Prevalence of Delirium in a Population of Elderly Outpatients with Dementia: A Retrospective Study

Abstract

Background:

Delirium is a multifactorial geriatric syndrome and often occurs in patients with cognitive impairment. It also remains under-recognized, specifically in elderly outpatients, because signs of delirium might overlap with symptoms of dementia.

Objective:

The aim of the present study is to retrospectively apply the chart-based delirium instrument on a cohort of elderly outpatients with dementia, to assess prevalence and features of delirium in this population.

Methods:

We randomly selected 650 medical records of outpatients referred to the “Neurogenetic Regional Centre” (CRN) of Lamezia Terme. Each evaluation included demographics, medical history, drugs, type and severity of dementia, and cognitive and functional status. Delirium was identified by the application of the chart-based delirium instrument.

Results:

The prevalence of delirium was 13.3%. The study population was divided, according to the presence of delirium, into two subgroups. Compared to the no delirium group, the delirium group was significantly older and had greater cognitive impairment with lower MMSE scores both at baseline and at the end of the follow up. They also had a significant lower score on the ADL and IADL. In this group, a higher intake of antihypertensive and antipsychotic drugs, together with a lower intake of cholinesterase inhibitors and memantine, was observed.

Conclusions:

In this study, the chart-based delirium instrument was applied to an outpatient population affected by dementia and followed for a long time. Our data confirm the importance that age and frailty play on the genesis of delirium and suggest attention should be paid to the pharmacological treatment of these patients.

Keywords

Delirium dementia prevalence retrospective studies

INTRODUCTION

Delirium is a transient, acute, and global disorder of cognition and consciousness with a complex multifactorial etiology. It is often associated with a change in general clinical condition and some authors have proposed to consider delirium as an additional vital sign [1]. It affects one in eight hospitalized patients and is associated with multiple adverse outcomes including increased length of stay and mortality, new institutionalization, and considerable distress for patients and families [2].

Definition of delirium is still not univocal [3], it often occurs in geriatric population with cognitive impairment, and in this condition, it is defined “Delirium superimposed on Dementia” (DSD). Its prevalence ranges from 22 to 89% in the different settings [4]. The Diagnostic and Statistical Manual of Mental Disorders fifth edition (DSM-5) states that delirium should not be explained by a preexisting dementia and it should always be excluded to allow the diagnosis of dementia [5]. This explains the difficulty to recognize a delirium superimposed on dementia, since each condition excludesthe other.

At the same time, it is important to consider that in many cases this condition remains under-recognized and under-reported, because signs of delirium might overlap with symptoms of dementia, or be mistaken for the fluctuations of cognitive function [6]. When delirium is super-imposed on dementia, it can worsen the clinical course of the disease and the trajectory of cognitive impairment [7]; therefore it is imperative to prevent and recognize thiscondition.

Diagnosis of delirium is essentially at the clinical level; multiple tools have been developed to detect it and most of them are based on observation of the patients. Therefore, a chart-based method for a retrospective detection of delirium has been validated [8] and has proved to be very useful to identify the appearance of delirium in patients with dementia, especially in those regularly followed by clinical and cognitive assessment, demonstrating an association with a faster cognitive decline [7]. In some studies, this method was applied to abstract delirium from medical records of hospitalized patients, focusing on daily nursing or physician notes, during the limited duration of hospitalization, with the aim to maximize sensitivity and not to miss potential delirium diagnoses [9].

We retrospectively applied the chart-based delirium instrument with the aim to identify the appearance of delirium in a cohort of outpatients with dementia during the course of the disease, and to assess prevalence and features of delirium in this population.

MATERIALS AND METHODS

Setting and patients

The study population consisted of Caucasian outpatients referred to the “Neurogenetic Regional Centre” (CRN) at Lamezia Terme General Hospital, in a period between 2002–2014 years. The “Neurogenetic Regional Centre” evaluates approximately 6,000 patients each year, and it is a regional referral center that has carried out both assistance and research activities in the field of dementia and neurodegenerative diseases since 1997.

We included in this study 650 clinical records of patients with different types of dementia (Alzheimer’s disease, frontotemporal dementia, dementia with Lewy bodies, Parkinson’s disease dementia, vascular dementia, and mixed dementia) diagnosed by neurologists with specific expertise in dementia. Every patient had a minimum of three observations at about 3–6 month intervals and were visited both at the CRN and, if necessary, at their home. The median of the follow-up was 3 years (range 1–16 years). The random selection of medical records was carried out using the RAND () function in Excel from the database.

