RESEARCH ARTICLE

Neuropsychiatric symptoms and Quality of Life in patients

with very mild and mild Alzheimers disease
Kristiina Karttunen1,2,3, Pertti Karppi4, Asta Hiltunen5, Matti Vanhanen6, Tarja Valimaki7, Janne Martikainen8,
Hannu Valtonen 9, Juhani Sivenius 10, Hilkka Soininen 1,2, Sirpa Hartikainen 8,11,12,
Jaana Suhonen 13 and Tuula Pirttila 1 for the ALSOVA study group
1
Department of Neurology, Kuopio University Hospital, Kuopio, Finland
2
School of Medicine, Institute of Clinical Medicine, Neurology, University of Eastern Finland, Kuopio Campus, Finland
3
Institute of Public Health and Clinical Nutrition, Department of Geriatrics, University of Eastern Finland, Kuopio Campus, Finland
4
Mikkeli Central Hospital, Mikkeli, Finland
5
Department of Neurology, North Karelia Central Hospital, Joensuu, Finland
6
Niuvanniemi Hospital, Kuopio, Finland
7
Department of Nursing Sciences, University of Eastern Finland, Kuopio Campus, Finland
8
School of Pharmacy, University of Eastern Finland, Kuopio Campus, Finland
9
Department of Health and Social Management, University of Eastern Finland, Kuopio Campus, Finland
10
Brain Research and Rehabilitation Center Neuron, Kuopio, Finland
11
Kuopio Research Centre of Geriatric Care, University of Eastern Finland, Kuopio Campus, Finland
12
Leppavirta Health Center, Leppavirta, Finland
13
Jokilaakso Hospital, Jamsa, Finland
Correspondence to: K. Karttunen, MD, E-mail: [email protected]

Background: Neuropsychiatric symptoms (NPS) are common manifestations of Alzheimer s disease

(AD).
Objective: To examine the prevalence and significance of NPS in very mild and mild AD patients with
emphasis on their influence on the well-being of the patients and their caregivers.
Methods: The participants were 240 patient-caregiver dyads who participated in a prospective, controlled
rehabilitation study (ALSOVA). Three Quality of Life (QoL) instruments were used; generic 15D,
disease-specific QoL-AD and Visual Analog Scale (VAS). The disease-specific QoL-AD was both self-
rated and caregiver rated. Other scales used were Mini-Mental State Examination (MMSE), Clinical
Dementia Rating (CDR), ADCS-ADL, Neuropsychiatric Inventory (NPI) and Beck Depression Inven-
tory (BDI).
Results: NPS were present in 76.5% of patients with very mild AD (CDR 0.5) and in 84.9% of patients
with mild to moderate AD (CDR 1). The most frequent symptoms were apathy, depression, irritability,
and agitation. The strongest predictor of self-reported QoL-AD scores was depressive symptoms whereas
functional decline and presence of NPS predicted poor caregiver ratings of patients QoL. However,
caregiver depression also influenced significantly their ratings.
Conclusion: NPS are common even in the early stages of AD. NPS were significantly associated with
caregiver assessment of the patients QoL but not with patients self-assessed QoL. Depression
decreases QoL, but may remain unrecognized in AD patients, emphasizing the need for careful and
structured assessment of NPS before deciding on the appropriate treatment. Copyright # 2010 John
Wiley & Sons, Ltd.
Key words: Quality of Life; Alzheimers disease; neuropsychiatric symptoms; depression; caregiver
History: Received 15 November 2009; Accepted 16 April 2010; Published online 29 October 2010 in Wiley Online Library
(wileyonlinelibrary.com).
DOI: 10.1002/gps.2550

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
474 K. Karttunen et al.

