The gender difference in the effects of air pollution on the risk of spinal osteoarthritis in Chinese middle-aged and older adults: a prospective cohort study in China

Osteoarthritis (OA) is a leading contributor to musculoskeletal disability worldwide, with an increasing prevalence driven in part by population aging and lifestyle changes (1, 2). Among its various forms, spinal OA is particularly concerning due to chronic back pain and disability, which can significantly impair quality of life in middle-aged and older adults (3). In China, the burden of OA has escalated alongside rapid socio-economic development, prompting a need to clarify its etiological risk factors and identify potential avenues for prevention (4, 5).

Air pollution has emerged as an important environmental determinant of health, implicated in cardiovascular, respiratory, and metabolic disorders (6). Accumulating evidence suggests that exposure to air pollutants such as particulate matter (PM) may also play a role in the onset or progression of joint diseases (7). For instance, a large cohort study in China linked prolonged PM_2.5 exposure to accelerated knee OA progression, underlining a possible pro-inflammatory and oxidative stress-mediated mechanism (8–10). Although investigations have begun to elucidate the relationship between air pollution and peripheral joint OA, fewer studies have addressed spinal OA, despite its substantial prevalence and impact on functional independence (1, 11).

Gender differences in OA incidence, disease severity, and risk factor profiles have been reported, with women generally exhibiting higher susceptibility and more severe structural changes (12, 13). Nonetheless, whether these gender disparities extend to air pollution-related spinal OA remains unclear. Biological factors such as hormonal regulation and differential immune responses could modulate inflammatory processes in joint tissues, potentially altering susceptibility to environmental stressors (14–16). Moreover, behavioral and occupational exposures may differ by gender, further modifying the relationship between air pollutants and spinal OA risk (17, 18).

Given the paucity of research on how air pollution influences spinal OA in Asian populations—and particularly how this association may vary by gender—we conducted a prospective cohort study using nationally representative data from middle-aged and older Chinese adults. By capturing longitudinal exposure to multiple air pollutants (PM₁, PM_2.5, PM₁₀, O₃, and NO₂), along with detailed demographic and health information, we aimed to determine the risk of spinal OA associated with ambient air pollution exposure, and assess potential differences in vulnerability between men and women. Our findings seek to inform targeted public health interventions that could mitigate OA-related disability in rapidly urbanizing settings.

Table 1 illustrates the baseline characteristics of the 7,663 participants stratified by spinal osteoarthritis status (no spinal osteoarthritis vs. spinal osteoarthritis). Of these individuals, 6,107 (79.7%) did not report spinal osteoarthritis, whereas 1,556 (20.3%) had spinal osteoarthritis. Compared to participants free of spinal osteoarthritis, those with spinal osteoarthritis were more likely to be female, older, hold a technical school or higher education, be unmarried, live in urban areas, be former smokers, abstain from alcohol, have UEMI/URRMI/URMI medical insurance, fall into the lowest income category, be unemployed, and engage in less physical activity (Table 1).

Based on the Kolmogorov–Smirnov test, the annual average concentrations of the five pollutants (PM₁, PM_2.5, PM₁₀, O₃, and NO₂) deviated from a normal distribution (p < 0.05). Figure 3 shows the annual average concentration of air pollutants by spinal osteoarthritis status. Hence, they are presented as median and interquartile range. Following a median follow-up of 7 years (IQR: 4–9 years), a total of 1,556 participants were classified as having spinal osteoarthritis. After adjusting for all potential confounders (Model IV), each 10 μg/m³ increment in annual mean exposure to PM₁, PM2.5, PM₁₀, O₃, and NO₂ was associated with a 13.8% (HR = 1.138, 95% CI 1.085–1.195), 6.8% (HR = 1.068, 95% CI 1.043–1.093), 5.1% (HR = 1.051, 95% CI 1.036–1.067), 1.1% (HR = 1.011, 95% CI 0.983–1.041), and 17.4% (HR = 1.174, 95% CI 1.113–1.239) higher risk of developing spinal osteoarthritis, respectively (Table 2). Figure 4 presents the restricted cubic spline plots for the association between the annual concentrations of each air pollutant and the risk of spinal osteoarthritis. The sensitivity analysis for two-years lag effect shows that there was no significant difference with the results of main analysis (Appendix).

