Prevalence and Correlates of Pica Among Adults with Intellectual Disability in Institutions

Abstract

Increased knowledge of complex behaviors such as pica is needed to improve the support and services in the community for individuals with intellectual disability (ID). Though the prevalence of pica has been documented extensively in institutionalized settings, few studies have explored its etiology. The aim of this study is to explore the correlates of pica among institutionalized adults with ID. Secondary data analysis was performed on census-level data on 1,008 persons with ID residing in Ontario's remaining specialized institutions. All persons had been assessed using the interRAI Intellectual Disability (interRAI ID; Martin, Hirdes, Fries, & Smith, 2007 Martin, L., Hirdes, J. P., Fries, B. E and Smith, T. F. 2007. Development and psychometric properties of an assessment for persons with intellectual disability—The interRAI ID. Journal of Policy and Practice in Intellectual Disabilities, 4(1): 23–29. [Crossref] , [Google Scholar]) assessment instrument—a comprehensive and standardized instrument that supports service planning. Bivariate and multivariate analyses were used to explore the relation between pica and demographic, functional, and clinical characteristics. The overall prevalence of pica was 21.8%. Logistic regression analysis showed that being male, a diagnosis of autism, and use of nonverbal means of communication were associated with a higher likelihood of pica, whereas impairment in activities of daily living reduced that likelihood. A curvilinear relation was observed between cognitive functioning and pica. The prevalence of pica is quite high in Ontario's institutions, and nonverbal communication emerged as the strongest correlate of the behavior. The implications for education, training, and interventions are discussed.