Presented at the American College of Emergency Physicians Research Forum, Chicago, IL, October 2008.
Delirium in Older Emergency Department Patients: Recognition, Risk Factors, and Psychomotor Subtypes
Article first published online: 20 JAN 2009
© 2009 by the Society for Academic Emergency Medicine
Academic Emergency Medicine
Volume 16, Issue 3, pages 193–200, March 2009
How to Cite
Han, J. H., Zimmerman, E. E., Cutler, N., Schnelle, J., Morandi, A., Dittus, R. S., Storrow, A. B. and Wesley Ely, E. (2009), Delirium in Older Emergency Department Patients: Recognition, Risk Factors, and Psychomotor Subtypes. Academic Emergency Medicine, 16: 193–200. doi: 10.1111/j.1553-2712.2008.00339.x
Dr. Han received support from the Vanderbilt Physicians Scientist Development Grant.
- Issue published online: 4 MAR 2009
- Article first published online: 20 JAN 2009
- Received July 7, 2008; revisions received September 30 and October 29, 2008; accepted October 30, 2008.
- emergency department;
- risk factors
Objectives: Missing delirium in the emergency department (ED) has been described as a medical error, yet this diagnosis is frequently unrecognized by emergency physicians (EPs). Identifying a subset of patients at high risk for delirium may improve delirium screening compliance by EPs. The authors sought to determine how often delirium is missed in the ED and how often these missed cases are detected by admitting hospital physicians at the time of admission, to identify delirium risk factors in older ED patients, and to characterize delirium by psychomotor subtypes in the ED setting.
Methods: This cross-sectional study was a convenience sample of patients conducted at a tertiary care, academic ED. English-speaking patients who were 65 years and older and present in the ED for less than 12 hours at the time of enrollment were included. Patients were excluded if they refused consent, were previously enrolled, had severe dementia, were unarousable to verbal stimuli for all delirium assessments, or had incomplete data. Delirium status was determined by using the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) administered by trained research assistants (RAs). Recognition of delirium by emergency and hospital physicians was determined from the medical record, blinded to CAM-ICU status. Multivariable logistic regression was used to identify independent delirium risk factors. The Richmond Agitation and Sedation Scale was used to classify delirium by its psychomotor subtypes.
Results: Inclusion and exclusion criteria were met in 303 patients, and 25 (8.3%) presented to the ED with delirium. The vast majority (92.0%, 95% confidence interval [CI] = 74.0% to 99.0%) of delirious patients had the hypoactive psychomotor subtype. Of the 25 patients with delirium, 19 (76.0%, 95% CI = 54.9% to 90.6%) were not recognized to be delirious by the EP. Of the 16 admitted delirious patients who were undiagnosed by the EPs, 15 (93.8%, 95% CI = 69.8% to 99.8%) remained unrecognized by the hospital physician at the time of admission. Dementia, a Katz Activities of Daily Living (ADL) ≤ 4, and hearing impairment were independently associated with presenting with delirium in the ED. Based on the multivariable model, a delirium risk score was constructed. Dementia, Katz ADL ≤ 4, and hearing impairment were weighed equally. Patients with higher risk scores were more likely to be CAM-ICU positive (area under the receiver operating characteristic [ROC] curve = 0.82). If older ED patients with one or more delirium risk factors were screened for delirium, 165 (54.5%, 95% CI = 48.7% to 60.2%) would have required a delirium assessment at the expense of missing 1 patient with delirium, while screening 141 patients without delirium.
Conclusions: Delirium was a common occurrence in the ED, and the vast majority of delirium in the ED was of the hypoactive subtype. EPs missed delirium in 76% of the cases. Delirium that was missed in the ED was nearly always missed by hospital physicians at the time of admission. Using a delirium risk score has the potential to improve delirium screening efficiency in the ED setting.
