If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
University of South-Eastern Norway, Drammen, NorwayDepartment of Internal Medicine, Telemark Hospital Trust, Skien, NorwayOld Age Psychiatry Research Network, Telemark Vestfold, Vestfold Hospital Trust, Tonsberg, Norway
Assess the frequency of delirium during any acute event, its risk factors, and the duration of delirium in nursing home patients.
Design
Prospective 2-month follow-up study.
Setting and Participants
145 nursing home patients living in 3 Norwegian nursing homes.
Methods
At baseline, known risk factors for delirium were obtained from medical records. During any acute events where the nurses decided that a physician had to be alerted, the Confusion Assessment Method was used to identify delirium on days 1, 2, 4, and 6 and thereafter weekly if delirium was present on day 6. The precipitating cause of delirium was registered based on diagnostics performed and treatment given.
Results
One or more acute events occurred in 57 patients, and 34 (60%) of these patients developed delirium. In 91% of the patients with delirium, the delirium was present when the physician was alerted about the acute change. Delirium lasted for more than 1 week in 15 of the 34 patients. In 18 of the 34 patients with delirium, an infection was its precipitating factor. Regular use of benzodiazepines and a diagnosis of vascular dementia were significantly associated with delirium in the logistic regression model adjusted for age, number of drugs, and comorbidity [adjusted odds ratio (95% confidence interval) 3.75 (1.44-9.74) and 5.59 (1.53-20.43), respectively].
Conclusions and Implications
Acute events and illness were common in nursing home patients, and in our study, 60% had delirium associated with the event. In 9 of 10 patients, the delirium was present when the physician was alerted about the acute change, and infection was the most frequent cause of the delirium. Regular use of benzodiazepines and a diagnosis of vascular dementia were independent predisposing factors for delirium.
Delirium is a syndrome characterized by an acute mental change, attention deficit, and altered arousal. It develops over a short period of time, fluctuates, and has an underlying cause.
Although patients of any age and robustness can develop delirium, patients with higher age, functional impairment, severe comorbidities, and dementia are at most risk.
Nursing home patients have a high burden of comorbidity, and most are dependent in activities of daily living. Patients in Norwegian nursing homes have a mean age of 84 years, and more than 80% suffer from dementia.
In Norway, the political initiatives are for older persons to live in their own home as long as possible and receive home-based care if dependent in activities of daily living. Consequently, long-term care is reserved for persons who need continuous assistance where home-based care is insufficient.
Even though most nursing home patients have many risk factors for delirium, few researchers have studied the prevalence, risk factors, and the course of delirium in nursing home patients. It is estimated that these patients experience 2 to 4 acute events yearly and run the same risk of delirium as hospitalized older patients, of whom more than 20% develop delirium.
Preliminary data: An adapted hospital elder life program to prevent delirium and reduce complications of acute illness in long-term care delivered by certified nursing assistants.
Our aim was to identify how often delirium occurred during acute events, determine its predisposing and precipitating factors, and to follow the course of delirium among patients living in 3 Norwegian nursing homes.
Method
Design
This prospective study took place in 3 Norwegian nursing homes from September 1 through October 2018. The nursing homes have 231 beds, and daily visits from physicians (not weekends). The municipal after-hours care assisted if acute events occurred in weekends. Norwegian municipal after-hours care is normally occupied by general practitioners who are able to both travel to the nursing home and provide consultation over the telephone, or the patients can be transferred to the out-of-hours medical center for further examination.
Participants
Patients of any age with permanent nursing home residence were included. Patients were excluded if they suffered from terminal illness (life expectancy <1 week).
Collection of Data
At inclusion, information regarding age, sex, and time in institution were registered by the project nurse specialized in geriatric medicine (W.H.S.). All regular medications were registered, and polypharmacy was defined as use of 5 or more regular medications. Based on the medical record and prescribed medications, chronic diseases were recorded in a standardized way as suggested by Barnett that allows diagnoses to be registered based on medications prescribed.
Diagnosis of dementia and its subtypes was extracted from medical records.
Screening for Delirium
Screening for delirium was initiated if an included patient had any acute change leading to contact with a physician. We used a wide definition of acute events, meaning any acute change where the nurse decided that a physician had to be involved. Screening for delirium was performed using the validated instrument Confusion Assessment Method (CAM) Short Form.
For a delirium to be present, there had to be an acute mental change or fluctuation, presence of inattention, and either disorganized thinking or altered level of consciousness. The nurse, health care worker, or physician who best knew the patient in their stable state performed the screening. The delirium screening was supported by cognitive testing adjusted to patients’ normal cognitive function, a formal test of attention, and the Observational Scale of Level of Arousal (OSLA).
