Forecasting Schizophrenia Incidence Frequencies Using Time Series Approach

Understanding the prevalence of schizophrenia has important implications for both health service planning and risk factor epidemiology. The aims of this study are to systematically identify and collate studies describing the prevalence of schizophrenia, to summarize the findings of these studies, and to explore selected factors that may influence prevalence estimates.
This historical cohort study was done on schizophrenia patients in Farshchian psychiatric hospital from April 2008 to April 2016. To analyze the data, the Holt-Winters Exponential Smoothing (HWES) method was applied. All the analyses were done by R.3.2.3. Software using the packages “forecast” and “tseries”. The statistical significant level was assumed as 0.05.
Our investigation show that a constant frequency of Schizophrenia incidence happens every month from August 2008 to February 2015 while a considerable increase occurs in March 2015. The high frequency of Schizophrenia incidence remains constant to the end of 2015 and a decrease is shown in 2016. Also, data demonstrate the development of Schizophrenia in the next 24 months with 95% confidence interval.
Our study showed that a significant increase happens in the frequency of Schizophrenia from 2016. Although the development is not constant and the same for all months, the amount of increase is considerably high comparing to before 2016.