Insights into associations between Life’s essential 8 and lung function from NHANES data

Baseline characteristics

This study comprised 10,400 participants, with an average age of 44 years, and males accounting for 48.75% of the cohort. Detailed demographic characteristics are provided in Table 1. Participants were categorized into three groups based on FEV1 Z-scores: Z1 representing normal lung function (n = 9,600), Z2 indicating mild lung function impairment (n = 618), and Z3 representing moderate to severe lung function impairment (n = 182). Differences between groups in age, race, poverty ratio, serum albumin, kidney function, vitamin D levels, chronic respiratory diseases, menopausal status and waist were all statistically significant (P < 0.001). In the LE8 health behavior domain, significant differences were found among the groups in HEI-2015 diet, physical activity, nicotine exposure, and sleep (P < 0.001). Notable characteristics of the Z3 cohort included a median age of 55 years, higher male prevalence (56.59%), older age distribution, increased proportion of Non-Hispanic Black individuals (46.70%), and a higher incidence of low-income participants (48.35%). The Z3 subgroup exhibited the lowest median scores in LE8 (53.50), health behaviors (47.50), and health factors (58.12). Additionally, the Z2/Z3 lung function impairment groups displayed lower median scores in diet (25.00), physical activity (0.00), and nicotine exposure (75.00) compared to the normal group; the Z3 group manifested the poorest sleep patterns (70.00). In the LE8 health factors domain, apart from no statistical difference in blood lipid levels among the groups (P = 0.533), significant differences were observed in BMI, blood glucose, and blood pressure scores (P < 0.001). The Z2/Z3 lung function impairment groups had lower blood glucose scores (60.00) compared to the normal group, and the Z3 group exhibited poorer blood pressure scores (52.50).

Association of LE8 score with lung function

As presented in Table 2, a linear regression model was employed to analyze the association between low, moderate, and high LE8 scores and lung function. Model 1 remained unadjusted, while Model 2 was adjusted for gender, age, race, and family poverty rate. Model 3 encompassed all covariates from Model 2 along with additional factors (albumin, ALT, AST, eGFR, vitamin D, alcohol, asthma, emphysema, chronic bronchitis and waist). In Model 1, devoid of covariate adjustments, the high LE8 score group exhibited a notable increase of 0.748 in FEV1 Z-score (95% CI 0.650–0.847, P < 0.001) compared to the low LE8 score group; similarly, the moderate group showed an increase of 0.498 in FEV1 Z-score (95% CI 0.413–0.583, P < 0.001). Upon adjusting for demographic factors (gender, age, race, family poverty ratio), the high score group displayed an adjusted increase of 0.552 in FEV1 Z-score (95% CI 0.454–0.650, P < 0.001) relative to the low score group, while the moderate group manifested an adjusted increase of 0.401 in FEV1 Z-score (95% CI 0.313–0.489, P < 0.001). In Model 3, after accounting for all covariates, the high score group demonstrated an adjusted increase of 0.413 in FEV1 Z-score compared to the low score group (95% CI 0.270–0.557, P < 0.001), whereas the moderate group exhibited an adjusted increase of 0.312 in FEV1 Z-score (95% CI 0.196–0.428, P < 0.001). The restricted cubic spline analysis reveals a significant non-linear relationship between LE8 total scores and lung function (P nonlinear < 0.001, Fig. 2A). Before the turning point at an LE8 score of 64.02, lung function increases significantly with higher LE8 scores. However, beyond this threshold, the improvement trend slows, indicating a plateau effect. In contrast, no non-linear relationship was observed between health behavior scores and lung function (P nonlinear = 0.883, Fig. 2B). Additionally, a significant non-linear association was found between health factor scores and lung function (P nonlinear < 0.001, Fig. 2C), with a turning point at 71.09. Before this point, improvements in health factors are strongly associated with enhanced lung function, while beyond it, the positive effects diminish but remain significant.

Nonlinear associations between LE8 (A), Health behaviors (B), Health Factors (C), and lung function.

