ABSTRACT
Background: The socioeconomic position (SEP) and educational level of individuals have an inverse correlation with mortality in developed societies. Aim To assess in a society undergoing a socioeconomic transition, the mortality risk associated to a low SEP (combination of education and income, scale 0-25 points, reference > 10 points) and low education (education years, reference > 8 years), adjusting for other known risk factors. Material and methods: In this prospective cohort study, a random sample of 920 subjects, living in San Francisco de Mostazal, Chile, aged more than 20years (395 males) was examined for the first time in 1997-1999 and re-examined in 2005-2006. All had information about economic household income and level of education. A Cox regression model was used to evaluate the association between mortality and socioeconomic measures. Results: The crude mortality hazard ratio (HR) was 3.34 (95 percent confidence interval (CI) 2.88-3.87) and 6.05 (95 percent CI 5.04-7.26) for low SEP and low educational level, respectively. After adjusting for age, gender, hypertension, diabetes, dyslipidemia, abdominal obesity, smoking, alcohol intake and family history of cardiovascular disease, the figures were 1.23 (95 percent CI 1.04-1.43) and 1.54 (95 percent CI 1.23-1.85) for low SEP and low educational level, respectively. Conclusions: In a society in socioeconomic transition, low SEP and especially low educational level are risk factors for mortality even after adjusting for known mortality risk factors.
Subject(s)
Adult , Aged , Aged, 80 and over , Female , Humans , Male , Middle Aged , Educational Status , Mortality , Socioeconomic Factors , Chile/epidemiology , Epidemiologic MethodsABSTRACT
Antecedentes: Medidas antropométricas de obesidad como el índice de masa corporal (IMC), circunferencia de cintura (CC), razón cintura-cadera (RCC) y razón cintura-estatura (RCE) son conocidas por estar asociadas a factores de riesgo metabólico. Sin embargo, es controversial cuál de ellas es mejor para predecir mortalidad. Método: En este estudio prospectivo observacional, reclutamos 920 adultos sanos (>20 años, 395 hombres, San Francisco de Mostazal, Chile) y examinamos la relación entre medidas antropométricas de obesidad y mortalidad general utilizando puntos de corte población-específicos (IMC≥28 kg/m2 ambos sexos; CC≥92cm hombres, ≥88cm mujeres; RCC≥0,94 hombres, ≥0,84 mujeres y RCE≥0,55 ambos sexos). El riesgo relativo (RR) fue calculado utilizando regresión logística controlando por factores de riesgo cardiovascular convencionales. Además, calculamos el área bajo la curva ROC (Receiving Operating Characteristic) para evaluar el desempeño de cada medida antropométrica para predecir mortalidad. Resultados: Durante 8 años de seguimiento ocurrieron 47 muertes. En hombres, el RR con intervalos de confianza al 95 por ciento (IC95 por ciento) para IMC, CC, RCC y RCE fue 1.21 (0.93-1.58), 2.52 (1.86-3.40), 1.11 (0.86-3.20) y 3.38 (2.31-4.96) respectivamente. En mujeres, el RR fue 0.78 (0.60-1.01), 1.44 (1.07-1.93), 1.54 (1.13-2.10), y 1.56 (1.13-2.26) respectivamente. En hombres, las áreas bajo la curva ROC (IC95 por ciento) fueron: IMC 0.66 (0.65-0.68), CC 0.72 (0.71-0.74), RCC 0.72 (0.71-0.74) y RCE 0.77 (0.75-0.78). En mujeres, estas áreas fueron: 0.59 (0.57-0.61), 0.65 (0.63-0.67), 0.58 (0.56-0.60) y 0.70 (0.68-0.71) respectivamente.
Background: Several obesity anthropometric measures like body-mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR) and waist-to-height ratio (WHtR) are known to be associated with metabolic risk factors. However, it remains controversial which of these markers is better to predict mortality. Methods: In this prospective observational study, we enrolled 920 healthy adults (> years, 395 men, 98.7 percent Chilean-Hispanics living in San Francisco de Mostazal, Chile). We examined the relation between obesity anthropometric measures and all-cause mortality using population-specific cutoffs (BMI ≥28 kg/m2 both genders; WC ≥92 cm men, ≥88 cm women; WHR ≥0,94 men, 0,84 women and WHtR ≥0,55 both genders). Multivariate risks were calculated with logistic regression models controlling for cardiovascular and metabolic risk factors. In addition we calculated area under ROC curve (Receiving Operating Characteristic) to evaluate performance of every anthropometric measure to predict mortality. Results: Forty seven deaths occurred during 8 years of follow-up. In men, multivariate risks with 95 percent CI) for BMI, WC, WHR, and WHtR were respectively 1.21 (0.93-1.58), 2.52 (1.86-3.40), 1.11 (0.86-3.20) and 3.38 (2.31-4.96). In women, multivariate risk were respectively 0.78 (0.60-1.01). 1.44 (1.07-1.93), 1.54 (1.13-2.10), and 1.56 (1.13-2.26). In men, areas under curve ROC (95 percent CI) were BMI 0.66 (0.65-0.68), WC 0.72 (0.71-0.74), WHR 0.72 (0.71-0.74) and WHtR 0.77 (0.75-0.78). In women these areas were respectively 0.59 (0.57-0.61), 0.65 (0.63-0.67), 0.58 (0.56-0.60) and 0.70 (0.68-0.71). Conclusions: In this Chilean-Hispanic cohort WHtR is the most accurate predictor of all-cause mortality in comparison with other anthropometric measures of adiposity.
