Baseline Characteristics of Participants
Table 1 shows the baseline characteristics of participants, categorized by their bone health. The NHANES cross-sectional study analyzed a total of 5,548 individuals aged 50 and above, focusing on demographic and clinical profiles to discern differences between those with and without osteoporosis. Among these participants, 313 were diagnosed with osteoporosis.
The average age of participants was 65.18 years, with a significant age difference between those without osteoporosis (64.69 years) and those with osteoporosis (73.51 years) (P < 0.001). Gender composition also varied markedly; females constituted 79.55% of the osteoporosis group compared to 44.83% of the non-osteoporotic group (P < 0.001).
Ethnic distribution revealed that individuals who were not of Hispanic descent had a higher likelihood of having osteoporosis, accounting for 70.93% of those impacted as opposed to 56.06% in the group without osteoporosis (P < 0.001). Socio-economic factors also differed, with a higher percentage of individuals with osteoporosis living below the poverty line. Moreover, more than half (51.12%) of individuals diagnosed with osteoporosis were unmarried or had lost a spouse, in contrast to 28.79% in the comparison group (P < 0.001).
In terms of health, individuals with osteoporosis had an average NHHR level of 2.66, which was lower than the healthier group's average of 3.04 (P < 0.001), indicating a potential link between lower NHHR levels and increased osteoporosis risk. BMI values further highlighted differences: a significant majority (64.54%) of those with osteoporosis had a BMI under 25, indicating a correlation between lower BMI and osteoporosis prevalence (P < 0.001). Differences were also observed in biochemical markers such as ALT and HDL-C, underscoring the complex nature of osteoporosis risks and impacts among the elderly. This comprehensive analysis reveals the intricate connections between demographic, socio-economic, and health variables in influencing the susceptibility to osteoporosis in the elderly.
Table 1
Characteristics of participants enrolled in NHANES cross-sectional study
Characteristic | Total (N = 5548) | Non-osteoporosis (N = 5235) | Osteoporosis (N = 313) | P value |
Age | 65.18 ± 9.81 | 64.69 ± 9.65 | 73.51 ± 8.69 | < 0.001 |
Gender | | | | < 0.001 |
Male | 2952 (53.21) | 2888 (55.17) | 64 (20.45) | |
Female | 2596 (46.79) | 2347 (44.83) | 249 (79.55) | |
Race | | | | < 0.001 |
Hispanic | 822 (14.82) | 792 (15.13) | 30 (9.58) | |
Non-Hispanic white | 3157 (56.90) | 2935 (56.06) | 222 (70.93) | |
Non-Hispanic black | 987 (17.79) | 965 (18.43) | 22 (7.03) | |
Other | 582 (10.49) | 543 (10.37%) | 39 (12.46) | |
Level of education | | | | 0.080 |
Under high school | 1573 (28.35) | 1467 (28.02) | 106 (33.87) | |
High school or equivalent | 1392 (25.09) | 1319 (25.20) | 73 (23.32) | |
Above high school | 2583 (46.56) | 2449 (46.78) | 134 (42.81) | |
Marital status | | | | < 0.001 |
Never married | 375 (6.76) | 355 (6.78) | 20 (6.39) | |
Married or living with partner | 3506 (63.19) | 3373 (64.43) | 133 (42.49) | |
Divorced, separated, or widowed | 1667 (30.05) | 1507 (28.79) | 160 (51.12) | |
Ratio of family income to poverty level | | | | < 0.001 |
<1.0 | 854 (15.39) | 791 (15.11) | 63 (20.13) | |
1.0 ~ 2.9 | 2375 (42.81) | 2208 (42.18) | 167 (53.35) | |
≥3.0 | 2319 (41.80) | 2236 (42.71) | 83 (26.52) | |
BMI(kg/m2) | | | | < 0.001 |
< 25 | 1505 (27.13) | 1303 (24.89) | 202 (64.54) | |
25 ~ 29 | 1738 (31.33) | 1666 (31.82) | 72 (23.00) | |
≥ 30 | 2305 (41.55) | 2266 (43.29) | 39 (12.46) | |
ALT (U/L) | 24.86 ± 19.15 | 25.13 ± 19.35 | 20.23 ± 14.56 | < 0.001 |
AST (U/L) | 26.75 ± 14.22 | 26.83 ± 14.44 | 25.45 ± 9.86 | 0.095 |
Serum calcium (mg/dL) | 9.46 ± 0.38 | 9.46 ± 0.38 | 9.46 ± 0.42 | 0.840 |
Alcohol consumption | | | | < 0.001 |
Yes | 3867 (69.70) | 3698 (70.64) | 169 (53.99) | |
No | 1681 (30.30) | 1537 (29.36) | 144 (46.01) | |
Smoking | | | | 0.002 |
Yes | 2974 (53.60) | 2833 (54.12) | 141 (45.05) | |
No | 2574 (46.40) | 2402 (45.88) | 172 (54.95) | |
Hypertension | | | | 0.030 |
Yes | 3628 (65.39) | 3405 (65.04) | 223 (71.25) | |
No | 1920 (34.61) | 1830 (34.96) | 90 (28.75) | |
Hyperlipidemia | | | | 0.673 |
Yes | 953 (17.18) | 896 (17.12) | 57 (18.21) | |
No | 4595 (82.82) | 4339 (82.88) | 256 (81.79) | |
Diabetes | | | | 0.107 |
Yes | 1373 (24.75) | 1308 (24.99) | 65 (20.77) | |
No | 4175 (75.25) | 3927 (75.01) | 248 (79.23) | |
HDL-C (mg/dL) | 53.96 ± 16.39 | 53.62 ± 16.25 | 59.66 ± 17.68 | < 0.001 |
Total Cholesterol (mg/dL) | 201.3 ± 42.92 | 201.27 ± 42.86 | 201.89 ± 43.95 | 0.800 |
NHHR | 3.02 ± 1.42 | 3.04 ± 1.42 | 2.66 ± 1.32 | < 0.001 |
Mean ± SD for continuous variables: the P value was calculated by the weighted linear regression model.
