AWOL (Age, WORLD backwards, Orientation to place, nurse-rated iLlness severity) is a 4-item bedside prediction rule to estimate risk of incident delirium during a general medical hospital admission.

Douglas and colleagues prospectively derived the score in 209 consecutive medical inpatients aged ≥50 years without delirium at admission, and validated it in a further 165 patients at two academic hospitals and a Veterans Affairs medical centre in the United States. Delirium was assessed daily for up to 6 days using the Confusion Assessment Method, and 1 point was assigned for each of age ≥80 years, inability to spell WORLD backwards, disorientation to place and nurse-rated illness severity of moderate or worse.

AWOL score	Incident delirium in combined derivation + validation cohorts
0	1/50 (2%)
1	5/141 (4%)
2	15/107 (14%)
3	10/50 (20%)
4	7/11 (64%)

The score showed good discrimination in the derivation cohort (area under the receiver operating characteristic curve 0.81, 95% CI 0.73–0.90) and more modest performance in validation (AUC 0.69, 95% CI 0.54–0.83).

On a neurosciences ward, AWOL has been incorporated into a multicomponent delirium care pathway, using AWOL ≥2 to define higher risk and trigger non-pharmacological prevention strategies. In that setting delirium incidence did not change, but length of stay among delirious patients decreased by around 2 days.

An electronic health record–based machine-learning model developed in the same health system used AWOL as a comparator in more than 18,000 adult non-ICU admissions, finding AWOL had an AUC of about 0.68 and was outperformed by complex models, although it remains a simple point-of-care bedside tool.

Use this score mainly in older adult, English-speaking, general medical inpatients without delirium on admission. Performance may be weaker in substantially different populations such as much younger adults, ICU or surgical patients, non-English speakers, or patients unable to complete brief cognitive testing.

AWOL Score for Delirium