Subject(s)
Humans , Male , Female , Adult , Middle Aged , Aged, 80 and over , Anthropometry , Cardiovascular Diseases/diagnosis , Cardiovascular Diseases/mortality , Waist-Hip Ratio , Body Mass Index , Chile/epidemiology , Cholesterol, HDL/blood , Cholesterol, LDL/blood , Cholesterol/blood , Dyslipidemias/epidemiology , Cardiovascular Diseases/blood , Follow-Up Studies , Longitudinal Studies , Obesity/epidemiology , Prospective Studies , Risk Factors , ROC Curve , Sensitivity and SpecificityABSTRACT
Antecedentes: Varios índices antropométricos (IA) de obesidad han mostrado ser predictores de morbilidad cardiovascular en estudios epidemiológicos internacionales. Objetivo: Evaluar el impacto del índice de masa corporal (IMC), circunferencia de abdomen (CA), razón cintura/cadera (RCC), razón cintura/estatura (RCE) e índice pulso-masa (IPM) sobre el riesgo de sufrir un evento cardiovascular (ECV) en una cohorte de población chilena. Diseño: Estudio longitudinal de 920 personas (edad 39,5 +/- 16,3 años, 382 hombres) que participan en el proyecto San Francisco (PSF). Método: La población fue examinada entre 1997 y 1999 consignándose nivel socioeconómico (NSE), hipertensión arterial (HTA), diabetes, colesterol total, triglicéridos, tabaquismo, IMC > 30 kg/m2, CA > 102 cm en hombres y > 85 cm en mujeres, RCC > 1,00 en hombres y > 0,85 en mujeres, IPM > 1,00 y RCE > 0,53. Análisis de riesgo (RR) con intervalos de confianza al 95 por ciento (IC95 por ciento) y regresión logística mediante procedimiento stepwise. Resultados: Luego de 5,3 +/- 0,3 años de seguimiento se registraron 26 ECV resultando predictores: diabetes RR=4,48 (IC95 por ciento 1,87-10,7); HTA RR=5,16 (IC84 por ciento 2,27-11,71); bajo NSE RR=1,81 (IC95 por ciento 1,38-2,37); IMC RR=2,43 (IC95 por ciento 1,10- 5,37); CA RR=3,10 (IC95 por ciento 1,39-6,91) y RCE RR=6,58 (IC95 por ciento 1,96-22,1). En el análisis multivariado que incluyo con todos los IA controlados por edad y sexo, sólo la RCE fue predictor independiente con riesgo de 3,14 (p<0,01). En el modelo completo que incluyó todos los factores de riesgo analizados, sólo la edad, HTA, diabetes, RCE y bajo NSE predicen un ECV no fatal. Conclusión: En la cohorte de San Francisco la RCE aparece como un predictor simple e independiente de enfermedad cardiovascular con mejor desempeño que otros IA de obesidad. Sin embargo, puntos de corte específicos para población chilena deberían ser determinados y evaluados.
Background: International studies have shown that several obesity related anthropometric indices are associated to increased cardiovascular risk. Aim: To evaluate the risk of non-fatal cardiovascular events associated to body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR) , pulse mass index (PMI) and waist-to-height ratio (WHtR) upon cardiovascular risk in a cohort of Chileans subjects. Methods: 920 subjects (age 39.5 +/- 16.3 years, 382 males) from the San Francisco Project were prospectively followed.From 1997 to 1999 we determined the socio-economic status (SES), the presence of hypertension, diabetes, cholesterol, and triglyceride levels, smoking status, obesity (BMI >30 kg/m2 , WC >102 for males and > 85 cm for females, WHR > 1 for males and > 0.85 for females, PMI >1 and WHtR > 0.53). Risk ratios and 95 percent confidence intervals (CI) along with stepwise logistic regression were used to assess statistical significance.Results: 26 cardiovascular events took place within 5.3+/-0.3 years of follow up. Significant predictive values were shown for diabetes (RR=4.8, CI 1.9-10.7), hypertension (RR=5.16, CI 2.27 11.7), low SEL(RR=1.81, CI 1.38-2.37), BMI (RR 2.43, CI 1.1-5.37), WC (RR=3.1, CI 1.39-6.91 and WHtR (RR = 6.58, CI 1.96-22.1). Logistic regression analysis indicated that only WHtR remained an independent predictor for cardiovascular events. After a different adjustment model, age, hypertension, diabetes, WHtR and low SES had predictive value for cardiovascular events. Conclusion: WHtR appears to be an independent marker for cardiovascular risk in the San Francisco cohort study. However, specific anthropometric cut-off points for chonic diseases in chilean subjets should be determined and tested.