Dement Neuropsychol 2013 September;7(3):258-262 Original Article

Correlation between neuropsychiatric

symptoms and caregiver burden
in a population-based sample from
São Paulo, Brazil
A preliminary report
Jefferson Cunha Folquitto1, Rita de Cássia Gomes Marques1,
Mariana Franciosi Tatsch1, Cássio Machado de Campos Bottino1

ABSTRACT. Neuropsychiatric symptoms and caregiver burden are highly prevalent in older adults with Alzheimer’s disease
(AD). Objective: To evaluate the correlation between neuropsychiatric symptoms and caregiver burden in a community-
based sample from São Paulo, Brazil. Methods: A total of 1,563 randomly-selected subjects were assessed by the Mini-
Mental State Examination, Fuld Object Memory Evaluation, Informant Questionnaire on Cognitive Decline in the Elderly and
Bayer – Activities of Daily Living Scale. Subjects considered screen-positives were submitted to a dementia workup and
diagnosis was determined according to ICD-10 criteria. The neuropsychiatric Inventory was applied to caregivers to evaluate
neuropsychiatric symptoms and the Zarit Burden Interview was also applied to assess caregivers’ burden. Results: Sixty-
one AD patients, 25 Cognitively Impaired Non Demented (CIND) and 79 healthy elderly subjects were evaluated. Zarit mean
scores for controls, CIND and AD were 2.32, 3.92 and 20.11, respectively. There was strong positive correlation between
total NPI and Zarit scores. Conclusion: In conclusion, neuropsychiatric symptoms showed a significant association with
higher rates of caregiver stress.
Key words: Alzheimer’s disease, cognitive impairment no dementia, neuropsychiatric symptoms, caregiver burden.

CORRELAÇÃO ENTRE SINTOMAS NEUROPSIQUIÁTRICOS E SOBRECARGA DO CUIDADOR EM UMA AMOSTRA COMUNITÁRIA DE

SÃO PAULO, BRASIL: UM RELATO PRELIMINAR
RESUMO. Sintomas neuropsiquiátricos e sobrecarga do cuidador apresentam alta prevalência em idosos com doença de
Alzheimer (DA). Objetivo: Avaliar a correlação entre sintomas neuropsiquiátricos e sobrecarga do cuidador em uma amostra
comunitária de São Paulo, Brasil. Métodos: Um total de 1.563 indivíduos selecionados aleatóriamente foram avaliados
através do Mini-Exame do Estado Mental, “Fuld Object Memory Evaluation”, “Informant Questionnaire on Cognitive Decline
in the Elderly” e da Escala Bayer – Atividades de Vida Diária. Indivíduos considerados suspeitos foram submetidos a
uma investigação para demência, sendo o diagnóstico feito com os critérios da CID-10. Inventário Neuropsiquiátrico foi
aplicado nos cuidadores para avaliar sintomas neuropsiquiátricos e a Escala Zarit de Sobrecarga no Cuidador para avaliar
sobrecarga nos cuidadores. Resultados: Sessenta e um pacientes com DA, 25 sujeitos com Comprometimento Cognitivo
não Demência (CIND) e 79 idosos saudáveis foram avaliados. A média dos escores da Zarit para controles, CIND e DA foi,
respectivamente, 2,32, 3,92 e 20,11 pontos. Houve correlação positiva boa entre os escores da NPI e da Zarit. Conclusão:
Sintomas neuropsiquiátricos mostraram uma associação significativa com escores mais elevados de estresse do cuidador.
Palavras-chave: doença de Alzheimer, comprometimento cognitivo não demência, sintomas neuropsiquiátricos, sobrecarga
do cuidador.

Old Age Research Group, Institute of Psychiatry, University of São Paulo, São Paulo, Brazil.
1

Jefferson Cunha Folquitto. Rua Dr. Ovídio Pires de Campos, 785 / 3º andar / sala 14 – 05403-903 São Paulo SP – Brazil. E-mail: [email protected]

Disclosure: The authors report no conflicts of interest.

Received May 08, 2013. Accepted in final form August 04, 2013.

