The study utilized data from the NHANES, a program designed to assess the health and nutritional status of adults and children in the United States. For this analysis, the study population included participants from the 2005–2010, 2013–2014, 2017–2018 NHANES cycles. Eligible participants were those aged 50 years and older, a demographic typically at increased risk for osteoporosis.

To be included in the study, participants needed to have complete data on dietary omega-3 fatty acid intake from 24-h dietary recall interviews and BMD measurements from dual-energy X-ray absorptiometry (DXA) scans. Both of these components are crucial for determining the association between dietary omega-3 intake and osteoporosis.

Exclusion criteria were missing data on dietary intake, incomplete bone density information, or any medical conditions affecting bone metabolism (e.g., chronic renal failure, hyperparathyroidism) as reported in the NHANES questionnaire. This approach ensured that the analysis was conducted on a cohort with reliable data that could accurately reflect the association between omega-3 intake and osteoporosis risk.

Total dietary omega-3 intake in this study was calculated by summing the intakes of ALA (18:3n-3), DPA (22:5n-3), eicosatetraenoic acid (ETA, 20:4n-3), EPA (20:5n-3), and DHA (22:6n-3). Omega-3 component intake data were obtained from diet-related questionnaires using a 24-h dietary review method and the food frequency questionnaire (FFQ), which follows the automated multiple-pass method (AMPM) and enables the obtained nutritional intake data to differ from the actual intake by within 10%. Food and beverages consumed during the previous 24 h were asked to be recalled during the interview. During the first dietary assessment, professionally trained visitors interviewed the subjects and collected detailed diet information in the previous 24 h. After days 3–10, a telephone call was made to schedule the second visit. Dietary omega-3 intake was calculated as the average intake from the two 24-h dietary review surveys, and only the results of the first survey were used if the subject did not complete both surveys. To correct for potential biases arising from dietary recall errors, we applied regression calibration in our data analysis. Additionally, sensitivity analyses were conducted to assess the robustness of our findings under varying assumptions, thereby enhancing the reliability of the results.

The DXA was used to examine the BMD of lumbar spine and femoral neck. The mean BMD for the scanned anteroposterior length of L1 to L4 was computed and used for lumbar spine BMD reporting. Osteoporosis and osteopenia were defined in the present study using the WHO criteria. Specifically, osteoporosis was defined as a T-score ≤ −2.5 at either the femoral neck or the lumbar spine. Among those without osteoporosis, osteopenia was defined as those with T-scores between −2.5 and −1.0, and normal was defined as T-score ≥ −1. T-scores were calculated as (mean BMD respondent group—mean BMD reference group) / SD reference group. The reference group for calculation of the femoral neck consisted of 20–29 white females from the NHANES III report. As there is no internationally recommended reference group for the lumbar spine, the reference group consisted of 24 normal women from the National Institutes of Health.

Bone loss accelerates with aging, especially in menopausal women; 40% of US White women and 13% of US White men aged > 50 years will experience at least one clinically apparent fragility fracture in their lifetime. Therefore, we set 50 years as the age division for analysis.

The covariates were selected from those previously reported factors to influence osteoporosis. Age, race, gender, and body mass index (BMI) are examples of demographic characteristics. The socio-economic covariates include education level, marital status. Covariates on health-related behaviors: smoking and alcohol consumption. Variables related to medical comorbidities: serum calcium, Alanine Aminotransferase (ALT; U/L), Aspartate Aminotransferase (AST; U/L), hypertension, and hyperlipidemia. Missing data in covariates were filled by interpolation.

To summarize the characteristics of the study population, we applied basic descriptive statistics. Continuous variables were described using means and standard deviations, while categorical variables were summarized through frequencies and percentages. Group differences among normally distributed continuous variables, non-normally distributed continuous variables, and categorical variables were evaluated using one-way ANOVA, Kruskal-Wallis tests, and chi-square tests, respectively. The analysis accounted for stratification and weighting to reflect the complex sampling design of NHANES. The relationship between dietary omega-3 intake and osteoporosis risk was analyzed using multivariate logistic regression. Initial adjustments in Model 1 included age, sex, and race. Model 2 extended adjustments to education level and marital status. The comprehensive Model 3 further included variables such as smoking status, BMI, alcohol consumption, serum calcium levels, ALT, AST, hyperlipidemia, and hypertension. Trends in omega-3 intake across quartiles were assessed by treating the median intake of each quartile as a continuous predictor in the models.

Subgroup analyses investigated potential modifying effects of age, sex, and smoking status on the omega-3 and osteoporosis association, with interaction terms tested for statistical significance. Additionally, to mitigate the influence of data extremes, the relationship between omega-3 intake and osteoporosis risk was modeled using restricted cubic splines with knots at the 5th, 35th, 65th, and 95th percentiles, excluding the top and bottom 5% of the data distribution. All statistical procedures were conducted using R statistical software (version 4.3.2), with a significance threshold set at a p-value less than 0.05.

A total of 8,889 study participants were included in this study, of whom 516 (5.80%) were osteoporosis patients and 8,373 (94.2%) were non-osteoporosis patients (Figure 1). Participants aged under 50 years (n = 6,227), lack of BMD status (n = 16,315), lack of data on dietary omega-3 intake (n = 5,805), and lack of individuals with covariates (n = 7,971) were excluded.

