Psychosomatics 40:193-204, June 1999
© 1999 The Academy of Psychosomatic Medine
The Delirium Rating Scale
Its Use In Consultation-Liaison Research
Paula T. Trzepacz , M.D.
Received August 26, 1998; revised November 24, 1998; accepted December 4, 1998. From the University of Mississippi Medical School, Jackson, Mississippi. Previously published in similar form in Japanese in Neuro-Psychiatric Review, Vol. 26, 1998, pp. 16–31. Address correspondence and reprint requests to Dr. Trzepacz, Department of Psychiatry and Human Behavior, University of Mississippi Medical Center, 2500 North State St., Jackson, MS 39216; e-mail: ptrzepacz{at}psychiatry.umsmed.edu
|
ABSTRACT
|
In addition to diagnostic criteria, delirium research requires standardized instruments to measure symptoms. This article reviews the literature about the Delirium Rating Scale (DRS), the most widely used scale to assess delirium that has been translated into at least seven other languages. The DRS has 10 items and is clinician-rated, but 7- or 8-item subscale adaptations have been used for repeated measurements. It has high scale characteristics, including internal consistency, validity, specificity, sensitivity and interrater reliability. The DRS distinguishes delirious from demented, schizophrenic, and depressed patients and is more accurate than cognitive tests in identifying delirium. Scores are sensitive to treatment of delirium. Principal components analyses find one underlying dimension that can be subdivided into two or three components. The DRS has been used in studies of phenomenology, physiology, treatment, outcome, and at-risk populations. Tables summarize details from various studies. The DRS is used clinically and in research. It is currently being revised to enhance its use in phenomenologic and treatment research.
Key Words: Delirium • Diagnostic Tools • Delirium Rating Scale
|
ASSESSMENT OF DELIRIUM
|
Delirium is a neuropsychiatric syndrome that involves a number of symptoms, including diffuse cognitive impairment, psychosis, sleep-wake cycle disturbance, perceptual disturbances, thought disorder, language impairment, and mood lability. Symptoms of delirium typically have an acute onset and tend to fluctuate in intensity throughout a 24-hour period. Its incidence averages about 18% to 20% of patients admitted to general hospitals, though this figure varies somewhat depending on the patient population and methodology of the research.1 Despite its prevalence and associated morbidity and mortality,1 delirium is understudied compared with dementias and other psychiatric disorders. One of the methodological difficulties in studying delirium had been a dearth of appropriate symptom-rating scales and lack of diagnostic criteria. DSM-III, DSM-III-R, and DSM-IV diagnostic criteria have been published as various revisions by the American Psychiatric Association since 1980 and have been beneficial in that they list specific inclusion and exclusion criteria for diagnosing cases of delirium.2,3
Several instruments have been devised for health care professionals or laypersons to screen for the presence of symptoms of delirium, such as the Confusion Assessment Method (CAM)4 and the Delirium Symptom Interview (DSI),5 respectively. The CAM and DSI are reviewed in more detail elsewhere.6,7 The CAM has four items that represent cardinal symptoms, three of which are required to be present for a diagnosis of delirium. The CAM was originally reported to have sensitivity ranging from 94% to 100% and specificity ranging from 90% to 95%, compared with diagnoses made by using DSM-III-R criteria. However, a subsequent study8 reported CAM sensitivity to be 68% when rated by nurses, compared with diagnoses made by research physicians who used DSM-III-R criteria (and the Delirium Rating Scale) to assess medically ill elderly patients.
There are a few instruments that include a broad range of delirium symptoms to detect and rate the severity of those symptoms. Three scales—the Saskatoon Delirium Checklist,9 the Organic Brain Syndrome Scale,10 and the Delirium Assessment Scale11—are checklists that operationalize DSM-III criteria, with items rated on a mild-to-moderate severe continuum. In contrast, three other scales rate symptoms by using descriptive choices for each item. These are the Delirium Rating Scale,12 the Nurses Delirium Rating Scale,13 and the Memorial Delirium Assessment Scale (MDAS).14 The MDAS is intended for rapid serial ratings of delirium severity but does not include items for diagnosis. When compared with the DRS in 51 delirious patients with AIDS or cancer, the MDAS showed a high correlation to DRS total scores (r=0.88, P<0.0001), while its individual item correlations with DRS total scores ranged from 0.50 to 0.78.14
The Delirium Rating Scale (DRS) is the most widely used instrument for rating the severity of delirium with validity, high interrater reliability, and substantial sensitivity and specificity. It is used for clinical and research purposes not only in the United States, but also in Canada, Europe, South America, and Asia. The DRS has been translated into seven languages other than English. It was translated into French by Darius Razavi in Brussels, Mandarin Chinese by Chia-Yih Liu in Taipei, Dutch by Herman Sno in Amsterdam, Spanish by A. Bulbena in Barcelona, Italian by Augusto Caraceni in Milan, Swedish by Walter Osika in Karlskoga, and Japanese by Kunihiro Isse in Tokyo. The DRS also may become available in Portugese and a language of India (being translated by Mahesh Tilwani in Gujarat).
