Neighbourhood socioeconomic characteristics and blood pressure among Jamaican youth: a pooled analysis of data from observational studies

Abstract

Introduction Neighbourhood characteristics are associated with several diseases, but few studies have investigated the association between neighbourhood and health in Jamaica. We evaluated the relationship between neighbourhood socioeconomic status (SES) and blood pressure (BP) among youth, 15–24 years old, in Jamaica. Methods A pooled analysis was conducted using data from three studies (two national surveys and a birth cohort), conducted between 2005–2008, with individual level BP, anthropometric and demographic data, and household SES. Data on neighbourhood SES were obtained from the Mona Geo-Informatics Institute. Neighbourhood was defined using community boundaries from the Social Development Commission in Jamaica. Community characteristics (poverty, unemployment, dependency ratio, population density, house size, and proportion with tertiary education) were combined into SES scores using principal component analysis (PCA). Multivariable analyses were computed using mixed effects multilevel models. Results Analyses included 2,556 participants (1,446 females; 1,110 males; mean age 17.9 years) from 306 communities. PCA yielded two neighbourhood SES variables; the first, PCA-SES1, loaded highly positive for tertiary education and larger house size (higher value = higher SES); while the second, PCA-SES2, loaded highly positive for unemployment and population density (higher value = lower SES). Among males, PCA-SES1 was inversely associated with systolic BP (β-1.48 [95%CI −2.11, −0.84] mmHg, p < 0.001, for each standard deviation unit increase in PCA-SES1 score) in multivariable model accounting for age, household SES, study, BMI, fasting glucose, physical activity and diet. PCA-SES1 was not significantly associated with systolic BP among females (β −0.48 [−1.62, 0.66], p = 0.410) in a similar model. Associations for PCA-SES2 was assessed using linear splines to account for non-linear effects. The were no significant associations between systolic BP and PCA-SES2 among males. Among females, higher PCA-SES2 (i.e. lower SES) was associated with higher systolic BP at spline 2 [z-score -1 to 0] (β4.09 [1.49, 6.69], p = 0.002), but with lower systolic BP at spline 3 [z-core 0 to 1] (β-2.81 [−5.04, −0.59], p = 0.013). There were no significant associations between diastolic BP and PCA-SES1, but PCA-SES2 showed non-linear associations with diastolic BP particularly among males. Conclusion Higher neighbourhood SES was inversely associated with systolic BP among male Jamaican youth; there were non-linear associations between neighbourhood SES and systolic BP among females and for diastolic BP for both males and females.