Objectives: Patients with dementia in the acute setting are generally considered to impose higher costs on the health system compared to those without the disease largely due to longer length of stay (LOS). Many studies exploring the economic impact of the disease extrapolate estimates based on the costs of patients diagnosed using routinely collected hospital discharge data only. However, much dementia is undiagnosed, and therefore in limiting the analysis to this cohort, we believe that LOS and the associated costs of dementia may be overestimated. We examined LOS and associated costs in a cohort of patients specifically screened for dementia in the hospital setting. Methods: Using primary data collected from a prospective observational study of patients aged≥70 years, we conducted a comparative analysis of LOS and associated hospital costs for patients with and without a diagnosis of dementia. Results: There was no significant difference in overall length of stay and total costs between those with (μ= 9.9 days, μ=€8246) and without (μ= 8.25 days, μ=€6855) dementia. Categorical data analysis of LOS and costs between the two groups provided mixed results. Conclusions: The results challenge the basis for estimating the costs of dementia in the acute setting using LOS data from only those patients with a formal dementia diagnosis identified by routinely collected hospital discharge data. Accurate disease prevalence data, encompassing all stages of disease severity, are required to enable an estimation of the true costs of dementia in the acute setting based on LOS.