Acute confusion indicators: risk factors and prevalence using MDS data.

Long-term care (LTC) Minimum Data Set (MDS) data from a Midwestern state were analyzed to validate whether components of a conceptual model developed from findings in acute care identified acute confusion risk variables in LTC. The prevalence of probable acute confusion in this sample was 13.98% (n = 324). Using a cross-sectional design, both univariate and unconditional stepwise logistic regression analyses were accomplished with presence or absence of probable acute confusion as the outcome variable (N = 2,318). Variables significantly related to acute confusion by univariate analysis were included in the logistic regression analysis. Inadequate fluid intake was the first variable to enter the stepwise equation and was highly significant (OR 3.40, 95% CI 2.99-3.81, p < .0001). Other significant variables included a diagnosis of dementia or a fall in the last 30 days. Implications for nursing practice, education and research are discussed.