How accurate is the CAM-ICU screening tool for delirium for dementia patients? What research is available for screening for delirium in dementia patients?

Comment by InpharmD Researcher

Studies evaluating the use of the CAM-ICU screening tool for delirium in dementia patients have observed relatively high sensitivities and specificities, typically ranging from 90 to 100%. However, available studies often did not exclusively include patients with dementia and only evaluated the screening tool for dementia patients within subgroup analyses, limiting the findings of these studies with respect to patients with dementia. Other delirium screening tools have been evaluated in studies with small samples of dementia patients and have observed moderate to high sensitivities and specificities, similar to CAM-ICU, but due to marked heterogeneity between studies and inconsistent results, it is currently unknown which method should be considered the most reliable for assessing delirium in dementia patients.
Background

A systematic review published in 2012 indicates that while several tools exist for detecting delirium, it is unclear which are valid in patients with delirium superimposed on dementia. Of 9 studies composed of 1,569 patients, 401 patients had dementia, and 50 patients had delirium superimposed on dementia; six delirium tools were evaluated. Within a study evaluating the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) in dementia patients, the sensitivity and specificity ranged from 94% to 100% and from 90% to 95%, respectively, but specific measures for those with dementia were not reported. In another validation study of a German translation of the CAM-ICU which included a high percentage (85%) of dementia patients, the assessment had a high specificity (96% to 100%) and moderate sensitivity (77%) in delirium detection, with a likelihood ratio for a positive test of 19.25. Two additional validation studies of CAM-ICU included a small number of dementia patients among 134 total patients (see Tables 1 and 2). The sensitivity and specificity in the entire sample were 98% to 100% and 93%, respectively; of note, patients with suspected severe dementia were excluded from the studies. In another study of 129 patients admitted to a stroke unit, 31.8% had probable dementia, but similar to the other studies, patients with severe dementia were excluded. In the entire sample, the sensitivity was moderate (76%) and the specificity was high (98%). [1]

Other delirium assessment methods including quantitative electroencephalography (EEG) and resting EEG, Short-Portable Mental Status Questionnaire (SPMSQ), Cognitive Test for Delirium (CTD), and Delirium Rating Scale (DRS) have observed moderate to high sensitivities and specificities in small populations of dementia patients. However, available studies for these other assessment methods in addition to studies that evaluated CAM-ICU are limited by marked heterogeneity and unaccounted confounding factors. Additionally, no studies examined the potential effects of dementia severity or subtype upon diagnostic accuracy. For these reasons, it was determined that meta-analysis to evaluate the accurate scale was not feasible. Additionally, it was concluded that further studies of existing or refined tools with larger samples and more detailed characterization of dementia are required to address the identification of delirium superimposed on dementia. [1]

References:

[1] Morandi A, McCurley J, Vasilevskis EE, et al. Tools to detect delirium superimposed on dementia: a systematic review [published correction appears in J Am Geriatr Soc. 2013 Jan;61(1):174. Ely, Wesley E [corrected to Ely, E Wesley]]. J Am Geriatr Soc. 2012;60(11):2005-2013. doi:10.1111/j.1532-5415.2012.04199.x
Literature Review

A search of the published medical literature revealed 3 studies investigating the researchable question:

How accurate is the CAM-ICU screening tool for delirium for dementia patients? What research is available for screening for delirium in dementia patients?

Please see Tables 1-3 for your response.

 

Delirium in Mechanically Ventilated Patients: Validity and Reliability of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU)

Design

Prospective cohort study

N=96

Objective

To validate a delirium assessment instrument that uses standardized nonverbal assessments for mechanically ventilated patients and to determine the occurrence rate of delirium in such patients

Study Groups

Suspected dementia (n= 12)

Unsuspected dementia (n= 84)

Inclusion Criteria

 Mechanically ventilated adult medical and coronary intensive care unit (ICU) patients

Exclusion Criteria

History of psychosis or neurologic disease that would confound the diagnosis of delirium, admitted to the ICU but extubated before study nurses' assessments

Methods

During the study interval, mechanically ventilated patients admitted to the ICU were included for analysis. Two critical care study nurses performed CAM-ICU ratings daily during the patient's stay, and were performed independently of the delirium experts who applied DSM-IV criteria. Patients were followed up daily using the Glasgow Coma Scale and Richmond Agitation Sedation Scale (RASS) to assess fluctuations in mental status.

