Evaluation of Patient Satisfaction Using a Random Forest Algorithm

  • Noura H. Alhashmi

    Student thesis: Master's Thesis


    Patient satisfaction is a widely used indicator in the evaluation of the quality of healthcare services. It is considered as a useful measure in assessing the quality of care as it embraces the patients' judgments about the received health service. The topic of patient satisfaction is considered a multidimensional concept with a range of determinants influencing it, such as patient-related and provider-related determinants. The concept is affected by such determinants but varied considerably across the literature. Therefore, it is still debatable which set of determinants that drive patient satisfaction in healthcare. It is also noted that the use of advanced statistical analysis is limited in this context. In particular, there is relatively little research that are dedicated to capture drivers affecting patient satisfaction level throughout the patient's entire journey in the hospital. In order to address this, we obtained patient satisfaction related data from one of the local hospitals in Abu Dhabi, UAE. The data was used to analyze the relationship between different determinants in every step of the patients' journey and patient satisfaction. Using a random forest algorithm, the results showed that ‘age' attribute which is a patient-related determinant was the leading driver of patient satisfaction in all models. While ‘time taken to answer your call', ‘total time taken for registration', ‘total time used for nurse assessment' and ‘attentiveness and knowledge of the doctor/physician while listening to your queries' were the provider-related determinants in each model developed. In terms of the most important type of question, the radar charts revealed that ‘demographics' were the most influential type of question in three of the models, whereas ‘behavior' was the most influential type in the consultation process model. In conclusion, the random forest model generated valuable results that can allow healthcare practitioners and researchers to understand the most important patient satisfaction drivers in any healthcare settings or areas of their concern.
    Date of AwardApr 2020
    Original languageAmerican English


    • Patient Satisfaction
    • Patient-Related Determinant
    • Provider-Related Determinant
    • Machine Learning
    • Random Forest Modelling
    • Predictive Analytics.

    Cite this