Missing delirium in the emergency department (ED) has been described as a medical error and an issue of quality of care.1 This form of organ dysfunction occurs in 1 of 10 older ED patients2 and is a major threat to their quality of life. Delirium is associated with higher death rates,3–5 prolonged hospitalization,6,7 increased health care costs,8 and accelerated long-term functional and cognitive impairment.9,10 Despite its frequent occurrence and negative consequences, delirium is missed by emergency physicians (EPs) in 57% to 83% of cases.5,11–15 There is some evidence to suggest that missing delirium in the ED portends higher-risk compared to patients whose delirium is detected by the EP.5
Delirium is missed at a high rate because EPs do not routinely screen for this diagnosis.16 The ED is a highly chaotic and demanding environment. Adding a delirium assessment to the traditional history and physical examination may not be feasible for all ED patients. However, performing delirium assessments on a subset of high-risk older patients may be practical. To the best of our knowledge, no study has adequately characterized delirium risk factors in the older ED patient. Most studies concerning delirium risk factors have been conducted in the inpatient setting and have included patients who developed delirium during hospitalization, limiting the generalizability of these studies to the ED patient.17,18
In addition, to the best of our knowledge, no ED study has characterized delirium by its psychomotor subtypes: hypoactive (“quiet”), hyperactive, and mixed.19 Hypoactive delirium is characterized by decreased psychomotor activity and has the appearance of depression and sedation. This subtype is most often missed by physicians and can be difficult to identify without a delirium assessment because of its subtle presentation.20 Hyperactive delirium is characterized by increased psychomotor activity, anxiety, and agitation. A patient with mixed-type delirium exhibits fluctuating levels of psychomotor activity over a period of time. Distinguishing delirium between its psychomotor subtypes also has important clinical ramifications, because each subtype has been associated with differential outcomes.21,22
To address these deficiencies, we performed a cross-sectional study that sought 1) to determine how often delirium is missed in the ED and how often these missed cases are detected by admitting hospital physicians at the time of admission, 2) to identify delirium risk factors in older ED patients, and 3) to characterize delirium by psychomotor subtypes in the ED setting.
We conducted a prospective cross-sectional study in a tertiary care, academic ED with an annual census of approximately 55,000. Approximately 10% of the annual census consists of patients who are 65 years and older. Because we wanted to study delirium’s natural course, emergency and hospital physicians were blinded to the study objectives and patients’ delirium status. Our local institutional review board reviewed and approved this study with these conditions, because performing delirium screening in the ED was not standard of care, and there was no evidence to suggest that early detection of delirium in the ED improved patient outcomes.
Study Setting and Population
This was a convenience sample of patients who were enrolled from May 2007 to July 2007 from 8AM to 10PM. ED patients who were 65 years and older and present in the ED for less than 12 hours at the time of enrollment were included. The purpose of the 12-hour limit was to maximize the number of patients that could be enrolled, while minimizing extraneous factors that would artificially cause the EP to not recognize delirium, such as physician shift change or new-onset delirium from prolonged exposure to known delirium precipitants (e.g., psychoactive medications). This limit was based upon our ED’s typical waiting room wait times and duration of an elder patient evaluation, EP shift duration, and research assistant (RA) availability. Patients who refused consent, were non–English-speaking, were previously enrolled, had severe dementia, were unarousable to verbal stimuli for all delirium assessments, or had incomplete data were excluded. Patients with incomplete data either withdrew from the study, or the prospective data collection could not be completed because they left the ED before the assessments could be completed. Patients who met inclusion and exclusion criteria were enrolled in the study after verbal consent was obtained from the patients or their authorized surrogates.
Delirium, dementia, and functional status were prospectively collected by two RAs. Prior to the start of the study, the RAs participated in an intensive 1-week training period where they studied training manuals, received didactic lectures, watched live patient demonstrations, and practiced administering the assessments using simulated patient scenarios. At the end of the training period, the primary investigator (JHH) observed the RAs performing these assessments with actual ED patients.
Delirium was determined using the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU).23 The CAM-ICU is a modification of the Confusion Assessment Method (CAM), but uses the exact same features as CAM: 1) acute onset of mental status changes or a fluctuating course, 2) inattention, 3) disorganized thinking, and 4) altered level of consciousness (Figure 1).24 Unlike the CAM, which requires clinical judgment to assess for all four features, the CAM-ICU uses objective assessments from objective neuropsychiatric tests to determine inattention (Feature 2) and disorganized thinking (Feature 3). In addition, the CAM-ICU is brief (less than 2 minutes) compared to the CAM (up to 10 minutes) and is easier to administer. This made the CAM-ICU ideal for the busy ED environment where interruptions frequently occur. The CAM-ICU has been validated in both mechanically ventilated and non–mechanically ventilated patients and has high sensitivity (93% to 100%) and specificity (98% to 100%) and excellent interrater reliability (κ = 0.77 to 0.95).23,25,26 Acute onset of mental status changes or fluctuating course (Feature 1) was determined by surrogate history. If the patient was from a nursing home and there was no documentation of altered mental status on the nurse’s triage assessment or nursing home transfer sheet, the nursing home staff was interviewed. Because of the waxing and waning nature of delirium, the CAM-ICU was performed at 0 and 3 hours. A patient was considered to have delirium if either the 0- or the 3-hour assessment was positive.
