Baseline characteristics of the study participants

The baseline characteristics of the study participants are shown in Table 1. A total of 11,324 individuals with obesity were included (weighted population: 456,457,366, median age 49.00 years [IQR, 35.00–61.00 years]). Of these, 47.48% were male, and 52.52% were female. The racial/ethnic composition was as follows: 10.34% were Mexican American, 65.5% were non-Hispanic White, 13.49% were non-Hispanic Black, and 10.68% were classified as “other” (including Hispanic, American Indian/Alaska Native, Pacific Islander, Asian, and multiracial). Regarding education, 58.48% of participants had completed college or higher, and 63.74% were married or living with a partner. Figure 1 illustrates the trends in the prevalence of depression and obesity among adults aged 18 years and older in the United States. From 2007–2008 to 2017–2018, the prevalence of PHQ-9 scores ≥ 10 ranged from 10.1% to 13.6%. Over this period, the prevalence of Class II obesity increased from 24.7% to 26.2%, while the prevalence of Class III obesity rose from 16.8% to 21.1% (Fig. 2).

Table 1 Weighted baseline characteristics of the study participantsFig.2

Trends in prevalence of depressive symptoms and obesity categories among US population, National Health and Nutrition Examination Survey 2007–2018

Participants were categorized into 3 groups based on obesity class: Class I (BMI < 35 kg/m2; n = 6,196), Class II (BMI 35–39.9 kg/m2; n = 2,909), and Class III (BMI ≥ 40 kg/m2; n = 2,219). Compared with those in Class I, individuals in Class III were younger, more likely to be female, non-Hispanic Black, unmarried, and had lower socioeconomic status. They also had significantly higher depression scores, engaged in less physical activity, had lower physical activity metabolic equivalent of task (MET) values, and lower hemoglobin levels. Furthermore, they exhibited higher counts of white blood cells, lymphocytes, monocytes, neutrophils, and platelets, along with elevated NLR, SIRI, and SII. Smoking, diabetes, and hypertension were also more prevalent in this group (all P < 0.05).

Association between obesity and the risk of depression

We employed multivariable logistic regression models to assess the associations between obesity and the risk of depressive symptoms, as presented in Table 2. When BMI was analyzed as a continuous variable, BMI was significantly associated with depressive symptoms (Total: OR = 1.02, 95% CI = 1.01–1.03, P < 0.001; Male: OR = 1.03, 95% CI = 1.01–1.05, P = 0.002; Female: OR = 1.02, 95% CI = 1.01–1.03, P < 0.001) after adjusting for multiple covariates, including age, sex, race, diabetes mellitus, hypertension, poverty income ratio, and serum creatinine levels. Further analysis stratified participants into three groups based on BMI classification: Class I obesity, Class II obesity, and Class III obesity, with Class I obesity serving as the reference group. In the unadjusted model, the odds ratios (ORs) and 95% confidence intervals (CIs) for depressive symptoms were 1.17 (1.02–1.34) for Class II obesity and 1.61 (1.40–1.86) for Class III obesity compared with Class I obesity. In the fully adjusted model (Model III), after accounting for potential confounders, the adjusted ORs (95% CIs) were 1.02 (0.88–1.18) for Class II obesity and 1.30 (1.12–1.51) for Class III obesity, compared with Class I obesity. Next, we aim to investigate the association between BMI and depressive symptoms, stratified by sex. For both sexes, higher BMI as a continuous variable is significantly associated with increased odds of depressive symptoms, with odds ratios (ORs) slightly above 1 across all models and p-values less than 0.001. When BMI is categorized into Obesity Class I, II, and III, distinct patterns emerge. In males, those in Obesity Class III (BMI ≥ 40 kg/m2) show a significant increase in the odds of depressive symptoms across all models, with ORs ranging from 1.35 to 1.53, indicating a clear relationship between higher obesity levels and depression risk, while the association is weaker in Class II obesity. A similar pattern is seen in females, particularly in Obesity Class III, where the odds of depressive symptoms are significantly higher, with ORs ranging from 1.26 to 1.44. However, for Obesity Class II, the relationship is weaker and not consistently significant. In summary, higher BMI, particularly in Obesity Class III, is associated with an increased risk of depressive symptoms, with a stronger association observed in males compared to females. These findings suggest that as BMI increases, particularly in higher obesity classes (e.g., Class III obesity), the risk of depressive symptoms significantly increases. This trend was observed consistently in both male and female participants. The consistent results across all models highlight the robustness of the association between obesity and depressive symptoms, with the strength of this association increasing in parallel with the severity of obesity.