This article reviews the literature involving the DRS, including scale characteristics, analyses of scale items, and studies on treatment, phenomenology, physiology, and outcome of delirium.
|
DELIRIUM RATING SCALE
|
Scale Description
The DRS is a 10-item rating scale (see Table 1) intended to be completed by a clinician with psychiatric training to more sensitively detect the range of psychiatric symptoms assessed by the scale. Thus, it most often has been used by psychiatrists and research-trained psychiatric or geriatric clinicians. Each item has specific descriptors that can be scored from 0 to a maximum either of 2, 3, or 4 points, depending on the item. The sum of all item scores comprises the total DRS score; the maximum possible score is 32 points.
It is suggested that symptoms be rated over a 24-hour period because of the fluctuating nature of delirium symptom severity and to better detect the disruption of the sleep-wake cycle. With some adaptation, the DRS can be rated for partial days. However, because the inherent waxing and waning of symptom severity might falsely lead the rater to conclude clinical improvement or worsening, smaller time intervals for repeated ratings must be chosen carefully. In addition, certain items are difficult to rate during repeated administrations. This includes Item #1, "temporal onset of symptoms," because a judgment is required to determine at what point during an episode to no longer take into account the characteristics of the initial symptom onset. For example, when the episode no longer meets DSM criteria for delirium, the rater might decide to rate Item #1 as "0" at that time. Though this possibility poses a rating dilemma, Item #1 does capture an important phenomenologic characteristic of delirium, which contributes to the validity of the scale and helps to differentiate delirium from other psychiatric disorders, including dementia. The same issue applies to the rating of Item #7, which requires a decision about when the underlying medical cause of delirium is no longer relevant. (A revised form of the DRS that is currently under study will address this problem for repeated ratings.) In addition, some researchers have used partial scale ratings when the DRS is used serially. For repeated measures in a treatment study, Uchiyama et al.15 used a subscale of 7 items (excluding temporal onset, physical disorder, and fluctuation of symptoms) that was used after the initial ratings. Nakamura et al.16 used an 8-item scale (excluding temporal onset and physical disorder) for repeated measurements in a drug treatment study of delirium. Koolhoven et al.17 used a 7-item modified scale as a symptom checklist in an observational study by excluding two items (temporal onset and physical disorder) and compressing perceptual disturbances and hallucinations into one item.
The DRS was devised intentionally to reflect a broader range of symptoms of delirium than was assessable by using only cognitive tests such as the Mini-Mental State Exam (MMSE). Bedside cognitive tests alone cannot distinguish between different disorders that involve cognitive impairment, for example, delirium and dementia. Instead, assessment of a broad range of symptoms is necessary to distinguish delirium from other cognitive disorders. This includes assessing the severity and type of cognitive deficits, because different dementias affect different cognitive functions at different stages of the illness (e.g., frontal lobe dementias vs. Alzheimer's dementia). One type of dementia — Lewy body dementia — has significant overlap with delirium in its presentation, as does Alzheimer's dementia at its end-stage.
It was originally intended that the DRS be supplemented by whatever cognitive tests that the rater chooses. The DRS has only one cognitive status item (Item #6) and does not rate separate items for different components of cognition (e.g., attention, memory, orientation). Though this allows relatively more emphasis of noncognitive symptoms, it may make the DRS less useful for phenomenological studies or repeated measurements.
Scale Attributes
The DRS has been studied in different settings and found to have very good construct validity based on multiple lines of evidence, which are detailed in Table 2A and Table 2B and Table 3A and Table 3A and in later sections of this article. These include comparisons to expert diagnoses by using DSM criteria, differences in DRS scores compared with other patient groups, differences in DRS scores following standard delirium treatment when patients no longer meet DSM criteria for delirium, and correlations with ratings on other scales (e.g., Brief Psychiatric Rating Scale [BPRS], cognitive tests, and global ratings) that assess some delirium symptoms.
The DRS has excellent internal consistency, with a Cronbach's alpha score of 0.90 in a study of elderly patients.18 In addition, principal components analysis of DRS scores has shown that it has one major underlying dimension,19,20 supporting that it is measuring one overall construct. Moekken analysis performed by van der Mast21 in postcardiotomy patients revealed a coefficient of reliability Rho P= 0.81, HI item coefficients that were each >0.40, and the H coefficient of scalability=0.61, indicating a strong hierarchical nature of the items of the scale. These authors felt that the results of their Moekken analysis implicated that each DRS item was related to the severity of delirium and, therefore, that the DRS can function as a symptom severity scale.
Correlations of DRS items to other items have been done in two studies. In 20 general medically ill delirious patients, the cognitive item correlated highly with hallucinations (r=0.44, P=0.027); delusions (r=0.63, P=0.001); psychomotor behavior (r=0.48, P=0.017); sleep-wake cycle disturbance (r=0.56, P=0.0005); and lability of mood (r=0.60, P=0.003) in one study.12 The cognitive item correlated highly with psychomotor behavior (r=0.41, P=0.014), sleep-wake cycle disturbance (r=0.39, P=0.02) and variability of symptoms (r=0.38, P=0.023) in a study of 36 delirious postcardiotomy patients.21 In both studies, temporal onset of symptoms correlated significantly with perceptual disturbances (r=0.44, P=0.028 in the former and r=0.53, P=0.001 in the latter).
