Meteorological Parameters and Hypertensive Crisis Risk: a Longitudinal Study for Prediction Model Developing

Abstract

INTRODUCTION. Integrating climatotherapy into health resort therapy for arterial hypertension in diverse landscapes has the potential to yield positive effects, if used in target groups and preventing the occurrence of meteopathic reactions, including a hypertensive crisis (HC). While the impact of natural healing factors on the human body has been previously studied, the utilization of modern mathematical approaches in developing HC models has enabled accurate predictions and timely prevention of HC during adverse weather periods.
 AIM. To analyze publicly available meteorological data time series to construct a mathematical model for predicting high-risk situations of HC based on the influence of climatic factors on patients with arterial hypertension. This model would identify unfavorable periods for hypertensive patients staying in health resorts throughout the year, allowing for timely therapeutic and preventive measures to prevent HC during these periods.
 MATERIALS AND METHODS. The study was conducted over a 22-month period, from January 1, 2019 to October 31, 2020, in Gelendzhik and Novorossiysk, renowned resort destinations located on the Black Sea coast of the Caucasus. These regions have a dry and subtropical climate. Meteorological data were obtained from Gelendzhik and Novorossiysk weather stations, and ambulance calls data were collected from Gelendzhik (12,268 calls) and Novorossiysk (12,226 calls), resulting in a total of 24,494 ambulance calls.
 The model was calculated using the maximum likelihood method through nonlinear logit regression. Key factors for the model included the main indicators of climate1 and geomagnetic conditions2. The logistic regression method exhibited a sensitivity of 56.0 % and a specificity of 77.3 %, with an overall accuracy of 76.0 %.
 RESULTS. According to the developed predictive model, the winter season has no more than 75.0 % of days associated with a low risk of hypertension, decreasing to 59.0 % in spring. However, the proportion increases to 89.0 % in summer and reaches 77.0 % in autumn. Model adequacy checks indicated a high degree of relevance, with Q (model quality) ranging between +0.64 and –0.117, and p 0.3.
 CONCLUSION. The developed logistic regression models provide more accurate calculations of individual risks for developing complications of hypertension and offer the opportunity to formulate individual strategies for patients. These models contribute to the field of climatotherapy and enhance the understanding of the impact of climatic factors on hypertensive patients, facilitating targeted interventions and improved management of hypertensive crises.