Introduction Study subjects and methods

Alzheimers disease (AD) is a chronic progressive Participants. The participants were 240 patient-care-
neurodegenerative disease that compromises all giver dyads living in three municipalities in Finland
aspects of functioning, impairing both cognitive and who were participating in a prospective, controlled
functional abilities, behavioral and social skills as well rehabilitation study (ALSOVA) evaluating the effec-
as causing emotional disturbances. Therefore, it tiveness of early rehabilitation on QoL and risk of
influences the well-being of affected individuals in institutionalization in mild AD. The participants were
very profound ways. Neuropsychiatric symptoms recruited during the first year after the AD diagnosis
(NPS) are common manifestations of AD. These and were randomized into an intervention group or
symptoms are closely related to the underlying brain control group at the baseline visit. After the AD
pathology (Cummins, 2000). Their prevalence is diagnosis, both groups were followed up regularly and
believed to range from 50 to 90% in AD patients over the intervention group also participated in the early
the course of the disease (Mega et al., 1996; Lyketsos rehabilitation planned especially for them.
et al., 2000, 2002; Hwang et al., 2004). Though there is Inclusion criteria for the patients were 65 years of
a belief that the prevalence of NPS increases as the age, very mild (Clinical Dementia Rating, CDR 0.5) or
disease progresses, recent studies indicate that a high mild AD (CDR 1), informed consent, and a family
proportion of subjects suffering from amnestic mild caregiver. The definition of caregiver was an individual
cognitive impairment (MCI) decline, the possible preferably in daily contact with the patients. A family
prodromal phase of AD, display clinically meaningful caregiver could be a spouse, sibling, child, or some
NPS (Lyketsos et al., 2002; Feldman et al., 2004; Geda other relative. The participants were recruited from
et al., 2004; Peters et al., 2006). These symptoms seem April 2002 to September 2006.
to be as common in MCI as in patients with mild AD The demographics of the participants are given in
(Hwang et al., 2004) and their presence is thought to Table 1. All but 10 participants were receiving drug
predict conversion from MCI to AD (Peters et al., treatment with either an anticholinesterase or mem-
2006; Copeland et al., 2003; Palmer et al., 2007). antine. There were some protocol violations i.e.,
The main focus in the care of patients suffering from younger age than 65 years (n 15), moderate dementia
AD is to promote their well-being and maintain an (CDR 2, n 2) and recruitment later than 1 year after
optimal Quality of Life (QoL). QoL is a multi- diagnosis (n 20).
dimensional and complex concept including aspects of
physical, psychological and social functioning. NPS Measures. AD was diagnosed by a geriatrician or
have many adverse consequences and cause serious neurologist using the criteria devised by the National
distress both in patients and caregivers. They are Institute of Neurological and Communicative Dis-
related to more severe impairment in activities of daily orders and Stroke and AD and Related Disorders
living (Tekin et al., 2001; Chan et al., 2003; Feldman Association (NINCDS-ADRDA) (McKhann et al.,
et al., 2004; Peters et al., 2006), worse QoL (Hoe et al., 1984). All patients went thru diagnostic evaluation
2007; Missotten et al., 2008) and greater caregiver including brain imaging. The patients were then
burden (Kaufer et al., 1998). Most studies have shown invited to a baseline study visit which included a
that depressive symptoms are associated with self- patient and caregiver interview performed by the study
reported QoL measures (Logsdon et al., 1999; Shin nurse and a neuropsychological examination of the
et al., 2005; Snow et al., 2005; Jonsson et al., 2006; Hoe patient conducted by a neuropsychologist. Data
et al., 2007; Karlawish et al., 2008a). There appears to collected during the interview included socio-demo-
be an inverse relationship between behavioral and graphic details (age, sex, education, living arrange-
psychological symptoms and caregiver-rated QoL of ments, income, and activities) and general health
the patient (Ready et al., 2004; Shin et al., 2005; Snow (other diseases, use of the drugs). Cognition was
et al., 2005; Vogel et al., 2006; Jonsson et al., 2006; evaluated using the Mini-Mental State Examination
Banerjee et al., 2006; Thomas et al., 2006; Hoe et al., (MMSE) (Folstein et al., 1975). The severity of the
2007). disease was rated using the CDR (Hughes et al., 1982;
The aim of the study was to examine the prevalence Morris, 1993). The ADCS-ADL interview of the
and significance of NPS in patients with either very caregiver was used to evaluate activities in daily living
mild or mild AD, concentrating on their influence on (Galasko et al., 1997). The 12-item Neuropsychiatric
the well-being of the patients and caregivers in a Inventory (NPI) (caregiver interview) was used to
prospective AD rehalibitation study. assess behavioral and psychological problems during

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
Neuropsychiatric symptoms and Quality of Life in patients with very mild and mild Alzheimers disease 475

Table 1 Characteristics of 240 patients with very mild (CDR 0.5) or mild (CDR 1) AD. Two patients had moderate disease (CDR 2) and they were included
in the mild group

All CDR 0.5 CDR 1 or 2

N 240 81 159
Age (years) 75.1 (6.6) 73.9 (6.9) 75.7 (6.3)
Women (%) 51.3 45.7 54.1a
Full-time education (years)
Mean 7.5 7.3 7.6
Range 1.020.0 1.018.0 1.020.0
Duration of symptoms (months)
Until diagnosis 39.1 (25.2) 37.5 (21.5) 39.9 (27.0)
Until randomization 44.3 (26.0) 42.7 (22.8) 45.1 (27.4)
Living
with spouse (%) 70 71.6 69.2
alone (%) 30 28.4 30.8
Drug treatment
AD drugs (%) 95.8 93.8 96.8
Total number of other drugs 3.7 (2.6) 3.5 (3.0) 3.7 (2.4)
Antipsychotics (%) 4.2 2.5 5.1
Antidepressives (%) 10.9 9.9 11.4
General health
cardiovascular diseases (%) 59.3 66.0 63.3
diabetes (%) 12.9 12.3 13.2
MMSE 21.5 (3.4) 21.7 (3.4) 21.3 (3.4)
ADCS-ADL 65.9 (8.8) 67.1 (8.7) 65.3 (8.7)
NPI 9.0 (9.8) 8.3 (9.2) 9.3 (10.1)
BDI
total score 10.0 (6.7) 10.1 (6.5) 10.0 (6.8)
>12 points (%) 32.5 29.6 34.0b
Time from diagnosis to randomization (months) 5.1 (6.2) 5.3 (7.5) 5.1 (5.6)

Results are expressed as mean (SD).

a
p 0.218.
b
p 0.498.

the 1 month prior to the examination (Cummings between 1028. A 10-cm VAS scale was a one-item
et al., 1994) and additionally symptoms of depression assessment of QoL including a single question of
were assessed by using the 21 item Beck Depression general well-being with a rating from 0 (worst possible)
Inventory (BDI) rated by the patients (Beck et al., to 100 (best possible).
1961).
Three QoL instruments were used; generic 15D,
disease-specific QoL-AD and Visual Analog Scale Data analysis. All statistical analyses were performed
(VAS) for general well-being. The 15D is a generic, with SPSS software for Windows release 11.5.1.
multidimensional, standardized, self-administered Missing data for 15D items (10 cases) were replaced
instrument, which has both profile and single index by median value of the particular item. Simple means
score properties (Sintonen, 1995). The dimensions of and proportions were computed for continuous and
15D are mobility, vision, hearing, breathing, sleeping, categorical variables. Paired t-tests and x2 test were
eating, speech, elimination, usual activities, mental used to compare group differences. The correlations
function, discomfort and symptoms, depression, between different variables were calculated by Pear-
distress, vitality and sexual activity. Each dimension sons correlation test or by Spearmans correlation test.
has five grades of severity and the 15D score is The factor analysis of NPI was carried out with
calculated using the multiattribute utility method. The principal component axis factoring with promax rota-
15D values range from 1 (full health) to 0 (dead). tion. A cut-off point of eigenvalues over 1.0 was used.
The QoL-AD is a 13-item measure of QoL that is Sum variables were formed based on compounded
specifically developed for use in AD patients and their factors and these were later used as continuous varia-
caregivers (Logsdon et al., 1999). The maximum score bles. Age and sex adjusted linear regression analysis was
is 52 indicating high QoL and the minimum is 13. It used to analyze the predictors of caregivers assessment
can be reliably used by individuals with MMSE scores of the patients QoL.

