A pilot volunteer reader program decreases delirium days in critically ill ICU patients

Delirium is a form of acute brain failure that alters mental status and affects proper memory, emotion, and behavior, developing over hours to days. Between 20-80% of ICU patients develop delirium, depending upon the tools used for diagnosis.

Delirium is commonly overlooked because only 25% of delirium patients present with hyperactivity, that is, showing evidence of hallucinations and agitation. More commonly patients experience hypoactive delirium with masked symptoms — delayed response and actions; occasionally, patients present both hyper-and hypo active symptoms. Delirium can have severe consequences, including increased ICU length of stay and mortality rate. Untreated delirium may cause irreversible cognitive failure. The precise causes of delirium remain unclear, furthering the incidence of misdiagnosed and undiagnosed cases. However, factors that seem to exacerbate delirium include sleep deprivation, sensory deprivation, and social isolation. All of these factors are common in ICUs and should be examined to better understand and prevent delirium.

As social isolation may be a delirium risk factor, we hypothesized that a program encouraging human interaction with patients could decrease its incidence. Specifically, we proposed reading to critically ill ICU patients on a daily basis could decrease delirium days—the number of days when patients present with delirium, in either the hypo-or hyperactive form.

This paper reported the preliminary results of the ICU Reader Program.

Materials and methods
The ICU Reader Program began as a service project in which we hoped to identify a “signal” if our hypothesis proved valid.

Volunteers underwent hospital-approved training related to patient/information confidentiality before enrolment as readers. The readers ranged in age from 16-40. These volunteers spent 5-10 hours in the ICU each week, sitting with 20-40 patients weekly at the height of the program.

Patients were chosen by the ICU Charge Nurse based upon their assessment of who was most suitable for reading/companionship. When the patients were conscious and alert, the readers confirmed interest — almost always given — before sitting and reading. Readers determined how much time was spent with patients, averaging 35 minutes.

Some patients — intubated or lightly sedated — preferred a conversation to reading. Notes were written and passed between the patient and reader. Others desired only companionship as they ate dinner. Typically, reading was the course of action taken. Books were chosen at the readers’ discretion; however, patients could request a religious text or newspaper in place of a novel.

Results
While our study timeframe was short, we noted a promising signal. Of 33 patients read to, 12 were diagnosed with delirium through the Confusion Assessment Method for the ICU (CAM-ICU), a tool relying on four diagnostic features: acute mental status change with fluctuating course (feature 1), inattention (feature 2), altered level of consciousness (feature 3), and disorganized thinking (feature 4). Patient responses and further observations are recorded by the examiner. If patients test positive for features 1 and 2 and either 3 or 4, delirium is diagnosed. This test was performed by the bedside nurse on every patient in the ICU, three times a day. The CAM-ICU scores were compared 72 hours before and 72 hours after the reading episode. Appropriate statistical analytical tests were performed.

Prior to reading, there were 27 delirium days experienced by the 33 patients (0.82 delirium days per patient). Total delirium days and delirium days per patient before and after the reading intervention were, respectively 27 versus 10, and 0.82 versus 0.33 (all statistically significant at p < 0.05) (Figure). As the program began as a service project, we have not yet accounted for other factors such as age, gender, days in the ICU, medications, comorbidities, and so forth.

Discussion
Delirium is known to increase costs of care, duration of mechanical ventilation, length of stay, re-intubation rate, long-term cognitive function, discharge to a long-term care facility, and patient mortality. Our ICU Reader Program was a controlled, non-randomized, study; patients served as their own controls pre-and post-reading. The presence of a signal showing a reduction of total delirium days in the cohort, as well as delirium days per patient, suggests that reading and talking with patients in the ICU may be an effective, simple, and low-cost intervention.

Due to our poor understanding of the pathophysiology, up to 50% of delirium episodes are missed by physicians. When diagnosed, 80% of delirious patients may receive inadequate treatment. Medications and environmental disturbances are the greatest causes of delirium. Recently started drugs or increased dosages are to be carefully observed and reversed, if possible. The risks versus benefits of medications should be considered, as they have been noted to cause, or increase, delirious states.

ICU patients need to be maintained, as much as is possible, with a normal sleep-wake cycle and with minimization of social isolation to avoid the onset of delirium; this also sustains their safety, comfort, and overall physical and psychological function. Patients should be allowed to use eyeglasses and hearing aids to preserve sensory input and be mobilized with the help of physical therapy to avoid immobility, falls, and pressure ulcers; delirium duration is decreased by 50% with early mobilisation. A clean hospital room with natural sunlight supports positive daytime living and behaviour, while a dark and quiet room with limited interruptions supports maintenance of a healthy sleep-wake cycle. Practitioners should consider not disturbing patients, barring an emergency, between the hours of 22:00 and 06:00. Reading and companion programs, like the ICU Reader Program, helps increase social interaction. The above measures help manage complications, maintain comfort and safety, and restore function while avoiding development of delirium.

Strictly hyperactive delirium, the conventional delirium image with patients exhibiting signs of hallucinations or inappropriate behaviour, is found in only a small proportion of ICU patients. Hypoactive or “quiet” delirium is characterized by lethargy and has been found to have a significantly higher six-month mortality rate of about 33%; the explanation for this is unclear. Misdiagnosed or mistreated delirium impedes communication between patients and physicians and between patients and families. Especially at the end of life, this barrier can cause significant stress and interfere with final decisions. Before death, typically within two days, delirium is present in most (about 88%) of patients; delirium presenting just before death is often irreversible. It is important to recognize and treat delirium in adults before it leads to readmission, institutionalization, and/or, potentially, irreversible cognitive decline. Delirium complicates the hospitalization of 20% of patients over the age of 65 years. With every day of delirium experienced, there is a 10% increase in relative risk of death. While delirium mortality rates vary broadly, from 10% – 76%, as of 2011, the yearly mortality rate for all delirium cases is 35% – 40%.

The presence of delirium in patients not only increases their chances of hospital complications but can leave them cognitively and emotionally impaired. Changes to medication prescriptions and especially environmental factors, including social isolation, provide possible treatment plans for consequential delirium.

As previously stated, this was a pilot project, and the signal we have noted must be taken as a preliminary. Additional work, carried out in a randomized, controlled, and multicenter manner, is warranted to determine if our findings hold up under scrutiny. Our data are limited and there is reason to approach our conclusions tentatively. The potential for bias in the data we report is not insignificant. The number of patients studied is small, and individuals considered “appropriate” for interaction with our Readers were chosen by the ICU charge nurse; both these issues may have inadvertently introduced bias. It will require further study and analysis to determine if the material read matters as much, or more, than the act of reading itself.

Nonetheless, these findings are of great interest as we may have identified a low cost and non-pharmacologic method to address delirium.