Experience in Delirium: Is It Distressing?

Delirium is an etiologically heterogenous syndrome, commonly seen among medically compromised patients. It is characterized by rapid onset, diurnal fluctuating course, and concurrent disturbance of consciousness, perception, thinking, memory, psychomotor behavior, and the sleep–wake cycle. Although it is considered to be short lasting, it has been shown to be associated with a high mortality rate, functional decline, prolonged hospital stay, increased rates of admission to long-term care, and increased costs of care and development of long-term cognitive decline.^1–6 Besides these consequences, although less evaluated, few studies suggest that experience of delirium is also associated with many psychological consequences in those who recover from it.^5,7–11

The psychological consequences in patients who recover from delirium include the inability to remember their experience of delirium and those who are able to recollect it describe it to be an unpleasant experience.⁷

Most of the available literature on the psychological consequences of delirium is in the form of qualitative studies that have evaluated patients in the intensive care/critical care setting. These studies suggest that a significant proportion of patients with delirium or confusional states recall their experience during the episode of delirium and that these experiences tend to be distressing and disturbing.^12–18

However, one of the major limitations of the research in this area is that these studies have generally not addressed the distress related to delirium in a quantitative manner. In view of the lack of data, Breitbart et al.¹⁹ designed the Delirium Experience Questionnaire, which can be used to assess the distress associated with delirium, both qualitatively and quantitatively. In their study, 53.5% of patients were able to recall their delirium experience, and in logistic regression analysis, short-term memory impairment and the presence of perceptual disturbances were significant predictors of delirium recall. Among the various symptoms, the presence of delusions was the most significant predictor of patient distress, and there was no difference in the distress level among those with hypoactive or hyperactive delirium.¹⁹

In one of the previous studies from our center, the authors evaluated 53 patients with delirium after recovery from the episode and reported that only 28.3% (N=15) remembered themselves as confused. Those who recollected their experience commonly described their experience as a state of fearfulness, anxiety, and a feeling of confusion and strange. Occasional patients described the presence of visual hallucination and illusions during the episode of delirium and reported distress due to the same. However, this study was limited by a small sample size and lack of estimation of distress associated with specific symptoms of delirium.⁵

In view of the limited literature in terms of the number of studies and the sample size of the available studies varying from 5 to 101, there is a need to expand the literature on this important topic.^{5,12,13,19–22} Keeping this in mind, the present study aimed to assess the patient’s experience of delirium and distress associated with the delirium experience. In contrast to the previous study from our center, this study included a relatively larger sample size and also assessed the level of distress associated with few specific symptoms of delirium.

Methodology

Setting

The study was carried out at a multispecialty teaching hospital in north India. The study sample was drawn from the patients referred to psychiatry consultation liaison services of the hospital and diagnosed to have delirium. The study was approved by the Ethics Review Board of the institute, and written proxy informed consent was obtained from the primary caregivers of the patients staying with the patient during the hospital stay.

All patients ≥18 years of age and diagnosed with delirium by the psychiatry consultation liaison team were eligible for this study. Caregivers of such patients were approached and provided an explanation of the study. Only those patients were recruited whose caregivers provided proxy informed consent. All patients with clinical diagnosis of delirium were evaluated on the DSM-IV-TR criteria to confirm the diagnosis of delirium. Those patients with a confirmed diagnosis of delirium were further evaluated on the Delirium Rating Scale-Revised 98 Version (DRS-R-98) for the phenomenology.²³ Patients were followed up daily by the consultation liaison psychiatry team to monitor the presence of delirium. After at least 24 hours of recovery from the episode of delirium, which was defined as having a DRS-R-98 severity score of ≤10, with a score of 0 on the diagnostic item of fluctuation of symptoms of the DRS-R-98, the patients were evaluated for their experience of delirium.

No interference in the management of patients was done as part of this study, and the consultation liaison psychiatry team managed these cases as per routine clinical practice.

