Математическая модель для прогнозирования продолжительности стационарного лечения при выполнении высокотехнологичных операций по устранению аритмий

Aim. To create a mathematical model for predicting the length of post-operative treatment after performing high-tech surgeries for arrhythmia treatment. Methods. To predict the in-patient treatment duration after performing high-tech surgeries for arrhythmia treatment, the data set was checked for normality of sample variance distribution and for variability, discriminant function analysis, variability analysis, Kolmogorov–Smirnov test, crosstab Pearson’s chi-squared test were performed. Normally distributed quantitative parameters are presented as M±m (m — standard error). Cross-sectional prospective study including the data of 345 patients aged 20 to 88 years (males 42.0%, females 58.0%) who underwent high-tech surgeries for arrhythmia treatment, was performed for modeling. Results. It was found that the main category of patients who require surgery for cardiac arrhythmia treatment were women aged 61 to 75 years (mean 68±7 years). Pacemaker implantation was the most common surgery type (47.0%, 162 patients), followed by radiofrequency ablation (31.0%, 107 patients) and encircling pulmonary veins isolation (22.0%, 76 patients). It was also found that the presence of postoperative complications after implantation of the pacemaker directly influencing treatment duration, increasing it almost twice fold (to an average of 14.2±5.1 days compared to 7.4±1.2 days in patients without complications, p=0.02). Statistical analysis allowed to identify five levels characte­rizing the duration of in-patient post-operative management. An automated software module for risk assessment in patients admitted for high-tech surgeries for arrhythmia treatment was created basing on the results of the study. The precision of the model reached 87% (mean value 84.7%). Conclusion. An automated software module for predicting the length of in-patient post-operative treatment allows to stratify the risk of post-surgical complications for patients and shows the influence of those risks on the use of hospital beds, medical aid management and funding of high-tech surgeries by obligatory health insurance funds.