Prediction of the hypertension risk in teenagers

Creation of a hypertension risk stratification model and development of an algorithm to detect hypertension in teenagers.

The study group consisted of 690 middle and high school students, aged 15-17 years, from the metropolitan area of Warsaw, Poland. Information concerning family history and presence of risk factors for cardiovascular disease was gathered. Three-time blood pressure measurements were taken during at least two separate visits, which were at least a week apart, using the auscultatory method, according to standard procedures. Anthropometric measurements included: body weight, height, arm, hip and abdominal circumference, skin-fold thickness measured on the rear surface of an arm, below the inferior angle of the scapula and at the belly. Following indexes were determined: body mass index, waist to hip ratio (WHR), waist to height ratio, hip to height ratio.

A logistic regression model, describing the risk of hypertension in adolescents aged 15-17 was invented. p(x) = (e^g(x))/(1+e^g(x)) where ĝ(x) = -0.097 × height+0.085 × weight+7.764 × WHR+1.312 × family hypertension. Family hypertension means presence of hypertension among members of the closest family. The formula was created, allowing the pre-selection of adolescents at risk of hypertension during screening. Next an algorithm for the detection of hypertension for practical use was proposed.

Body weight, WHR and incidence of hypertension in the family are the strongest predictors of hypertension in teenagers. Proposed screening algorithm can be a useful tool for selecting teenagers at risk of hypertension and in need of specialized diagnostics and care.