In patients who were CAM-ICU positive, the Richmond Agitation and Sedation Scale (RASS) was used to categorize the psychomotor subtype of delirium.27,28 As previously reported, patients with a RASS score between +1 and +4 were considered to have hyperactive delirium. Patients with a RASS score between 0 and −3 were considered to have hypoactive delirium. Patients exhibiting both positive and negative RASS scores at 0 and 3 hours were considered to have the mixed type.
The determination of whether or not delirium was recognized by emergency and hospital physicians was performed by medical record review using previously established criteria.11,14,15 Any reference to acute or new confusional state or disorder, acute mental status change, encephalopathy, toxic-metabolic state, or acute organic brain syndrome in the physician’s impression and diagnosis indicated provider recognition of delirium.11,14,15 Documentation of a delirium assessment performed by the emergency and hospital physician was also abstracted from the history and physical examination. For the hospital physician, only the initial history and physical examination were used, which were typically performed several hours after the initial ED assessment. Delirium recognition was reassessed by the same chart reviewer 3 months after the initial review; no discrepancies were found. Physician interview was not performed to maintain feasibility (e.g., physician shift change and high volume of patients) of the study. In addition, a previous study conducted in the ED showed that adding a physician interview to medical record review only increased the delirium recognition by 11.8%.15 The chart review was performed by a single investigator (JHH) who was blinded to the patient’s CAM-ICU status, but not to the study hypothesis.
Dementia was determined by the Mini-Mental State Examination (MMSE)29, or the short form of the Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE)30, or from the medical record. The MMSE was only performed in patients who were CAM-ICU negative, because it would not have accurately reflected a delirious patient’s premorbid cognition. Patients who had a MMSE score less than 24 or an IQCODE score greater than 3.38 or had dementia documented in the medical record were considered to have dementia. Functional status was measured using the Katz Activities of Daily Living (Katz ADL).31 Patients with a Katz ADL ≤ 4 were considered to be functionally dependent.
Patient demographics, medical history, number of home medications, residence, visual or hearing impairment, and recent hospitalization were obtained from the patients, their surrogates, and the medical record. Visual and hearing impairment were not objectively measured, but were assessed by history and the presence of corrective lenses or hearing aids. Chief complaint, EP diagnosis, vital signs, and emergency severity index (ESI) at triage were also obtained from the medical record. The Charlson Comorbidity Index was used to measure comorbid burden.32 The presence of systemic inflammatory response syndrome (SIRS) was used as a surrogate for severity of illness. Patients were considered to have SIRS if they had two or more of the following criteria: 1) heart rate > 90 beats/min, 2) body temperature <36 or >38°C, 3) respiratory rate > 20 breaths/min, or 4) white blood cell (WBC) count <4 × 109 or >12 × 109 cells/L.33 All data abstraction from the medical record occurred after patient enrollment and were double checked for accuracy.
Proportions with their 95% confidence intervals (CIs), medians, and interquartile ranges (IQRs) were reported where appropriate. For simple comparisons, chi-square analyses were performed for categorical data, and Wilcoxon rank sum tests were performed for continuous data. Multivariable logistic regression was performed to determine which clinical variables were independently associated with delirium in the ED. Age, gender, race, dementia, Katz ADL ≤ 4, visual impairment, hearing impairment, Charlson Comorbidity Index, number of home medications, triage ESI, SIRS, ED diagnosed infection, nursing home residence, and hospitalization within the past week were considered for the model and were based on literature review and expert opinion.34 Given the number of patients with events (CAM-ICU positive), only three covariates were selected for the multivariable model in order to avoid overfitting.35 We utilized a forward selection process and first considered covariates that were biologically plausible and were consistently found to be associated with delirium in the hospital literature. We selected a combination of covariates that had the highest discriminatory power (c-statistic). If two or more models had similar discriminatory power, we chose the model that used covariates that would potentially be more readily available to the EP. All covariates included in the model were reported in odds ratios (ORs) with their 95% CIs. Pearson chi-square test was performed on the logistic regression model to test for goodness of fit. We also performed a secondary analysis to determine how the number of delirium risk factors affected the likelihood of delirium; each risk factor was weighted according to its effect size. A receiver operating characteristic (ROC) curve was constructed, and the area under the curve (AUC) was calculated. Using ROC curve analysis, an optimal cut point was chosen. Sensitivity, specificity, and positive and negative likelihood ratios with their 95% CIs were determined for that cut point.36 A p-value less than 0.05 was considered statistically significant. All statistical analyses were performed with SAS 9.1 (SAS Institute, Cary, NC) and Microsoft Excel 2003 (Microsoft Corp., Seattle, WA).
