|
Abstract
|
OBJECTIVES: In high-income countries, there is a positive relationship between neighborhood-level crime and cardiovascular risk factors (CRFs) in adolescence. A few studies that examined this association in low- and middle-income countries yielded inconsistent results. The aim of our study was to analyze the associations between homicide rates and prevalence rates of CRFs in adolescents living in Brazilian municipalities.
STUDY DESIGN: Cross-sectional ecologic study.
METHODS: An ecologic study was conducted to evaluate associations between homicide rates and CRFs in 122 municipalities included in the Study of Cardiovascular Risk Factors in Adolescents (Portuguese acronym, ERICA). The following CRFs were evaluated: adiposity (overweight/obese), sedentary behavior (screen time ≥3 h per day), high waist circumference (between 10 and 16 years, percentile ≥90; 16-17 years, ≥90 cm for men and ≥80 cm for women), low high-density lipoproteins (HDL) levels (<45 mg/dl), alcohol use (≥1 drink in the last 30 days) and hypertension (≥percentile 95th). In addition to these risk factors, we also evaluated the association of homicides with physical inactivity. Neighborhood homicide data were obtained from the Ministry of Health Mortality Information System (Portuguese acronym, SIM).
RESULTS: The 3-year homicide (115 per 100,000; interquartile range (IQR): 66.1, 150.5.) and CRFs prevalence rates were high (eg, obesity = 7.7%; IQR:6.2,9.5, overweight = 24.7%; IQR:21.6,27.9, sedentary behavior = 54.4%; IQR:47.9,58.9). Crude and adjusted Spearman correlation coefficients between 3-year homicide rate and CRFs were weak, and with the exception of HDL level, in directions contrary to expectation. Prevalence rate ratios (PR) tended toward the null value, even after multiple adjustments. Although the upper limit of the PR expressing the association between the highest homicide category (65-95% percentile range) and physical inactivity in boys barely overlapped the null hypothesis, the point estimate was also very close to one (boys, PR = 0.98, 95% confidence interval 0.95, 1.0).
CONCLUSIONS: In ERICA municipalities, although high prevalence rates of homicides and CRFs are important public health concerns, no significant associations were found between these variables.
Language: en |