Assessment and Prediction of Incidence Trend of Stomach, Colorectal and Esophagus Cancers Using Time-Series Models in Kurdistan, West of Iran

The cancers of the gastrointestinal tract, because of their high prevalence and fatality, are of great importance in most countries like Iran. In terms of prevalence, stomach, esophagus and colorectal cancers in Iran are ranked first, second and eighth, respectively. Therefore, this study aimed to model the incidence of the frequency of new cases of these cancers and their prediction in Kurdistan province, Iran.

In this cross-sectional study, the incidence of stomach, esophagus and colorectal cancers were analyzed from April 2000 to March 2017. In order to model time series data of the cancers, the autoregressive integrated moving average (ARIMA) and seasonal autoregressive integrated moving average (SARIMA) with Box-Jenkins approach were applied. After determining the suitable models, the frequencies of new cases for the cancers were predicted until March 2020. Data analysis was performed using R-3.4.2 statistical software package.

A total of 6439 gastrointestinal cancers were recorded during the study period, among which stomach, esophagus and colorectal cancers had the highest frequencies with 2548, 1722 and 989 new cases, respectively. The best fited model for stomach, esophagus and colorectal cancers were SARIMA (0, 1, 1) (0, 0, 1)4، SARIMA (3, 1, 0) (0, 0, 1)4 and ARIMA (0, 1, 1), respectively. It is predicted that the incidence pattern of these cancers have an upward trend with moderate slope by March 2020.

The high incidence of stomach, colorectal and esophagus cancer, as well as their increasing trend in the predicted time can be a serious alarm. Therefore, in order to prevent and reduce the frequency of these cancers, it is vital to design and implement the effective programs such as screening, especially in high risk groups, early diagnosis of the cancers, increasing public awareness and controlling the risk factors to prevent and reduce of these increasing trends.