Table 2 The association between obesity and depressive symptomsAssociation between obesity and systemic inflammation markers

Table 3 demonstrate the association between BMI and inflammatory markers (NLR, SIRI, and SII) in the obese population. Across all markers, BMI as a continuous variable is positively correlated with higher inflammatory levels. For NLR, each 1-unit increase in BMI is associated with a 0.01 to 0.02 unit increase in NLR (P < 0.001). In the categorical analysis, Class III obesity (BMI ≥ 40 kg/m2) shows a significant elevation in NLR, with adjusted β values ranging from 0.19 to 0.26 (P < 0.001). Both males and females demonstrate a positive correlation between BMI and NLR, though the association is weaker and not significant in females with Class II obesity (P = 0.124). Similarly, for SIRI, a 1-unit increase in BMI is linked to a 0.01 to 0.02 unit rise in SIRI (P < 0.001). Class III obesity is associated with a significant increase in SIRI, with β values between 0.18 and 0.26 (P < 0.001). While both sexes exhibit a positive relationship between BMI and SIRI, the association is weaker and non-significant in females with Class II obesity (P = 0.154). For SII, BMI is significantly associated with elevated SII levels (P < 0.001). In the categorical analysis, Class III obesity shows a marked increase in SII, with β values ranging from 75.03 to 78.95 in males and 56.05 to 61.42 in females (P < 0.001). Overall, the higher the BMI, the greater the increase in SII, with stronger effects observed in individuals with higher obesity grades. In summary, higher BMI is consistently associated with increased levels of systemic inflammatory markers (NLR, SIRI, and SII), particularly in individuals with Class III obesity, with more pronounced effects seen in males. These findings highlight the robust link between obesity and systemic inflammation.

Table 3 The association between obesity and inflammatory markersAssociation of systemic inflammation markers with the risk of depression

Table 4 shows the association between inflammatory markers (NLR, SIRI, SII) and depressive symptoms in the obese population. In the total population, each 1-unit increase in NLR and SIRI is associated with an 8% (OR = 1.08, 95% CI = 1.02–1.14, P = 0.005) and 12% (OR = 1.12, 95% CI = 1.05–1.20, P = 0.001) higher risk of depression, respectively, while each 100-unit increase in SII raises depression risk by 3% (OR = 1.03, 95% CI = 1.01–1.05, P = 0.001). In males, the associations are stronger: NLR increases depression risk by 17% (OR = 1.17, 95% CI = 1.08–1.26, P < 0.001), SIRI by 19% (OR = 1.19, 95% CI = 1.08–1.31, P < 0.001), and SII by 6% (OR = 1.06, 95% CI = 1.03–1.10, P < 0.001). In females, these associations are weaker and not statistically significant for NLR (OR = 1.02, 95% CI = 0.95–1.10, P = 0.526), SIRI (OR = 1.09, 95% CI = 0.99–1.19, P = 0.067), and SII (OR = 1.01, 95% CI = 0.99–1.04, P = 0.2). In summary, systemic inflammatory markers are more strongly associated with depression risk in males, while the associations in females are weaker or non-significant.

Table 4 The association between inflammatory markers and depressive symptoms in obese populationRestricted Cubic Spline Analysis

We used restricted cubic spline (RCS) curves to evaluate potential nonlinearity in the association between BMI and the risk of depressive symptoms in the total population, as well as in males and females separately, as illustrated in Fig. 3. The results showed that higher BMI is significantly associated with an increased risk of depressive symptoms across all groups, particularly at BMI levels above 40 kg/m2. (Total: P = 0.003, P-nonlinear = 0.388; Male: P < 0.001, P-nonlinear = 0.427; Female: P = 0.002, P-nonlinear = 0.092). The association is mostly linear across groups, with a steady increase in the odds of depressive symptoms as BMI rises.

Fig.3

The restricted cubic spline (RCS) analysis between BMI and the risk of depression in obese population

We also applied RCS curves to assess the association between BMI and systemic inflammation markers among obese individuals. Figure 4 demonstrate a positive association between BMI and systemic inflammatory markers (NLR, SIRI, and SII) across the total population, males, and females. In all groups, increasing BMI is significantly correlated with higher levels of each marker (P < 0.001). The relationship between BMI and NLR is linear, with NLR increasing steadily as BMI rises (Overall: P < 0.001, P-nonlinear = 0.055; Male: P < 0.001, P-nonlinear = 0.193; Female: P < 0.001, P-nonlinear = 0.213). For SIRI, the pattern is similar, with a consistent increase in SIRI levels as BMI increases (Overall: P < 0.001, P-nonlinear = 0.057; Male: P < 0.001, P-nonlinear = 0.576; Female: P < 0.001, P-nonlinear = 0.120). For SII, the relationship is particularly strong, showing a steep linear increase with BMI across all groups (Overall: P < 0.001, P-nonlinear = 0.440; Male: P < 0.001, P-nonlinear = 0.154; Female: P < 0.001, P-nonlinear = 0.485). These findings suggest a consistent and significant link between higher BMI and increased systemic inflammation, as indicated by NLR, SIRI, and SII, across both sexes.

