Abstract
Background
Delirium is a common and serious complication of inpatient hospital care in older patients. The current approaches to prevention and treatment followed in German hospitals are inconsistent. The aim of this study was to test the effectiveness of a standardized multiprofessional approach to the management of delirium in inpatients.
Methods
The patients included in the study were all >65 years old, were treated for at least 3 days on an internal medicine, trauma surgery, or orthopedic ward at Münster University Hospital between January 2016 and December 2017, and showed cognitive deficits on standardized screening at the time of admission (a score of ≤=25 on the Montreal Cognitive Assessment [MoCA] test). Patients in the intervention group received standardized delirium prevention and treatment measures; those in the control group did not. The primary outcomes measured were the incidence and duration of delirium during the hospital stay; the secondary outcomes measured were cognitive deficits relevant to daily living at 12 months after discharge (MoCA and Instrumental Activities of Daily Living [I-ADL]).
Results
The data of 772 patients were analyzed. Both the rate and the duration of delirium were lower in the intervention group than in the control group (6.8% versus 20.5%, odds ratio 0.28, 95% confidence interval [0.18; 0.45]; 3 days [interquartile range, IQR 2–4] versus 6 days [IQR 4–8]). A year after discharge, the patients with delirium in the intervention group showed fewer cognitive deficits relevant to daily living than those in the control group (I-ADL score 2.5 [IQR 2–4] versus 1 [IQR 1–2], P = 0.02).
Conclusion
Structured multiprofessional management reduces the incidence and duration of delirium and lowers the number of lasting cognitive deficits relevant to daily living after hospital discharge.
In patients aged over 65, the commonest complication of inpatient treatment is delirium (1–3): in this population its prevalence is between 40% and 80%. Apart from age (> 65 years), the greatest risk factor for the occurrence of delirium is a preexisting cognitive deficit (4– 6).
Delirium has serious consequences. Not only does it prolong the hospital stay, it also leads to increased mortality (adjusted hazard ratio 1.95; mean follow-up 22.7 ± 15.5 months), worsens functional and cognitive outcome, and increases the risk that patients will go on to need long-term care (7– 20). Delirium leads to persistent cognitive deficits of dementia-like severity that are relevant to daily living, and these may occur even in the absence of preexisting dementia-like symptoms (21). In addition to this, longer hospital stays, increased nursing requirements, and the resulting consumption of resources are a financial burden on the treating hospital (3, 22– 24).
Although delirium in hospital is of pressing relevance for both patients and and those treating them, it is rare in German hospitals to find any standardized approach to its prevention and treatment – despite the fact that the population is aging and cases of delirium may be expected to rise. One reason for this is that in the current hospital landscape delirium does not fall clearly within one particular specialty or profession, because its symptoms vary widely and its pathophysiology is multifactorial. Against this background, in 2015 the concept of a “dementia-sensitive hospital” was introduced at Münster University Hospital (Universitätsklinikum Münster, UKM) (box). Through the use of a standardized approach based on this concept, and a multiprofessional delirium prevention and treatment team, it is hoped that the frequency, severity, and duration of delirium during a hospital stay will be reduced. The efficacy of this approach has been tested in a randomized controlled trial. The aim was to achieve a significant reduction in delirium rates and durations and to prevent permanent cognitive deficits relevant to everyday living resulting from the stay in hospital. The study was approved by the ethics committee of the Westphalia–Lippe Medical Association and the University of Münster (December 2015, 2015–531-f-S). It was subsequently registered in the German Register of Clinical Trials (DRKS) on 10 January 2019 (registration number DRKS 00016352).
BOX. The concept of the “dementia-sensitive hospital” at Münster University Hospital.
A full-time multidisciplinary team consisting of neurologists, psychogeriatricians, clinical pharmacists, specialized nursing staff and nurse scientists, social workers, and volunteers provides close care for all patients aged >65 years during their stay on the medical and surgical wards. The aim is a pragmatic approach to put the concept into practice as part of everyday clinical routine.
Immediately after admission, patients are first screened for cognitive deficits by trained nurses using the Montreal Cognitive Assessment (MoCA), adjusted for educational level. Patients admitted overnight who do not undergo an emergency intervention are screened the following morning. If cognitive impairment is found (≤=25 points), the patient is considered to be at risk. At-risk patients receive special primary nursing with regular cognitive reorientation and both physical and cognitive activation. In addition, regular delirium screening (at least once a day) is carried out using the Confusion Assessment Method (CAM) on the first three postoperative days (surgical wards) or days after admission (medical wards). On admission, a clinical pharmacist reviews the patient’s home medication and checks it for indication, tolerability, dosage level, adjustment to age, liver or kidney dysfunction, drug interactions, and anticholinergic side effects. If necessary, targeted optimization of the home medication is carried out as early as the first day in hospital. Patients continue to be under the care of a pharmacist throughout their hospital stay so that potentially inappropriate medication can be avoided.
At-risk patients are accompanied by specialized nursing staff and care assistants to operations, complex investigations, and diagnostic or therapeutic interventions. In addition, interdisciplinary rounds (with a clinical pharmacist, nurse scientist, geriatrician, neurologist, and psychiatrist) are carried out once a week jointly with the treating departments, during which the cases of at-risk patients and patients with delirium are discussed on an interdisciplinary basis. The focus is on delirium-triggering factors or medications as well as on the required diagnostic procedures.