A total of 376 patients were screened, and 303 met inclusion but not exclusion criteria (Figure 2). The median (IQR) age was 74 (69–80) years old, 169 (55.8%) were females, 50 (16.5%) were nonwhite, and 20 (6.6%) were from a nursing home. Of the patients who had 0-hour CAM-ICUs performed, 21 (6.9%) were CAM-ICU positive. Two patients were initially not assessable because they were in a stupor or coma. Eighty-two (27.1%) patients had the 3-hour assessment performed; an additional 4 patients were CAM-ICU positive. Combining the 0- and 3-hour CAM-ICU assessments, 25 (8.3%) of our cohort had delirium.
Of the 25 patients with delirium, 19 (76.0%, 95% CI = 54.9% to 90.6%) were not recognized to be delirious by the EP. Four patients with delirium were discharged home, and of these, only 1 patient was determined to be delirious by the EP. No ED patient had a delirium assessment documented in the history and physical examination by the EP.
Twenty-one delirious patients were admitted to the hospital, and 15 (71.4%, 95% CI = 47.8% to 88.7%) were not recognized to have delirium by the admitting hospital physician. In the 5 admitted patients in whom delirium was recognized by the EP, all were recognized by the hospital physician. Of the 16 admitted patients whose delirium was undiagnosed by EPs, only 1 patient (6.3%, 95% CI = 0.2% to 30.2%) was recognized by the hospital physician at the time of admission. None of the hospitalized patients had a delirium assessment documented in the history and physical examination.
Patient demographics, history, and clinical variables (Table 1) were compared between older ED patients with and without delirium. Patients with delirium were more likely to be older, reside in a nursing home, have dementia or a Katz ADL ≤ 4, have visual or hearing impairment, be on more home medications, meet SIRS criteria, and have an infectious etiology diagnosed by the EP. No differences in Charlson Comorbidity Index, hospitalization within the past week, and triage ESI were observed between the delirium and nondelirium groups. In the multivariable logistic regression model, dementia (adjusted OR = 3.3, 95% CI = 1.2 to 8.9), a Katz ADL ≤ 4 (adjusted OR = 4.4, 95% CI = 2.1 to 9.4), and hearing impairment (adjusted OR = 3.8, 95% CI = 1.4 to 10.0) were independently associated with delirium in the ED. The c-statistic for the model was 0.83 and the Pearson chi-square test was 0.204, indicating that there is no proof of lack of fit.
|Variable||Delirium Positive (n = 25)||Delirium Negative (n = 278)||p-value|
|Median age, years (IQR)||80 (72, 85)||74 (69, 79)||0.009|
|Male||10 (40.0)||124 (44.6)||0.657|
|Non-white||43 (15.5)||7 (28.0)||0.106|
|Home alone||2 (8.0)||75 (27.0)||<0.001|
|Home with others||11 (44.0)||185 (66.6)|
|Assisted living||2 (8.0)||3 (1.1)|
|Nursing home||9 (36.0)||11 (4.0)|
|Rehabilitation||1 (4.0)||0 (0.0)|
|Homeless||0 (0.0)||1 (0.4)|
|Other||0 (0.0)||3 (1.1)|
|Median Charlson (IQR)||2 (1, 4)||2 (1, 3)||0.165|
|Median total medications (IQR)||10 (8, 13)||7 (4, 11)||0.013|
|Dementia||19 (76.0)||106 (38.1)||<0.001|
|Katz ADL ≤ 4||16 (64.0)||46 (16.6)||<0.001|
|Visual impairment||11 (44.0)||61 (21.9)||0.013|
|Hearing impairment||9 (36.0)||36 (13.0)||0.002|
|Hospitalized within past week||4 (16.0)||17 (6.1)||0.062|
|SIRS criteria||18 (72.0)||120 (43.2)||0.006|
|1||0 (0.0)||0 (0.0)||0.815|
|2||18 (72.0)||177 (63.7)|
|3||7 (28.0)||94 (33.8)|
|4||0 (0.0)||6 (2.2)|
|5||0 (0.0)||1 (0.4)|
|EP-diagnosed infection||12 (48.0)||41 (14.8)||<0.001|
From the multivariable model, a delirium risk score was developed. Dementia, Katz ADL ≤ 4, and hearing impairment were weighed equally. The proportion of older ED patients with delirium was stratified by the delirium risk score (Figure 3). As the delirium risk score increased, the proportion of older ED patients with delirium increased. The AUC was 0.82. Using a cutoff of one or more points, the sensitivity was 96.0% (95% CI = 88.0% to 100.0%), the specificity was 49.3% (95% CI = 43.4% to 55.2%), the positive likelihood ratio was 1.9 (95% CI = 1.6 to 2.2), and the negative likelihood ratio was 0.08 (95% CI = 0.01 to 0.56) for delirium. If older ED patients with one or more delirium risk factors were screened for delirium, 165 (54.5%, 95% CI = 48.7% to 60.2%) would have required a delirium assessment at the expense of missing 1 patient with delirium, while screening 141 patients without delirium.