Fig.4

The restricted cubic spline (RCS) analysis between BMI and systemic inflammation markers (NLR, SIRI, SII) in obese population

We further investigated the potential nonlinear associations between systemic inflammation markers and depression risk in obese individuals using RCS models. Figure 5 show the associations between three inflammatory markers—NLR, SIRI, and SII—and the odds of depressive symptoms across the total population, males, and females. For all three markers, higher levels are significantly associated with an increased risk of depressive symptoms. NLR shows a linear association across all groups (Overall: P < 0.001, P-nonlinear = 0.935; Male: P < 0.001, P-nonlinear = 0.066; Female: P = 0.088, P-nonlinear = 0.033), while SIRI demonstrates a consistent positive trend, particularly in males and the total population, with females showing a weaker but still positive association (Overall: P < 0.001, P-nonlinear = 0.490; Male: P < 0.001, P-nonlinear = 0.144; Female: P = 0.056, P-nonlinear = 0.055). SII is also strongly associated with increased odds of depressive symptoms, especially in females, where the relationship appears more nonlinear (Overall: P < 0.001, P-nonlinear = 0.376; Male: P < 0.001, P-nonlinear = 0.546; Female: P = 0.001, P-nonlinear < 0.001). Overall, these findings suggest that elevated levels of systemic inflammation, as indicated by NLR, SIRI, and SII, are consistently linked to a higher risk of depressive symptoms across both sexes and the general population.

Fig.5

The restricted cubic spline (RCS) analysis between systemic inflammation markers (NLR, SIRI, SII) and the risk of depression in obese population

Subgroup Analysis

Stratified analyses and interaction tests were conducted to evaluate the association between obesity severity and depression risk across subgroups defined by sex, age (< 65 years vs. ≥ 65 years), race, marital status, education level, alcohol consumption, diabetes, hypertension, smoking status, coronary heart disease, and chronic obstructive pulmonary disease (COPD), as shown in Fig. 6. The results demonstrated that higher BMI was consistently associated with an increased risk of depression across all subgroups. Interaction tests (P for interaction) revealed no significant interactions between BMI and any of the stratified variables (P for interaction > 0.05), except for diabetes (P for interaction = 0.029), where a significant interaction was observed. This finding suggests that the association between obesity and depression risk is more pronounced among individuals with diabetes, indicating that diabetes may amplify the relationship between BMI and depression risk.

Fig.6

Subgroup analysis for the association between BMI and the risk of depression in obese population

Mediating effect of systemic inflammatory markers on the association between obesity and depressive symptoms

Given the observed interrelationships among obesity, systemic inflammatory markers, and depressive symptoms, we conducted a mediation analysis to further explore these associations. As presented in Table 5 and Fig. 7, NLR, SIRI, and SII each served as partial mediators in the relationship between BMI and depressive symptoms, with all mediation effects being statistically significant (P < 0.001). Specifically, NLR accounted for 5.2% of the mediation effect, SIRI for 5.9%, and SII for 6.1%, indicating that SII plays the largest mediating role in this relationship. These findings suggest that inflammatory markers (NLR, SIRI, and SII) may function as biological intermediaries, helping to explain the link between obesity (BMI) and the increased risk of depression.

Table 5 Inflammatory markers mediating the association between obesity and depressive symptomsFig.7

Analysis of the mediation by systemic inflammation markers (NLR, SIRI, SII) of the associations of BMI with the risk of depression

When participants were stratified by sex and adjusted for all covariates, SIRI significantly mediated the association between obesity and depressive symptoms in both men and women (SIRI: Male, percentage mediated [PM] = 9.39%, P < 0.0001; Female, PM = 4.40%, P = 0.018). However, NLR and SII significantly mediated the association between obesity and depressive symptoms in men but not in women (NLR: Male, PM = 10.64%, P < 0.0001; Female, PM = 2.06%, P = 0.18. SII: Male, PM = 11.51%, P < 0.0001; Female, PM = 3.52%, P = 0.072).