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
476 K. Karttunen et al.

The variables showing possible effect ( p < 0.2) on Table 3 The loadings of NPI items into four factors
caregiver assessment of the patients QoL in the
Psychotic/ Affective Frontal Other
univariate linear analysis were also added to the linear agitation type
regression model. In the final analyses, the significance
p < 0.05 was used. Irritability 0.800
Agitation 0.695
Apathy 0.401
Delusions 0.336 0.235
Results Eating behavior 0.213
Depression 0.666
Anxiety 0.608
The mean age of the participants was 75.1 years Euphoria 0.933
(Table 1) and 51.3% were female. Most patients (70%) Disinhibition 0.504 0.202
Hallucinations 0.743
were married and living with a spouse and the Sleep disorders 0.310
remainder lived alone in their own home. The mean Aberrant motor 0.202 0.243
MMSE was 21.5 and 34% had very mild AD (CDR 0.5). behavior
NPS were common at baseline (Table 2). At least one
symptom was present in 76.5% of patients with very
mild AD (CDR 0.5) and in 84.9% of patients with mild There was no correlation between BDI scores and
to moderate AD. The caregivers reported three or more MMSE or ADCS-ADL scores.
NPS in 49.3% of patients with very mild AD and in Table 3 shows the factor analysis of NPI. This factor
47.3% with mild AD. The most frequent symptoms solution explained 39.4% of the total variance of NPI.
were apathy, depression, irritability, and agitation The first factor (psychotic/agitation) accounted for the
(Table 2/Figure). Spouse-caregivers reported sleep major part of the explained variance (21.6%). The
disturbances significantly more often than other four-factor model fitted to the data well as confirmed
caregivers (16.9% vs. 6.9%, p < 0.05). Delusions were by KMO (0.704) and Bartletts test (x2 513.376,
more common in women (28.5% vs. 16.2%, p 0.017) p < 0.001). The item content of the four compounded
and aberrant motor behavior was more common in factors was labeled as Psychotic/agitation (factor 1),
men (25.6% vs. 12.2%, p 0.008) but otherwise there Affective (factor 2), Frontal type behavior (factor 3),
were no sex related differences in the prevalence of the and Other (factor 4). Apathy loaded in all four factors
NPS. High scores for depressive symptoms (BDI over though most prominently in the psychotic/agitation
12 points) were found in 32.5% of patients. BDI scores factor. Sleeping disturbances and hallucinations loaded
were significantly higher in women than in men together in the fourth factor. Aberrant motor behavior
(10.9
6.9 vs. 9.0
6.4, p 0.031). The correlation and eating behavior did not load prominently into any
between patients self-rated BDI scores and caregivers of the factors.
assessed NPI depression score was statistically signifi- NPI scores correlated inversely with ADCS-ADL
cant though rather modest (r 0.141, p 0.029). scores ( p 0.344, p < 0.001) but not with MMSE
scores. All four factors correlated significantly with
ADCS-ADL scores (data not shown). Since previous
Table 2 Prevalence of neuropsychiatric symptoms studies have suggested that spouse caregivers and other
All CDR 0.5 CDR 1 or 2
caregivers may differ in their ratings (Novella et al.,
(n 240) % (n 81) % (n 159) % 2001), we included only 168 subjects with spouse
caregivers for the analyses of the relationship between
At least one 82.1 76.5 84.9
symptom
NPS and caregiver well-being. NPI total scores
Delusions 22.5 17.3 25.2 correlated significantly with caregiver GHQ scores
Hallucinations 15.4 16.0 15.1 (r 0.258, p < 0.01) and with depression (r 0.191,
Agitation 29.6 23.5 32.7
Depression 37.1 32.1 39.6
p 0.013), and were inversely associated with caregiver
Anxiety 25.8 28.4 24.5 QoL (VAS scores, r 0.269, p < 0.001). In particular,
Euphoria 5.8 4.9 6.3 psychotic, and mood factors were related to caregiver
Apathy 47.9 49.4 47.2
Disinhibition 14.6 18.5 12.6
GHQ scores (psychotic r 0.261, p 0.001, mood
Irritability 34.2 33.3 34.6 r 0.262, p 0.001), whereas frontal type behavior
Aberrant motor 18.8 16.0 20.1 and the fourth factor showed no significant correlation
behavior
Sleep 13.9 11.1 15.3
with GHQ.
Eating 25.8 22.2 27.7 The patients self-rated QoL did not differ signifi-
cantly between patients with very mild AD and those

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
Neuropsychiatric symptoms and Quality of Life in patients with very mild and mild Alzheimers disease 477

Table 4 Quality of life of the patients

All CDR 0,5 CDR 1 or 2

a
15D scores (max 1) (n 240) 0.8577 (0.0828) 0.8645 (0.0770) 0.8543 (0.0857)
QoL-AD (self-rated, n 228)
total (max 52) 33.5 (6.1) 32.6 (6.5) 33.9 (5.8)
without marriage (max 48) 30.4 (5.4) 30.0 (5.8) 30.6 (5.2)
QoL-AD (spouse-rated, n 168)
total 30.1 (5.2) 30.4 (5.3) 30.0 (5.3)
without marriage 27.3 (4.9) 27.7 (4.9) 27.1 (4.9)
VAS score (max 100) (n 240) 79.05 (16.6) 80.2 (15.6) 78.4 (17.1)

a
10 subjects had a missing value in one dimension of 15D. Missing values were replaced by the median value of the whole group for that particular dimension.
Results are expressed as mean (SD).