Clinical evaluation

Every patient had a baseline evaluation including: age, sex, education, family history of dementia, type and severity of dementia, presence of other diseases (hypertension, diabetes, thyroid disorders, cardiovascular diseases, hypercholesterolemia, and history of cancer), and drug therapy.

Dementia severity was assessed across every evaluation, using the Clinical Dementia Rating Scale (CDR) [10], of which the score ranges from 0 (no cognitive decline), 0.5 (mild cognitive impairment), 1 (mild dementia), 2 (moderate dementia), 3 (severe dementia), 4 (very severe dementia), and 5 (terminal dementia). This study did not involve patients with a CDR score of 0.5. Cognitive tests were conducted at baseline and approximately every 3–6 months using the Mini-Mental State Examination (MMSE) [11]. Furthermore, ADL (Activities of Daily Living) and IADL (Instrumental Activities of Daily Living) were performed to assess the impact of dementia on patients’ autonomy [12].

Identification of delirium

Delirium was identified through the application of the chart-based delirium instrument [8] that demonstrated a sensitivity, specificity, likelihood ratio, and overall agreement for delirium diagnoses of 74%, 83%, 4.4% and 82%, respectively, when compared with the Confusion Assessment Method (CAM) [13]. In this study, the method was adapted to our ambulatory setting and the clinical records were analyzed through the review of the entire clinical history, focusing on neurologists notes and on the collected clinical documentation. The most important aspect of this work was to identify the baseline mental status of the patient. It was performed through the observation of clinical evaluations, reading the information provided by caregivers or physicians and noted in medical records. Then, we focalized on the description of any sudden change in mental status (occurred at home or during hospitalization), duration of these episodes, type and possible causes of these “changes”, and reversibility or improvement of the acute confusion. In particular, symptoms of delirium were abstracted focusing on “key words” in the medical record, such as “change in mental status or in behavior”, “acute confusion”, “disoriented”, “agitated/drowsy”, and “unresponsive”. Then each of these symptoms was related to DSM-5 [5] criteria for the diagnosis (Table 1).

Table 1

Comparison among DSM-5 criteria for delirium and corresponding clinical notes extracted from medical records

DSM-5	Clinical Picture Described in the Medical Record
A. Disturbance in attention (i.e., reduced ability to direct, focus, sustain, and shift attention) and awareness (reduced orientation to the environment).	Research of key words such as “agitated”, “lethargic”, “drowsiness”, “unresponsive”, “acute confusion”, “disoriented”.
B. The disturbance develops over a short period of time (usually hours to a few days), represents an acute change from baseline attention and awareness, and tends to fluctuate in severity during the course of a day.	Evidence, derived from the clinical history, of a sudden change in cognitive/behavioral status from the previous assessments (obtained through MMSE scores, information provided by the caregiver) and resolution of the framework with a substantial return to premorbid state.
C. An additional disturbance in cognition (memory deficit, disorientation, language, visuospatial ability, or perception).	Description about appearance of visual/auditory hallucinations, disorientation, sudden worsening of other cognitive performances (memory, language).
D. The disturbances in Criteria A and C are not better explained by a pre-existing, established or evolving neurocognitive disorder and do not occur in the context of a severely reduced level of arousal such as coma.	Evidence, derived from the clinical history, of a sudden change of the functional status from the clinical course of the underlying dementia (obtained through the analysis of previous assessments noted in clinical records).
E. There is evidence from the history, physical examination or laboratory findings that the disturbance is a direct physiological consequence of another medical condition, substance intoxication or withdrawal (i.e., due to a drug of abuse or to a medication), or exposure to a toxin, or is due to multiple etiologies.	Evaluation of the possible organic condition responsible for the acute framework and obtained through the analysis of the entire documentation collected in the medical record (discharge, blood tests, modifications in drug therapy) or according to the report of family members.

To identify the possible trigger of delirium, the closest clinical documentation referred to metabolic and physiological parameters, blood tests, hospital medical records/discharge sheet, or changes in drug therapy was assessed.

Finally, the motor subtype of delirium was determined according to the presence of increased psychomotor activity and agitation (hyperactive form) or drowsiness, lethargy, and psychomotor retardation (hypoactive form). The third motoric subtype was defined a mixed form [14].