Correlation analysis of LE8 subcomponents with lung function

In this study, linear regression analyses were performed to examine the association between the eight components of the LE8 score (HEI-2015 diet, physical activity, nicotine exposure, sleep, blood pressure, blood lipids, blood glucose, and BMI scores) and lung function (FEV1 Z-scores) under different adjustment models (Table 3). Among the LE8 components, higher HEI-2015 diet scores were significantly associated with improved lung function. In Model 3, the high diet score group demonstrated an increase of 0.225 (95% CI 0.046–0.404, P = 0.017) in FEV1 Z-scores compared to the low score group, highlighting the beneficial impact of dietary quality on pulmonary health. Physical activity scores also showed a significant positive association with lung function. In Model 3, the high PA score group exhibited a β coefficient of 0.131 (95% CI 0.035–0.227, P = 0.012), while the moderate PA group did not demonstrate statistical significance, suggesting that higher levels of physical activity contribute more substantially to lung function enhancement. Nicotine exposure scores, where higher values indicate reduced exposure risk, were strongly correlated with better lung function. The high nicotine exposure score group presented a significant increase of 0.380 (95% CI 0.281–0.480, P < 0.001) in FEV1 Z-scores in Model 3, indicating the protective role of minimized nicotine exposure against pulmonary impairment. Sleep quality also emerged as a significant factor. Participants with high sleep scores exhibited an increase of 0.165 (95% CI 0.062–0.267, P = 0.004) in FEV1 Z-scores compared to the low sleep score group, emphasizing the positive contribution of better sleep quality to lung health.

Blood glucose and pressure scores were positively associated with lung function in the fully adjusted model. The high blood glucose score group demonstrated an increase of 0.365 (95% CI 0.242–0.488, P < 0.001), while the high blood pressure score group showed a β coefficient of 0.159 (95% CI 0.055–0.263, P = 0.006). These findings underline the importance of managing metabolic and cardiovascular risk factors for maintaining lung function. In contrast, BMI scores, which reflect lower obesity risk with higher values, exhibited a complex and seemingly adverse association with lung function. In Model 3, both the moderate (β = −0.190, 95% CI −0.283 to −0.098, P = 0.001) and high (β = −0.491, 95% CI −0.632 to −0.349, P < 0.001) BMI score groups were negatively associated with FEV1 Z-score. This suggests that even with lower obesity risk, as indicated by higher BMI scores, there may still be a detrimental effect on lung function. It is worth noting that the inclusion of waist circumference data in the model could introduce collinearity, potentially influencing these results. Waist circumference, as an indicator of central obesity, may overlap with BMI in capturing adiposity-related health risks, warranting careful interpretation of these findings. Further analyses are needed to disentangle the independent effects of BMI and waist circumference on lung function. Blood lipids, however, showed no significant associations in the fully adjusted model. Although moderate and high scores were associated with slight positive trends in the unadjusted models, these associations were not maintained after covariate adjustments, highlighting a potentially limited role of blood lipids in lung function within this population.

Subgroup and sensitivity analysis

The subgroup analysis demonstrated a consistent positive association between LE8 scores and lung function across most subgroups (Fig. 3). No significant interactions were found for sex (P = 0.913), race (P = 0.842), and poverty ratio (P = 0.929), indicating that these factors did not significantly influence the relationship between LE8 and lung function. However, a significant interaction with age was observed (P = 0.011), with a stronger association in the 60–79 age group (β = 0.016, 95% CI 0.010–0.022) compared to the 40–59 age group (β = 0.013, 95% CI 0.010–0.017). Sensitivity analyses further confirmed the robustness of these findings (Table 4). After excluding participants with asthma, emphysema, and chronic bronchitis, the positive correlation between LE8 and lung function remained significant. In a fully adjusted model, the high LE8 group (β = 0.521, 95% CI 0.418–0.624, P < 0.001) and moderate LE8 group (β = 0.394, 95% CI 0.301–0.487, P < 0.001) showed significantly better lung function compared to the low LE8 group. After adjusting for menopause, these values remained significant, at 0.494 (95% CI 0.340–0.647, P < 0.001) and 0.389 (95% CI 0.239–0.539, P < 0.001), respectively, further confirming the robustness of the positive association between LE8 scores and lung function.

Subgroup analysis of LE8 and lung function.