with mild to moderate AD (Table 4). Generic 15D There was a statistically significant correlation
scores correlated very moderately with disease-specific between self-rated and spouse caregiver rated patients
QoL-AD scores (0.139, p 0.039) but not with VAS QoL-AD scores (r 0.223, p < 0.01). Spouses rated all
scores. The 15D scores correlated inversely with age dimensions of patients QoL lower than the patients
(r 0.156, p 0.016) and general health (the themselves (Table 4). Among all patients, linear
number of drugs) (r 0.129, p < 0.05) but not with regression analysis showed that patients functional
sex, education, income, marital status, MMSE, ADCS- capacity (ADCS-ADL), neuropsychiatric symptoms
ADL, NPI, or BDI scores. (NPI), total amount of drugs and caregivers depres-
There was a significant correlation between self- sion were significant predictors for the caregivers
rated disease-specific QoL-AD and VAS (r 0.451, assessment of patients QoL. We also conducted the
p < 0.001). Self-rated QoL-AD correlated inversely linear regression analysis by dividing the patients into
with BDI scores (r 0.421, p < 0.001). There was two groups according to the presence of NPS as seen in
also a minor but statistically significant correlation Table 5. In the patients with at least one neuropsy-
between self-rated QoL-AD and education and income chiatric symptom (NPI 1), the significant ( p < 0.05)
but not with age (data not shown). There was no predictors for their QoL assessed by the caregivers were
correlation between self-rated QoL-AD scores and the patients functional capacity (ADCS-ADL), severity
MMSE, ADCS-ADL, or NPI scores. of disease (CDR), total amount of drugs and

Table 5 Predictors influencing in the spouse caregivers (n 168) rating of the patients Quality of Life. The patients were divided into the two groups
according to the presence of neuropsychiatric symptoms

Unstandardized Standardized Standardized Significance

B Error Beta

NPI 0 (17.9%)a,c
Patients age 0.241 0.079 0.441 0.005
Patients
ADCS-ADL 0.178 0.080 0.325 0.033
NPI 1 (82.1%)b,d
Caregiver depression 2.264 0.725 0.214 0.002
Patients CDR 0.597 0.260 0.177 0.023
Patients
ADCS-ADL 0.196 0.044 0.343 0.000
Patients total amount of drugs 0.281 0.134 0.139 0.038

Age, sex, and the following factors with p < 0.2 in the univariate analysis were fitted in the linear regression model: patients CDR, patients depression, patients
ADCS-ADL, patients total amount of drugs and caregiver depression. Other than the variables mentioned in the Table 5, did not influence significantly in the
patients QoL estimated by the spouse caregivers.
a
The percentage of the patients without any neuropsychiatric symptoms.
b
The percentage of the patients with at least one neuropsychiatric symptom.
c 2
R for the model is 0.393.
d 2
R for the model is 0.314.

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
478 K. Karttunen et al.

caregivers depression whereas in the patients without been reported in previous studies (Mega et al., 1996).
neuropsychiatric symptoms (NPI 0), the significant These results suggest that the underlying pathophy-
predictors for QoL were patients age and functional siology behind these symptoms may be different in
capacity (ADCS-ADL, Table 5). early AD. Hallucinations and REM sleep disturbances
are characteristic features of Lewy body dementia
(McKeith et al., 2005). Although none of our study
Discussion subjects had other clinical features that are character-
istic for Lewy body dementia, we cannot exclude the
We found that NPS are common in the early stages of possibility that there was some co-morbid Lewy body
AD in line with previous studies. Over 80% of our brain pathology in a few patients. However, previous
patients exhibited at least one symptom and almost studies in AD have also detected this association
half of the patients displayed three or more symptoms. between hallucinations and REM sleep behavior
In our cohort, the most frequent symptoms were disorder (McCurry et al., 1999; Sinforiani et al., 2007).
apathy, anxiety, agitation, irritability, and depression. Many studies have reported that cognitive decline
A high prevalence of these kinds of symptoms in MCI and NPS seem to represent distinct and partially
or mild AD have been consistently found in other independent manifestations of the disease, since the
studies (Reisberg et al., 1989; Jost and Grossberg, 1996; severity of the general cognitive decline displays only a
Mega et al., 1996; Lyketsos et al., 2000, 2002; Feldman weak or no correlation with NPS (Tekin et al., 2001;
et al., 2004; Hwang et al., 2004). NPS were significantly Chan et al., 2003; Tractenberg et al., 2005). Our results
associated with caregiver assessment of the patients showed that NPS correlated with functional capacity
QoL but not with the patients self-assessed QoL. but not with MMSE scores. Previous studies have also
However, the presence of depression was strongly reported that NPS are associated more closely with
associated with the self-assessed QoL. functional capacity than with cognitive decline
Previous studies have shown that many NPS co-exist (Cullum et al., 2001; Tractenberg et al., 2002).
but they can be categorized into three or more clusters The assessment and predictors of QoL in patients
(Lyketsos et al., 2001; Hollingworth et al., 2006; Aalten with AD have become an important focus of dementia
et al., 2007; Archer et al., 2007). In our cohort, the research. Generic measures of health-related QOL are
symptoms loaded into four factors; psychotic/agita- used to estimate the quality-adjusted life years
tion, affective, frontal type behavior, and a fourth (QALYs) allowing comparisons across different disease
factor that included sleep disturbances and hallucina- groups but their content validity in relation to
tions. The different results in the recent published dementia has been questioned (Silberfeld et al.
studies may be due to differences in patient cohorts 2002). However, many studies indicate that prefer-
such as severity of the disease in study subjects and the ence-based QoL measures such as EQ-5D may be used
clustering methodology. Most studies have included to measure health utilities in AD (Coucill et al., 2001;
more advanced AD cases and classification methods Ankri et al., 2003; Jonsson et al., 2006; Karlawish, et al.,
other than factor analysis have been used. Psychotic 2008b). Our results showed that self-rated health-
and mood clusters have been found in most studies. related QoL as assessed by 15D was as good in mild AD
Psychotic clusters have often included delusions and as QoL of the age-matched general population in
hallucinations (Jost and Grossberg, 1996; Hwang et al., previous studies (Saarni et al., 2007). However, this
2004). Delusions seem to appear earlier than halluci- study cohort represents otherwise healthy AD patients
nations and we also found delusions more frequently i.e., the presence of other severe diseases was an
than hallucinations in our patients with very mild or exclusion criterion for this longitudinal intervention
mild AD. In accordance with previous studies study. The significant predictors of health-related QoL
(Bassiony et al., 2002), delusions were also associated in this study i.e., age and general health, were similar to
with depression in our cohort. A frontal type cluster those previously found in the general population
has been reported in some previous studies (Frisoni (Saarni et al., 2007). These results suggest that patients
et al., 1999; Starr and Lonie, 2007) and specialist with very mild and mild AD assess their health-related
clinicians are familiar with a group of patients who QoL in a similar way as non-demented subjects.
exhibit prominent frontal type behavior even in the However, our results also suggest that generic measures
early stage of AD. One interesting finding in our study evaluate different aspects of QoL than the disease
was that hallucinations were not linked with delusions specific instruments. Disease specific instruments
but did associate with sleep disturbances. No contain items that are specific for the problems
association between hallucinations and delusions has associated with a particular disease and may therefore