Statistical analysis

The number of clinical records to be randomly extracted by archives of our Institution was derived by assuming a prevalence rate of delirium of 13% with a standard error of 3%. This prevalence considered the results of previous retrospective studies [4], but also the different care setting and the delirium measure instrument of our study. According to these criteria we calculated that a sample size of 523 clinical records produced a 95% confidence interval (CI) equal to the sample proportion plus or minus 0.03 when the estimated proportion is 0.13. The sample size also included a 10% attrition rate due to potentially incomplete clinical folders. Data were reported as mean±SD or percentage as appropriated. Not normal data were reported as median and interquartile range. They were processed by unpaired t-test, χ², and Mann-Whitney test as appropriated. Logistic regression was performed to evaluate the relationship between the stage of dementia and delirium. In the model we also introduced as covariates age, years of education and gender. Differences were assumed to be significant at p < 0.05.

All comparisons were performed using the statistical package SPSS 21.0 for Windows (SPSS Inc., Chicago, IL, USA).

RESULTS

Population of the study (n = 650) was divided according to the presence of delirium into two subgroups: delirium group (n = 87) and no delirium group (n = 563). The characteristics of the study population are shown in Table 2. The prevalence of delirium was 13.3%.

Table 2

Characteristics of the study population according to the presence of delirium

Variables	All	Delirium	No Delirium	p
	(No. 650)	(No. 87)	(No. 563)
Sex, male (%)^**	246 (38)	38 (44)	208 (37)	0.277
Age, y ^*	79±9	81±9	78±9	0.013
Educations, y^***	5 (3–8)	5 (2–5)	5 (3–8)	0.341
Duration of dementia, y^***	7 (5–10)	7 (5–10)	7 (5–9)	0.269
MMSE baseline ^***	19 (14–23)	18 (12–21)	19 (14–23)	0.021
MMSE end of follow-up ^***	12 (3–18)	8 (1–15)	13 (16–18)	< 0.001
ADL baseline ^***	6(4–6)	5 (3–6)	6 (5-6)	< 0.001
ADL end of the follow-up ^***	3(1–5)	1 (0–4)	3 (1–5)	0.009
IADL baseline ^***	4 (2–6)	3 (1–5)	4 (2–6)	0.011
IADL end of the follow-up ^***	1 (0–2)	0 (0-1)	1 (0–3)	0.003
Hypertension, % ^**	430 (66)	61 (70)	369 (66)	0.473
Dyslipidemia, % ^**	266 (41)	29 (33)	237 (42)	0.153
Diabetes, % ^**	133 (20)	16 (18)	117 (21)	0.710
Heart diseases, % ^**	171 (26)	27 (31)	144 (26)	0.345
Thyroid disease, % ^**	133 (20)	21 (24)	112 (20)	0.441
Cancer, % ^**	54 (8)	13 (15)	41 (7)	0.028
Antiplatelet agents, % ^**	426 (66)	64 (74)	362 (64)	0.116
Benzodiazepines, % ^**	264 (41)	33 (74)	232 (41)	0.644
Antipsychotic, % ^**	266 (41)	52 (53)	214 (38)	< 0.001
Start before delirium, %^**		23 (26)	214 (38)	0.049
Antidepressants, % ^**	388 (60)	46 (38)	342 (61)	0.202
Anti-cholinesterase, % ^**	378 (58)	42 (48)	336 (60)	0.059
Memantine, % ^**	391 (60)	42 (48)	349 (62)	0.021
Antihypertensive, % ^**	456 (70)	69 (79)	387 (69)	< 0.001
No RAS, %^**	332 (51)	56 (64)	276 (49)	0.011
RAS, % ^**	381 (59)	57 (67)	324 (58)	0.198
Dementia
Alzheimer, % ^**	349 (54)	40 (46)	309 (55)	0.151
Frontotemporal, % ^**	10 (11)	10 (12)	60 (11)	0.961
Lewy, % ^**	14 (0.2)	7 (8)	22 (4)	0.144
Mixed, % ^**	175 (27)	24 (28)	151 (27)	0.984
Parkinson, % ^**	14 (2)	3 (3)	11 (2)	0.619
Vascular, % ^**	13 (2)	3 (3)	10 (2)	0.532

^*Data expressed as mean±SD. ^**Data reported as percentage. ^***Data reported as median and interquartile range. MMSE, Mini-Mental State Examination; ADL, Activities of Daily Living; IADL, Instrumental Activities of Daily Living; RAS, renin-angiotensin system.