Instruments

DRS-R-98.

The severity and frequency of various symptoms of delirium was assessed by the DRS-R-98 scale. It consists of 16 items, which are divided into two subgroups: those assessing the severity of delirium and those of diagnostic significance. The 13 severity items are rated on four points from 0 to 3, with higher scores suggesting a higher severity of symptoms. Those items that are rated as 1 or more are considered to be present and are used for studying the frequency of symptoms. The scale has been shown to have good validity, sensitivity, and interrater reliability (intraclass correlation coefficient=0.98).²³

Amended delirium motor symptom scale.

The original Delirium Motor Symptom Scale, which is an 11-item scale, was developed by Meagher et al.²⁴ In a later analysis, two more items were added to make it more comprehensive and was named the amended Delirium Motor Symptom Scale. It is a 13-item checklist designed to classify the subjects with delirium into various subtypes based on motor activity. Each item is rated on the basis of the presence of definite evidence of the same in the previous 24 hours, which is a deviation from the predelirious baseline. To be classified as a hyperactive subtype, the subject has to have at least two of five items from hyperactive items, and for hypoactive subtype, at least two of the eight items from hypoactive items must be present. A subject is considered to have a mixed subtype when there is evidence of both the above subtypes, and a subtyping of no motor subtype is made when evidence of neither hypoactive nor hyperactive is present.^24,25

Short informant questionnaire on cognitive decline in the elderly.

This was used to screen patients for the presence of dementia. It consists of 16 items that evaluate the cognitive status of the patient during the 6 months prior to their current assessment based on the input of a key relative. Each item is scored on a five-point scale, with a score of 3 as no change (higher scores, worsening cognition; lower scores, improved cognition). The scores obtained on all items are added and divided by 16 to get a mean score per item. The suggested cutoff for suspected dementia is a mean score of 3.31–3.38. It is a well-validated instrument.²⁶

Delirium experience questionnaire.

The Delirium Experience Questionnaire was designed by Breitbart et al.¹⁹ to assess the experience of delirium in patients who have recovered from delirium and includes the following six questions. 1) Do you remember being confused? (responses: yes or no). 2) If the answer to question 1 is no, then the patient is asked the following: are you distressed that you can't remember?’ (responses: yes or no). 3) Question 3 inquires about the level of distress: how much distressed are you that you cannot remember your experience (level of distress is rated on a five-point scale, 0–4, with 0 equal to not at all and 4 equal to extremely). 4) Question 4 inquires about the following: of you do remember being confused, was the experience distressing? (yes or no). 5) Question 5 inquires about associated level of distress on a five-point scale. 6) Question 6 allows for a qualitative assessment of the delirium experience through the verbatim transcription of patients' description of the experience to the open-ended question in which the patient is asked to describe her/his expereince. The Delirium Experience Questionnaire has been reported to have adequate face validity for the parameters it assesses.¹⁹

Besides the Delirium Experience Questionnaire, a questionnaire was devised specifically to assess the experience of the patient and level of distress associated with specific symptoms of delirium in those who remembered their symptoms. This was done by substituting the open-ended qualitative assessment of delirium experience as assessed on item 6 of the Delirium Experience Questionnaire. The items for this questionnaire were based on the commonly reported symptoms in patients and included 15 symptoms: delusions, visual hallucinations, auditory hallucinations, tactile hallucinations, pulling out the tubing’s, abusing people, talking about things of past, trying to run away, uncooperativeness for treatment, drowsiness/confusion, night time insomnia, disorientation, attention deficit/distractibility, short-term memory deficits, and long-term memory deficits. Using a semistructured interview, the responses for these items were rated as “yes(present)/no (absent)” and the level of distress was rated on a five-point scale, varying from “not at all (rated as 0)” to “very severe (rated as 4).” A total distress score was calculated based on the severity of distress.