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
Neuropsychiatric symptoms and Quality of Life in patients with very mild and mild Alzheimers disease 479

be more sensitive at detecting changes. In this study, we result in a less consistent assessment of QoL (Ready
used the QoL-AD that has been specifically developed et al., 2006). Therefore self-reporting of QOL by AD
for use in AD patients and their caregivers (Logsdon patients has been criticized and caregiver reporting has
et al., 1999). The correlation between 15D and the been used in many studies. Our results showed that
QoL-AD scores was weak and the predictors of QoL caregivers reported lower QoL than patients and the
were different. Depressive symptoms (BDI) correlated correlation between self-reported and caregiver-
inversely with the QoL-AD scores but not with 15D reported QoL of the patient was only weak. Similar
scores whereas age and general health correlated with findings have been reported in many previous studies
15D but not with the QoL-AD scores. However, (Logsdon et al., 1999; Novella et al., 2001; Snow et al.,
MMSE, functional capacity or NPI scores showed no 2005; James et al., 2005; Shin et al., 2005; Ready et al.,
correlation with either measures of self-reported QoL. 2006; Naglie et al., 2006; Jonsson et al., 2006; Vogel
Depression is commonly present in all stages of AD et al., 2006). These studies suggested that the
(Zubenko et al., 2003; Starkstein et al., 2005). divergences between patients and caregivers in the
However, the consequences of depression in AD are perception of patient QoL are not associated with a
not well-known. Many studies have suggested that lack of insight or the severity of cognitive decline
depression increases the degree of functional impair- (Ready et al., 2006; Vogel et al., 2006). It is possible
ment (Payne et al., 1998; Starkstein et al., 2005; Lam that patient and caregiver reports about patients QoL
et al., 2007; Benoit et al., 2008). Our results did not may represent two unique but different perspectives
show a correlation between BDI scores and ADCS- both of which are valid.
ADL scores. The discrepant results between our It has been hypothesized that patients define their
findings and previous studies may be attributable to QoL in terms of emotional well-being and general
differences in the study cohorts and the measures used health whereas caregivers emphasize physical and
to assess depression and functional capacity. Our functional aspects and our results support this
subjects had early AD with mild functional impairment hypothesis. The strongest predictor of self-reported
whereas the subjects examined in previous studies have QoL-AD scores was depressive symptoms whereas the
been more heterogeneous including patients with presence of functional decline and NPS predicted a
advanced disease. The assessment of ADL functions has poor caregiver rating of the patients QoL. The strong
varied with the severity of depressive symptoms being association between caregiver rating of patients QoL
estimated with informant-based NPI rather than self- and ADL functioning and particularly NPS has been
rated depression measure. However, most studies found in many previous studies (Ready et al., 2004;
including this present one have shown that depressive Snow et al., 2005; Naglie et al., 2006; Vogel et al., 2006;
symptoms are strongly associated with self-reported Banerjee et al., 2006; Jonsson et al., 2006). However,
QoL measures (Logsdon et al., 1999; Shin et al., 2005; our results showed that caregiver depression was
Snow et al., 2005; Jonsson et al., 2006; Naglie et al., associated with the presence of NPS and also
2006; Hoe et al., 2007; Karlawish et al., 2008a). Our influenced significantly their rating of the patients
results suggest that depression may be under- or QoL. Other studies have also shown that caregiver
mistreated in AD patients. Depressive symptoms were related factors can influence their assessment ratings
also associated with agitation in our patients and the (Logsdon et al., 1999; Snow et al., 2005; Karlawish
use of sedatives was more frequent in those subjects et al., 2008a). This limits the use of proxy measures of
with high depressive scores. However, a mere 20% of patients QoL as an outcome measure in clinical trials.
these patients were receiving antidepressant treatment. The main strengths of our study are the thorough
The results suggest that depression had remained diagnostic evaluation of the patients and the prospec-
unrecognized emphasizing the need for careful and tive set up nature of the study. However, the study has
structured assessment of NPS before deciding on the some weaknesses that limit the generalization of the
appropriate treatment. results. The study cohort was drawn from referral
Most authors agree that QoL is a subjective issue, memory clinics and subjects with other severe illnesses
unique to each individual. Lack of insight and decline that could have impacted on survival were excluded
in cognitive abilities may influence the patients from this intervention study. The study cohort is thus
awareness of his/her situation and functional deficits. not representative of the general population. This may
Approximately 50% of AD patients rate themselves as partially explain the good health related QoL values.
having perfect health (Naglie et al., 2006; Karlawish Another source of selection bias relates to the
et al., 2008b) even though, by definition, AD patients willingness to participate in a 6 year-long study.
are disabled. This lack of insight has been shown to However, our results are consistent with those of many