At-risk patients also receive special primary nursing care at least twice a day (interaction with the nurse caring for the patient and with the physiotherapist; examination of the patient; specific education of relatives) with regular cognitive reorientation and both physical and cognitive activation.
If delirium cannot be prevented despite preventive measures, a standardized diagnostic procedure is followed, principally to avoid overdiagnosis (which is common), and rapid, standardized pharmacological, nonpharmaceutical treatment is given. A pharmaceutical guideline based on the current literature was developed for the purpose (efigure), with treatment suggestions adapted to the geriatric patient population and to the structures of an acute hospital. The team can be reached daily by other departments via a telephone “dementia hotline.” In addition, training for physicians and nursing staff and also for relatives has been developed and is carried out at regular 6-month intervals.
Patients and methods
Study participants
The study included older patients (>65 years of age) who received treatment as an inpatient for at least 3 days in one of the surgical or medical pilot wards (Departments of Orthopedics, Nephrology, and Trauma Surgery) between January 2016 and December 2017 and who scored ≤=25 points (at least mild cognitive deficits) on the Montreal Cognitive Assessment (MoCA) (25), adjusted for education, that was administered on admission. Alcohol withdrawal delirium was an exclusion criterion. Further details are given in the eMethods.
Study design
Patients were assigned to the intervention or the control group (figure 1). Details of the randomization procedure are given in the eMethods section.
Figure 1.
Flow diagram of the study design
I-ADL, Instrumental Activities of Daily Living; MoCA, Montreal Cognitive Assessment
The intervention group received delirium prevention measures and – if required – standardized treatment for delirium (26) (efigure) from the multiprofessional team in accordance with the concept of the “dementia-sensitive hospital” (box). The control group received standard therapy without specific delirium prevention or treatment.
eFigure.
Flow chart for pharmacological and nonpharmaceutical treatment of delirium
Patients in both the intervention and the control group were assessed for symptoms of delirium on the first 3 postoperative days or days in hospital using the Confusion Assessment Method (CAM) (27). In both groups the CAM was always carried out at a fixed time in the afternoon. The testers were trained in its use. The CAM is regarded as a time-efficient and valid screening tool for the detection of delirium and is recommended in the relevant guidelines (27– 30). All patients (intervention and control groups) underwent a second MoCA 1–2 days before discharge. All study participants who developed delirium during their hospital stay were again reassessed by the MoCA 12 months after discharge. In addition, the Instrumental Activities of Daily Living (I-ADL) scale (31) was administered to patients or relatives to quantify the extent of limitations relevant to daily living. Those who administered the CAM, MoCA, and I-ADL tests were blinded as to treatment group (intervention versus control).
Endpoints
The primary endpoints of the study were:
Incidence of delirium, objectified using the CAM;
Duration of delirium, objectified using the CAM.
The secondary endpoints of the study were:
Cognitive deficits at discharge (MoCA score);
Cognitive deficits relevant to daily living 12 months after discharge (MoCA and I-ADL test results).
Statistical analysis
All statistical analyses were performed using IBM SPSS software, version 27.0 (IBM Corp., Armonk, NY, USA) and R, version 4.0.3 (R Foundation, Vienna, Austria). A significance level of 5% was set. For details of the analysis, see eMethods.
Results
A total of 1472 patients > 65 years of age were assessed on admission to the pilot wards using the MoCA. Of these, 64% showed at least mild cognitive deficits on admission (≤= 25 points [n = 936]), and of these patients, 785 were included and randomized (figure 1).
Data from a total of 772 patients (55.4% female) were analyzed. Details are shown in Table 1.
Primary endpoints
During their hospital stay, 27 of 397 patients (6.8%) in the intervention group developed delirium, compared with 77 of 375 patients (20.5%) in the control group (odds ratio: 0.28; 95% confidence interval: [0.18; 0.45]; P < 0.001; relative risk: 0.33). Delirium in patients in the intervention group lasted an average of 3 days (interquartile range [IQR]: 2–4) and in patients in the control group an average of 6 days (IQR: 4–8) (p < 0.001).
Secondary endpoints
Cognitive deficits at discharge and after 12 months
In 69% of the study participants (273 patients in the intervention group and 256 patients in the control group), repeat cognitive testing (MoCA) was successfully performed at the time of discharge (figure 1). At this timepoint the median MoCA score had changed in the intervention group (P < 0.001) but not in the control group (P = 0.45). Using a hierarchical linear model, in the next step it was shown that the MoCA score rose over time in patients from the intervention group, while in patients from the control group it fell (P < 0.001), so long as the patients had not developed delirium. In both groups, developing delirium meant a drop in MoCA score by 5.3 points (P < 0.001) (eMethods).
Regarding mortality rate, after 12 months there was no difference between the two groups (59 patients in the intervention group versus 64 patients in the control group; P = 0.40), and no difference between patients with delirium in the intervention group (n = 10) and those with delirium in the control group (n = 22) (P = 0.41).
In a subgroup of 45 of the 104 patients (figure 1) who developed delirium during their hospital stay, it was possible to carry out a repeat MoCA approximately 12 months after discharge (15 patients in the intervention group, 31 patients in the control group).