There are a few studies about DRS items. One study found individual DRS item scores to be significantly higher in delirious than in moderately demented patients for all items, except cognitive status, which was similar (P=0.022 for hallucinations and P<0.001 for other items).19 In a comparison between elderly patients with delirium and those with both delirium and dementia,20 only the cognitive status item differentiated the groups, with more impairment in the dual diagnosis group. Wada and Yamaguchi22 reported that the presence of certain DRS items predicted the duration of delirium episodes. Meagher et al.23 subgrouped DRS items into subscales for "core disturbances" representing either cognitive impairment or behavioral management problems in a study of environmental manipulation as a delirium treatment.
Because the DRS does not rely on only cognitive items, it is has been found to be superior to cognitive tests alone in distinguishing delirium from other disorders. The DRS, unlike the MMSE, distinguished delirium from dementia patients, and the DRS also distinguished delirium from schizophrenia patients, unlike the Trailmaking Tests.12 Receiver operating characteristic curve analyses comparing the DRS with the MMSE in a study of 104 elderly delirious patients found the DRS to be significantly more accurate than the MMSE in classifying delirium (area under the curve=0.87 for the DRS, area test=-1.93, P=0.03).18
Though the DRS measures more than just cognition, DRS scores have correlated with cognitive test scores in general hospital samples. The DRS correlated with the MMSE scores (r=-0.43, P=0.033) and with Trailmaking Test part B scores (r=0.66, P=0.007) in 20 delirious patients and with the MMSE scores (r=-0.78, P<0.01) in 104 elderly delirious patients.18 In contrast, in a study of 61 elderly delirious patients who were admitted to a psychiatric hospital instead of a general hospital, there was no correlation between the DRS and the MMSE (r=0.16).20 However, in that same study, the DRS was significantly correlated with the BPRS in delirious (r=0.57, P=0.017) and delirious-demented patients (r=0.35, P= 0.04). This finding suggests possible differences in the presentation of patients whose identified medical morbidity is lower and who are admitted to a psychiatric hospital instead of a general hospital.
Two studies have compared delirious, demented, schizophrenic, and noncognitively impaired medically ill patients and found little or no overlap in total DRS scores among these groups when graphing scores on boxplots.18,20 Similarly, a study of postcardiotomy patients found very little overlap in DRS scores graphed on boxplots between the delirious and nondelirious patients;21 overlap was attributed to a methodological issue in the timing of the ratings. By using ROC analysis, Rockwood18 found the DRS to be more accurate than the MMSE in classifying delirium among groups of elderly patients with either delirium, dementia, or other psychiatric disorders.
Rosen et al.24 studied 791 consecutive geriatric psychiatry inpatients who were divided into three study groups: delirium, dementia/organic mental disorder not otherwise specified, and noncognitively impaired patients. Rosen et al. reported that the DRS had a high predictive negative value of 99%, but its predictive positive value was only 33%, largely related to its difficulty differentiating the other organic mental group from the delirium group due to overlapping symptoms of psychosis and cognitive impairment. There were only 4 out of 70 false-negatives among the delirium group, but 94 out of 291 other organic mental disorder patients were rated as false-positives when using a DRS cutoff of  10 points. However, these false-positives in the other organic group were more likely to have significant electroencephalogram (EEG) abnormalities, with Grades II and III dysrhythmias, higher BPRS psychosis scores, and lower MMSE scores than the true-negatives in that group. Abnormal EEGs of that severity are usually common in delirium and uncommon in dementia until very late stages, which opens the possibility of misassignment of patients who may have really been delirium cases and were not truly "false-positive" cases. Their DRS ratings were in the high 8-point range. In addition, differential diagnosis of organic mental disorders is especially difficult in the elderly. This study used nonphysician DRS raters whose interrater reliabilities were lower than in other studies that used physicians (for example, in Rockwood et al.,18 1996).
The DRS generally has high values for sensitivity and specificity and high interrater reliability (see Table 2A and Table 2B). Interrater reliabilities vary according to who does the ratings. Highest ratings are for psychiatric and geriatric medical physicians and range from 0.86 to 0.97. For nonphysicians who are trained to use the DRS for research purposes, interrater reliability ranges from 0.59 to 0.99. Sensitivity and specificity depends on the cutoff value used for the DRS. In usage of a cutoff score of 10 points, sensitivity has been reported at 0.94 and 0.82, and specificity at 0.94 and 0.82. In usage of a cutoff score of 7.5 points, sensitivity has been reported at 0.90 and specificity at 0.82.
Mean DRS scores for delirious patients are considerably higher than for demented or other diagnostic groups of patients in many studies (see Table 3A and Table 3B) across a broad range of ages of adults who have a wide variety of delirium etiologies, medical and surgical. Mean scores decrease with treatment of delirium.15,16,25–27 Mean DRS scores also do not change when treatment is ineffective.15,26 DRS scores can distinguish different motoric presentations of delirium, that is, hyperactive and hypoactive subtypes, with mixed presentations having intermediate scores between them.23 Total DRS scores differentiated patients whose delirium episodes lasted more or less than 1 week,22 suggesting the score has predictive value as an outcome measure.