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
480 K. Karttunen et al.

other studies, which indicate that our cohort does not Key Points
significantly differ from those examined previously.
One limitation was that the mood scale BDI which was  Neuropsychiatric symptoms (NPS) are common
utilized in our study is not a specific test to examine the even in the early stages of Alzheimers disease
depressive symptoms in Alzheimer patients. Another (AD). The most frequent symptoms are apathy,
limitation is that we did not directly evaluate the depression, irritability, and agitation.
participants function and behavior but relied on the  Patients define their QoL in terms of emotional
caregiver interview. Caregiver burden is closely related well-being and general health. Depressive symp-
with their rating of the patients symptoms. We cannot toms are the strongest predictor of poor self-
exclude the possibility that burdened caregivers are reported QoL-AD.
 Caregivers define the QoL-AD of the patients with
more prone to report NPS and over-estimate the extent
the emphasis on physical and functional aspects.
of the functional decline in the patient. This bias is
Functional decline and presence of NPS predicts
unavoidable in all studies that rely on interviews
poor caregiver ratings of patients QoL.
instead of objective observations.
 Caregiver depression is associated with the
We conclude that NPS are common in the early
presence of NPS and it also significantly
stages of AD and contribute to the patients functional
influences their rating of the patients QoL.
capacity and well-being as well as to caregivers depres-
sion. Careful recognition and appropriate treatment of
these symptoms may confer substantial benefits on the
patient and alleviate the burden on the caregiver.
Anna-Maija Pietila, PhD, Professor, Department of
Nursing Sciences, University of Eastern Finland,
Kuopio Campus, Finland.
List of contributors Hilkka Soininen, MD, Professor, Department of
Neurology, Kuopio University Hospital and School of
ALSOVA study group Medicine, Institute of Clinical Medicine, Neurology,
Tuula Pirttila, MD, Professor, project leader, Depart- University of Eastern Finland, Kuopio Campus, Finland.
ment of Neurology, Kuopio University Hospital and Juhani Sivenius, MD, Professor, Brain Research and
School of Medicine, Institute of Clinical Medicine, Rehabilitation Center Neuron, Department of Neurol-
Neurology, University of Eastern Finland, Kuopio ogy, Kuopio University Hospital and School of
Campus, Finland. Medicine, Institute of Clinical Medicine, Neurology,
Ilona Hallikainen, psychologist, Brain Research University of Eastern Finland, Kuopio Campus, Finland.
Unit, Clinical Research Centre, University of Eastern Hannu Valtonen, PhD, Professor, Department of
Finland, Kuopio Campus, Finland. Health and Social Management, University of Eastern
Asta Hiltunen, MD, Department of Neurology, Finland, Kuopio Campus, Finland.
North Karelia Central Hospital, Finland. Matti Vanhanen, PhD, neuropsychologist, Niuvan-
Janne Martikainen, PhD, research director, School niemi Hospital, Kuopio, Finland.
of Pharmacy, University of Eastern Finland, Kuopio Katri Vehvilainen-Julkunen, Professor, Department
Campus, Finland. of Nursing Sciences, University of Eastern Finland,
Helena Makela, study nurse, Brain Research Unit, Kuopio Campus, Finland.
Clinical Research Centre, University of Eastern Fin- Tarja Valimaki, MNSc, Department of Nursing Sciences,
land, Kuopio Campus, Finland. University of Eastern Finland, Kuopio Campus, Finland.
Pertti Karppi, MD, geriatrician, Mikkeli Central
Hospital, Finland.
Kristiina Karttunen, MD, geriatrician, Institute of
Public Health and Clinical Nutrition, Department of Acknowledgements
Geriatrics, University of Eastern Finland, Kuopio
Campus, Finland and School of Medicine, Institute This study was supported by grants from Yrjo-
of Clinical Medicine, Neurology, University of Eastern Jahnsson Foundation, Hospital Neuron, Kuopio
Finland, Kuopio Campus, Finland. University Hospital (EVO-grant 5220/5772728) and
Teemu Paajanen, psychologist, Brain Research Unit, the Social Insurance Institute of Finland (Kela). The
Clinical Research Centre, University of Eastern Fin- authors thank study nurse Helena Makela for her
land, Kuopio Campus, Finland. invaluable work in organizing and performing the