While the median MoCA score changed for patients with delirium in the intervention group compared to the score at admission (P = 0.04), this was not the case for patients with delirium in the control group (P = 0.14); using the hierarchical linear model, in the next step it was shown that the cognitive deficits of patients with delirium who were cared for in line with the specific dementia-sensitive approach and by the team (the intervention group) were not permanent (rise in MoCA by 2.85 points during the study) (P = 0.01), whereas cognitive impairments in patients with delirium who were not treated in line with the dementia-sensitive approach during their hospital stay persisted even at 12 months after discharge (fall in MoCA by 1.62 points) (eMethods).
Deficits relevant to everyday living at 12 months after discharge
Out of a total of 772 patients, 525 patients or their relatives were successfully contacted for a telephone interview on the I-ADL scale. The other 247 patients were not reached or had died during the study period (figure 1).
Patients with delirium in the intervention group (group G2), patients with delirium in the control group (group G4), and patients who did not show delirium during their hospital stay (intervention group: group G1; control group: group G3) differed as to their I-ADL scores 12 months after discharge (figure 2). Differences were shown between groups G1 and G2 (P = 0.04), G1 and G4 (P < 0.001), G2 and G4 (P = 0.02), and G3 and G4 (P < 0.001) (table 2). Patients without delirium (G1 and G3, median 5 points) in the hospital showed higher I-ADL scores than patients with delirium (G2 and G4), and also patients with delirium in the intervention group (G2, median 2.5 points) showed higher I-ADL scores than those in the control group (G4, median 1 point) (figure 2); thus, they had significantly fewer persisent impairments relevant to daily living (I-ADL score = 0: “completely dependent on the support of others”; I-ADL score = 8: “independent, able to carry out necessary activities of daily living”). Persistent cognitive deficits relevant to everyday living were more frequent and more pronounced after delirium in the hospital without the specific dementia-sensitive treatment approach (intervention group) than with this concept (I-ADL 1 point in the control group versus I-ADL 2.5 points in the intervention group) (figure 2).
Figure 2.
I-ADL values from patients without and those with delirium in the intervention group
(G1, G2) and in the control group (G3, G4)
I-ADL, Instrumental Activities of Daily Living
Table 2. I-ADL values of patients with and those without delirium in the intervention and control groups.
| I-ADL value | Pairwise comparison | P value | |
| Intervention group | |||
| G1: Patients without delirium (n = 253) | |||
| Median I-ADL (IQR) | 5 (2– 7) | G1 – G2 | P = 0.04* |
| G2: Patients with delirium (n = 14) | |||
| Median I-ADL (IQR) | 2.5 (2– 4) | G2 – G4 | P = 0.02* |
| Control group | |||
| G3: Patients without delirium (n = 214) | |||
| Median I-ADL (IQR) | 5 (2– 6) | G3 – G4 | P < 0.001* |
| G4: Patients with delirium (n = 44) | |||
| Median I-ADL (IQR) | 1 (1– 2) | G4 – G1 | P < 0.001* |
* Post-hoc test
Significance level set at 5%.
I-ADL, Instrumental Activities of Daily Living; IQR, interquartile range
Discussion
The results of the study show that structured delirium management in hospital can significantly reduce the incidence and duration of delirium in elderly patients and also likely prevent permanent cognitive deficits.
Implementing this kind of concept of delirium management in a German hospital requires preliminary work in setting up a multiprofessional team to develop and put in place a procedural approach adapted to that hospital without having other tasks to attend to. For the approach described here to become accepted, and for it to succeed in medical and economic terms, it is important that strategies are developed both for delirium prevention and for a standardized treatment for delirium – because not every instance of delirium in a hospital is preventable. Achieving this requires a multidisciplinary approach, i.e., collaboration between different professional groups (e.g., neurologists, psychogeriatricians, nursing staff and nurse scientists, clinical pharmacists).
Approaches to preventing delirium in older patients in hospital have already been published. Although evidence supporting them has been demonstrated in numerous studies (32– 35), to date the implementation of standardized delirium management in German hospitals has not been consistent enough. One reason for this is no doubt the low impact of delirium during a hospital stay on the billing of care. In addition, most studies were conducted in countries with differing health care systems and models of inpatient care, so that it is difficult to simply extrapolate the results to German hospitals with a heterogeneous patient population. For example, the Hospital Elder Life Program (HELP) put forward by S. K. Inouye in 1999 (https://help.agscocare.org/) (22, 36) is now being applied in quite a uniform way in more than 200 hospitals in 32 states in the USA and in 11 other countries, with both economic and medical success. In German hospitals, there has been little experimentation, and that mostly with modified models (37).
Comparison with the data presented here is therefore difficult. However, compared with, for example, existing geriatric-based approaches, the approach to managing delirium presented here is clearly more specifically aimed at preventing and treating delirium. Symptoms of delirium have a multifactorial etiology. Today, the assumption that delirium merely reveals a preexisting dementia that was undiagnosed until the patient was admitted to hospital needs to be revised. People with preexisting neurodegenerative diseases are, it is true, at significantly increased risk of developing delirium, but recent study results show that delirium is an independent risk factor for permanent cognitive deficits, irrespective of the etiology of the delirium or other comorbidities (21).
Despite all the limitations, the results of this study support these observations. Regardless, older patients are fundamentally at higher risk of developing delirium (38). Therefore, to ensure that a management approach can be effectively implemented in practical terms, it seems sensible to focus on older at-risk patients in hospital. In the present study, screening for less obvious cognitive deficits in older patients for the purpose of delirium prevention proved to be effective and feasible, and the etiology of the impairments was deliberately not included. Since implementing this approach was cost-neutral, as calculated by the control department of UKM (chiefly because the hospital stay was shorter), it has since been expanded: currently, all specialties at UKM are taking part.
