Abstract
Individuals with intellectual disabilities (ID) have considerable health inequalities including higher levels of unmet health needs and a shorter life expectancy compared to the general population. The prevalence of obesity, a commonly accepted measure of health inequalities, is higher in people with ID than in the general population, and the factors leading to the increased prevalence among people with ID have not been well understood yet. This has become worse during the COVID-19 pandemic due to nationwide full and partial curfews. In this study, based on a dataset that comprises a set of parameters related to eating habits and physical conditions of a number individuals, the use of classification algorithms for predicting obesity levels of individuals with ID is proposed, and a performance analysis is made using well-known performance metrics. The results could be used by researchers and practitioners in this field to choose the best classifier for their mobile application solutions. Opportunities, research challenges, and future research directions in this topic are also presented.
TopIntroduction
Intellectual Disabilities (ID) is an equivalent term for mental retardation onset before 18 years of age (American Psychiatric Association, 2013), and comprises additional support needs in at least two of the following domains: functional academic skills, communication, social and interpersonal skills, self‐direction, self‐care, home living, safety, health, work, leisure and use of community resources. The level of ID can be categorized into four classes as profound, severe, moderate and mild. People with ID suffer from considerable health inequalities compared to the general population and have significant levels of unmet health needs (Lennox, Diggens, & Ugoni, 1997; Royal College of General Practitioners Working Party, 1990), a shorter life expectancy (Patja et al., 2000; Bittles et al., 2002) and higher mortality rates (McGuigan, Hollins, & Attard, 1995; Decouflé & Autry, 2002). Obesity and being overweight are independent risk factors for chronic diseases, which could result in reduced life expectancy (Hubert et al., 1983).
As it has been shown, obesity is one of the internationally agreed health indicators for people with ID (Walsh, Kerr, & Van Schrojenstein Lantman‐de Valk, 2003; Melville et al., 2007) and one of the reasons of the reduced life expectancy and higher health needs of people with ID (Janicki et al., 2002). Therefore, the International Association for the Scientific Study of Intellectual Disabilities clearly emphasized the impact and management of obesity on health and quality of life of people with ID (Lennox et al., 2002). Body Mass Index (BMI) is a measure of body fat and applies to adults. It is calculated using Equation (1). Obesity is classified based on BMI value as listed in Table 1. 
(1)
Table 1. | BMI (kg/m2) | Weight Status | Risk of Co-morbidities |
| <18.5 | Underweight | Low |
| 18.5-24.9 | Normal | Average |
| 25.0-29.9 | Pre-obese | Increased |
| 30.0-34.9 | Obese I | Moderate |
| 35.0-39.9 | Obese II | Severe |
| ≥40 | Obese III | Very severe |
(adopted from (World Health Organization, 1998))
Key Terms in this Chapter
Obesity: A chronic condition involving an excess amount of body fat.
Classification Accuracy: The rate of correct classifications, either for an independent test set, or using cross-validation.
Classification Algorithm: It assigns a given input data to a specific category.
Electronic Health Records: Patient-centered records that can be securely and instantly obtained by authorized users.
Intellectual Disability: It is a disability characterized by impaired intellectual and adaptive functioning.