The vast majority (92.0%, 95% CI = 74.0% to 99.0%) of delirious ED patients had the hypoactive psychomotor subtype of delirium. Two patients had hyperactive or mixed-type delirium and both were recognized by EPs as having altered mental status. Of those with hypoactive delirium (n = 23), 18 (78.3%, 95% CI = 56.3% to 92.5%) were not recognized by EPs.
Our cross-sectional study provides a comprehensive investigation of delirium in the ED and observed three key findings not previously reported in the ED literature: 1) delirium that was unrecognized by EPs was most likely to be missed by hospital physicians at the time of admission, 2) delirium risk factors were characterized in older ED patients and may help identify patients at high risk for having delirium, and 3) the vast majority of delirious older patients presented to the ED with the hypoactive subtype.
Delirium is a significant problem in the ED, and a large proportion of patients with delirium are unrecognized. Similar to previous reports, we observed that 76.0% of the cases of ED delirium were not recognized by EPs.5,11–15 Adding to the existing body of literature, we observed that over 90% of admitted patients whose delirium was missed by the EP was also missed by the hospital physician at the time of admission. This suggests that if delirium is missed in the ED, there is potential delay in diagnosing delirium in the hospital setting. The consequences of missed delirium in the ED are unclear. However, Kakuma et al.5 studied older patients discharged from the ED and observed that patients whose delirium was unrecognized by the EP had the highest death rate, compared to ED patients whose delirium was recognized and patients without delirium.5 Other potential consequences for missing delirium exist; ED patients with underlying life-threatening illnesses may receive inappropriate diagnostic evaluations and be discharged home. If discharged, delirious patients may not be able to comprehend their discharge instructions, and this may lead to noncompliance and recidivism. As a result, improved detection and earlier recognition of delirium in the ED has the potential to improve patient outcomes.
We observed that underrecognition of delirium by EPs is secondary to the absence of routine delirium screening in the ED. There are several potential reasons for this. EPs have little didactic training in geriatric medicine, especially in assessing for delirium.37 Several bedside delirium assessments are available,38,39 and the CAM is the most widely used in the clinical and research setting.24 However, the CAM may be not feasible in the busy ED environment, as it requires up to 10 minutes to perform.40 The CAM-ICU used in our study is a modification of the CAM and requires 30 seconds to 2 minutes to perform. However, spending even an additional 30 seconds to 2 minutes on the typical patient assessment may be difficult in the ED setting, especially during periods of high demand.
Therefore, we attempted to identify patients who are at high risk for presenting to the ED with delirium. Performing delirium assessments on selected high-risk patients may potentially improve ED delirium screening. We identified dementia, functional dependence, and hearing impairment as independent risk factors for presenting to the ED with delirium. Our findings are consistent with studies conducted in hospitalized patients. Dementia is the strongest and most consistently observed risk factor for delirium.18,34,41,42 Similarly, independent associations between functional or hearing impairment and delirium have been reported.34,43
To the best of our knowledge, our study is the first ED study to characterize delirium by its psychomotor subtypes. The vast majority of our older ED patient population presented with hypoactive symptomatology, whereas hyperactive delirium was rarely observed. Our findings are consistent with hospital-based studies.28,44,45 Distinguishing delirium by its psychomotor subtypes is essential for several reasons. The etiology and pathophysiology of delirium may differ between the various psychomotor subtypes.46 In addition, delirious patients with hypoactive symptomatology are significantly more likely to be unrecognized and misdiagnosed for psychiatric illnesses.19 In our study, 78.3% of patients with hypoactive delirium were not recognized by EPs. However, our study sample was too small to make any firm conclusions.