Based on the bone health status of participants, Table 1 shows the baseline characteristics of these individuals. This cross-sectional study evaluated 8,889 individuals, stratified into three categories based on their bone density findings: normal bone density (n = 4,421), osteopenia (n = 3,952), and osteoporosis (n = 516). The average age progressively increased across these categories from 62.28 years in the normal group to 71.97 years in the osteoporosis group, demonstrating a clear association between advancing age and deteriorating bone health (p < 0.001).The analysis revealed that osteoporosis was predominantly found in females, who made up 80.0% of the osteoporosis category, compared to 20.0% of males, suggesting significant gender disparities in bone degradation risks (p < 0.001). Racial distribution showed that Non-Hispanic whites were the most affected by osteoporosis, followed by Hispanic, Non-Hispanic black, and other races, indicating varying susceptibilities among racial groups. Body mass index (BMI) analyses indicated that individuals with osteoporosis were more likely to have a BMI under 25 kg/m² (57.9%), a figure significantly higher than in other bone health categories (p < 0.001). Lifestyle behaviors also varied, with osteoporotic individuals less likely to consume alcohol (59.7%) compared to those with normal bone density or osteopenia (p < 0.001). A significant decrease in ALT and AST levels was observed in the osteoporosis group (p < 0.001). Clinical evaluations showed a decline in liver enzyme levels (ALT and AST) correlating with poorer bone health. However, serum calcium levels remained stable across all groups. Notably, conditions such as hypertension and hyperlipidemia were more prevalent in the osteoporosis group, underscoring their possible influence on bone health deterioration. Regarding dietary habits, omega-3 fatty acid intake was inversely related to bone health status, with the highest intake observed in the normal group (1.86 ± 1.08 g/d) and the lowest in the osteoporosis group (1.66 ± 0.97 g/d), suggesting a potential protective role of omega-3 intake against bone density loss (p < 0.001).

To explore the relationship between dietary Omega-3 intake and BMD, a linear regression analysis was performed. The results indicated a modest positive association between Omega-3 intake and BMD, suggesting that higher Omega-3 intake may be linked to slightly improved bone density. As shown in Figure 2, the scatter plot reveals a slight upward trend in BMD with increasing Omega-3 intake, with a regression line indicating the general trend and a 95% confidence interval capturing the variability. While this association was statistically significant (p < 0.05), the effect size was relatively small, implying that while Omega-3 intake might contribute to bone health, it is likely one of many factors involved in BMD regulation.

The association between dietary omega-3 intake and osteoporosis was rigorously analyzed using data from the NHANES dataset, focusing on a range of omega-3 intake levels divided into quartiles (Table 2). The findings indicate a statistically significant inverse relationship between higher omega-3 intake and the risk of developing osteoporosis, as evidenced by the odds ratios (ORs) across several adjusted models. In the simplest model (Model 1), which adjusted for age, sex, and race, participants in the highest quartile of omega-3 intake (Q4) exhibited a 32% reduction in the odds of osteoporosis compared to those in the lowest quartile (Q1) with an OR of 0.68 (95% CI: 0.52–0.89). This trend was consistent and statistically significant across more comprehensively adjusted models: Model 2, which included adjustments for education level and marital status, showed a similar pattern, with the highest quartile having a 30% reduction in osteoporosis risk (OR = 0.70, 95%CI: 0.54–0.92).Model 3, the most adjusted model incorporating additional factors such as smoking, BMI, alcohol consumption, and various health indicators like serum calcium, liver enzymes (ALT, AST), hyperlipidemia, and hypertension, still showed a protective effect of high omega-3 intake with an OR of 0.71 (95% CI:0.53–0.93).A significant p-value for trend (<0.05) across the quartiles in all models suggests a dose–response relationship, where increasing omega-3 intake is associated with progressively lower odds of osteoporosis. Further explorations using restricted cubic splines indicated (Figure 3) a nonlinear relationship between omega-3 intake and osteoporosis risk, though this nonlinear component was not statistically significant (P for nonlinear = 0.366). However, the overall effect of omega-3 intake on reducing osteoporosis risk remained significant (P for overall = 0.046).

In a comprehensive subgroup analysis using data from NHANES, the relationship between dietary omega-3 intake and osteoporosis was assessed across various demographic and lifestyle factors, including age, sex, race, education, marital status, alcohol intake, smoking status, BMI, hypertension, and hyperlipidemia (Table 3). The analysis, detailed in Model 3, controlled for all variables except the primary stratification factor in each subgroup. Results indicated no significant variations in the association between omega-3 intake and osteoporosis risk across most subgroups (p > 0.05 for interaction), suggesting that the protective effects of omega-3 were consistent regardless of sex, race, marital status, and other lifestyle factors. However, age stood out as a significant modifier of the omega-3 and osteoporosis relationship. Younger participants under 60 showed a marked decrease in osteoporosis risk with higher omega-3 intake, whereas this association did not hold for those aged 60 and above, where increased omega-3 intake did not translate to reduced osteoporosis risk (p < 0.001 for age interaction). This indicates that the beneficial effects of omega-3 on bone health may be more pronounced in younger populations, highlighting the importance of early dietary intervention.