Factor Analysis of DRS Items
There are three studies of DRS items using principal components analyses to better understand the relationship between items of the scale and whether they reflect one or more groupings.19–21 In two studies, there seems to be a single underlying dimension consistent with the DRS measuring a single construct, that is, the syndrome of delirium. Trzepacz and Dew19 found one strong underlying dimension that could be further subdivided into two components after Varimax rotation, in a study of 20 delirious adults seen in a general hospital. One component was composed of delusions, psychomotor behavior, cognition, sleep-wake cycle disturbance, and mood lability; the other composed of temporal onset of symptoms, perceptual disturbances, hallucinations, and fluctuation of symptoms. Van der Mast21 studied 40 postcardiotomy delirious patients and found three underlying factors after Varimax rotation: 1) temporal onset of symptoms, perceptual disturbances, delusions, and psychomotor behavior; 2) psychomotor behavior, cognitive dysfunction, sleep-wake cycle disturbance, and fluctuation of symptoms; and 3) physical disorder and absence of lability of mood. Van der Mast, however, felt the third was an artifact because the most severely physically ill patients were not ratable on the mood lability item. Trzepacz et al.20 studied a psychogeriatric population, comparing delirious with delirious-demented patients, and identified two core factors on the DRS after Varimax rotation. A total of seven DRS items were common to both groups, while three items did not cluster similarly (hallucinations, cognition, and fluctuation of symptoms). In addition, these delirious-only patients' factor structure was remarkably similar to that from a general hospital delirium sample from a general hospital sample.19 In both of these delirium samples, Factor 1 loaded sleep-wake cycle disturbance, psychomotor behavior, and mood lability, while Factor 2 loaded temporal onset, symptom variability, and perceptual disturbances.20 The three items that loaded onto Factor 2 have been considered by several experts to be highly associated with delirium, suggesting that the DRS analyses were detecting information applicable to delirium phenomenology.
|
STUDIES USING THE DRS
|
In addition to reports about scale characteristics or DRS item analyses, the DRS has been used in research studies and as an assessment instrument in at-risk populations. These include studies about phenomenology, physiology, outcome, and treatment.
Phenomenology Studies
Platt et al.27 studied motoric subtypes of 24 hospitalized delirious AIDS patients by using DSM-III-R criteria and repeated DRS measurements. All had DRS scores >12 points prior to treatment with either haloperidol or chlorpromazine. Nine were considered hyperactive and 11 hypoactive in presentation based on clinical judgment and a score of at least >1 point on the DRS psychomotor item. Both the hyperactive and hypoactive groups responded within hours of treatment with either antipsychotic drug (even before the underlying medical problems were corrected), as evidenced by significant declines in their DRS scores (P<0.001 for each subtype). Platt et al. suggested that both motoric subtypes respond equally well to dopamine blockade.
Rockwood28 studied the occurrence and duration of symptoms in 173 elderly medical inpatients in Camp Hill Hospital, Nova Scotia. Many had concurrent dementia. DSM-III and DSM-III-R criteria and DRS ratings were applied on admission. Delirious patients' mean±standard deviation (SD) age was 82±8 years and mean±SD DRS score was 17±4 points; nondelirious patients' mean±SD age was 79±8. Rockwood28 found that DSM-III-R had better sensitivity and DSM-III had better specificity in delirium diagnosis, compared with a clinical gold standard. In addition, some delirium symptoms were found to persist at discharge after the episode had ended.
Uchiyama et al.15 used the DRS to compare drug response between hyperactive and hypoactive presentations of delirium in their study of 62 delirious elderly. The researchers found that 80.6% were hyperactive and 19.4% were hypoactive on the basis of clinical observation. The hyperactive patients responded better to mianserin than did the hypoactive patients (P<0.05). Medication response was defined as improvement on item scores for 4 DRS items with worsening on  2 items. DRS scores for each motoric subtype were not reported.
Meagher et al.23 studied hypoactive, mixed, and hyperactive subtypes in 46 delirious inpatients. By using International Classification of Diseases (ICD)-10 criteria and Liptzin and Levkoff's definition of motoric subtypes, the researchers found significantly higher (P<0.01) DRS scores in hyperactive and lower (P<0.01) scores in hypoactive than in mixed subtypes (see Table 3A and Table 3B). In addition, the hyperactive group required significantly more nursing management interventions (P<0.05) and hypoactive less (P<0.01) than the mixed group. The hyperactive group (93%) also received more psychotropic drugs prior to consultation.
Physiology Studies
In a drug treatment study of delirious patients, Nakamura et al.16 measured plasma-free 4-methoxy-3-hydroxy-phenylethylene glycol (MHPG) and homovanillic acid levels. The researchers found a significant reduction in plasma free-MHPG over time in all delirious patients, regardless of drug treatment using mianserin, oxypertine, or haloperidol. This reduction paralleled a decrease in DRS scores, indicating that reducation in delirium severity was associated with decreased plasma-free MHPG.
Tanaka et al.29 used the DRS in a study of the role of cerebrospinal fluid (CSF) prostaglandin D2 and E2 levels in the sleep-wake mechanism of delirium. Seven inpatients at Tokyo Metropolitan Tama General Hospital (mean age: 78.4) who had delirium were compared with 7 age-matched controls (mean age: 74.9) without neurologic or psychiatric problems. DSM-III-R criteria were used for diagnosis, and the DRS was used serially. Mean DRS score=24 points. The researchers found that CSF prostaglandin D2 was significantly higher in the delirious patients than in the control subjects.