N (%) for categorical variables: the P-value was calculated by the weighted chi-square test.
Association between NHHR and Osteoporosis
The relationship between the NHHR and osteoporosis was rigorously analyzed using logistic regression and RCS methods across NHHR quartiles in a NHANES dataset. Table 2 categorizes participants into four NHHR quartiles: Q1 (0.31–2.04), Q2 (2.04–2.74), Q3 (2.74–3.74), and Q4 (3.74–24.56), with Q1 serving as the reference group.
As NHHR increased across the quartiles, the ORs for osteoporosis decreased, indicating an inverse relationship between NHHR and osteoporosis risk. In Model 1, after accounting for age, gender, and ethnicity, displayed a decline in ORs from Q2 to Q4, where Q4 had an OR of 0.68 (95% CI: 0.47–0.67, P = 0.014).
In Model 2, further modifications for educational background, marital status, and income ratios showed a more robust connection, where Q4 had an odds ratio of 0.63 (95% CI: 0.44–0.90, P = 0.004).
In Model 3, which adjusted for BMI, smoking, alcohol use, serum calcium, liver enzymes, and other health conditions, the protective effect of higher NHHR quartiles was not observed, with Q4 showing an OR of 0.96 (95% CI: 0.63–1.43, P = 0.669), indicating no significant trend.
Table 2
Odds Ratios for associations between NHHR and osteoporosis
Model | Quartiles of NHHR | P for trend |
| Q1 | Q2 | Q3 | Q4 | |
Model 1 | Ref | 0.84 (0.61–1.13) | 0.72 (0.51–1.01) | 0.68 (0.47–0.67) | 0.014* |
Model 2 | Ref | 0.82 (0.60–1.11) | 0.70 (0.49–0.98) | 0.63 (0.44–0.90) | 0.004* |
Model 3 | Ref | 0.97 (0.67–1.34) | 0.93 (0.64–1.33) | 0.96 (0.63–1.43) | 0.669 |
Model 1 was adjusted for age, gender, and ethnicity. Model 2 was adjusted for age, gender, ethnicity, educational level, marital status, and the proportion of household income compared to the poverty threshold. Model 3 was adjusted for various demographic and health factors including age, gender, race, education, marital status, income, body mass index, smoking and drinking consumption, serum calcium, ALT, AST, diabetes, hyperlipidemia, hypertension. *P < 0.05
As shown in Fig. 2, the estimated risk of osteoporosis is shown by the solid line, with the fitted 95% CI represented by the shaded area. The analysis demonstrates a significant overall relationship (P for overall < 0.001) and a non-linear trend (P for nonlinear = 0.008) between NHHR and osteoporosis risk
Subgroup Analysis of the Association between NHHR and Osteoporosis
Figure 3 illustrates the subgroup analysis, indicating that the protective effect of higher NHHR against osteoporosis is more pronounced among certain subgroups. Specifically, the forest plot reveals that higher NHHR is associated with a significantly reduced risk of osteoporosis [OR: 0.77 (95% CI: 0.68–0.86)]. The analysis indicates a stronger protective effect in males [OR: 0.72 (95% CI: 0.56–0.92)] compared to females, although the interaction by gender is not statistically significant (P = 0.163). Among the elderly (≥ 60 years), the effect is also significant [OR: 0.82 (95% CI: 0.72–0.92)], suggesting that age might enhance the protective benefits of higher NHHR (P = 0.66). Non-smokers also have a lower risk of osteoporosis than smokers [OR: 0.82 (95% CI: 0.70–0.96)], and drinkers show a protective effect [OR: 0.68 (95% CI: 0.59–0.80)] with a significant interaction (P = 0.025). These findings underscore the importance of considering lifestyle factors, as the benefits of higher NHHR are more noticeable in males, elder, non-smokers, and drinkers.
This forest plot illustrates the subgroup analysis, showing the protective effect of higher NHHR against osteoporosis in various demographic and clinical subgroups. The analysis includes adjustments for age, gender, BMI, smoking status, alcohol consumption, hypertension, diabetes, and hyperlipidemia. The stratified variable is omitted from the model.