Exploring the antecedents of the quality of life of patients with sickle cell disease: using a knowledge discovery and data mining process model-based framework. Issue 1 (1st March 2016)

• Record Type: Journal Article
• Authors: Mansingh, Gunjan; Osei-Bryson, Kweku-Muata; Asnani, Monika
• Is Part Of: Health systems. Volume 5:Issue 1(2016)
• Page Start: 52
• Page End: 65
• Publication Date: 2016-03-01
• Subjects: sickle cell disease -- quality of life -- data mining -- decision tree induction -- regression -- regression splines; 610.285
• DOI: 10.1057/hs.2015.3
• Languages: English
• ISSNs: 2047-6965

Abstract

Abstract: Sickle cell disease (SCD) is the most common single-gene disorder worldwide and has multiple and variable manifestations. The many medical complications associated with SCD such as acute chest syndrome and painful crises, along with a lack of normal functioning, may lead to various psychosocial problems such as depression, loneliness and impaired quality of life (QOL). A few studies have sought to examine the relationships between demographics, disease severity, depression, loneliness and the QOL of patients with SCD. In this paper we apply an integrated knowledge discovery and data mining (IKDDM) process to explore the factors that impact the QOL of patients with SCD in Jamaica to explicate knowledge that can be used by medical professionals. Following the IKDDM process provides several benefits: (1) it ensures that adequate experimentation is done to ensure that the best model will be generated and (2) it provides guidance in generating and evaluating models. We use different data mining techniques such as Decision Trees Induction, Regression and Regression Splines to analyze the data and multiple performance measures to evaluate the models in order to identify the best set of models to present to the medical professionals. This allows the medical professionals to select model(s) that will assist them in the decision-making process. The results of this study confirm prior hypotheses regarding the variables predictive of the QOL of SCD patients and additionally provide new insights by identifying the values for these variables.