At-Risk Population Studies
The DRS has been used to assess the presence of delirium symptoms in at-risk medical populations. DiMartini and colleagues30 used the DRS in studies of liver transplant patients at the University of Pittsburgh Medical Center. In a comparison of two immunosuppressant drugs, cyclosporine A and FK506 (tacrolimus), DRS scores were in the nondelirious range (see Table 3A and Table 3B), despite the presence of neuropsychiatric symptoms known to be associated with toxicity of these drugs.30 Some patients scored below 24 points on the MMSE and above 80 seconds on Trailmaking Test part B. This finding suggests some specificity for the DRS regarding the symptoms it detects. In a randomized comparison study of FK506 and cyclosporine A, DiMartini et al.31 described 31 orthotopic liver transplant patients cross-sectionally 1 week postop (mean±SD age: 40±9). DRS scores were in the nondelirious range in both drug groups, with only 2 patients scoring above 12 points (19 and 22 points), both in the FK506 group. MMSE scores were also in the unimpaired range, though some patients had impairment on the Trailmaking tests.
Goldstein and Fogel32 used the DRS to assess 82 elderly patients (mean±SD age: 67±7) undergoing elective surgery at State University of New York, Buffalo, School of Medicine, as part of a longitudinal study. Patients whose MMSE score was <23 points preoperatively were excluded from the study. Mean±SD DRS score at Day 3 postop was 2.17±2.73 (range: 0–16) and only one patient's DRS score was in the delirium range. The researchers found that a combination of preop MMSE scores and an MMSE change score (between preop and postop) accounted for most of the variance in MMSE scores that were measured at a 10-month follow-up. This finding suggests that the decline in cognitive function following surgery in these nondemented elderly was not related to delirium, but rather to some other vulnerability for the later decline.
Outcome Studies
Albronda et al.25 studied 43 elderly patients (mean age: 78.5) hospitalized on a geriatric unit at Ziekenhuis Rijnstate in Arnhem, Netherlands, who had a recent change in mental status. By using DSM-III-R criteria for diagnosis, 28 had delirium, 23 dementia, 6 depression, and 21 multiple problems. The mean DRS was 14.5 points within 3 days of admission and 6.6 at discharge (P<0.001 for this difference in scores). Despite improvement in most patients' conditions they found that those admitted for recent mental status changes were at increased risk for nursing home placement after discharge — 53%, compared with 13%, for geriatric admissions in general. The researchers did not report any predictive data for the DRS, however.
Wada and Yamaguchi22 studied 28 elderly patients who met DSM-III-R criteria for delirium (mean±SD age: 67.7±9.1, range: 50–83 years) who were referred to the Neuropsychiatric Unit of Kanazawa University Hospital. The DRS was administered the first day of admission. The patients were followed throughout their hospitalization and divided into groups according to a clinical assessment of duration of the delirium episode. In 16 patients, delirium lasted more than 1 week. There was no difference in age or gender for the patients whose delirium lasted 1 week or >1 week, but DRS scores differentiated these groups (14.0 vs. 18.27, respectively; P<0.001). In addition, the group with shorter episodes had significantly lower scores for three DRS items: cognitive status (P<0.05), sleep-wake cycle disturbance (P<0.005), and mood lability (P<0.005). No comment was made about treatment for delirium, though "usual care" was implied. These authors suggested that the DRS may predict one dimension of delirium outcome (i.e., episode duration).
Rockwood et al.8 studied the outcome of educational interventions to recognition of delirium by housestaff at Victoria General Hospital in Dalhousie, Nova Scotia. Medical records were reviewed to compare 187 elderly control patients consecutively admitted before the educational intervention with 247 patients admitted thereafter. Housestaff were educated about delirium and in how to use the CAM. Recognition of delirium or acute confusion, as documented in the record, was the outcome measure. An independent assessment was done by research nurses who used the CAM and by research physicians who completed both the CAM and the DRS. The CAM had a low sensitivity (0.68), compared with research physicians' diagnosis of delirium using DSM-III-R. The educational intervention had a significant impact on housestaff recognition of delirium. However, when compared with diagnosis by research physicians, housestaff recognition had a sensitivity of 0.64 and specificity of 0.93. DRS scores were not reported.
Rudberg et al.33 studied 432 elderly consecutively admitted to medical and surgical wards of the University of Chicago Hospitals to assess characteristics of presentation, course, and duration of delirium. After using the CAM and attention tests to screen cases daily to identify possible delirium, research clinicians independently used DSM-III-R criteria for definitive diagnosis. Those identified as being delirious (about 15%) were rated daily by using the DRS to describe attributes of the delirium. The researchers found that 69% had delirium only for 1 day and that DRS scores were significantly higher for the first day (mean±SD: 25.4±3.6) if the delirium lasted multiple days than if it lasted only 1 day (mean±SD: 22.6±4.4), suggesting that more severe delirium episodes last longer. The researchers found no patterns of change among mean scores for individual DRS items over the multiple days, instead finding much interindividual variation for symptoms that was attributed to physiological differences from multiple medical diagnoses (mean±SD: 6.3±2.3 per patient) and multiple medications (mean±SD: 6.9±2.6 per patient). Effects on DRS items related to whether and how delirium was treated were not discussed.
