Reliability of normative tables in assessing elevated blood pressure in children

Epidemiological studies indicate that elevated blood pressure (BP) is common among children. Early detection of elevated BP in children is important in the prevention of cardiovascular disease (CVD). Difficulties exist in the diagnosis of "elevated" BP in children of 8-12 years of age. Normative tables of sex, age, and height specific BP percentile levels are required. Reports provide normative tables of BP percentile levels in children. Elevated BP is defined as systolic and diastolic BP >= 90th age, sex, and height specific BP percentile. The study tests the internal consistency and reliability between the normative tables of BP in children. The main objective of the study is to assess the reliability of existing normative tables of BP in children by using ad hoc analysis of a prior survey. A cross-sectional survey was performed in elementary school children (8-12 years of age, n = 818) from Istanbul, Turkey. The survey was conducted in 2007-2008. Four different normative tables of BP were used to determine children with elevated BP. The reliability and internal consistency between the normative tables of BP in children were assessed. The prevalences of elevated systolic and diastolic BP in children differed significantly between the normative tables of BP. The internal consistency between the normative tables was poor (Cronbach alpha coefficient = 0.429). Kuder-Richardson 21 test indicated poor reliability between the local and international normative tables of BP. In conclusion, the study indicates that there are significant inconsistencies and poor reliability between the national and international normative tables of BP in children.