Associations of socioeconomic status with infectious diseases mediated by lifestyle, environmental pollution and chronic comorbidities: a comprehensive evaluation based on UK Biobank

Abstract

Background: Socioeconomic status (SES) inequity was recognized as a driver of some certain infectious diseases. However, few studies evaluated the association between SES and the burden of overall infections, and even fewer identified preventable mediators. This study aimed to assess the association between SES and overall infectious diseases burden, and the potential roles of factors including lifestyle, environmental pollution, chronic disease history.

Methods: We included 401,009 participants from the UK Biobank (UKB) and defined the infection status for each participant according to their diagnosis records. Latent class analysis (LCA) was used to define SES for each participant. We further defined healthy lifestyle score, environment pollution score (EPS) and four types of chronic comorbidities. We used multivariate logistic regression to test the associations between the four above covariates and infectious diseases. Then, we performed the mediation and interaction analysis to explain the relationships between SES and other variables on infectious diseases. Finally, we employed seven types of sensitivity analyses, including considering the Townsend deprivation index as an area level SES variable, repeating our main analysis for some individual or composite factors and in some subgroups, as well as in an external data from the US National Health and Nutrition Examination Survey, to verify the main results.

Results: In UKB, 60,771 (15.2%) participants were diagnosed with infectious diseases during follow-up. Lower SES [odds ratio (OR) = 1.5570] were associated with higher risk of overall infections. Lifestyle score mediated 2.9% of effects from SES, which ranged from 2.9 to 4.0% in different infection subtypes, while cardiovascular disease (CVD) mediated a proportion of 6.2% with a range from 2.1 to 6.8%. In addition, SES showed significant negative interaction with lifestyle score (OR = 0.8650) and a history of cancer (OR = 0.9096), while a significant synergy interaction was observed between SES and EPS (OR = 1.0024). In subgroup analysis, we found that males and African (AFR) with lower SES showed much higher infection risk. Results from sensitivity and validation analyses showed relative consistent with the main analysis.

Conclusions: Low SES is shown to be an important risk factor for infectious disease, part of which may be mediated by poor lifestyle and chronic comorbidities. Efforts to enhance health education and improve the quality of living environment may help reduce burden of infectious disease, especially for people with low SES.

Keywords: Chronic comorbidities; Environmental pollution; Healthy lifestyle; Infectious diseases; Socioeconomic status.

Fig. 1

Flowchart of the participants selection in the UK Biobank (a) and US NHANES (b). SES socioeconomic status

Fig. 2

Bar plots indicating socioeconomic, lifestyle, environmental pollution, and chronic comorbidity factors on infectious diseases in participants from UK biobank. Odds ratios (ORs) were adjusted for age, sex, ethnic and assessment center. Dashed line represents no significant association. TDI Townsend deprivation, SES socioeconomic status, EPS environment pollution score, APS air pollution score, PM_2.5 particulate matter ≤ 2.5 μm, PM_2.5–10 particulate matter 2.5–10 μm, PM₁₀ particulate matter ≤ 10 μm, NO_x nitrogen oxides, NO₂ nitrogen dioxide, CVD cardiovascular disease, CAD cardiovascular diseases, AF atrial fibrillation

Fig. 3

Mediation effects of SES on infectious diseases by Lifestyle scores (a), CVD (b), diabetes (c), psychiatric disorders (d), and cancer (e), and TDI by EPS (f). Regression analyses of SES on mediators, and mediators on infection were all adjusted for age, sex, ethnic and assessment center. TDI Townsend deprivation, SES socioeconomic status, CVD cardiovascular disease

Fig. 4

Forest plot indicating lifestyle scores on infectious diseases in different SES subgroups from UK biobank. The group with low SES and poor lifestyle scores (0–1) was selected as the overall control group (a), or for each SES subgroup individually, that with poor lifestyle scores (0–1) was selected as the control group (b). Odds ratios (ORs) were adjusted for age, sex, ethnic and assessment center. Dashed line represents no significant association. SES socioeconomic status

Fig. 5

Forest plot indicating environmental pollution score (EPS) groups on infectious diseases in different SES subgroups from UK biobank. The group with high SES and low EPS (top fifth, Q1) was selected as the overall control group (a), or for each SES subgroup individually, that with low EPS (Q1) was selected as the control group (b). Odds ratios (ORs) were adjusted for age, sex, ethnic and assessment center. Dashed line represents no significant association. SES socioeconomic status

Fig. 6

Forest plot indicating chronic comorbidity factors on infectious diseases in different socioeconomic status (SES) subgroups from UK biobank. Odds ratios (ORs) were adjusted for age, sex, ethnic and assessment center. Dashed line represents no significant association. SES socioeconomic status, CVD cardiovascular disease, CAD cardiovascular diseases, AF atrial fibrillation

Fig. 7

Forest plot indicating risk of infectious diseases in different sex (a) or ethnic (b) by SES subgroups from UK biobank. Odds ratios (ORs) were adjusted for age, sex (analysis on ethnic), ethnic (analysis on sex) and assessment center. Dashed line represents no significant association. SES socioeconomic status, EUR European, AFR African, ASA Asian