H Tireh, R Yousefi, Sb Mazloum Shahri , Mt Shakeri,
Volume 14, Issue 3 (Vol.14, No.3, 2018)
Abstract
Background and Objectives: Hypertension is a major global issue due to its consequences. Ordinary regression models have limitations in assessment of hypertension since the reference graph derived from a specific population may not be appropriate for another population. The polynomial quantile regression model is considered as a possible alternative. Hence, this study was conducted with the aim of determining reference values as well as blood pressure percentile curves in Mashhad.
Methods: This cross-sectional study was carried out in a random sample of 6949 individuals attending Samen health centers for diabetes screening in 2010. Different percentiles were analyzed using some variables such as gender, age, BMI, WHR, and systolic and diastolic blood pressure. The R software (version 3.0.1) was used for data analysis.
Results: In this study, 70.58% and 29.42% of subjects were men and women, respectively. The results of the quantile regression model showed that with an increase in age, BMI, and WHR, blood pressure increased in all percentiles. In all variables, subjects in the 75th and 95th percentiles had moderately high and high blood pressure while they had a normal blood pressure in other percentiles.
Conclusion: The model provided more information about blood pressure and its related patterns. According to the results, it seems that more attention should be paid to elderly and overweight individuals in the 75th and 95th percentiles.
