Hospital discharge data under-reports delirium occurrence: results from a point prevalence survey of delirium in a major Australian health service

Background: Delirium in hospitalised patients is common, and a risk factor for adverse outcomes. Health services require accurate delirium data to monitor the impact of initiatives designed to improve detection and prevention of delirium.

Aim: To determine the extent to which International Classification of Disease codes represent delirium occurrence.

Methods: A cross-sectional point prevalence survey was used to audit delirium occurrence in 25 inpatient wards of an Australian health service. All adult patients were eligible. Exclusion was for coma, end of life or behaviour that posed a risk to delirium assessors. Specially trained nurses and allied health professionals (AHP) screened patients for any cognitive impairment using the 4 A's Test (4AT). Those with abnormal screen test results were assessed using the '3-Minute Diagnostic Interview for the Confusion Assessment Method' (3D-CAM). Delirium detected by 3D-CAM was the reference standard.

Results: Of potentially eligible patients, 559 of 667 (83.8%) patients were assessed. The mean age was 73 years (±16.4), 54.5% were female and 43.8% (245/559) had cognitive impairment (4AT score ≥1). The occurrence of delirium during hospitalisation as identified by ICD codes was 10.4% (58/559; 95% confidence interval (CI), 7.9-12.7) compared with a point prevalence of 16.2% (91/559; 95% CI, 13.2-19.1). Only 31 of 91 (34.1%) of those with delirium had ICD delirium codes assigned.

Conclusion: ICD coding is inadequate to determine in-hospital delirium incidence. Instead, a point prevalence detection of delirium using the methods described above could be used. Health services could apply the described survey method to evaluate their local initiatives for the improvement of delirium detection and prevention.