Analysis

Data were analyzed using SPSS, 14th version (SPSS, Chicago). Continuous variables were studied in the form of mean and standard deviation. The ordinal and nominal variables were studied in terms of frequency and percentages. Comparisons were done by using t test, analysis of variance, the chi-square test, and Fischer exact tests.

Results

Demographic and Clinical Profile

For this study, 322 consecutive patients ≥18 years of age and diagnosed as having delirium as per DSM-IV criteria were assessed at the baseline. Of the 322 patients, 40 patients died during the inpatient stay while being delirious and hence could not be assessed for distress. Seventy-nine patients could not be assessed for distress because they were discharged/left against medical advice in a delirious state (75 patients) or could not be traced (four patients). The final sample that completed both the assessments was comprised of 203 participants.

The mean age of the final study sample was 47.3 (SD=18.9) years, with a range of 18–104 years. The majority (N=156; 76.84%) of the patients were <65 years of age. The mean number of years of education was 9.5 (SD=4.8; range, 0–18) years, and men formed two-thirds (68%; N=138) of the study sample. There was a slight predominance of patients being referred from medical specialties (N=110; 54.2%) compared with those referred from surgical specialties (N=93; 45.8%). The mean duration of delirum at the time of baseline assessment was 2.3 days (SD=2.3), with a range of 1–20 days, and about fourth-fifths (N=162; 78.8%) of the patients developed delirium after being admitted to the hospital, i.e., had hospital-emergent delirium. Only a few patients (N=43; 21.2%) had delirium at the time of admission to the hospital. The mean Informant Questionnaire on Cognitive Decline in the Elderly score for the study group was 3.05 (SD=0.19), and only 10 patients (4.9%) had a mean Informant Questionnaire on Cognitive Decline in the Elderly score >3.38, suggesting possible underlying dementia. The most common etiology associated with delirium was that of metabolic disturbances, which was present in more than three-fourths (N=157; 77.3%) of cases. However, in all patients, delirium was associated with multiple etiologies.

There was no statistically significant difference on the above-described variables, among those who could be followed up (N=203) and those who could not be followed up (N=119), suggesting that the profile of patients evaluated at the follow-up was representative of the whole study sample.

As per DRS-R-98, the frequency of various symptoms of delirium varied from 27.8% to 100%, with the lowest frequency for delusions. The most common symptoms were sleep–wake cycle disturbances (97.5%), poor attention (96.6%), disorientation (92.6%), and short-term memory disturbances and motor agitation (89.7%). In terms of the severity of symptoms, slightly more than two-fifths (42.4%) of the patients scored <15 on DRS-R-98, and one-third (32.5%) scored <18 for the DRS-R-98 total score, indicating possible subsyndromal delirium. The mean DRS-R-98 score at first assessment was 22 (7.3), the mean DRS-R-98 noncognitive score was 9.0 (3.8), and the mean DRS-R-98 cognitive score was 8.0 (3.4).

Subtypes of Delirium

Most of the patients were found to have the hyperactive motor subtype (131; 64.5%) of delirium. This was followed by a mixed subtype (47; 23.2%) and hypoactive subtype (N=13; 6.4%), and 12 patients were categorized as having no subtype (N=12; 5.9%).

Follow-Up Assessment

The mean DRS-R-98 severity score at the time of assessment for delirium was 3.5 (SD=2.3; range, 1–10). Three-fourths (N=151; 74.6%) of the patients had a DRS-R-98 severity score ranging from 1 to 4, another one-sixth (N=33; 16.25%) had a DRS-R-98 score of 5–6, and about 1/10th (N=19; 9.4%) had a DRS-R-98 severity score of 7–10.

Assessment of Recall and Level of Distress

Of 203 patients who were assessed at the follow-up, about one-third (N=71; 35%) could remember being confused, and the majority (N=132; 65%) could not recall that they had being confused recently. The majority of those who could not remember themselves to be confused were not distressed, and most of those patients who remembered that they were confused had a mild level of distress at not being able to remember their experience. The majority of those who remembered that they were confused were distressed (61 of 71), and half of them had a moderate level of distress, and another one-fifth had a severe or very severe level of distress. These results are shown in Table 1.