The “dementia-sensitive hospital” approach is deliberately designed to be practical. Necessary specific treatments, most of which are not focused on cognitive symptoms or symptoms of dementia or delirium (and this is the case on both surgical and medical wards), are accompanied by an interdisciplinary team of delirium specialists using established approaches to delirium prevention and treatment. Given the incidence of delirium and its intensely negative consequences for the patient, the patient’s family, and the health care system, this seems both reasonable and necessary.
One limitation of the study that must be mentioned is a high drop-out rate. Twelve months after hospital discharge, a repeat MoCA was possible in only 43% of all patients who had developed delirium during their hospital stay. Moreover, the repeat measurement of cognitive deficits was limited to patients who had developed delirium while they were in hospital. This restricted use of resources can be justified by the body of recent research, which shows delirium as the greatest single risk factor for persistent deficits in the patient population studied (21). Nevertheless, it cannot be ruled out that other illnesses during the intervening period account for some of the deficits found at follow-up. Because of the small case numbers at follow-up, the results must be considered as only indicative, not representative. The rise in MoCA score in patients in the intervention group is plausible. It is known and to be expected that, because of their limited cognitive reserve, even healthy aged patients in the acute situation of hospital admission will achieve a low MoCA score which is not seen at follow-up after 12 months.
Although mortality is a competing event for long-term outcome, the lack of difference between mortality rates in the control group and the intervention group in the 12 months after discharge makes it highly unlikely that it affected the results.
Another limitation was the structural impossibility of blinding the staff on the wards where the approach was implemented. The raters for the CAM, MoCA, and I-ADL assessments were blinded, but this was not possible for all team members. However, this confounding factor tends to lead to more delirium-preventive treatment for patients in the control group as well, meaning that if anything significant results would be underestimated.
I-ADL score was determined in patients 12 months after discharge but not on admission. Most patients were not admitted electively. For this reason, it is not possible to conclude whether impairments shown in I-ADL scores were delirium-related or were preexisting.
Patients who unexpectedly remained in hospital for less than 3 days were not included. Whether the established approach to delirium management can also prevent delirium in short-stay patients must be investigated in further studies.
To summarize, it was possible to achieve effective delirium reduction with a standardized approach adapted to the hospital. We recommend that a multiprofessional approach to the prevention and also treatment of delirium in hospital should be implemented rigorously and in a coordinated way, so that it can be sufficiently effective and so become permanently accepted. In future, it will certainly make sense and also be economically efficient to reimburse costs for preventive approaches of this kind for patients who have to undergo hospital treatment despite having cognitive impairments.
On this latter point, various initial steps have been taken – for example, the service categories determined by the Joint Federal Committee that are suitable for quality contracts to be signed between hospitals and health insurance companies, and health policy strategies (for example, point 3.3 of the German government’s National Dementia Strategy) – although these need wider implementation. Certainly improvement is needed here at present. However, with the approach shown, the many existing concepts could be developed into a uniform, practical, and effective strategy for German hospitals, which in medical terms already appears worthwhile.
Supplementary Material
eMethods
Sample size calculation and registration
Before the planned start of the clinical trial, the Institute of Biometry and Clinical Research, University of Münster, carried out a power analysis. A statistical sample size calculation was performed based on previous publications in the anglophone countries with the same focus and similar study design. The power analysis/sample size calculation was carried out by means of a simulation study related to the primary research question of the incidence of delirium, which involved using a two-tailed Fisher’s exact test for 2 × 2 contingency tables. Based on information in the literature, the underlying assumption was that there was a delirium incidence of 60.0% in the control group and 38.4% in the intervention group, and that both groups consisted of patients with cognitive deficits in an acute care hospital. The calculation was limited by the highly heterogeneous and variable reporting of delirium rates in the recent literature. The significance level was set at α = 5%. This initially resulted in a sample size calculation of 190. Six months after the study began, it became apparent that because the patient population was more heterogeneous (surgical and medical patients), there were fewer delirium events to record and analyze than assumed in the literature. The sample size was therefore recalculated by the Institute of Biometry and Clinical Research, University of Münster. The result was an increase of the total study population to >600 patients.
Randomization
Randomization was by computer using a randomization list stratified by age and sex. Randomized allocation to the two treatment groups (“standard” versus “innovative”) was by stratified block randomization. The possible block lengths were 4/6/8. There were four strata:
Stratum A: male, over 50 years old;
Stratum B: female, over 50 years old;
Stratum C: male, up to 50 years old;
Stratum D: female, up to 50 years old.
Statistical analysis
Bonferroni adjustment was carried out for the two primary analyses (significant result at P ≤= 2.5%). All other analyses should be interpreted as explorative or descriptive. Pearson’s χ2 test was used to compare the incidence of delirium, mortality risk, sex, and preexisting dementia between the intervention group and the control group, and to compare mortality risk between patients with delirium in the intervention group and in the control group.
The two-tailed Mann-Whitney U test was used to compare delirium duration, Montreal Cognitive Assessment (MoCA) scores on admission and discharge, and 12 months after discharge, between the intervention and control groups and, in the case of MoCA scores, also between patients with delirium in the intervention and control groups. The Kruskal-Wallis test was used to analyze differences in deficits relevant to activities of daily living (median Instrumental Activities of Daily Living [I-ADL]) among the groups.