The delirium screening was repeated on day 1 (contact with physician), day 2, day 4, and day 6. If a patient had symptoms of delirium on day 6, the delirium screening was repeated once a week until symptoms resolved. On day 6, the staff at the nursing home registered what procedures and diagnostic tests that had been performed to reveal the precipitating cause of the acute change. Also, what conditions the physician found that had provoked the change was registered. If an included patient was hospitalized during the project period, the project nurse visited the patient in hospital and performed the delirium screening following the same procedure as in the nursing homes.
In August 2018, before the start of the project, the staff at the nursing homes had lectures and training on delirium and the use of the CAM. Also, the project nurse visited each ward twice a week during the project period to be available for any questions. If uncertainty existed on any scoring of the CAM, the staff were instructed to write a description of the patient in text, and the case was discussed with the project nurse and project leader (M.K.) until consensus was reached.
To assess reliability, the CAM questionnaire was performed simultaneously by 2 of the staff in 9 patients; agreement of the presence of delirium was found in all patients (kappa score: 1.00).
Ethics
The Regional Committee for Medical and Health Research Ethics approved the project (ID 2018/1099). Inclusion was based on written consent to participate. In patients with reduced ability to sign, written informed consent was obtained from both the patients and their next of kin.
Statistics
Statistical analyses were performed using IBM SPSS Statistics for Windows, version 25.0 (IBM Corp, Armonk, NY). Continuous data is presented as mean and standard deviation (SD), whereas categorical variables are reported as frequencies and percentages. When exploring risk factors for delirium, known predisposing factors for delirium were selected before onset of the project based on previous literature. First, we performed bivariate analyses comparing patients with and without delirium. For categorical variables, the chi-square test was used, whereas t test was used in normally distributed continuous variables. In skewed continuous variables, the Mann-Whitney test was used. We then performed a multivariate logistic regression including preselected potential risk factors for delirium. “Delirium any time” was used as the dependent variable, and all subjects were included in these analyses.
Cohen kappa statistics was used to assess the reliability of the delirium assessments. Kappa score range from 0.00 till 1.00, and a score above 0.75 indicates excellent reliability.
Among 182 patients available for inclusion, 145 patients gave consent to participate. Inability to contact a family caregiver for consent or staff recruitment time limitations led to omission of 32 potential subjects, whereas 5 patients refused participation. Descriptives of the included patients are given in Table 1. Two in 3 patients were women, and 75% of the patients were aged ≥80 years. Nine patients (6%) were aged ≤67 years. Polypharmacy was present in 81% of the patients, and 54% of the patients had lived in an institution for less than 1 year. The chart review revealed that 80% of the patients had a diagnosis of dementia at inclusion.
Table 1Patient Characteristics on Inclusion (N = 145)
During the project period of 2 months, 77 acute events demanding examination by a physician occurred. These 77 events were distributed among 57 of the 145 patients, and 5 patients were hospitalized. Delirium screening revealed that 34 of the 57 patients (60%) with acute events had delirium. Accordingly, among all included patients, 23% (34 of 145 patients) experienced delirium during the project period. Predisposing factors for delirium are presented in Table 1. In bivariate analysis, higher number of regular medications, more chronic diseases, a diagnosis of vascular dementia, and regular use of benzodiazepines were associated with an increased risk of delirium. In the adjusted model, regular use of benzodiazepines and vascular dementia, were significantly associated with risk of delirium, adjusted odds ratio: 3.75 [95% confidence interval (CI) 1.44-9.74] and 5.59 (95% CI 1.53-20.43) respectively.
Course of Delirium
Thirty-one of the 34 patients (91%) with delirium had delirium on the day the acute illness was registered. In the 3 remaining patients, first day of delirium was registered on day 2 or 4 of their acute illness. Ten of the 34 patients with delirium (29%) had resolution of their delirium after the first day (Figure 1). In 15 of the 34 patients with delirium (44%), a full syndrome of delirium was present for more than 1 week. In 2 patients, the delirium lasted for 2 weeks or more. A total of 126 days with delirium were registered in the 34 patients during these 2 months, 5 patients were hospitalized, and 11 of the 145 included patients died.
Figure 2 presents the precipitating causes for delirium. In 16 patients, more than 1 precipitating cause was identified. In 2 patients, no direct cause of the delirium could be found.
Fig. 2Precipitating cause of delirium (n = 34). The figure illustrates the 50 identified precipitating factors for delirium among included patients. This exceeds the number of patients with delirium as more than 1 precipitating cause of delirium was found in 16 patients.