Distress Due to Delirium

Distress was assessed by using a qualitative method, per the spontaneous reporting of their experience of delirium on question 6 of the Delirium Experience Questionnaire. Table 2 shows the answers of a subgroup of patients who remembered their experience. In terms of common themes extracted from the qualitative data, the experiences commonly remembered included that of hallucinations, most commonly visual hallucinations, and this was associated with a moderate to severe level of distress. In terms of an emotional reaction, the common emotions were those of fear/fright (49.2%) and anxiety (14.8%).

Distress With Specific Symptoms

We assessed the distress associated with specific symptoms in those patients (N=61) who remembered their experience and reported distress on Delirium Experience Questionnaire question 5, by using a self-designed questionnaire that was completed by using a semistructured interview with the patient. As shown in Table 3, among the various symptoms of delirium that could be recollected and were associated with distress, distress was most commonly associated with insomnia, followed by visual hallucinations, uncooperativeness for treatment, pulling out tubes, being abusive, and drowsiness/confusion. Of all the symptoms associated with distress, most led to a moderate to severe level of distress, except for insomnia, which was most commonly associated with a mild level of distress, as depicted in Table 3. In terms of frequency, of 61 patients, one-third (N=20; 32.8%) reported distress associated with at least three of the 15 symptoms and another one-third reported distress associated (N=20; 32.8%) with four or more symptoms.

The total distress score was 5.36 (SD=3.2; range, 0–13), and mean total number of symptoms associated with distress was 2.95 (SD=1.53; range, 0–7).

Comparison of Profiles of Those Who Recalled Their Delirium Experience (N=71) Versus Those Who Did Not Recall Their Delirium Experience (N=132)

There was no significant difference between the two groups, those who recalled their delirium experience (N=71) versus those who did not recall their delirium experience (N=132), with respect to age, years of education, and gender distribution. Similarly, there was no difference in the duration of delirium at the time of baseline assessment and onset of delirium (i.e., hospital emergent or prior to hospitalization). Those who recalled the delirium experience and were distressed were more frequently admitted in the surgical wards than the other participants (59.2% versus 38.6%; chi-square value, −7.3; p=0.007).

In terms of the DRS-R-98 profile, those who recalled their delirium experience had a higher baseline severity score for the DRS-R-98 items of perceptual disturbances, language and fluctuation in symptoms, DRS-R-98 severity score, DRS-R-98 total score, DRS-R-98 noncognitive items severity score, and DRS-R-98 severity score at the time of assessment for the delirium experience. However, there was no difference in the DRS-R-98 cognitive items severity score. In terms of frequency of symptoms, those who remembered also had a significantly higher frequency of disturbance in language. Further, those who recalled had a lower frequency of subsyndromal delirium at baseline per the DRS-R-98 total score. All the variables that had a significant difference between the two groups are shown in Table 4.

In terms of motor subtypes, for comparison, the no subtype group was not included in the analysis. As shown in Table 4, those who remembered their experience of delirium more frequently had a hyperactive or mixed subtype of delirium.

Comparison of Profiles of Those Who Recalled and Were Distressed (N=61) Versus Other Study Participants (N=142)

When the study sample was divided into those who recalled and were distressed versus the other study participants, in terms of a sociodemographic profile, clinical profile, and the DRS-R-98, the differences persisted for the same variables, which differed significantly between those who recalled and those did not recall. Besides these variables, in terms of the frequency of the symptoms of delirium as assessed on DRS-R-98, those who recalled and were distressed had a higher frequency of perceptual disturbances, lability of affect, and long-term memory. In terms of severity of symptoms as assessed on DRS-R-98, those who recalled and were distressed had higher severity of long-term memory. However, the difference in the DRS-R-98 severity score at the time of assessment between the two groups became nonsignificant. Those who recalled and were distressed had a lower frequency of subsyndromal delirium at baseline. All the variables that had a significant difference between the two groups are shown in Table 4. Additionally, in terms of the frequency of the symptoms of delirium, those who recalled and were distressed had a higher frequency of perceptual disturbances, lability of affect, and long-term memory. In terms of severity, those who recalled and were distressed had higher severity of long-term memory. However, the difference in the DRS-R-98 severity score at the time of assessment between the two groups became nonsignificant. Those who recalled and were distressed had a lower frequency of subsyndromal delirium at baseline.