Treatment Studies
Nakamura et al.16 studied 53 delirious patients referred for psychiatric consultation at Kurume University Hospital. Alcohol withdrawal patients were excluded. Their mean±SD age was 69.5±12.4 years. The patients were rated serially using a modified DRS (without items #1 and 7) and treated with either mianserin (n=46), oxypertine (a phenylpiperazine antipsychotic) (n=17), or haloperidol (n=13). Mean±SD DRS scores (see Table 3A and Table 3B) decreased in each drug group over time, though no statistical comparisons were made between treatment groups nor were survival analyses performed. It was not mentioned whether patients were randomized to drug treatments or if the DRS was rated blind to treatment status. Each evening throughout the study period, a single dose of drug was given. DRS scores correlated with plasma mianserin levels (r=0.67, P<0.01) on Day 3 of treatment. The calculated difference between baseline DRS scores and Day 3 scores was inversely correlated with mianserin plasma levels (r=-0.67, P<0.01), suggesting that mianserin was related to delirium reduction.
Uchiyama et al.15 did an open trial of mianserin (10–60 mg) treatment of delirium in 62 elderly patients hospitalized on a psychogeriatric unit of a general hospital, whose mean age was 80.7 years. DSM-IV categories for delirium etiology were used. Twenty-five of these patients (40.3%) also had a preexisting dementia, and 8 were continued on benzodiazepine hypnotics or morphine. Of the 31 patients who had not responded to psychotropic treatment of their delirium episode, only 25 had received an antipsychotic (drug or dose was not reported). A 7-item DRS was used for repeated measurements, with clinical improvement determined by improvement on at least four items and worsening on two or fewer items. Baseline±SD DRS scores did not predict response to mianserin (21.9±3.2 and 22.2±3.3). Rate of improvement on DRS items included psychomotor behavior (85.5%), hallucinations (80.6%), sleep-wake cycle disturbance (82.3%), lability of mood (81.4%), delusions (79.6%), perceptual disturbances (67.7%), and cognitive status (48.4%). It was hypothesized that mianserin's H1 and 5-HT2 blocker activity might differentially affect certain delirium symptoms.
Breitbart et al.26 did a double-blind, randomized comparison of haloperidol, chlorpromazine, and lorazepam for delirium in hospitalized AIDS patients at Memorial-Sloan Kettering Cancer Center in New York City. DSM-III-R criteria and a score of 13 on the DRS were used to diagnose delirium in a prospectively assessed cohort of 244 patients, 30 of whom (12%) became delirious. MMSE scores were used to guide the rating of the DRS cognitive status item. The patients were rated hourly on the DRS during drug treatment and, when scores exceeded 13 points, the next drug dosage was administered. The randomization to lorazepam was stopped halfway through the study because all those patients developed treatment-limiting adverse effects (e.g., increased confusion, sedation). They also showed no clinical improvement over the study period. DRS scores (see Table 3A and Table 3B) significantly improved in each of the haloperidol and chlorpromazine groups between baseline and Day 2, producing DRS scores below the delirium cutoff value within the first 24 hours of treatment. The mean dose of drugs given during the first 24 hours was 2.8 mg for haloperidol, 50 mg for chlorpromazine, and 3.0 mg for lorazepam.
Masand et al.34 retrospectively completed the DRS based on chart reviews in a nonrandomized naturalistic comparison of delirious psychiatric patients treated with either haloperidol (n=11, mean±SD age: 63.5±18.3) or olanzapine (n=11, mean±SD age: 63.5±23.2). Mean±SD pretreatment DRS scores were 20.1±5 for haloperidol and 17.9±4.4 for olanzapine. Mean±SD posttreatment DRS scores were 11.1±7.1 for haloperidol and 10.3±4.8 for olanzapine. Clinical improvement was similar in each group, though there were methodological issues, including nonrandomization, simultaneous use of other neuroleptics in the olanzapine group, and delirium being comorbid with other Axis I diagnoses that share some of its symptoms.
Meagher et al.23 examined the pattern and frequency of use of environmental interventions and psychotropic drug use in delirium management. Consecutive referrals to a consultation-liaison (C-L) psychiatry service in a large general hospital in Dublin were prospectively assessed for delirium by using ICD-10 criteria for diagnosis and the DRS for symptom severity. Mean±SD age of the sample was 60.1±19.5 years, and mean±SD DRS was 20.17±4.98. The patients received a mean±SD of 4.04±1.87 nursing interventions. When separated into high and low intervention groups (using cutoff of four interventions), the high group was found to have greater delirium severity (P<0.01), though the DRS score was not reported. The DRS was also divided into two core subscales: "cognitive," which included cognitive status, delusions, and hallucinations, and "behavioral management," which included psychomotor behavior, sleep-wake cycle disturbance, and mood lability. Nursing intervention scores were significantly associated (P<0.01) with the DRS "behavioral management core" scores but not with "cognitive core" scores.
|
REVISION OF THE DRS
|
The DRS is very useful clinically, and its routine use can enhance the detection of delirium symptoms as well as assist trainees to appreciate the breadth of the delirium syndrome. Use of the DRS for research purposes has highlighted some of its shortcomings. To be more useful as a research tool, the DRS needs to be revised by separating items more relevant for diagnosis from items more related to symptom severity, which will allow its use during repeated serial ratings in intervention studies. To be used to advance phenomenologic research of delirium, the DRS needs to be more comprehensive in its ability to detect abnormalities of cognitive, behavioral, thinking, and language functions that have been described in the clinical literature.