To analyze the cognitive outcome at the time of discharge in the intervention group compared with the control group, the statistical test chosen was the hierarchical linear model. The following variables were included: treatment: intervention (= 1) or control (= 0); delirium: present (= 1) or absent (= 0); timepoint: admission (= 1), discharge (= 2); hospital stay in days; and interaction term: intervention: (= 1) * timepoint: admission (= 1), discharge (= 2).
For this analysis, multiple imputation was carried out in order to use the entire patient data base (“intention to treat” [ITT] analysis) (N = 772).
Analysis of cognitive outcome at discharge
The estimated average expected MoCA value at discharge can be calculated using the following formula:
MoCA = 20.3 -- 1.3 * intervention -- 5.3 * delirium -- 0.2 * timepoint + 0.02 * days in hospital + 1.3 * timepoint * intervention
The result shows that the MoCA value rose over time in patients in the intervention group, whereas in patients in the control group it fell (P < 0.001), so long as the patient had not developed delirium. Delirium led to a drop in MoCA value by 5.3 points in both groups (P < 0.001) (etable 1).
Analysis of cognitive outcome 12 months after discharge
To analyze cognitive outcome 12 months after discharge in patients with delirium in the intervention group compared with patients with delirium in the control group, the statistical test chosen was the hierarchical linear model. The following variables were included: treatment: intervention (= 1) or control (= 0), timepoint (admission = 1, discharge = 2, 12 months after discharge = 3), delirium duration in days, hospital stay in days, and interaction term: intervention (= 1) * timepoint (admission = 1, discharge = 2, 12 months after discharge = 3).
For analysis of cognitive deficits 12 months after discharge, “completer” analysis was performed (N = 45) because the structure of the missing data meant that data were “not missing at random,” so imputation could not be carried out.
The estimated average expected MoCA value 12 months after discharge can be calculated using the following formula:
MoCA = 19.6 -- 3.4 * intervention -- 0.8 * timepoint -- 0.3 * delirium duration in days -- 0.01 * days in hospital + 2.2 * timepoint * intervention
While MoCA value rose in patients in the intervention group over a timeframe of approximately 12 months, in patients in the control group it fell (P = 0.01). Thus, after 12 months patients with delirium in the intervention group had improved their cognitive deficits compared to admission.
The reason that patients with delirium in the intervention group show better MoCA scores after 12 months is that the first MoCA was carried out after nonelective hospital admission, so the initial MoCA score often overestimates cognitive deficits and gives worse results than a MoCA performed after time spent in familiar surroundings and without acute illness (etable 2).
+ 20.3 (intercept)
- 1.3 (for patients in the intervention group)
- 5.3 (if delirium occurred)
- 0.2 per timepoint
+ 0.02 per days in hospital
+ 1.3 per timepoint (for patients in the intervention group)
+ 19.6 (intercept)
- 3.4 (for patients in the intervention group)
- 0.8 per timepoint
- 0.3 per day of delirium duration
- 0.01 per day in hospital
+ 2.2 per timepoint (for patients in the intervention group)
Table 1. Patient characteristics, MoCA scores, and incidence and duration of delirium.
| Intervention group (n = 397) | Control group (n = 375) | P value | |
| Age, years | |||
| Median (IQR) | 77 (73–81) | 78 (72–82) | P = 0.77*1 |
| Sex | |||
| Female – n (%) | 220 (55) | 208 (55) | P = 1*2 |
| Male – n (%) | 177 (45) | 167 (45) | |
| Preexisting dementia | |||
| n (%) | 37 (9.3) | 34 (9.1) | P = 0.12*2 |
| Polymedication/number of medications | |||
| Median (IQR) | 8 (5– 10) | 7 (5– 10) | P = 0.27*1 |
| MoCA score on admission | |||
| Median (IQR) | 22 (18– 24) | 21 (17– 23) | P = 0.08*1 |
| MoCA score on discharge | |||
| Median (IQR) | 22 (17– 25) | 20 (16– 23) | P < 0.001*1 |
| MoCA score 12 months after discharge | |||
| Median (IQR) | 20 (17– 24) | 15 (11– 19) | P = 0.01*1 |
| MoCA score (patients with delirium) on admission | |||
| Median (IQR) | 18 (16– 21) | 14 (10– 19) | P = 0.03*1 |
| MoCA score (patients with delirium) on discharge | |||
| Median (IQR) | 17 (14– 21) | 15 (10– 19) | P = 0.01*1 |
| MoCA score (patients with delirium) after 12 months | |||
| Median (IQR) | 20 (17– 24) | 15 (11– 19) | P = 0.01*1 |
| Delirium | |||
| n (%) | 27 (6.8) | 77(20.5) | P < 0.001*2 |
| Delirium duration, days | |||
| Median (IQR) | 3 (2– 4) | 6 (4– 8) | P < 0.001*1 |
*1 Mann-Whitney U-test
*2 Pearson’s χ 2 test
IQR, Interquartile range; MoCA, Montreal Cognitive Assessment
eTable 1. Results of the hierarchical linear model.
| Variable | MoCA value – points | P value |
| Intercept | 20.3 | P < 0.001 |
| Intervention | −1.3 | P = 0.03 |
| Delirium – present | −5.3 | P < 0.001 |
| Timepoint | −0.2 | P = 0.38 |
| Days in hospital | + 0.02 | P = 0.14 |
| Interaction term | ||
| Intervention * timepoint | +1.3 | P < 0.001 |
MoCA, Montreal Cognitive Assessment
eTable 2. Results of the hierarchical linear model.
| Variable | MoCA value – points | P value |
| Intercept | 19.6 | p < 0.001 |
| Intervention | −3.4 | p = 0.14 |
| Timepoint | −0.8 | p = 0.09 |
| Duration of delirium (in days) | −0.3 | p = 0.13 |
| Days in hospital | −0.01 | p = 0.79 |
| Interactions term | ||
| Intervention * timepoint | + 2.2 | p = 0.01 |
MoCA, Montreal Cognitive Assessment
Acknowledgments
Translated from the original German by Kersti Wagstaff, MA.