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
Neuropsychiatric symptoms and Quality of Life in patients with very mild and mild Alzheimers disease 481

examinations of the study subjects, PhD Vesa Jonsson L, Andreasen N, Kilander L, et al. 2006. Patient- and proxy-reported Utility in
Alzheimer disease using the EuroQoL. Alzheimer Dis Assoc Disord 20(1): 4955
Kiviniemi for his statistical expertise and the patients Karlawish JH, Zbrozek A, Kinosian B, et al. 2008. Caregivers assessment of preference-
and their caregivers for their time and effort. based quality of life in Alzheimer disease. Alzheimers Dement 4: 203211.
Karlawish JH, Zbrozek A, Kinosian B, et al. 2008. Preference-based quality of life in
patients with Alzheimers disease. Alzheimers Dement 4: 193202.
Kaufer DI, Cummings JL, Christine D, et al. 1998. Assessing the impact of neuro-
psychiatric symptoms in Alzheimers disease: the Neuropsychiatric Inventory
Conflicts of interest Caregiver Distress scale. J Am Geriatr Soc 46(2): 2105
Lam LC, Tam CW, Chiu HF, et al. 2007. Depression and apathy affect functioning in
communiy active subjects with questionable dementia and mild Alzheimers
None declared. disease. Int J Geriatr Psychiatry 22: 431437.
Logsdon RG, Gibbons LE, McCurry SM, et al. 1999. Quality of life in Alzheimers
disease: patient and caregiver reports. J Ment Health Aging 5: 2132.
Lyketsos CG, Steinberg M, Tschanz JT, et al. 2000. Mental and behavioral disturbances
in dementia: findings from the Cache County study on memory in aging. Am J
References Psychiatry 157: 708714.
Lyketsos GB, Sheppard JME, Steinberg M, et al. 2001. Neuropsychiatric disturbances
in Alzheimers disease clusters into three groups: the Cache County study. Int J
Aalten P, Verhey FR, Boziki M, et al. 2007. Neuropsychiatric syndromes in dementia. Geriatr Psychiatry 16: 10431053.
Results from the European Alzheimer Disease Consortium: part I. Dement Geriatr Lyketsos CG, Lopez O, Jones B, et al. 2002. Prevalence of neuropsychiatric symptoms
Cogn Disord 24: 457463. in dementia and mild cognitive impairment. Results from the cardiovascular health
Ankri J, Beaufils B, Novella JL, et al. 2003. Use of the EQ-5D among patients suffering study. JAMA 288: 14751483.
from dementia. J Clin Epidemiol 56: 10551063. McCurry SM, Logsdon RG, Teri L, et al. 1999. Characteristics of sleep disturbance in
Archer N, Brown RG, Reeves SJ, et al. 2007. Premorbid personality and behavioral and community-dwelling Alzheimers disease patients. J Geriatr Psychiatry Neurol 12:
psychological symptoms in probable Alzheimer disease. Am J Geriatr Psychiatry 15: 5359.
202213. McKeith IG, Dickson DW, Lowe J, et al. 2005. Diagnosis and management of dementia
Banerjee S, Smith SC, Lamping DL, et al. 2006. Quality of life in dementia: more than with Lewy bodies: third report of the DLB consortium. Neurology 65: 1863
just cognition. An analysis of associations with quality of life in dementia. J Neurol 1872.
Neurosurg Psychiatry 77: 146148. McKhann G, Drachman D, Folstein M, et al. 1984. Clinical diagnosis of Alzheimers
Bassiony MM, Warren A, Rosenblatt A, et al. 2002. The relationship between delusions disease: Report of the NINCDS-ADRDA work group under the auspices of
and depression in Alzheimers disease. Int J Geriatr Psychiatry 17: 549556. Department of Health and Human Services Task Force on Alzheimers Disease.
Beck A, Ward C, Mendelson M, et al. 1961. An inventory for measuring depression. Neurology 34: 939944.
Arch Gen Psychiatry 4(561): 571. Mega MS, Cummings JL, Fiorello T, et al. 1996. The spectrum of behavioral changes in
Benoit M, Andrieu S, Lechowski L, et al. 2008. Apathy and depression in Alzheimers Alzheimers disease. Neurology 46: 130135.
disease are associated with functional deficit and psychotropic prescription. Int J Missotten P, Squelard G, Ylieff M, et al. 2008. Relationship between quality of life and
Geriatr Psychiatry 23: 409414. cognitive decline in dementia. Dementia Geriatr Cogn Disord 25: 564572.
Chan DC, Kasper JD, Black BS, et al. 2003. Prevalence and correlates of behavioral and Morris J. 1993. The CDR: current version and scoring rules. Neurology 43: 24122413.
psychiatric symptoms in community-dwelling elders with dementia or mild Naglie G, Tomlinson G, Tansey C, et al. 2006. Utility-based quality of life measures in
cognitive impairment: the memory and medical Care study. Int J Geriatr Psychiatry Alzheimers disease. Qual Life Res 15: 631643.
18: 174182. Novella JL, Jochum C, Jolly D, et al. 2001. Agreement between patients and proxies
Copeland MP, Daly E, Hines V, et al. 2003. Psychiatric symptomatology and reports of quality of life in Alzheimers disease. Qual Life Res 10: 443452.
prodromal Alzheimers disease. Alzheimer Dis Assoc Disord 17: 18. Palmer K, Berger AK, Monastewro R, et al. 2007. Predictors of progression from mild
Coucill W, Bryan S, Bentham P, et al. 2001. EQ-5D in patients with dementia. Med cognitive impairment to Alzheimer disease. Neurology 68: 15961602.
Care 39: 760771. Payne JL, Lyketsos CG, Steele C, et al. 1998. Relationship of cognitive and functional
Cullum CM, Saine K, Davis-Chan L, et al. 2001. A performance-based instrument to impairment to depressive features in Alzheimers disease and other dementias.
assess functional capacity in dementia: the Texas functional living scale. Neuro- J Neuropsychiatry Clin Neurosci 10: 440447.
psychiatr Neuropsychol Behav Neurol 14: 103108. Peters KR, Rockwood K, Black SE, et al. 2006. Characterizing neuropsychiatric
Cummins JL. 2000. Cognitive and behavioral heterogeneity in Alzheimers disease: symptoms in subjects referred to dementia clinics. Neurology 66: 523528.
seeking the neurobiological basis. Neurobiol Aging 21: 845861. Ready RE, Ott BR, Grace J. 2004. Patient versus informant perspectives of quality of life
Cummings J, Mega M, Gray K, et al. 1994. The neuropsychiatric inventory: in mild cognitive impairment and Alzheimers disease. Int J Geriatr Psychiatry 19:
comprehensive assessment of psychopathology in dementia. Neurology 44(12): 256265.
23082314. Ready RE, Ott BR, Grace J. 2006. Insight and cognitive impairment. Effects on quality-
Feldman H, Scheltens P, Scarpini E, et al. 2004. Behavioral symptoms in mild cognitive of-life reports from mild cognitive impairment and Alzheimers disease patients.
impairment. Neurology 62: 11991201. Am J Alzheimers Dis Other Demen 21: 242248.
Folstein M, Folstein S, McHugh P. 1975. Mini-mental state. A practical method for Reisberg B, Franssen E, Sclan SG, et al. 1989. Stage specific incidence of potentially
grading the cognitive state of patients for the clinician. J Psychiatr Res 12: 189198. remediable behavioral symptoms in aging and Alzheimers disease: a study of 120
Galasko D, Bennet D, Sano M, et al. 1997. An inventory to assess activities of daily patients using the BEHAVE-AD. Bull Clin Neurosci 54: 95112.
living for clinical trials in Alzheimers disease. Alzheimer Dis Assoc Disord 11 Saarni SI, Suvisaari J, Sintonen H, et al. 2007. The health-related quality-of-life impact
(Suppl 2): S33S39. of chronic conditions varied with age in general population. J Clin Epidemiol 60:
Geda YE, Smith GE, Knopman DS, et al. 2004. De novo genesis of neuropsychiatric 12881297.
symptoms in mild cognitive impairment (MCI). Int Psychogeriatr 16: 5160. Shin IS, Carter M, Masterman D, et al. 2005. Neuropsychiatric symptoms and quality
Frisoni GB, Rozzini L, Gozzetti A, et al. 1999. Behavioral syndromes in Alzheimers of life in Alzheimer disease. Am J Geriatr Psychiatry 13: 469474.
disease: description and correlates. Dement Geriatr Cogn Disord 10: 130138. Silberfeld M, Rueda S, Krahn M, et al. 2002. Content validity for dementia of three
Hoe J, Katona C, Orrell M, et al. 2007. Quality of life in dementia: care recipient and generic preference based health related quality of life instruments. Qual Life Res 11:
caregiver perceptions of quality of life in dementia: the LASER-AD study. Int J 7179.
Geriatr Psychiatry 22: 10311036. Sinforiani E, Terzaghi M, Pasotti C, et al. 2007. Hallucinations and sleep-wake cycle in
Hollingworth P, Hamshere ML, Moskvina V, et al. 2006. Four components describe Alzheimers disease: a questionnaire-based study in 218 patients. Neurol Sci 28: 96
behavioral symptoms in 1120 individuals with late-onset Alzheimers disease. J Am 99.
Geriatr Soc 54: 13481354. Sintonen H. 1995. The 15D measure of health related quality of life. II Feasibility,
Hughes C, Berg L, Danziger W, et al. 1982. A new clinical scale for the staging of reliability and validity of its valuation system. Working paper 42, Melbourne: Centre
dementia. Br J Psychiatry 140(566): 572. for health program evaluation, Monash University.
Hwang TJ, Masterman DL, Ortiz F, Fairbanks LA, Cummings JL. 2004. Mild cognitive Snow AL, Dani R, Souchek J, et al. 2005. Comorbid psychosocial symptoms and
impairment is associated with characteristic neuropsychiatric symptoms. Alzheimer quality of life in patients with dementia. Am J Geriatr Psychiatry 13: 393401.
Dis Assoc Disord 18: 1721. Starkstein SE, Jorge R, Mizrahi R, et al. 2005. The construct of minor and major
James BD, Xie SX, Karlawish JHT. 2005. How do patients with Alzheimers disease rate depression in Alzheimers disease. Am J Psychiatry 162: 20862093.
their overall quality of life ? Am J Geriatr Psychiatry 13: 484490. Starr JM, Lonie J. 2007. Relationship between behavioural and psychological symp-
Jost BC, Grossberg GT. 1996. The evolution of psychiatric symptoms in Alzheimers toms of dementia and cognition in Alzheimers disease. Dement Geriatr Cogn Disord
disease: a natural history study. J Am Geriatr Soc 44: 10781081. 24: 343347.

