implications for the treatment of depression. Drugs Aging. 2016;33:21–5. [DOI] [PubMed] [Google Scholar]
- [9].Kishimoto T, De Hert M, Carlson HE, Manu P, Correll CU. Osteoporosis and fracture risk in people with schizophrenia. Curr Opin Psychiatry. 2012;25:415–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Rauma PH, Koivumaa-Honkanen H, Williams LJ, Tuppurainen MT, Kröger HP, Honkanen RJ. Life satisfaction and bone mineral density among postmenopausal women: cross-sectional and longitudinal associations. Psychosom Med. 2014;76:709–15. [DOI] [PubMed] [Google Scholar]
- [11].Michelson D, Stratakis C, Hill L, et al. Bone mineral density in women with depression. N Engl J Med. 1996;335:1176–81. [DOI] [PubMed] [Google Scholar]
- [12].Wu Q, Liu B, Tonmoy S. Depression and risk of fracture and bone loss: an updated meta-analysis of prospective studies. Osteoporos Int. 2018;29:1303–12. [DOI] [PubMed] [Google Scholar]
- [13].Ozsoy S, Eşel E, Turan MT, et al. Is there any alteration in bone mineral density in patients with depression? Turk Psikiyatri Derg. 2005;16:77–82. [PubMed] [Google Scholar]
- [14].Oh SM, Kim HC, Ahn SV, Rhee Y, Suh I. Association between depression and bone mineral density in community-dwelling older men and women in Korea. Maturitas. 2012;71:142–6. [DOI] [PubMed] [Google Scholar]
- [15].Lorentzon M, Johansson H, Harvey NC, et al. Osteoporosis and fractures in women: the burden of disease. Climacteric. 2022;25:4–10. [DOI] [PubMed] [Google Scholar]
- [16].Parker G, Brotchie H. Gender differences in depression. Int Rev Psychiatry. 2010;22:429–36. [DOI] [PubMed] [Google Scholar]
- [17].Liang J, Huang S, Jiang N, et al. Association between joint physical activity and dietary quality and lower risk of depression symptoms in US adults: cross-sectional NHANES study. JMIR Public Health Surveill. 2023;9:e45776. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Selman A, Dai J, Driskill J, Reddy AP, Reddy PH. Depression and obesity: focus on factors and mechanistic links. Biochim Biophys Acta Mol Basis Dis. 2025;1871:167561. [DOI] [PubMed] [Google Scholar]
- [19].Coskun Benlidayi I, Gupta L, Parihar J, Levy AL, Alexanderson H. Exercise for improving bone health in patients with AIRDs: understanding underlying biology and physiology. Int J Rheum Dis. 2024;27:e15402. [DOI] [PubMed] [Google Scholar]
- [20].Zhu C, Ding X, Chen M, Feng J, Zou J, Zhang L. Exercise-mediated skeletal muscle-derived IL-6 regulates bone metabolism: a new perspective on muscle–bone crosstalk. Biomolecules. 2025;15:893. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Movassagh EZ, Vatanparast H. Current evidence on the association of dietary patterns and bone health: a scoping review. Adv Nutr. 2017;8:1–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Whooley MA, Kip KE, Cauley JA, Ensrud KE, Nevitt MC, Browner WS. Depression, falls, and risk of fracture in older women. Arch Intern Med. 1999;159:484–90. [DOI] [PubMed] [Google Scholar]
- [23].Forsén L, Meyer HE, Søgaard AJ, Naess S, Schei B, Edna TH. Mental distress and risk of hip fracture. Do broken hearts lead to broken bones? J Epidemiol Community Health. 1999;53:343–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Spangler L, Scholes D, Brunner RL, et al. Depressive symptoms, bone loss, and fractures in postmenopausal women. J Gen Intern Med. 2008;23:567–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Ahmed LA, Schirmer H, Berntsen GK, Fønnebø V, Joakimsen RM. Self-reported diseases and the risk of non-vertebral fractures: the Tromso study. Osteoporos Int. 2006;17:46–53. [DOI] [PubMed] [Google Scholar]
- [26].Williams LJ, Pasco JA, Jackson H, et al. Depression as a risk factor for fracture in women: a 10 year longitudinal study. J Affect Disord. 2016;192:34–40. [DOI] [PubMed] [Google Scholar]
