Abstract

Introduction: Investigating and forecasting the changes in the frequency of referrals to the treatment centres of the patients with drug poisoning can aid the assessment of the burden of health problems and planning of appropriate intervention programmes. The aim of this study is to predict the trend of drug poisoning case referrals to the Iranian hospitals up to 2021. Material and methods: The Box & Jenkins model was applied in a longitudinal study to forecast the frequency of drug and alcohol poisoning case referrals. The number of cases referred to the hospitals in each month from March 2011 to February 2015 was provided. After data processing to gain stationary time series and investigation of the stability assumption with the Dickey-Fuller test, ARIMA model parameters were determined using ACF and PACF graphs. Using Akaike statistics, ARIMA (0, 1, 1) was selected as the best fit model. Results: The number of referrals increased from 400 in 2011 to 735 in 2015. The highest referrals in 2011-2015 were 56 cases in January 2011, 43 in October 2012, 59 March 2013, 66 April 2014 and 80 in January 2015. The trend was linear without a regular seasonal pattern. The mean monthly poisoning referrals to the selected hospitals up to 2021 was predicted as 58.2. Discussion: The estimated increased trend of referral cases due to poisoning showed in this study is parallel to the estimated increase of illegal drug use prevalence. Recent surveys, which showed that drug abuse prevalence in Iran is higher than the global mean, were also confirmed by our research. The limitation of our study is the lack of access to the full information on psychoactive substance consumption as well as classified information like police and national registry system databases. Conclusions: Our study revealed a rising trend for drug poisoning patients referred to the studied hospitals. The forecasting also suggested that the trend will continue up to 2021 without suitable intervention.