OBJECTIVES: Predicting the length of stay (LOS) of patients in a hospital is important in providing them with better services and higher satisfaction, as well as helping the hospital management plan and managing hospital resources as meticulously as possible. We propose applying data mining techniques to extract useful knowledge and draw an accurate model to predict the LOS of heart patients. METHODS: Data were collected from patients with coronary artery disease (CAD). The patient records of 4,948 patients who had suffered CAD were included in the analysis. The techniques used are classification with three algorithms, namely, decision tree, support vector machines (SVM), and artificial neural network (ANN). LOS is the target variable, and 36 input variables are used for prediction. A confusion matrix was obtained to calculate sensitivity, specificity, and accuracy. RESULTS: The overall accuracy of SVM was 96.4% in the training set. Most single patients (64.3%) had an LOS < or =5 days, whereas 41.2% of married patients had an LOS >10 days. Moreover, the study showed that comorbidity states, such as lung disorders and hemorrhage with drug consumption have an impact on long LOS. The presence of comorbidities, an ejection fraction <2, being a current smoker, and having social security type insurance in coronary artery patients led to longer LOS than other subjects. CONCLUSIONS: All three algorithms are able to predict LOS with various degrees of accuracy. The findings demonstrated that the SVM was the best fit. There was a significant tendency for LOS to be longer in patients with lung or respiratory disorders and high blood pressure.
Comorbidity ; Coronary Artery Disease ; Coronary Vessels ; Data Mining ; Decision Trees ; Heart ; Hemorrhage ; Humans ; Hypertension ; Insurance ; Length of Stay ; Lung ; Sensitivity and Specificity ; Social Security ; Support Vector Machine

Purpose::Child injuries are a public health concern globally. Injury surveillance systems (ISSs) have beneficial impact on child injury prevention. There is a need for evidence-based consensus on frameworks to establish child ISSs. This research aims to investigate the key components of a child ISS for Iran and to propose a framework for implementation.Methods::Data were gathered through interview with experts using unstructured questions from January 2017 to December 2018 to identify child ISS functional components. Qualitative data were analyzed using content analysis method. Then, modified Delphi method was used to validate the functional components. Based on the outcomes of the content analysis, a questionnaire with closed questions was developed and presented to a group of experts. Consensus was achieved in two rounds.Results::In round I, 117 items reached consensus. In round II, 5 items reached consensus and were incorporated into final framework. Consensus was reached for 122 items comprising the final framework and representing 7 key components: goals of the system, data sources, data set, coalition of stakeholders, data collection, data analysis and data distribution. Each component consisted of several subcomponents and respective elements.Conclusion::This agreed framework will assist in standardizing data collection, analysis and distribution, which help to detect child injury problems and provide evidence for preventive measures.

PURPOSE: Child injuries are a global public health problem and injury surveillance systems (ISS) can be beneficial by providing timely data. However, ISS implementation has challenges. Opinions of stakeholders of ISS implementation barriers and facilitators are a good source to understand this phenomenon. The aim of this study is to investigate barriers and facilitators of implementing ISS in Iran. METHODS: This is a qualitative study. Data were gathered through interviews with 14 experts in the field of child injury and prevention from Iranian Ministry of Health and Medical Education (MOHME), medical universities, pediatrics hospitals, general hospitals and health houses during January 2017 to September 2017. Data collection and analysis continued until data saturation. Data were analyzed using content analysis through identifying meaning units. RESULTS: Barriers were classified in three main categories and nine subcategories including management barriers (including performance, coordination and cooperation, supervision and attitude), weakness in data capture and usage (including data collection, data recording and data dissemination) and resource limitation (including human and financial resources). Facilitators identified in three areas of policy making (including empowerment and attitude), management (including organization, function and cooperation and coordination) and data recording and usage (including data collection/distribution and data recording). CONCLUSION The most important barrier is lack of national policy in child injury prevention. The most important facilitator is improving MOHME function through passing supportive regulations. Effective data usage and dissemination of information to those requiring data for policy making can help reduce child injuries. Coalition of stakeholders helps overcome existing barriers.