- [27].Harlow SD, Burnett-Bowie SM, Greendale GA, et al. Disparities in reproductive aging and midlife health between Black and White women: the Study of Women’s Health Across the Nation (SWAN). Womens Midlife Health. 2022;8:3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Yang D, Li X, Qu C, Yi J, Gao H. Comparison of triglyceride-glucose related indices in prediction of cardiometabolic disease incidence among US midlife women. Sci Rep. 2025;15:19359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401. [Google Scholar]
- [30].Weissman MM, Sholomskas D, Pottenger M, Prusoff BA, Locke BZ. Assessing depressive symptoms in five psychiatric populations: a validation study. Am J Epidemiol. 1977;106:203–14. [DOI] [PubMed] [Google Scholar]
- [31].Shin J, Kim KY, Park JH, et al. Fracture risk in Korean postmenopausal women: the influence of BMI, age, and bone density. Osteoporos Sarcopenia. 2025;11:65–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Crandall CJ, Miller-Martinez D, Greendale GA, Binkley N, Seeman TE, Karlamangla AS. Socioeconomic status, race, and bone turnover in the Midlife in the US Study. Osteoporos Int. 2012;23:1503–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [33].Stokes G, Herath M, Samad N, Trinh A, Milat F. ‘Bone health – across a woman’s lifespan’. Clin Endocrinol (Oxf). 2025;102:389–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [34].Valentin G, Ravn MB, Jensen EK, et al. Socio-economic inequalities in fragility fracture incidence: asystematic review and meta-analysis of 61 observational studies. Osteoporos Int. 2021;32:2433–48. [DOI] [PubMed] [Google Scholar]
- [35].Ma M, Liu X, Jia G, et al. The association between depression and bone metabolism: a US nationally representative cross‑sectional study. Arch Osteoporos. 2022;17:113. [DOI] [PubMed] [Google Scholar]
- [36].Mollard EM, Bilek L, Waltman N. Emerging evidence on the link between depressive symptoms and bone loss in postmenopausal women. Int J Womens Health. 2017;10:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [37].Wu Q, Liu J, Gallegos-Orozco JF, Hentz JG. Depression, fracture risk, and bone loss: a meta-analysis of cohort studies. Osteoporos Int. 2010;21:1627–35. [DOI] [PubMed] [Google Scholar]
- [38].Qiu L, Yang Q, Sun N, et al. Association between depression and the risk for fracture: a meta-analysis and systematic review. BMC Psychiatry. 2018;18:336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Ganesan K, Teklehaimanot S, Tran T, Asuncion M, Norris K. Relationship of C-reactive protein and bone mineral density in community-dwelling elderly females. J Natl Med Assoc. 2005;97:329–33. [PMC free article] [PubMed] [Google Scholar]
- [40].Atteritano M, Lasco A, Mazzaferro S, et al. Bone mineral density, quantitative ultrasound parameters and bone metabolism in postmenopausal women with depression. Intern Emerg Med. 2013;8:485–91. [DOI] [PubMed] [Google Scholar]
- [41].Gkiatas I, Lykissas M, Kostas-Agnantis I, Korompilias A, Batistatou A, Beris A. Factors affecting bone growth. Am J Orthop (Belle Mead NJ). 2015;44:61–7. [PubMed] [Google Scholar]
- [42].Alghadir AH, Gabr SA, Al-Eisa E. Physical activity and lifestyle effects on bone mineral density among young adults: sociodemographic and biochemical analysis. J Phys Ther Sci. 2015;27:2261–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [43].Li C, Palka JM, Abdullah N, et al. Link between depression and bone mineral density in Cooper Center Longitudinal Study: indirect effects of vitamin D, inflammation, and physical activity. J Affect Disord. 2024;344:277–83. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.