Thus, the revised DRS will include separate items for each cognitive function (disorientation, attention, memory, etc.), irrespective of whether formal, standardized cognitive tests are administered as an adjunct. The revised DRS will separate psychomotor items into one for hyperactive and one for hypoactive states. It will add items that each rate the severity of thought process and language abnormalities, to address what constitutes the problems that are traditionally represented by the terms "confusion" or "clouding of consciousness."
The revised DRS is now being validated against the original DRS and will become available for translation and international crossvalidation during 1999.
|
CONCLUSIONS
|
C-L research requires standardized rating instruments to measure symptoms and improve methodological quality.35 Delirium is still an understudied disorder and is of major clinical importance to C-L psychiatrists. The DRS is the most widely used and best studied of instruments designed specifically for use in delirium. It has been shown to have excellent scale characteristics and reliability, to serve as a symptom severity scale, and to be responsive to change in delirium status during treatment. Recommended cutoff score for the DRS is about 12 points, though choosing a cutoff score also depends on the study design, in which sensitivity or specificity may be affected by context or comparison groups.
Though the revised DRS will be better suited for phenomenological and treatment studies of delirium, the original DRS will still be useful and valid for clinical and research purposes in C-L psychiatry.
|
REFERENCES
|
-
Trzepacz PT: Delirium: Advances in diagnosis, pathophysiology, and treatment. Psychiatr Clin North Am 1996;19:429–448
-
American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric Association, 1987
-
American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association, 1994
-
Inouye SK, van Dyke CH, Alessi C, et al: Clarifying confusion: the confusion assessment method. Ann Intern Med 1990; 113:941–948
-
Albert MS, Levkoff SE, Reilly C, et al: The Delirium Symptom Interview: an interview for the detection of delirium symptoms in hospitalized patients. J Geriatr Psychiatry Neurol 1992; 5:14–21
-
Musselman DL, Hawthorne CN, Stoudemire A: Screening for delirium: a means to improved outcome in hospitalized elderly patients. Reviews in Clinical Gerontology 1997; 7:235–56
-
Trzepacz PT: A review of delirium assessment instruments. Gen Hosp Psychiatry 1994; 16:397–405[Medline]
-
Rockwood K, Cosway S, Stolee P, Kydd D, et al: Increasing the recognition of delirium in elderly patients. J Am Geriatr Soc 1994; 42:252–256[Medline]
-
Miller PS, Richardson JS, Jyu CA, et al: Association of low serum anticholinergic levels and cognitive impairment in elderly pre-surgical patients. Am J Psychiatry 1988; 145:342–5[Abstract/Free Full Text]
-
Berrgren D, Gustafson Y, Eriksson B, et al: Postoperative confusion following anesthesia in elderly patients treated for femoral neck fractures. Anesth Analg 1987; 1394:497–504
-
O'Keeffe ST: Rating the severity of delirium: the Delirium Assessment Scale. Int J Geriatr Psychiatry 1994; 9:551–956
-
Trzepacz PT, Baker RW, Greenhouse J: A symptom rating scale for delirium. Psychiatr Res 1988; 23:89–97
-
Rutherford L, Sessler CN, Levenson JL, et al: Prospective evaluation of delirium and agitation in a medical intensive care unit. Crit Care Med 1991; 19(suppl):S81
-
Breitbart W, Rosenfeld B, Roth A, et al: The Memorial Delirium Assessment Scale. J Pain Symptom Manage 1997; 13:128–137[Medline]
-
Uchiyama M, Tanaka K, Isse K, et al: Efficacy of mianserin on symptoms of delirium in the aged: an open trial study. Progress in Neuro-Psychopharmacology & Biological Psychiatry 1996; 20:651–656[Medline]
-
Nakamura J, Uchimura N, Yamada S, et al: Effects of mianserin hydrochloride on delirium: comparison with effects of oxypertine and haloperidol. Japanese J Psychopharmacology 14:269–277;1994
-
Koolhoven I, Tjon-A-Tsien MRS, van der Mast PC: Early diagnosis of delirium after cardiac surgery. Gen Hosp Psychiatry 1996; 18:448–451[Medline]
-
Rockwood K, Goodman J, Flynn M, et al: Cross-validation of the Delirium Rating Scale in older patients. J Am Geriatr Soc 1996; 44:839–842[Medline]
-
Trzepacz PT, Dew MA: Further analyses of the Delirium Rating Scale. Gen Hosp Psychiatry 1995; 17:75–79[Medline]
-
Trzepacz PT, Mulsant BH, Dew MA, et al: Is delirium different when it occurs in dementia: A study using the Delirium Rating Scale. J Neurospsychiatry Clin Neurosci 1998; 10:199–204
-
