Considering the rapid growth and ethnic and racial diversity of the aging population in the United States, fracture rates will concurrently increase. Identifying risk factors for the loss of bone mineral density (BMD) and associated fractures among diverse postmenopausal women is a research priority. My colleagues and I conducted a prior investigation that discovered that stress within one’s social environment is a risk factor for decreased BMD. Whether this risk factor for low BMD increases fracture risk has not been established, however. We hypothesized that the biological embodiment of stress from the social environment is one pathway through which social inequity and structural racism create health disparities, which may contribute to fractures through degradation of BMD.
For a follow-up study published in the Journal of Epidemiology & Community Health, we tested our hypothesis on whether social stress was prospectively associated with fracture incidence among racially and ethnically diverse postmenopausal women. The study included more than 160,000 postmenopausal women in the Women’s Health Initiative (WHI), a long-term US health study launched in 1993 to assess strategies for preventing chronic conditions, including fractures, in postmenopausal women. Participants completed a questionnaire at enrollment on their social environment, specifically on social strain, social support, and social functioning. Fracture incidence and location information was collected annually during 16 years of follow-up. Cox proportional hazards regression models were calculated to estimate hazards ratios (HRs) for social stress with time to any body part (total) fracture and hip fracture incidence.
Greater Social Strain Linked With Fracture Risk
We found that greater social strain, indicating greater stress, was associated with an increased risk of fractures, and that greater social support and social functioning, indicating lower stress, were associated with lower risk of fractures, after adjusting for other risk factors. Women were 13% less likely (HR = 0.87, 95% confidence internal [CI]: 0.84-0.89) to experience a hip fracture with each one standard deviation (SD) increase in social functioning, which translates to 35% lower hip fracture risk in women reporting the highest level (three SDs higher) of social functioning.
In addition, our study team discovered that associations were modified by age and race/ethnicity. The magnitude of association between social stress and total fracture risk was greatest among the youngest group (aged 50-59 at baseline) and diminished with increasing age (Figure). Each one SD increase in social strain was associated with 17% greater total fracture risk among Native American women (HR = 1.17, 95% CI: 1.02-1.34), 9% among Latina/Hispanic women (HR = 1.09, 95% CI: 1.03-1.15), 7% among White women (HR = 1.07, 95% CI: 1.06-1.09), and 5% among Black women (HR = 1.05, 95% CI: 1.02-1.09).
Race & Age Play Key Role
These findings suggest that, among postmenopausal women in the US, stress from the social environment is associated with fractures differently within intersectional race and age groups. Interpreted with evidence from our previous BMD study, the biological pathway between social environments and fractures may be through stress, which decreased BMD. Our study found racial differences in social stress-associated fracture risk, with Native-American women at particularly high risk. While race is a social, and not a biological, categorization, systemic exclusion from advantaged social environments by race contributes to greater stress. Identifying population patterns of fracture incidence as biological expressions of social environments reveals potential for systemic intervention in fracture risk. The longitudinal study design established temporality in this pathway, with stress occurring before changes in bone outcomes, and aids in elucidating the complex pathophysiology of osteoporosis.
Fracture risk is multifactorial, and our study provides evidence that equitable access to healthy social environments, along with recommended guidelines, may lower risk of fractures among postmenopausal women. As an observational study, however, we cannot determine a cause-and-effect relationship. However, we hope these findings lead to informed understanding of social stress and its role in patterning health disparities.