Development and validation of a delirium predictive score in older people

Abstract

This simple predictive model highlights functional status and a proxy for dehydration as a useful tool for identifying older patients that may benefit from close monitoring and preventive care for early diagnosis of delirium.

Risk Factors for Delirium in Older Medical Inpatients in Tanzania

“…Raised serum urea or urea to creatinine ratio (used as a proxy for dehydration) was found to be significantly predictive of incident delirium, and authors have argued that these could form part of a simple screening tool [26,27]. Due to study resource constraints, we were only able to assess blood results for the admissions with delirium.…”

“…Functional dependency has been measured in a number of ways, including through assessment of dependency using the Katz Activities of Daily Living (ADL) questionnaire [26,28,29]. However, by reporting functioning by difficulties with ADL, (dressing, feeding, etc.)…”

Predicting delirium: a review of risk-stratification models

“…Given the well-established importance of this particular risk factor, it is somewhat surprising that it was not included in all models. Three other studies (27,28,31) included either advanced age or dependence in activities of daily living but not cognitive impairment. In addition, Douglas et al [28] found that brief screening for cognitive impairment on admission, that is, inability to spell WORLD backwards or answer orientation questions accurately, predicted delirium risk.…”

“…As expected, model accuracy was lower in validation cohorts than in the populations from which they were derived. In the validation cohorts, Carrasco et al [31] pursued a different strategy, developing an equation for a "Delirium Predictive Score" that, when used with a particular cut point, produces positive and negative likelihood ratios thought to be clinically useful. For instance, with a pretest probability of 20%, the posttest probabilities of delirium for a negative and positive result are 3% and 45%, respectively.…”

“…Kobayashi [19] pointed out that Inouye's "prediction tool entailed a blood test and calculation according to… APACHE II." Carrasco [31] asserted that, in addition to requiring special assessments, earlier models relied on "cognitive assessments [that], if not done properly, may misdiagnose delirium or previous cognitive impairment." Finally, Douglas [28] also suggested that "existing prediction rules…have been limited by their complexity.…”