Development of a delirium risk screening tool for long-term care facilities

Abstract

Objective: The aim of this study is to develop a delirium risk screening tool for use in long-term care (LTC) facilities.

Methods: The sample comprised residents aged 65 years and over of seven LTC facilities in Montreal and Quebec City, Canada, admitted for LTC. Primary analyses were conducted among residents without delirium at baseline. Incident delirium was diagnosed using multiple data sources during the 6-month follow-up. Risk factors, all measured at or prior to baseline, included the following six groups: sociodemographic, medical, cognitive status, physical function, agitated behavior, and symptoms of depression. Variables were analyzed individually and by group using Cox regression models. Clinical judgment was used to select the most feasible among similarly performing factors.

Results: The cohort comprised 206 residents without delirium at baseline; 69 cases of incident delirium were observed (rate 7.6 per 100 person weeks). The best-performing screening tool comprised five items, with an overall area under the curve of 0.82 (95% CI 0.76, 0.88). These items included brief measures of cognitive status, physical function, behavioral, and emotional problems. Using cut-points of 2 (or 3) over 5, the scale has a sensitivity of 90% (63%), specificity of 59% (85%), and positive predictive value of 52% (66%).

Conclusions: This brief screening tool allows nurses to identify LTC residents at increased risk for delirium. These residents can be targeted for closer monitoring and preventive interventions.