The two subgroups did not significantly differ in gender, years of formal education, disease duration, type of dementia, cardiovascular risk factors (hypertension, hypercholesterolemia, diabetes), heart disease, thyroid diseases, drug therapy with antiplatelet agents, benzodiazepines, antidepressants, and cholinesterase inhibitors.

Delirium group was significantly older and showed a greater cognitive impairment documented by lower scores at MMSE both at baseline and at the end of the follow up. It also had a significant lower score at ADL and IADL. Patients with delirium were more frequently treated with antipsychotic drugs (52.9% versus 38.0%). A significant increased administration of memantine was also observed in the group that did not develop delirium and the same trend was observed for cholinesterase inhibitors, even if they did not reach the level of significance (p = 0.059). Regarding the antihypertensive therapy, patients in delirium group were more frequently treated, with a neutral role of modulators of the renin angiotensin system.

In a subgroup of 221 patients (34% of the sample) of which 32 in the delirium group (36.8%) and 189 in the no delirium group (33.6%), the genotype of apolipoprotein E was available. No significant differences according to the presence of ɛ4 allele were found in the two groups.

Table 3 shows the distribution of delirium episodes according to the severity of dementia. After adjustment for age, education, and gender, the moderate stage of the disease (CDR 2) increased nearly twice the risk to develop delirium, and the advanced stage of dementia (CDR 3) significantly increased the risk of 3.3 fold.

Table 3

Onset of delirium according to the stage of dementia adjusted for gender, age and education

	Odds Ratio	95% CI	p
Stage of dementia
CDR 1 (mild)
CDR 2 (moderate) versus CDR1	1.730	0.818–3.660	0.152
CDR 3-4 (advanced stages) versus CDR1	3.303	1.654–6.592	0.001

CI, confidence interval; CDR, Clinical Dementia Rating.

In 7 patients (8%), delirium preceded the diagnosis of dementia. These cases were all hyperkinetic forms. Also in the mild (CDR 1) and moderate (CDR 2) stages of the disease, the hyperkinetic form was predominant (90.3% and 95.4%, respectively). The hypoactive form was more common in the advanced stages of disease (48.2%) (Table 4).

Table 4

Stage of dementia and motor subtype of delirium

	Motor subtype
Stage	Hyperkinetic	Hypokinetic	Mixed
Before diagnosis of dementia	7 (100)	–	–
Mild (CDR 1), %	28 (90)		3 (10)
Moderate (CDR 2), %	21 (95)		1 (5)
Severe (CDR 3-4), %	9 (33)	13 (48)	5 (19)

CDR, Clinical Dementia Rating.

Table 5 shows the likely underlying trigger factors of delirium in the study population.

Table 5

Possible underlying illness

Causes
Hospitalization, %	58 (67)
Orthopedic surgery, %	18 (31)
Other surgery, %	5 (9)
Infections, %	11 (19)
Cardiovascular diseases, %	10 (17)
Anemia, %	4 (7)
Other causes, %	10 (17)
Drugs, %	13 (15)
Metabolic/endocrine diseases, %	7 (8)
Other causes, %	6 (7)
Unknown, %	3 (3)

DISCUSSION

In this study, the chart-based delirium instrument was applied to an outpatient population affected by different types of dementia and regularly followed at the CRN. To the best of our knowledge, few studies have described and analyzed the occurrence and the features of delirium in this setting. Recently, a study has investigated characteristics, symptom profiles, and underlying somatic disorders of delirium in elderly patients referred for cognitive evaluation to a center for old age psychiatry, demonstrating a prevalence of delirium of 16% [15].

The similar prevalence of 13.3% attested in our population could suggest that delirium is still underestimated and not recognized in the clinical practice and above all when it is superimposed on dementia, due to the overlap of symptoms and to the difficulty in performing a differential diagnosis among the two syndromes. Delirium plays a pivotal pathogenic role in the natural history of a patient with dementia and many studies have demonstrated that it is an independent predictor of adverse outcomes and faster cognitive decline [16]. Accordingly, it became very informative to identify delirium in different care settings, even when delirium measures are retrospectively derived, without neglecting the possible onset of delirium in outpatients with dementia. In this perspective, this work, conducted in ambulatory setting and especially in home-living patients, has an important value because it provides further information about a condition that is normally poorly studied. This is helpful both for further knowledge derived from this analytical approach and for training general practitioners and caregivers to early recognize the occurrence of delirium.