A subgroup analysis was performed in patients who have suspected dementia which was defined as having any of the three following criteria: rated by delirium expert as having dementia, modified Blessed Dementia Rating scale score of at least 3, or rating by the surrogate of at least 3 out of 5 as possibly having dementia.

Duration

February 1, 2000 through July 15, 2000

Outcome Measures

Inter-rater reliability of delirium assessment using CAM-ICU compared with delirium experts via sensitivity and specificity

Baseline Characteristics

  

Study patients (N= 96)

  

Mean Age, years

 55.3  

Male

 47.9%  

Race

White

Black

Hispanic

 

79.2%

19.8%

1%

  

APACHE II score

 22.9  

Glasgow Coma Scale

 7.6  

Blessed Dementia Rating Scale

 0.25  

Results

Endpoint

 Suspected dementia (n= 12) Unsuspected dementia (n= 84)

Nurse 1

Sensitivity (95% confidence interval [CI])

Specificity (95% CI)

 

100 (66 to 100)

100 (3 to 100)

 

98 (88 to 99)

100 (69 to 100)

Nurse 2

Sensitivity (95% CI)

Specificity (95% CI)

 

100 (63 to 100)

100 (40 to 100)

 

100 (92 to 100)

91 (59 to 100)

Study Author Conclusions

 The CAM-ICU appears to be rapid, valid, and reliable for diagnosing delirium in the ICU setting and may be a useful instrument for both clinical and research purposes.

InpharmD Researcher Critique

Analysis of dementia patients is limited to a subgroup of patients which is not powered to detect a statistically significant difference.



References:

Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated patients: validity and reliability of the confusion assessment method for the intensive care unit (CAM-ICU). JAMA. 2001;286(21):2703-2710. doi:10.1001/jama.286.21.2703

 

Evaluation of delirium in critically ill patients: Validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU)

Design

Prospective cohort study

N= 38

Objective

To develop and validate an instrument for use in the intensive care unit to accurately diagnose delirium in critically ill patients who are nonverbal

Study Groups

Suspected dementia (n= 11)

Inclusion Criteria

Adults admitted to the medical intensive care unit (ICU)

Exclusion Criteria

 History of severe dementia, psychosis, or neurologic disease that would confound the diagnosis of delirium; patient or family refusal to participate; admission to ICU after predefined cap of ten study patients per day

Methods

Assessment of cognitive functioning of the patients was performed independently in a blinded manner by delirium experts, nurses and intensivists on a daily basis. Daily ratings from ICU admission to hospital discharge by two study nurses and an intensivist who use the CAM-ICU were compared against the reference standard, a delirium expert that used delirium criteria from the Diagnostic and Statistical Manual of Mental Disorders (fourth edition). A subgroup analysis was conducted for patients with suspected dementia.

Duration

 June through August 1999

Outcome Measures

Reliability, sensitivity, specificity

Baseline Characteristics

  

All patients (N= 38)

Age, years

  60 ± 19  

Male

  23 (60%)  

Race

White

Black

Hispanic


84%

14%

2%

APACHE II score

17.1 ± 8.7

Mechanical ventilation

  22 (58%)  

Blessed Dementia Rating Scale

  0.76 ± 1.26  

Visual or hearing deficit by patient or family report

  71% 

Results

Endpoint

 Suspected dementia (n= 11)

Nurse 1 (95% CI)

Sensitivity

Specificity


100 (56 to 100)

100 (56 to 100)

Nurse 2 (95% CI)

Sensitivity

Specificity


100 (56 to 100)

100 (56 to 100)

Intensivist (95% CI)

Sensitivity

Specificity


100 (56 to 100)

100 (56 to 100)

Interrater reliability (kappa) ranged from 0.88 to 1.0 (p< 0.0001).