In terms of motor subtypes as per the amended Delirium Motor Symptom Scale, those who recalled and were distressed more frequently had a mixed subtype; however, this difference was not statistically significant.

Relationship of Distress With Other Variables

Correlation analysis was done to study the relationship between the levels of distress in terms of the severity of the delirium distress score (as assessed by the self-designed questionnaire) and total number of symptoms in the self-designed questionnaire in those who recalled and reported distress (N=61). There was no significant relationship that emerged for distress severity score and the total number of symptoms associated with distress, demographic variables (age and years of education), or clinical variables (duration of delirium at the baseline). A higher level of distress was seen among men, both in terms of distress severity score (Spearman rank correlation coefficient=0.254; p=0.05) and number of symptoms associated with distress (Spearman rank correlation coefficient=0.276; p=0.03). In terms of severity of the DRS-R-98 symptoms, the distress severity score had a significant positive correlation with the severity score on the items of perceptual disturbances (Pearson correlation coefficient=0.345; p=0.007).

Discussion

Although psychological distress after recovery from delirium is an important consequence of delirium, it has not received much attention. Most of the previous studies that evaluated this aspect have been limited by small sample.^{5,12,13,18,20–22} These studies have mostly focused on patients with cancer^19,26 or patients admitted to the intensive or critical care units.^15,17,18 These factors limit the generalization of findings to all the patients admitted to medical and surgical wards and are not necessarily terminally ill or very severely ill, requiring life support in the intensive care unit setting. Further, previous studies have not attempted to quantify the level of distress associated with various symptoms of delirium and have not looked at the relationship of severity of delirium with distress. The present study was an attempt to overcome some of the limitations of the existing literature.

Previous studies have reported that 28.3%−94% of patients can recollect their delirium experience. In the present study, slightly more than one-third of the patients could recollect their experience, and this is within the reported range. The majority of those who could not remember themselves to be confused were not distressed, and most of those patients who remembered that they were confused had a mild level of distress of not being able to remember. This is in contrast to the findings of our previous study in which patients who could not remember reported moderate to severe distress.^5,26 It is difficult to understand these differences, and it is possible that this difference could be due to other clinical and psychological factors, such as severity of primary physical illness, type of associated etiology, and the personality of the patient. Hence, further studies are required in this area to reach to a conclusion. The majority of those who remembered that they were confused were distressed (61 of 71; 85.9%), and half of them had a moderate level of distress and another one-fifth had a severe or very severe level of distress. This profile of distress is more akin to the findings of a previous study from our center and highlights the fact that those who remember their experience of delirium perceive a high level of distress.⁵

In the present study, among the various symptoms of delirium, in those who recalled their experience and reported distress about their experience, the commonly recalled symptoms included those of insomnia, followed by visual hallucinations, uncooperativeness for treatment, pulling out tubes, and being abusive. These symptoms were associated with a moderate to severe level of distress, except for insomnia, which was associated with a mild level of distress. This profile of recollection of symptoms and association with distress is somewhat similar to the existing literature, which suggests that, among the various symptoms, a perceptual disturbance such as illusions and hallucinations is more frequently recollected and is associated with distress.^18,26