In our study of 145 nursing home patients followed for 2 months, we found that 57 patients had 1 or more acute events, provoking an examination by a physician, and 60% developed delirium in relation to acute events. In 9 of 10, the delirium was already present the day the physician was alerted, and in 44% the delirium lasted for more than 1 week. Infection was the most common precipitating factor for delirium, and regular use of benzodiazepines and a diagnosis of vascular dementia were independent predisposing factors for delirium.
As nursing home patients hold many predisposing factors for delirium, the high prevalence of delirium in our study confirms previous research.
However, our finding that more than 20% of included patients had 1 or more episodes of delirium during 2 months, and that more than half of the patients developed delirium during acute events, implies that delirium might be even more common among nursing home patients than previously reported. As long-term care in Norway is reserved for those in need of continuous care and the majority suffer from dementia, our included patients might be both older and less robust than patients in comparable studies from other countries.
Among included patients, both the number of chronic diseases and regular medications were high, which may have hampered these factors from reaching statistical significance as risk factors for delirium. The high rate of dementia in nursing home patients can make the diagnostic of delirium challenging, as neuropsychiatric symptoms in dementia can be difficult to differentiate from delirium. In our study, the precipitating factors for delirium were identified in all patients, except 2, and illustrate the importance of searching for acute illness or troublesome conditions in patients who develop any sudden change.
In our study, almost all patients with delirium had delirium already when the physician was alerted about the acute change. This illustrates that atypical presentation of acute disease is common in nursing home patients. Unfortunately, this may limit the transferability of the well-proven nonpharmacologic prevention of delirium in acutely ill hospitalized patients
toward nursing home patients. However, as nursing home patients run a high risk of both acute events, and delirium during any acute event, we believe this illustrates that nonpharmacologic delirium prevention must be integrated in daily care in nursing homes. Review of medications is a part of delirium prevention,
we are the first to report this in nursing home patients. Whether deprescribing benzodiazepines may reduce the risk of delirium in nursing home patients should be an aim for future research. A higher prevalence of delirium in patients with vascular dementia was also found by others.
Preliminary data: An adapted hospital elder life program to prevent delirium and reduce complications of acute illness in long-term care delivered by certified nursing assistants.
and supports the need for qualified personnel making good observations and close cooperation with physicians with knowledge on the patients' comorbidities and levels of function. Norwegian nursing home patients have a life expectancy of less than 2 years
The Norwegian Directorate of Health. Botid i sykehjem og varighet av tjenester til hjemmeboende. Length of stay in nursing homes and the duration of home care services. The Norwegian Directorate of Health,
2017
and have a poor prognosis if severe illness strikes. It is important to plan how any acute illness should be treated by integrating medical knowledge about each patient's prognosis with the wishes of the patient and their caregivers. Few of our patients were hospitalized during acute events; this might imply that the acute events were less serious. However, in Norwegian nursing homes, most patients are only hospitalized for conditions that cannot be treated outside of hospitals, like hip fractures or ileus. Internationally, there exists interventions on reducing hospitalizations from nursing homes.
Our study has several limitations, and the low number of included patients and the short follow-up period are the most important. Also, risk factors were registered from medical records only, and we did not assess cognitive or physical function at baseline. Also, the validity of CAM in the nursing home setting with high prevalence of severe dementia is unknown. Strengths of the study are the wide inclusion criteria and the use of validated tools. None of the patients were lost from follow-up, and all acute events were recorded.
Conclusions and Implications
In our study of 145 nursing home patients followed for 2 months, delirium and acute events were common. Most patients with delirium had delirium when the physician was alerted about an acute illness or change, and almost half of the patients had delirium for a week or longer. This illustrates the importance of integrating the well-proven nonpharmacologic prevention of delirium into the daily care in nursing homes. Regular use of benzodiazepines at inclusion and a diagnosis of vascular dementia were independent risk factors for delirium, and infection was the most common precipitating cause of delirium.
Acknowledgments
The authors thank all the patients and employees at Modumheimen, Eikertun, and Solberglia nursing homes. Thanks to professor Leiv Sandvik for statistical support.
References
American Psychiatric Association
Diagnostic and Statistical Manual of Mental Disorders.
Preliminary data: An adapted hospital elder life program to prevent delirium and reduce complications of acute illness in long-term care delivered by certified nursing assistants.
The Norwegian Directorate of Health. Botid i sykehjem og varighet av tjenester til hjemmeboende. Length of stay in nursing homes and the duration of home care services. The Norwegian Directorate of Health,
2017
00909-9&id=gr1.jpg){kind=link}
00909-9&id=gr2.jpg){kind=link}