The Risk Factors of Hypertension Using Penalized Spline

Share on facebook
Share on google
Share on twitter
Share on linkedin
Illustration by Feri Fenoria

UNAIR NEWS – Hypertension is an increase in blood pressure that can increase the risk of attacks on organs, such as stroke, coronary heart disease, and right ventricular hypertrophy.

Hypertension is a disease caused by many factors including obesity, unhealthy eating patterns, lack of physical activity, psychological stress conditions, habits of consuming alcoholic beverages, smoking and excessive coffee consumption patterns.

“The criteria used to determine whether someone has hypertension or not. If the blood pressure is greater or equal to 140 mmHg or the diastolic blood pressure is greater or equal to 90 mmHg,” said the lecturer in Mathematics, Faculty of Science and Technology (FST) UNAIR, Dr. Nur Chamidah, S.Si, M.Si

According to WHO,50% of hypertensive patients recover, only 25% receive treatment, and only 12.5% can be adequately treated. The results of the 2014 household health survey also showed that the prevalence of hypertension in Java was 41% with a range in each province of 36.6 – 47.7%, while the prevalence in urban areas was 39.9 % and in rural areas was 44,1%

A study revealed that a person’s risk of hypertension pressure by using the parametric regression model approach based on the complementary and probit linkage function resulted in a classification accuracy of 81.5 % and 85.2 %, respectively.

Together with UNAIR Statistics student, Tati Adiwati, Nur Chamidah explained that by writing a journal entitled Modeling of Hypertension Risk Factors Using Penalized Spline to Prevent Hypertension in Indonesia, published by the IOP Conference Series: Materials Science and Engineering.

In that study, Nur Chamidah explained the risk hypertension could be done with a nonparametric logistic regression approach, which is based on a penalized spline estimator that results in a higher classification accuracy than using the parametric logistic regression model approach that functions as complementary and probit linkage.

A spline is a piece of polynomial which is a combination of segments or fragments of different curves. Therefore, the spline can adequately accommodate the localized properties of a function or data.

“The data used in my study are primary data from the results of questionnaires and interviews from August to September 2018 of 59 respondents who were heart disease patients treated at the heart polyclinic of Haji Hospital Surabaya,” explained the Surabaya-born lecturer.

“In this article, only examples of hypertension risk estimation for the 34th (respondent) observation are given, while other observations are carried out using a similar process,” she continued.

Hypertension risk based on age factors, body mass index, heart rate, and stress scores using the nonparametric logistic regression model approach. Based on the penalized spline estimator is better than using the parametric regression model, with the connective function of complementary and probit, because it can increase classification accuracy from 81.5 % and 85.2 % to 96.6 %.

“Hopefully, people are aware with healthy lifestyle to manage health to control blood pressure and reduce the number of cases of hypertension in Indonesia,” she concluded.

Penulis : Fariz Ilham Rosyidi

Editor : Khefti Al Mawalia

Link        :

Tati Adiwati And Nur Chamidah. 2019. Modelling of Hypertension Risk Factors Using Penalized Spline to Prevent Hypertension in Indonesia. IOP Conf, Series: Materials Science and Engineering,  546 052003

Berita Terkait



Media komunikasi dan informasi seputar kampus Universitas Airlangga (Unair).