The method used in our study resumes that already validated by Inouye et al. [8], giving the possibility to detect and describe the onset of delirium associated with different pathogenic noxae, even outside the medical setting and to follow the subsequent clinical course of the disease. In fact, the possibility to verify the reversibility of the clinical picture after removing the precipitating factors or treating the acute underlying illness is very helpful. In our population, the risk of developing delirium was related to older age and to greater cognitive and functional impairments. These findings are in agreement with previous data, showing that for each point lost to the MMSE, the risk of developing delirium increases 5% [17]. Interestingly, patients in treatment with memantine and cholinesterase inhibitors showed a lower incidence of delirium and this could corroborate the neurotransmitter hypotheses of delirium. Glutamate excitotoxicity is one of the underlying mechanism of behavioral and psychological symptoms of dementia [18], and the treatment with memantine is able to ameliorate these symptoms, beyond the modest effect on cognition. At the same time, delirium and dementia seem to share the same neurotransmitter imbalances and they may represent different points along a continuum, with delirium reflecting underlying brain vulnerability [19].

In addition, drug therapy should always be evaluated when an acute confusion occurs, especially in elderly fragile patients. Our data show a prevalence of drug-induced delirium in 15% of cases, in agreement with literature evidence in which drugs (especially those with anticholinergic properties) may be precipitating factors in 12–39% of cases [20]. The use of antihypertensive drugs (except for modulators of the renin-angiotensin system) was more frequent in the delirium group, in accordance with previous data that reported a significant association between delirium and beta-blockers, dihydropyridine calcium channel-blockers, and diuretics [21, 22]. According to the motor subtype, it is interesting that the hyperkinetic form was predominant in patients in a mild and moderate stage of the disease, while the hypoactive subtype characterized patients with a severe cognitive impairment in agreement with previous data [23].

Finally, in our population, hospitalization was associated with delirium in 66.8% of cases, with a higher prevalence in patients hospitalized for orthopedic surgery (especially hip fracture), followed by those hospitalized for infectious and cardiovascular diseases. This supports the hypothesis that systemic inflammation might trigger the onset of delirium by neuro-inflammation mechanisms [24, 25].

We are aware that this work has several limits. Delirium is an acute, transient, and reversible disorder, which makes it difficult to retrospectively recognize it within the cognitive-behavioral framework of a dementia. This could result in a limitation of the retrospective methodology and in an underestimation of the diagnosis. Furthermore, the chart-based delirium instrument has not been validated in this setting and the chart abstractor was only one examiner (DA).

However, the accuracy of the collected clinical histories allowed us to apply the chart-based delirium instrument in an outpatient setting. In particular, we focused on the baseline evaluations, on the accurate description of the acute episode, and on the reversibility of confusion state after removing the underlying condition. This represents the key point that allowed us to differentiate the behavioral and psychological symptoms of dementia from delirium, thus avoiding the mistake of viewing it as a simple fluctuation of behavioral symptoms.

Finally, this methodology is not intended as a diagnostic instrument in clinical practice, but as an epidemiological measure whose purpose is to implement the detection of delirium on a large scale to improve the quality of care in patients with dementia through projects and preventive strategies.

Footnotes

ACKNOWLEDGMENTS

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors’ disclosures available online ().

References

Flaherty

, Rudolph

, Shay

, Kamholz

, Boockvar

, Shaughnessy

, Shapiro

, Stein

, Weir

, Edes

(2007) Deliriumis a serious and under-recognized problem: Why assessment ofmental status should be the sixth vital sign. J Am Med DirAssoc 8, 273–275.

Inouye

, Westendorp

, Saczynski

, Kimchi

, Cleinman

(2014) Delirium in elderly people. Lancet 383, 911–922.

Morandi

, Pandharipande

, Trabucchi

, Rozzini

, Mistraletti

, Trompeo

, Gregoretti

, Gattinoni

, Ranieri

, Brochard

, Annane

, Putensen

, Guenther

, Fuentes

, Tobar

, Anzueto

, Esteban

, Skrobik

, Salluh

, Soares

, Granja

, Stubhaug

, de Rooij

, Ely

(2008) . Understanding international differences in terminology for delirium and other types of acute brain dysfunction in critically ill patients. Intensive Care Med 34, 1907–1915.

