Acta Biomedica Scientifica (Jul 2020)

Development and Application of a Mathematical Model in the Prediction of Stroke in People Exposed to Toxic Dust Factors in the Workplace

  • M. V. Yashnikova,
  • E. L. Poteryaeva,
  • B. M. Doronin,
  • V. N. Maksimov,
  • E. L. Smirnova

DOI
https://doi.org/10.29413/ABS.2020-5.3.4
Journal volume & issue
Vol. 5, no. 3
pp. 29 – 35

Abstract

Read online

Background. The problem of studying the development of stroke in men of working age working in conditions of high professional risk is relevant. Aim of the research. To develop a method for predicting the development of stroke in men working under conditions of exposure to toxic dust factors. Materials and methods. The study included 111 men aged 30 to 65 years who worked for 5 or more years in the following professions: electric gas welder, gas welder, smelter. The study participants were divided into two groups: the first group – 45 men who had a stroke; the second group – 66 men who did not have stroke history. The subjects of both groups were comparable in terms of occupational structure, age and length of service. Using the method of sequential inclusion of the studied risk factors for stroke development (Forward Stepwise), we determined those that made a significant contribution to the possible development of stroke. Results. Based on multivariate analysis, a logistic regression formula was derived that determines the likelihood of a stroke in patients working under the influence of the toxic dust factor, which included the following factors: the presence of dyslipidemia factor (β-coefficient = 3.7; p = 0.001), the patient’s diagnosis is “stage III arterial hypertension” (β-factor = 5.6; p = 0.002), blood pressure control (β-factor = –3.4; p = 0.005). The patient determines the value of these three parameters, substitutes the appropriate formula for calculating the value of p(X). A value of p(X) in excess of 0.5 indicates a high risk of stroke in men working under conditions of exposure to toxic dust factors. The sensitivity of the proposed method was 84.4 %, specificity – 92.4 %. Conclusion. The multivariate model is informative for the individual prediction of the development of stroke in men working under the influence of the toxic-dust factor, since it reflects the degree of influence of one or another factor on the development of the disease.

Keywords