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.
482 K. Karttunen et al.

Tekin S, Fairbanks LA, OConnor S, et al. 2001. Activities of daily living in Alzheimers Tractenberg RE, Weiner MF, Patterson MB, et al. 2002. Emergent psychopathology in
disease: neuropsychiatric, cognitive, and medical influences. Am J Geriatr Psychiatry AD patients over 12 months associated with functional, not cognitive, changes.
9: 8186. J Geriatr Psychiatry Neurol 15: 110117.
Thomas P, Lalloue F, Preux PM, et al. 2006. Dementia patients caregivers quality of life Vogel A, Mortensen EL, Hasselbach SG, et al. 2006. Patient versus informant reported
: the PIXEL study. Int J Geriatr Psychiatry 21: 5056. quality of life in the earliest phases of Alzheimer disease. Int J Geriatr Psychiatry 21:
Tractenberg RE, Weiner MF, Cummings JL, et al. 2005. Independence of changes in 11321138.
behavior from cognition and function in community-dwelling persons with Zubenko GS, Zubenko WN, McPherson S, et al. 2003. A collaborative study of the
Alzheimers disease: a factor analytic approach. J Neuropsychiatry Clin Neurosci emergence and clinical features of the major depressive syndrome of Alzheimers
17: 5160. disease. Am J Psychiatry 160: 857866.

Copyright # 2010 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2011; 26: 473482.