Adverse Events

 N/A

Study Author Conclusions

 ICU nurses and physicians who use the CAM-ICU can detect delirium reliably and with a high degree of sensitivity and specificity

InpharmD Researcher Critique

The sample size for patients with suspected dementia was small, thus results may be hindered by a lack of power and additional confounding factors that are unaccounted for. Sensitivity and specificity were 100% for both nurses and the intensivist, which most likely would not be observed had the sample size been larger.

 

References:

Ely EW, Margolin R, Francis J, et al. Evaluation of delirium in critically ill patients: validation of the confusion assessment method for the intensive care unit (Cam-icu): Critical Care Medicine. 2001;29(7):1370-1379.

 

Detection of Delirium in Hospitalized Older General Medicine Patients: A Comparison of the 3D-CAM and CAM-ICU

Design

Cross-sectional comparative effectiveness study

N= 101

Objective

To conduct a comparative effectiveness study of the Confusion Assessment Method for the ICU (CAM-ICU) and the newly developed 3-minute diagnostic assessment for delirium using the Confusion Assessment Method (3D-CAM) in general medicine inpatients

Study Groups

 Dementia patients (n= 26)

Inclusion Criteria

Aged ≥ 75 years, admitted to general or geriatric medicine inpatient services, able to communicate effectively in English, without imminently terminal conditions, expected hospital stay ≥ 2 days

Exclusion Criteria

Not explicitly stated

Methods

Eligible patients were initially assessed using a reference standard assessment for delirium that included patient interviews, family interviews, and review of medical records. Experts determined the presence or absence of delirium using DSM-IV criteria. Then, in a randomized fashion, patients were each assessed using both the CAM-ICU and 3D-CAM, with results compared to the reference standard.

Duration

 Not explicitly stated

Outcome Measures

Sensitivity, specificity

Baseline Characteristics

  

All patients (N= 101)

Age, years

 84 ± 5.5

Female

 62 (61%)

White

 92 (91%)

Length of stay at assessment, days

 3.4 ± 2.9

Montreal Cognitive Assessment score

 19.5 ± 6.2

Dementia by panel diagnosis

Mild

Moderate 

Severe

26 (26%)

16 (16%)

8 (8%)

2 (2%)

Mild cognitive impairment

 26 (26%)

Subysndromal delirium by reference standard

 11 (11%)

Delirium based on reference standard

Mild

Moderate

Severe

19 (19%)

10 (10%)

4 (4%)

5 (5%)

Delirium by CAM-ICU

 10 (10%)

Delirium by 3D-CAM

 24 (24%)

Results

Endpoint

 Dementia patients (n= 26)

3D-CAM

Sensitivity (95% confidence interval [CI])

Specificity (95% CI


12/13 (92%; 64% to 100%)

10/13 (77%; 46% to 95%)

CAM-ICU

Sensitivity (95% CI)

Specificity (95% CI)


8/13 (62%; 32% to 86%)

13/13 (100%; 75% to 100%)

Adverse Events

 N/A

Study Author Conclusions

In conclusion, in our study of very old, non-critically ill general medicine patients with a high prevalence of underlying dementia, the 3D-CAM performed better than the CAM-ICU as a case identification assessment for delirium. These findings do not diminish the utility of the CAM-ICU as a case identification tool in the ICU. However, our findings suggest that generalizing this instrument to general medicine patients where the spectrum of delirium is different may be suboptimal. Further research will focus on translating the 3D-CAM into the routine care of older general medicine patients, testing how well it performs when administered by clinicians, and whether improved detection of delirium can result in improved outcomes for vulnerable hospitalized elders.

InpharmD Researcher Critique

This study was limited to its small cross-sectional design. Due to only 26 dementia patients being evaluated in this study, extrapolation of results is limited. 

 

References:

Kuczmarska A, Ngo LH, Guess J, et al. Detection of Delirium in Hospitalized Older General Medicine Patients: A Comparison of the 3D-CAM and CAM-ICU. J Gen Intern Med. 2016;31(3):297-303. doi:10.1007/s11606-015-3514-0