Acknowledgment
We thank Eric Rost of IU International University of Applied Sciences for his assistance in the statistical analysis calculations..
Footnotes
Conflict of interest statement
Laura Zupanc has received fees from Schlütersche Verlagsgesellschaft GmbH & Co. KG.
Prof. Hempel and Dr. Nolte have received lecture fees from the Westphalia–Lippe Chamber of Pharmacists.
The other authors declare that no conflict of interest exists.
References
- 1.Fong TG, Tulebaev SR, Inouye SK. Delirium in elderly adults: diagnosis, prevention and treatment. Nat Rev Neurol. 2009;5:210–220. doi: 10.1038/nrneurol.2009.24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Inouye SK. Delirium in hospitalized older patients: recognition and risk factors. J Geriatr Psychiatry Neurol. 1998;11:118–25. doi: 10.1177/089198879801100302. discussion 57-8. [DOI] [PubMed] [Google Scholar]
- 3.Inouye SK, Westendorp RG, Saczynski JS. Delirium in elderly people. Lancet. 2014;383:911–922. doi: 10.1016/S0140-6736(13)60688-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Singler K, Fruhwald T. [Delirium] Internist (Berl) 2014;55:775–781. doi: 10.1007/s00108-014-3464-3. [DOI] [PubMed] [Google Scholar]
- 5.Bickel H, Hendlmeier I, Hessler JB, et al. The prevalence of dementia and cognitive impairment in hospitals. Dtsch Arztebl Int. 2018;115:733–740. doi: 10.3238/arztebl.2018.0733. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Feinkohl I, Winterer G, Spies CD, Pischon T. Cognitive reserve and the risk of postoperative cognitive dysfunction. Dtsch Arztebl Int. 2017;114:110–117. doi: 10.3238/arztebl.2017.0110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kratz T, Diefenbacher A. [Acute and long-term cognitive consequences of treatment on intensive care units] Nervenarzt. 2016;87:246–252. doi: 10.1007/s00115-016-0078-0. [DOI] [PubMed] [Google Scholar]
- 8.Fick D, Foreman M. Consequences of not recognizing delirium superimposed on dementia in hospitalized elderly individuals. J Gerontol Nurs. 2000;26:30–40. doi: 10.3928/0098-9134-20000101-09. [DOI] [PubMed] [Google Scholar]
- 9.Isaia G, Astengo MA, Tibaldi V, et al. Delirium in elderly home-treated patients: a prospective study with 6-month follow-up. Age (Dordr) 2009;31:109–117. doi: 10.1007/s11357-009-9086-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Robinson TN, Raeburn CD, Tran ZV, Angles EM, Brenner LA, Moss M. Postoperative delirium in the elderly: risk factors and outcomes. Ann Surg. 2009;249:173–178. doi: 10.1097/SLA.0b013e31818e4776. [DOI] [PubMed] [Google Scholar]
- 11.Bickel H, Gradinger R, Kochs E, Forstl H. High risk of cognitive and functional decline after postoperative delirium. A three-year prospective study. Dement Geriatr Cogn Disord. 2008;26:26–31. doi: 10.1159/000140804. [DOI] [PubMed] [Google Scholar]
- 12.Pandharipande PP, Girard TD, Jackson JC, et al. Long-term cognitive impairment after critical illness. N Engl J Med. 2013;369:1306–1316. doi: 10.1056/NEJMoa1301372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ely EW, Shintani A, Truman B, et al. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA. 2004;291:1753–1762. doi: 10.1001/jama.291.14.1753. [DOI] [PubMed] [Google Scholar]
- 14.Pisani MA, Kong SY, Kasl SV, Murphy TE, Araujo KL, Van Ness PH. Days of delirium are associated with 1-year mortality in an older intensive care unit population. Am J Respir Crit Care Med. 2009;180:1092–1097. doi: 10.1164/rccm.200904-0537OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Girard TD, Jackson JC, Pandharipande PP, et al. Delirium as a predictor of long-term cognitive impairment in survivors of critical illness. Crit Care Med. 2010;38:1513–1520. doi: 10.1097/CCM.0b013e3181e47be1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.McCusker J, Cole M, Abrahamowicz M, Primeau F, Belzile E. Delirium predicts 12-month mortality. Arch Intern Med. 2002;162:457–463. doi: 10.1001/archinte.162.4.457. [DOI] [PubMed] [Google Scholar]
- 17.Schubert M, Schürch R, Boettger S, et al. A hospital-wide evaluation of delirium prevalence and outcomes in acute care patients - a cohort study. BMC Health Serv Res. 2018;18 doi: 10.1186/s12913-018-3345-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Émond M, Boucher V, Carmichael PH, et al. Incidence of delirium in the Canadian emergency department and its consequences on hospital length of stay: a prospective observational multicentre cohort study. BMJ Open. 2018;8 doi: 10.1136/bmjopen-2017-018190. e018190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Witlox J, Eurelings LS, de Jonghe JF, Kalisvaart KJ, Eikelenboom P, van Gool WA. Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA. 2010;304:443–451. doi: 10.1001/jama.2010.1013. [DOI] [PubMed] [Google Scholar]
- 20.Khan BA, Zawahiri M, Campbell NL, et al. Delirium in hospitalized patients: implications of current evidence on clinical practice and future avenues for research–a systematic evidence review. J Hosp Med. 2012;7:580–589. doi: 10.1002/jhm.1949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Goldberg TE, Chen C, Wang Y, et al. Association of delirium with long-term cognitive decline: a meta-analysis. JAMA Neurol. 2020;77:1373–1381. doi: 10.1001/jamaneurol.2020.2273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Inouye SK, Bogardus ST Jr, Charpentier PA, et al. A multicomponent intervention to prevent delirium in hospitalized older patients. N Engl J Med. 1999;340:669–676. doi: 10.1056/NEJM199903043400901. [DOI] [PubMed] [Google Scholar]
