A machine learning analysis of predictors of future hypertension in a young population.

BACKGROUND: Early diagnosis of hypertension (HT) is crucial for preventing end-organ damage. This study aims to identify the risk factors for future HT in young individuals through the application of machine learning (ML) models. METHODS: The study included individuals aged 18-40 years who had not been diagnosed with HT through ambulatory blood pressure monitoring (ABPM). These participants were monitored for hypertension diagnosis from the date of ABPM application until the date of data collection. Hypertension prediction was carried out using three distinct ML methods: Support Vector Machine, Random Forest, and Least Absolute Shrinkage and Selection Operator. The identification of variables significant for future HT was based on the outcomes of these models. RESULTS: This study comprised 516 patients, with a mean follow-up duration of 793.4±58.6 days. Following the integration of demographic data, laboratory results, and ABPM findings into the ML models, age, high-density lipoprotein cholesterol, triglycerides, and the standard deviation of systolic blood pressure (SDsis) were identified as predictors for future HT. A logistic regression with the selected variables (age, diabetes mellitus history, HDL, triglycerides, white blood cell count, and SDsis) using the full data set gave the following log odds 0.0737 (P<0.001), 0.7146 (P<0.001), -0.0160 (P=0.071), 0.0026 (P=0.002), 0.0857 (P=0.069), and 0.0850 (P=0.005), respectively. The corresponding probability values of age, diabetes mellitus history, HDL, triglycerides, white blood cell count, and SDsis were 0.5184, 0.6714, 0.4960, 0.5006, 0.5214, and 0.5212, respectively. This indicates a unit increase in all factors, except diabetes mellitus history, increases the probability of future HT by 50%. A history of diabetes, however, increases the probability of future HT by more than two thirds. The history of diabetes mellitus emerged as the most crucial predictor of future HT across all applied methods. CONCLUSIONS: ML methods appear to be valuable tools for predicting future HT. The widespread adoption of these methods and the refinement of more comprehensive models will lay the groundwork for future studies.

Clinical usability of morning surge blood pressure for predicting future hypertension in a young population.

Objective    Early diagnosis of hypertension (HT) is a critical issue for physicians. This study was conducted to determine if morning surge blood pressure (MSBP) could be used to predict future HT. The study also examined which demographic data in a regression model might help to detect future HT without any invasive procedure.Material and methods    A young population between 18 and 40 yrs of age was included in the study. MSBP and demographic data were used to determine an optimal model for predicting future HT by using Bayesian information criteria and binary logistic regression.Results    1321 patients with 24 hr ambulatory blood pressure monitoring were included in this study. The odds ratio of 10 units of increase in diastolic MSBP was 1.173511 in the model, which indicates that a 10 mmHg increase in diastolic MSBP increases the odds of future HT in the patient by 17.4 %. The odds ratio of age was 1.096365, meaning that at each age above 18 yrs, the patients' odds of future HT rise by 9.6 %. The odds ratios for gender (male) and previous HT were 1.656986 and 3.336759, respectively. The odds of future HT in males were 65 % higher than for females, and a history of HT implies that the odds of future HT were higher by 230 %.Conclusion    Diastolic MSBP can be used to predict HT in young individuals. In addition, age, male gender, and previous HT add more predictive power to diastolic MSBP. This statistically significant, predictive model could be useful in lessening or preventing future HT.

On the performance of the variance ratio unit root tests with flexible Fourier form.

This article introduces a new unit root test that combines the variance ratio framework with the Flexible Fourier Form under the generalized least squares detrending mechanism. The advantage of the proposed method against its alternatives can be listed as: (1) it suggests a non-parametric procedure that does not require any parametric or semi-parametric model to remove serial correlation in the innovation process; (2) it can reasonably adapt itself to deal with the multiple structural breaks with various functional specifications. In the simulation exercises, we show that the proposed method exhibits satisfactory performance in the size and size-adjusted power analysis.