This study compares the proportions of surgical intensive care unit (ICU) patients with delirium detected using the Confusion Assessment Method for the ICU (CAM-ICU) who received administrative documentation for delirium using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes, stratified by delirium motoric subtypes. This retrospective cohort study was conducted at a surgical ICU from 06/2012 to 05/2013. Delirium was assessed twice daily and was defined as having ≥1 positive CAM-ICU rating. Delirious patients were categorized into hyperactive/mixed and hypoactive subtypes using corresponding Richmond Agitation Sedation Scales. Administrative documentation of delirium was defined as having ≥1 of 32 unique ICD-9-CM codes. Proportions were compared using Pearson's Chi-square test. Of included patients, 40 % (423/1055) were diagnosed with delirium, and 17 % (183/1055) had an ICD-9-CM code for delirium. The sensitivity and specificity of ICD-9-CM codes for delirium were 36 and 95 %. ICD-9-CM codes for delirium were available for 42 % (95 % CI 35-48 %; 105/253) of patients with hyperactive/mixed delirium and 27 % (95 % CI 20-34 %; 46/170) of patients with hypoactive delirium (relative risk = 1.5; 95 % CI 1.2-2.0; p = 0.002). ICD-9-CM codes yielded a low sensitivity for identifying patients with CAM-ICU positive delirium and were more likely to identify hyperactive/mixed delirium compared with hypoactive delirium.
Keywords: CAM-ICU; Delirium; Delirium motoric subtypes; ICD-9-CM; Surgical ICU.