Table 3 displays the outcomes for models incorporating multiple pollutants. Under single-pollutant conditions, O₃ showed no discernible link to spinal osteoarthritis, and this pattern persisted even after accounting for additional pollutants. In contrast, the associations of the remaining pollutants with spinal osteoarthritis held steady when O₃ was included, each with a slightly increased effect size. Once NO₂ was introduced, the relationships involving PM₁ and PM_2.5 ceased to be statistically significant, whereas PM₁₀ exhibited the opposite trend. Moreover, NO₂‘s connection to frailty remained largely independent of the other pollutants.

The interaction tests for PM₁, PM_2.5, PM₁₀ and NO₂ with gender were significant (all P for interaction less than 0.05), and only the O₃ was not significant (P for interaction: 0.19). Further subgroup analysis showed that for male participants, the association between PM₁, PM_2.5, PM₁₀ and NO₂ exposure and the spinal osteoarthritis risk was not significant (p > 0.05). In contrast, for female subgroup, PM₁, PM_2.5, PM₁₀ and NO₂ exposure significantly increased the risk of spinal osteoarthritis by 12.9% (HR = 1.129, 95% CI 1.071–1.189), 6.0% (HR = 1.060, 95% CI 1.032–1.088), 5.0% (HR = 1.050, 95% CI 1.034–1.066), 17.0% (HR = 1.170, 95% CI 1.103–1.241) (Figure 5).

Our prospective study of 7,663 community-dwelling Chinese adults has demonstrated a positive relationship between elevated ambient concentrations of PM₁, PM_2.5, PM₁₀, and NO₂ and increased susceptibility to spinal osteoarthritis over a median follow-up of 4 years. In contrast, O₃ exposure displayed a weaker correlation with spinal OA onset. Notably, participants with spinal osteoarthritis were more likely to be older, female, living in urban areas, and less physically active, suggesting that both demographic and behavioral factors may synergistically exacerbate the detrimental effects of airborne pollutants. These findings offer a nuanced perspective on the environmental determinants of musculoskeletal health, extending beyond the conventional emphasis on cardiovascular and respiratory diseases.

Recent epidemiological and experimental evidence supports the plausibility of pollutant-induced degenerative changes in joint tissues. In particular, mounting data suggest that fine particulate matter and traffic-related emissions may provoke systemic oxidative stress, chronic low-grade inflammation, and impaired microvascular function (21, 22). Our results are congruent with those of one study (3), who reported that increased exposure to PM_2.5 and PM₁₀ was associated with a heightened prevalence of osteoarthritis among Chinese middle-aged and older adults. Additionally, another study identified PM_2.5 as a potential risk factor for rapid knee OA progression, proposing that deposited particulate matter within the joint environment could contribute to cartilage breakdown (23, 24). Although fewer studies have focused specifically on spinal OA, our findings highlight its potential susceptibility to similar inflammatory pathways. Future mechanistic investigations—potentially using imaging modalities or biomarkers of inflammation—could further elucidate how airborne toxins drive molecular changes in intervertebral discs and facet joints (25, 26).

A key observation in our analysis is the notable gender disparity, with female participants appearing more vulnerable to pollution-related joint deterioration. This aligns with previous studies based on the biological and behavioral and socio-environmental perspectives (27, 28). From a biological standpoint, hormonal regulation and immunological factors in women may contribute significantly to this heightened susceptibility. Estrogen, for instance, has been shown to exert protective effects on cartilage by modulating inflammatory cytokines (e.g., IL-1β, TNF-α) and promoting collagen synthesis. After menopause, the decline in estrogen levels may create a more pro-inflammatory and oxidative environment, potentially exacerbating the adverse impact of environmental toxins on joint structures (29–31). Furthermore, emerging research suggests that women could exhibit distinct immunological responses to particulate matter, including greater expression of reactive oxygen species and inflammatory mediators, which could accelerate cartilage and subchondral bone damage (32, 33).