- 23.Rizzo JA, Bogardus ST Jr, Leo-Summers L, Williams CS, Acampora D, Inouye SK. Multicomponent targeted intervention to prevent delirium in hospitalized older patients: what is the economic value? Med Care. 2001;39:740–752. doi: 10.1097/00005650-200107000-00010. [DOI] [PubMed] [Google Scholar]
- 24.Akunne A, Murthy L, Young J. Cost-effectiveness of multi-component interventions to prevent delirium in older people admitted to medical wards. Age Ageing. 2012;41:285–291. doi: 10.1093/ageing/afr147. [DOI] [PubMed] [Google Scholar]
- 25.Nasreddine ZS, Phillips NA, Bedirian V, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53:695–699. doi: 10.1111/j.1532-5415.2005.53221.x. [DOI] [PubMed] [Google Scholar]
- 26.Zoremba N, Coburn M. Acute confusional states in hospital. Dtsch Arztebl Int. 2019;116:101–106. doi: 10.3238/arztebl.2019.0101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Inouye SK, van Dyck CH, Alessi CA, Balkin S, Siegal AP, Horwitz RI. Clarifying confusion: the confusion assessment method. A new method for detection of delirium. Ann Intern Med. 1990;113:941–948. doi: 10.7326/0003-4819-113-12-941. [DOI] [PubMed] [Google Scholar]
- 28.Maschke M., et al. Delir und Verwirrtheitszustände inklusive Alkoholentzugsdelir, S1-Leitlinie, 2020, in: Deutsche Gesellschaft für Neurologie (Hrsg), Leitlinien für Diagnostik und Therapie in der Neurologie. dgn.org/leitlinien/ll-030-006-delir-und-verwirrtheitszustaende-inklusive-alkoholentzugsdelir/ (last accesed on 24 February. 2022) [Google Scholar]
- 29.Müller A, Markewitz A, Waydhas C, Weiß B, Spies C. Analgesie, Sedierung und Delirmanagement in der Intensivmedizin (DAS-Leitlinie) AWMF 2021. https://www.awmf.org/leitlinien/detail/ll/001-012.html (last accessed on December 1. 2021) [Google Scholar]
- 30.Green JR, Smith J, Teale E, et al. Use of the confusion assessment method in multicentre delirium trials: training and standardisation. BMC Geriatr. 2019;19 doi: 10.1186/s12877-019-1129-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9:179–186. [PubMed] [Google Scholar]
- 32.Hshieh TT, Yue J, Oh E, et al. Effectiveness of multicomponent nonpharmacological delirium interventions: a meta-analysis. JAMA Intern Med. 2015;175:512–520. doi: 10.1001/jamainternmed.2014.7779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Siddiqi N, Harrison JK, Clegg A, et al. Interventions for preventing delirium in hospitalised non-ICU patients. Cochrane Database Syst Rev. 2016;3 doi: 10.1002/14651858.CD005563.pub3. CD005563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Oh ES, Fong TG, Hshieh TT, Inouye SK. Delirium in older persons: advances in diagnosis and treatment. JAMA. 2017;318:1161–1174. doi: 10.1001/jama.2017.12067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Trogrlić Z, van der Jagt M, Bakker J, et al. A systematic review of implementation strategies for assessment, prevention, and management of ICU delirium and their effect on clinical outcomes. Crit Care. 2015;19 doi: 10.1186/s13054-015-0886-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Inouye SK, Bogardus ST Jr, Baker DI, Leo-Summers L, Cooney LM Jr. The hospital elder life program: a model of care to prevent cognitive and functional decline in older hospitalized patients. Hospital Elder Life Program. J Am Geriatr Soc. 2000;48:1697–1706. doi: 10.1111/j.1532-5415.2000.tb03885.x. [DOI] [PubMed] [Google Scholar]
- 37.Kratz T, Heinrich M, Schlauss E, Diefenbacher A. Preventing postoperative delirium. Dtsch Arztebl Int. 2015;112:289–296. doi: 10.3238/arztebl.2015.0289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bugiani O. Why is delirium more frequent in the elderly? Neurol Sci. 2021;42:3491–3503. doi: 10.1007/s10072-021-05339-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods
Sample size calculation and registration
Before the planned start of the clinical trial, the Institute of Biometry and Clinical Research, University of Münster, carried out a power analysis. A statistical sample size calculation was performed based on previous publications in the anglophone countries with the same focus and similar study design. The power analysis/sample size calculation was carried out by means of a simulation study related to the primary research question of the incidence of delirium, which involved using a two-tailed Fisher’s exact test for 2 × 2 contingency tables. Based on information in the literature, the underlying assumption was that there was a delirium incidence of 60.0% in the control group and 38.4% in the intervention group, and that both groups consisted of patients with cognitive deficits in an acute care hospital. The calculation was limited by the highly heterogeneous and variable reporting of delirium rates in the recent literature. The significance level was set at α = 5%. This initially resulted in a sample size calculation of 190. Six months after the study began, it became apparent that because the patient population was more heterogeneous (surgical and medical patients), there were fewer delirium events to record and analyze than assumed in the literature. The sample size was therefore recalculated by the Institute of Biometry and Clinical Research, University of Münster. The result was an increase of the total study population to >600 patients.