The different psychomotor subtypes of delirium also have important prognostic implications. In hospitalized medical patients, hypoactive delirium is associated with prolonged hospital length of stay7 and higher mortality.22 However, in older patients who receive hip fracture repairs, hypoactive delirium is associated with lower rates of death, and the patient is less likely to be placed in a nursing home, compared to patients with any hyperactivity.45 Given the heterogeneity of these findings and the limited size of our study sample, it is unclear how specific psychomotor subtypes of delirium affect ED patient outcomes; future research in the ED setting is required to clarify this relationship.
Deficiencies in our understanding of delirium in the ED patient still remain. It is unclear if missing delirium in the ED is associated with worse outcomes, and if early detection and treatment of delirium in the ED will improve delirium’s adverse effects on long-term mortality, cognitive and functional impairment, and quality of life. Furthermore, there is no universally accepted treatment for delirium in the ED, and hospital interventions for delirium have not proven to be successful in thwarting delirium’s negative consquences.47 As a result, a multifaceted line of future investigations must be conducted to address this dearth of knowledge and improve the quality of care delivered to the older ED patient.48
Given the constraints of the busy ED environment and limited lengths of stay (∼5 hours), we had to balance the amount of prospective data collected against feasibility. Consequently, we used the CAM-ICU to assess for delirium, because of its ease of use, brevity (less than 2 minutes), and high reliability. However, the CAM-ICU has not been formally validated in the ED setting. McNicoll et al.49 compared the CAM-ICU with the CAM and found that the CAM-ICU was 73% sensitive and 100% specific compared to the CAM. It is possible that the proportion of older patients with delirium was underestimated. However, the proportion of our cohort with delirium is comparable to those in prior studies using other instruments.5,11–15
Only 27.0% had their 3-hour CAM-ICUs performed. Patients with missing 3-hour assessments were more likely to be discharged from the ED. Because our definition of delirium was a positive CAM-ICU assessment at either 0 or 3 hours, we may have underestimated the proportion of patients with delirium.
This was a convenience sample, as we were only able to enroll patients during the daytime and early evening hours, potentially introducing selection bias. Older patients who present to the ED during the early morning or late evening hours may have been missed; these patients may have had a higher acuity of illness, and may have been more likely to present with delirium. Therefore, the proportion of delirious ED patients may have been underestimated.
The proportion of delirium that was unrecognized by emergency and hospital physicians was obtained from the medical record. It is possible that physicians may have recognized delirium, but failed to document their findings, possibly overestimating the proportion of missed delirium. However, interviewing the physician would most likely have modestly affected our delirium recognition rate. Elie et al.15 performed chart review and physician interview to determine how often EPs recognized delirium; the physician interview increased the absolute recognition rate by 11.8%. Even if our proportion of unrecognized delirium was overestimated by 25%, a 50% miss rate would still be clinically significant.
Our analysis of the usefulness of the number of delirium risk factors to improve screening efficiency was exploratory in nature and has not been validated. Future multicenter studies will be performed to validate these findings.
Our sample size was relatively small. We limited the number of covariates included in the multivariable model to avoid overfitting. As a result, certain delirium risk factors may have been excluded from the multivariable model. However, our small model achieved a c-statistic of 0.83, indicating that our multivariable model had very good predictive ability. We were also unable to adequately address how the psychomotor subtypes affect recognition of delirium in the ED. This study was performed at a single center, and our findings may not be generalizable to rural, nonacademic, or non–tertiary care centers.
Delirium commonly occurs in older ED patients, and the vast majority are of the hypoactive subtype. EPs miss delirium at a high rate because they do not routinely screen for this diagnosis. For the subset of admitted patients, if delirium is missed in the ED, there is a high likelihood that it will go unnoticed by hospital physicians at the time of admission. Delirium screening in the ED may be best focused on patients with one of the following risk factors: dementia, a Katz ADL ≤ 4, and any hearing impairment.
The authors acknowledge Karen Miller, RN, MPA, for coordinating this study and Amanda K. Laun, RN, for her assistance in editing the manuscript.
- 33American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med. 1992; 20:864–74.
- 35Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis. New York, NY: Springer, 2001..
- 40The Confusion Assessment Method (CAM): Training Manual and Coding Guide. Available at: http://www.hospitalelderlifeprogram.org/pdf/The_Confusion_Assessment_Method.pdf. Accessed Dec 15, 2007..