In addition to these biological factors, behavioral and socio-environmental dimensions may also contribute to women’s higher susceptibility. In many cultures, older women often assume more indoor responsibilities, such as cooking, which can increase exposure to household air pollutants (e.g., from biomass fuels or inadequate ventilation) (31). Occupational factors may also play a role: women in certain urban service or informal jobs may face prolonged outdoor or semi-enclosed exposures to traffic-related pollutants, especially if protective measures are lacking. Moreover, caregiving and domestic duties can limit leisure-time physical activity, reducing muscular support around the spine and potentially heightening the detrimental effects of pollution on weight-bearing structures (32, 34).

Despite the robustness of our longitudinal design and adjustment for a wide array of demographic, socioeconomic, and lifestyle covariates, several limitations warrant discussion. First, our reliance on city-level pollutant data might underestimate inter-individual variation in exposure, as local traffic patterns or building environments could produce marked differences in actual inhaled doses (35). Efforts to incorporate personal monitoring devices or satellite-based micro-scale estimations may refine future risk assessments (36, 37). Second, the identification of spinal osteoarthritis was based on self-reported physician diagnoses rather than imaging or clinical examination, raising concerns about potential misclassification—particularly in early or asymptomatic cases (16, 38). Third, while we included known confounders, unmeasured factors such as dietary habits, vitamin D status, or coexisting metabolic diseases may also influence osteoarthritis outcomes. Lastly, the omission of multi-pollutant modeling means we have a limited view of interactive or additive effects among different pollutants (35, 39, 40).

From a public health perspective, our findings underscore the urgent need for policies and interventions aimed at reducing ambient concentrations of PM and NO₂ in rapidly urbanizing regions. The link between these pollutants and spinal OA risk represents an underrecognized domain of morbidity that may add to healthcare burdens, given the chronic and disabling nature of back pain. In particular, more focused interventions on traffic-related pollutants could involve the implementation of low-emission zones, stricter tailpipe standards, catalytic converter mandates, and congestion-pricing policies to curb vehicular emissions. Additionally, expanding community-based air quality monitoring networks in heavily polluted districts would allow for more precise identification of high-risk areas and facilitate timely public health alerts. Such efforts should be accompanied by improved public transit systems and the strategic development of green spaces, which not only reduce pollution but also encourage physical activity (41, 42). Furthermore, heightened surveillance and targeted prevention initiatives—especially for older women—could mitigate pollution’s detrimental impact on skeletal health. These programs might incorporate routine musculoskeletal screenings in clinics situated in high-emission neighborhoods, along with educational campaigns to increase awareness of air pollution’s potential role in joint degeneration. As many older women may have additional exposures from household cooking or occupational settings, multi-sector collaboration is crucial to minimize total pollutant loads. Investigating the joint influences of occupational exposures, household air pollution, and broader environmental factors remains critical in developing effective, context-specific strategies to preserve spinal health in aging societies (43, 44). Future studies should also incorporate advanced clinical imaging techniques (e.g., MRI, CT, or cartilage compositional imaging) and molecular profiling (e.g., inflammatory biomarkers, epigenetic analyses) to elucidate the precise pathophysiological mechanisms by which air pollutants may accelerate degenerative changes in the spine. Moreover, continued exploration of these specific mechanisms using more diverse data sources—such as wearable monitoring devices, high-resolution regional pollution metrics, and longitudinal biobanks—will be essential to further validate and expand upon our current research findings, ultimately guiding more targeted and effective public health interventions.

In this nationwide longitudinal analysis, exposure to elevated levels of PM₁, PM_2.5, PM₁₀, and NO₂ was consistently linked to a higher incidence of spinal osteoarthritis, while O₃ demonstrated a comparatively weaker association. Our data also indicate that women may be especially sensitive to the harmful effects of air pollution on spinal structures. These observations contribute to a growing body of literature suggesting that environmental pollutants, long implicated in cardiopulmonary conditions, also pose significant risks to musculoskeletal integrity. By pinpointing high-risk populations and key pollutant sources, policymakers and clinicians can collaborate on interventions—ranging from emission controls to health education campaigns—that may reduce the overall burden of osteoarthritis and promote healthier aging. Further research integrating advanced exposure assessment, clinical imaging, and molecular profiling is warranted to fully characterize the mechanisms linking air pollutants to spinal joint degeneration.