In the study by Breitbart et al.,¹⁹ delirium severity (as measured by the Memorial Delirium Assessment Scale total score) was found to be negatively associated with patient delirium recall. They did not find any difference in the delirium experience between the hyperactive and hypoactive subtypes. In contrast, in the present study, recall of the delirium experience was associated with a higher baseline severity score, DRS-R-98 severity score, DRS-R-98 total score, DRS-R-98 noncognitive items severity score, and DRS-R-98 severity score at the time of assessment for the delirium experience. Similarly, those who remembered their experience more frequently had a hyperactive or mixed subtype of delirium. However, there was no difference in the DRS-R-98 cognitive items severity score. There was a difference in the scale used to assess the symptomatology and the type of patients included in the present study and that in the study by Breitbart et al.,¹⁹ which could explain why the findings are contradictory. Several hypotheses can be suggested to understand these differences. First, there could be a difference in the overall severity of the physical illness between the two studies. The study of Breitbart et al.¹⁹ was limited to those with various types of cancer, whereas in the present study, the most common etiology was that of metabolic disturbances. Second, in the study by Breitbart et al.,¹⁹ about one-fifth of the patients had a history of dementia, whereas in the present study, only 5% of patients had a history of recent cognitive decline. Third, most of the patients in the sample of Breitbart et al.¹⁹ had moderate to severe delirium, whereas in the present study, about one-third of the patients had subsyndromal delirium at baseline. Fourth, in the present study, patients were assessed for distress after at least 24 hours of recovery from the delirium, and most of the patients had very few residual symptoms of delirium. However, similar comparative data were not presented by Breitbart et al.¹⁹ Fifth, overall, the age of the study sample in the present study was 47 years, in contrast with the mean age of the study population as 58 years in the study sample of Breitbart et al.¹⁹ Sixth, the whole experience of delirium in terms of recall could be influenced by the total duration of symptoms, which was not taken into account in both the studies. It is possible that some of these factors could have influenced the recall pattern in both these studies and could explain the differences. Further studies are required to address this issue further. Seventh, in the present study, motor subtyping was based on a well-validated scale compared with Brietbart et al.,¹⁹ who categorized patients into different subtypes on the basis of rating of a particular item of the Memorial Delirium Rating Scale.

The themes of the verbatim description were those of remembering the delirium experience in terms of hallucinations most frequently, and in terms of the emotional experience, it was that of fearfulness and anxiety. Similar descriptions in terms of an emotional experience of fear, anxiety, and feeling threatened have been described in the previous literature, and in terms of symptoms, a previous study from our center also suggested that patients most frequently remembered hallucinations and illusions.^5,7

Findings in relation to the distress associated with the DRS-R-98 profile were also similar to the recall, except that there was no difference in the DRS-R-98 severity score at the time of assessment for the delirium experience. These findings can have overall management implications. The association of a higher severity of delirium with a higher level of distress suggests that it is very important to identify and manage delirium as early as possible to not only reduce the severity of the delirium but to reduce the overall distress associated with the delirium experience.

The relationship of perceptual disturbances with distress and recall is supported by the findings of some of the studies that suggest that severe disturbances in these parameters can lead to posttraumatic stress disorder.¹⁰

The present study has certain limitations. The study was limited to a patient population referred to psychiatry consultation liaison services, and the findings should be interpreted accordingly. We assessed the recall and distress in a quantitative manner mainly limited to certain features. It is quite possible that the assessment might not be comprehensive, and there would be other areas of recall and distress. We also did not look at the relationship of the delirium experience with the total duration of delirium, which can possibly influence delirium experience.

To conclude, findings in the present study suggest that delirium is associated with distress in a significant proportion of patients. It is also evident that a higher severity of delirium and a higher severity of noncognitive symptoms (i.e., mean score of item 1–8 on the DRS-R-98) are associated with higher distress in those who recall their delirium experience. Because of this, it is desirable to take steps to prevent the occurrence of delirium, and when it occurs, it should be treated at the earliest time possible to reduce the severity of delirium, which is associated with distress. This might have a preventive effect in mitigating the potential psychological consequences in the future.