Fick

, Agostini

, Inouye

(2002) Delirium superimposed on dementia: A systematic review. J Am Geriatr Soc 50, 1723–1732.

American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders, 5th ed. American Psychiatric Association, Arlington, VA.

Fick

, Foreman

(2000) Consequences of not recognizing delirium superimposed on dementia in hospitalized elderly individuals. J Gerontol Nurs 26, 30–40.

Gross

, Jones

, Habtemariam

, Fong

, Tommet

, Quach

, Schmitt

, Yap

, Inouye

(2012) Delirium and long-term cognitive trajectory among persons with dementia. Arch Intern Med 172, 1324–1331.

Inouye

, Leo-Summers

, Zhang

, Bogardus

Jr , Leslie

, Agostini

(2005) A chart-based method for identification ofdelirium: Validation compared with interviewer ratings using theconfusion assessment method. J Am Geriatr Soc 53, 312–318.

Kuhn

, Du

, McGrath

, Coveney

, O’Regan

, Richardson

, Teodorczuk

, Allan

, Wilson

, Inouye

, MacLullich

, Meagher

, Brayne

, Timmons

, Davis

(2014) Validation of a consensus method for identifying delirium from hospital records. PLoS One 9, e111823.

10.

Morris

(1993) The Clinical Dementia Rating (CDR):Current version and scoring rules. Neurology 43, 2412–2414.

11.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

12.

Katz

, Ford

, Moskowitz

, Jackson

, Jaffe

(1963) Studies of illness in the aged: The index of ADL, a standardized measure of biological and psychosocial function. JAMA 185, 914–919.

13.

Inouye

, Van Dyck

, Alessi

, Balkin

, Siegal

, Horwitz

(1990) Clarifying confusion: The Confusion Assessment Method. A new method for detection of delirium. Ann Intern Med 113, 941–948.

14.

Lipowski

(1990) Delirium: Acute confusional state, Oxford University Press, Oxford.

15.

Stroomer-van Wijk

, Jonker

, Kok

, van der Mast

, Luijendijk

(2016) Detecting delirium in elderly outpatients with cognitive impairment. International. Psychogeriatrics 28, 1303–1311.

16.

Fong

, Jones

, Shi

, Marcantonio

, Yap

, Rudolph

, Yang

, Kiely

, Inouye

(2009) Delirium accelerates cognitive decline in Alzheimer disease. Neurology 72, 1570–1575.

17.

Davis

, Skelly

, Murray

, Hennessy

, Bowen

, Norton

, Brayne

, Rahkonen

, Sulkava

, Sanderson

, Rawlins

, Bannerman

, MacLullich

, Cunningham

(2015) Worseningcognitive impairment and neurodegenerative pathology progressivelyincrease risk for delirium. Am J Geriatr Psychiatry 23, 403–415.

18.

Huang

, Lin

, Lanea

, Tsai

(2012) NMDA Neurotransmission dysfunction in behavioural and psychological symptoms of Alzheimer’s disease. Curr Neuropharmacol 10, 272–285.

19.

Hshieh

, Fong

, Marcantonio

, Inouye

(2008) Cholinergic deficiency hypothesis in delirium: A synthesis of current evidence. J Gerontol A Biol Sci Med Sci 63, 764–772.

20.

Alagiakrishnan

, Wiens

(2004) An approach to drug induced delirium in the elderly. Postgrad Med J 80, 388–393.

21.

Van der Mast

, Roest

(1996) Delirium after cardiac surgery: Acritical review. Psychosom Res 41, 13–30.

22.

Clegg

, Young

(2011) Which medications to avoid in people at risk of delirium: A systematic review. Age Ageing 40, 23–29.

23.

Koponen

, Riekkinen

(1993) A prospective study of delirium in elderly patients admitted to a psychiatric hospital. Psychol Med 23, 103–109.

24.

Adamis

, Lunn

, Martin

, Treloar

, Gregson

, Hamilton

, Macdonald

(2009) Cytokines and IGF-1 in delirious and non delirious acutely ill older medical inpatients. Age Ageing 38, 326–332.

25.

Maclullich

, Ferguson

, Miller

, de Rooij

, Cunningham

(2008) Unravelling the pathophysiology of delirium: A focus on therole of aberrant stress responses. J Psychosom Res 65, 229–238.