Randomization
Randomization was by computer using a randomization list stratified by age and sex. Randomized allocation to the two treatment groups (“standard” versus “innovative”) was by stratified block randomization. The possible block lengths were 4/6/8. There were four strata:
Stratum A: male, over 50 years old;
Stratum B: female, over 50 years old;
Stratum C: male, up to 50 years old;
Stratum D: female, up to 50 years old.
Statistical analysis
Bonferroni adjustment was carried out for the two primary analyses (significant result at P ≤= 2.5%). All other analyses should be interpreted as explorative or descriptive. Pearson’s χ2 test was used to compare the incidence of delirium, mortality risk, sex, and preexisting dementia between the intervention group and the control group, and to compare mortality risk between patients with delirium in the intervention group and in the control group.
The two-tailed Mann-Whitney U test was used to compare delirium duration, Montreal Cognitive Assessment (MoCA) scores on admission and discharge, and 12 months after discharge, between the intervention and control groups and, in the case of MoCA scores, also between patients with delirium in the intervention and control groups. The Kruskal-Wallis test was used to analyze differences in deficits relevant to activities of daily living (median Instrumental Activities of Daily Living [I-ADL]) among the groups.
To analyze the cognitive outcome at the time of discharge in the intervention group compared with the control group, the statistical test chosen was the hierarchical linear model. The following variables were included: treatment: intervention (= 1) or control (= 0); delirium: present (= 1) or absent (= 0); timepoint: admission (= 1), discharge (= 2); hospital stay in days; and interaction term: intervention: (= 1) * timepoint: admission (= 1), discharge (= 2).
For this analysis, multiple imputation was carried out in order to use the entire patient data base (“intention to treat” [ITT] analysis) (N = 772).
Analysis of cognitive outcome at discharge
The estimated average expected MoCA value at discharge can be calculated using the following formula:
MoCA = 20.3 -- 1.3 * intervention -- 5.3 * delirium -- 0.2 * timepoint + 0.02 * days in hospital + 1.3 * timepoint * intervention
The result shows that the MoCA value rose over time in patients in the intervention group, whereas in patients in the control group it fell (P < 0.001), so long as the patient had not developed delirium. Delirium led to a drop in MoCA value by 5.3 points in both groups (P < 0.001) (etable 1).
Analysis of cognitive outcome 12 months after discharge
To analyze cognitive outcome 12 months after discharge in patients with delirium in the intervention group compared with patients with delirium in the control group, the statistical test chosen was the hierarchical linear model. The following variables were included: treatment: intervention (= 1) or control (= 0), timepoint (admission = 1, discharge = 2, 12 months after discharge = 3), delirium duration in days, hospital stay in days, and interaction term: intervention (= 1) * timepoint (admission = 1, discharge = 2, 12 months after discharge = 3).
For analysis of cognitive deficits 12 months after discharge, “completer” analysis was performed (N = 45) because the structure of the missing data meant that data were “not missing at random,” so imputation could not be carried out.
The estimated average expected MoCA value 12 months after discharge can be calculated using the following formula:
MoCA = 19.6 -- 3.4 * intervention -- 0.8 * timepoint -- 0.3 * delirium duration in days -- 0.01 * days in hospital + 2.2 * timepoint * intervention
While MoCA value rose in patients in the intervention group over a timeframe of approximately 12 months, in patients in the control group it fell (P = 0.01). Thus, after 12 months patients with delirium in the intervention group had improved their cognitive deficits compared to admission.
The reason that patients with delirium in the intervention group show better MoCA scores after 12 months is that the first MoCA was carried out after nonelective hospital admission, so the initial MoCA score often overestimates cognitive deficits and gives worse results than a MoCA performed after time spent in familiar surroundings and without acute illness (etable 2).
+ 20.3 (intercept)
- 1.3 (for patients in the intervention group)
- 5.3 (if delirium occurred)
- 0.2 per timepoint
+ 0.02 per days in hospital
+ 1.3 per timepoint (for patients in the intervention group)
+ 19.6 (intercept)
- 3.4 (for patients in the intervention group)
- 0.8 per timepoint
- 0.3 per day of delirium duration
- 0.01 per day in hospital
+ 2.2 per timepoint (for patients in the intervention group)