van der Mast RC: Detecting and measuring the severity of delirium with the symptom rating scale for delirium, in Delirium After Cardiac Surgery. Amsterdam, The Netherlands, Benecke Consultants, 1994, pp 78–89
-
Wada Y, Yamaguchi N: Delirium in the elderly: relationship of clinical symptoms to outcome. Dementia 1993; 4:113–116
-
Meagher D, O'Hanlon D, O'Mahony E, et al: The use of environmental strategies and psychotropic medication in the management of delirium. Br J Psychiatry 1996; 168:512–515[Abstract/Free Full Text]
-
Rosen J, Sweet RA, Mulsant BH, et al: The DRS in a psychogeriatric inpatient setting. J Neuropsychiatry Clin Neurosci 1994; 6:30–35[Abstract/Free Full Text]
-
Albronda T, Bruijns E, Willekens FL: Diagnosis and discharge destination of patients admitted for mental dysfunctioning in a geriatric department. Nederlands Tijdschrift voor Geneeskunde 1996; 140:1993–1996
-
Breitbart W, Marotta R, Platt MM, et al: A double-blind trial of haloperidol, chlorpromazine, and lorazepam in the treatment of delirium in hospitalized AIDS patients. Am J Psychiatry 1996; 153:231–237[Abstract/Free Full Text]
-
Platt MM, Breitbart W, Smith M, et al: Efficacy of neuroleptics for hypoactive delirium. J Neuropsychiatry Clin Neurosci 1994; 6:66[Free Full Text]
-
Rockwood K: The occurrence and duration of symptoms in elderly patients with delirium. J Gerontol Med Sci 1993; 48:162–166
-
Tanaka K, Mitsushio H, Isse K, et al: A clinical study of prostaglandin D2 in cerebrospinal fluid of elderly patients with delirium. Ann Rep Pharmacopsychiat Res Found 1993; 24:14–21
-
DiMartini AF, Pajer K, Trzepacz PT, et al: Psychiatric morbidity in liver transplant patients. Transplant Proc 1991; 23:3179–3180
-
DiMartini AF, Trzepacz PT, Pajer KA, et al: Neuropsychiatric side effects of FK506 vs. cyclosporine A: first week postoperative findings. Psychosomatics 1997; 38:565–569[Abstract/Free Full Text]
-
Goldstein MZ, Fogel BS: Cognitive change after elective surgery in nondemented older adults. Am J Geriatric Psychiatry 1993; 1:118–125[Abstract/Free Full Text]
-
Rudberg MA, Pompei P, Foreman MD, et al: The natural history of delirium in older hospitalized patients: a syndrome of heterogeneity. Age Aging 1997; 26:169–174[Abstract/Free Full Text]
-
Masand P, Sipahimalani A: Olanzapine in the treatment of delirium. Psychosomatics 1998; 39:422–429[Abstract/Free Full Text]
-
Trzepacz PT: Neuropathophysiology of delirium: a need to focus our research. Psychosomatics 1994; 35:374–391[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
C.-K. Fang, H.-W. Chen, S.-I. Liu, C.-J. Lin, L.-Y. Tsai, and Y.-L. Lai
Prevalence, Detection and Treatment of Delirium in Terminal Cancer Inpatients: A Prospective Survey
Jpn. J. Clin. Oncol.,
January 31, 2008;
(2008)
hym155v1.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. R. Fann, C. M. Alfano, S. Roth-Roemer, W. J. Katon, and K. L. Syrjala
Impact of Delirium on Cognition, Distress, and Health-Related Quality of Life After Hematopoietic Stem-Cell Transplantation
J. Clin. Oncol.,
April 1, 2007;
25(10):
1223 - 1231.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Bosisio, A. Caraceni, L. Grassi, and the Italian Delirium Study Group (IDSG)
Phenomenology of Delirium in Cancer Patients, as Described by The Memorial Delirium Assessment Scale (MDAS) and The Delirium Rating Scale (DRS)
Psychosomatics,
December 1, 2006;
47(6):
471 - 478.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. Kelly, S. McClement, and H. M. Chochinov
Measurement of psychological distress in palliative care
Palliative Medicine,
December 1, 2006;
20(8):
779 - 789.
[Abstract]
[PDF]
|
|
|
|
|
|
|
E. Kozora, C. F. Emery, M. C. Ellison, F. S. Wamboldt, P. T. Diaz, and B. Make
Improved Neurobehavioral Functioning in Emphysema Patients Following Lung Volume Reduction Surgery Compared With Medical Therapy
Chest,
October 1, 2005;
128(4):
2653 - 2663.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. B. Turkel, K. Braslow, C. J. Tavare, and P. T. Trzepacz
The Delirium Rating Scale in Children and Adolescents
Psychosomatics,
April 1, 2003;
44(2):
126 - 129.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. T. Trzepacz, D. Mittal, R. Torres, K. Kanary, J. Norton, and N. Jimerson
Validation of the Delirium Rating Scale-Revised-98: Comparison With the Delirium Rating Scale and the Cognitive Test for Delirium
J Neuropsychiatry Clin Neurosci,
May 1, 2001;
13(2):
229 - 242.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. J. Meagher, D. O'Hanlon, E. O'Mahony, P. R. Casey, and P. T. Trzepacz
Relationship Between Symptoms and Motoric Subtype of Delirium
J Neuropsychiatry Clin Neurosci,
February 1, 2000;
12(1):
51 - 56.
[Abstract]
[Full Text]
|
|
|
|
TOP
ABSTRACT
ASSESSMENT OF DELIRIUM
