Assessing Some Risk Factors of Recurrent Times to Childhood Seizures Using Parametric Copula Marginal Model applying Bayesian Approach

Seizure is a neurological disorder in children. Children have experienced seizures at least once in the first 16 years of life. The aim of this study is to determine some risk factors of recurrent times to childhood seizures using Parametric Copula Marginal Model applying Bayesian Approach. In this descriptive-longitudinal study, data from a case filed in 300 children who were admitted in four hospitals between 2014 and 2016 due to recurrence of seizure disease were collected. The seizure times of children were recorded after each visit to these centers and the time intervals between seizure recurrence events were calculated. For modeling of time to recurrent seizures in children, the joint distribution of the Parametric Copula Marginal Model applying Bayesian Approach was fitted to the data and analyzed by Bayesian approach. WinBugs and R versions 3.4.1 were used for data analysis. 15.0% of the children had seizure and 64.3% of these children were hospitalized at least once due to seizure recurrence. 25.0% of the mothers of these children had a hard labor. The result of the electrolytic test was 40% normal. 27.0% had a family history of seizure and 78.4% of the children had normal evolutionary status. The 95% confidence interval for the dependency parameter (detailed parameter) was estimated with the parametric marginal risk of the Weibull-Clayton equation (0.434, 0.227). There was a significant relationship between the history of hospitalization and drug use. There was a correlation between the recurrence events of children with seizure. It is suggested that doctors consider the clinical steps, drug use factors, and hospitalization history, which are among the factors associated with frequent recurrence of seizure in children.