A systematic review of Canadian and American cost-of-illness studies in dementia

Abstract

Estimates of the annual cost of dementia range into the billions of dollars. The objective of this study was to conduct a systematic review of published cost-of-illness (COI) studies in Canada and the United States, as well as to synthesize the conclusions of these studies into a set of common themes about the cost of dementia. Medline, CINAHL, EconLit, AMED, and the Cochrane Library were searched from inception to October 2009 for English-language COI studies. Other types of economic studies (cost-benefit, cost-effectiveness, and cost-utility analyses) were excluded from the review. Articles retrieved in the literature search were screened for inclusion using two levels of screening: (1) title and abstract and (2) full text. Data were extracted from articles that passed both levels of screening. Extracted data included information on study author(s), design, study participants, cost data, and sponsorship. Content analysis was used to synthesize the major conclusions of the extracted COI studies into a set of common themes about dementia cost. Nine hundred fifty-one articles were retrieved in the literature search and data were extracted from 46 articles. All except three articles reported American data. Most articles pertained to AD. Cost estimates varied considerably across articles. Sources of variation were differences in study populations and heterogeneity of COI methodologies (e.g., types of included costs, methods of calculating indirect costs). Despite cost variability, the extracted studies contained several common themes about the cost of dementia. Themes included general observations about dementia cost, cost drivers in dementia, caregiver cost, items that may lower dementia cost, social service cost, Medicare and Medicaid cost, and cost comparisons with other diseases. The content of each of the common themes could serve as the basis for policy-oriented research. However, authors of COI articles tended to focus on reporting the cost of dementia as an end in itself. Given methodological heterogeneity, there was no way to ascertain which (if any) of the articles actually reported the true cost of dementia.