258 Neuropsychiatric symptoms and caregiver burden     Folquitto JC, et al.

 Dement Neuropsychol 2013 September;7(3):258-262

INTRODUCTION ly, drawn from three districts of the urban area of São

N europsychiatric symptoms are highly prevalent,

affecting 10% to 73% of dementia patients,1 with
this variation being due to differences in assessment
Paulo representing high, medium and low socioeco-
nomic classes, was evaluated. The following screening
algorithm was applied to identify subjects suspected
methods and in population samples studied.2 The neu- of being demented: Mini-Mental State Examination
ropsychiatric symptoms include delusions, hallucina- (MMSE)9 scores for illiterate <20; 1-4 years of schooling
tions, agitation (physical and verbal), depression, anxi- <25. 5-8 years of schooling <27; and ≥9 years of school-
ety, euphoria/elation, apathy, disinhibition, irritability, ing <28; or Fuld Object Memory Evaluation score <3510
aberrant motor behavior, night-time disturbances and and Informant Questionnaire on Cognitive Decline in
eating changes.2 In Brazil, Tatsch et al.2 found a preva- the Elderly11 score >3.40 or Bayer – Activities of Daily
lence of neuropsychiatric symptoms in AD and CIND Living12 score >3.19. This screening method was previ-
(cognitive impairment no dementia) of 78.3% and 36%, ously tested in an outpatient sample of 93 elderly out-
respectively. The most frequent symptoms in AD were patients (34 AD and 59 controls), showing a sensitivity
apathy (53.3%), depression (38.3%), sleep disturbances and specificity of 100%.13
(38.3%) and anxiety (25%), whereas in CIND these were In the second phase, screen positives were submit-
anxiety (24%), night-time disturbances (24%) and de- ted to a work-up for dementia, entailing physical and
pression (16%). In a comprehensive literature review, neurological examination, Cranial Computed Tomogra-
apathy and depression were considered the most preva- phy or Brain Magnetic Resonance Imaging, application
lent neuropsychiatric symptoms in subjects with AD of the Cognitive Section (CAMCOG) of the Cambridge
and CIND.3 Examination for mental Disorders (CAMDEX),14,15 Clin-
Neuropsychiatric symptoms in patients with de- ical Dementia Rating Scale (CDR),16 Neuropsychiatric
mentia are associated with worse prognoses, higher Inventory (NPI)17 and Zarit Caregiver Burden Interview
health care costs, greater impairment in daily function- (which includes health of caregivers, psychological well-
ing and quality of life, faster cognitive decline, earlier in- being, finances, social life and relationship between
stitutionalization, higher mortality, and increased care- caregivers and patients. NPI and Zarit were applied us-
giver burden.2-4 These serious consequences call for the ing their standardized Brazilian versions.18,19
development of new strategies for the prevention, early Diagnoses of Alzheimer’s Disease (AD) and Cogni-
recognition and intervention to deal with neuropsychi- tive Impairment No Dementia (CIND) were determined
atric symptoms in dementia.3 based on criteria of the DSM-IV20 and Ebly et al.,21
A caregiver can be defined as a person who helps respectively.
with the basic and daily instrumental activities of daily This study included subjects with a diagnosis of de-
living of a patient for most of the time, without receiv- mentia at the end of the second phase, and 79 elderly
ing payment for this activity.5 Caregiver burden is de- considered screen-negative in the first phase (random-
fined as the sum of physical, psychological, social and ized by SPSS 16.0 for Windows).
financial problems which arise among members of the The study was approved by the Ethics Research Com-
family or people who assist the diseased elderly.6 mittee from the Clinicas Hospital of the Medical School
The presence of neuropsychiatric symptoms in pa- of the University of São Paulo and all subjects evaluated
tients with dementia and subjects with mild cognitive gave consent to participate in the study.
impairment is associated with greater caregiver bur- Statistical analyses were performed using SPSS ver-
den.7 The symptoms that more frequently cause care- sion 16.0 for Windows. For continuous variables, the
giver burden are aggression and delusions.8 Kolmogorov-Smirnov test was applied to test for nor-
The aim of the present study was to evaluate the in- mality. The Chi-square test was used to compare sex
fluence of neuropsychiatric symptoms on caregiver bur- and the Kruskal-Wallis test to compare age and educa-
den in a community-based sample of elderly subjects tion. The Mann-Whitney test was applied to analyze the
with Alzheimer’s disease or cognitive impairment no groups using two-by-two comparisons. The Spearman
dementia (CIND). rank correlation coefficient was used to evaluate the cor-
relation between NPI and Zarit.
METHODS
The present study evaluated subjects aged 60 years or RESULTS
over from an epidemiological survey conducted in São A total of one hundred and sixty-five caregivers were
Paulo, Brazil. In the first phase, a total of 1,563 elder- evaluated: 61 caregivers of patients with AD, 25 caregiv-

Folquitto JC, et al.     Neuropsychiatric symptoms and caregiver burden 259

 Dement Neuropsychol 2013 September;7(3):258-262

Table 1. Demographic characteristics of sample.

Groups Control CIND AD Statistical analysis
Age Mean 72.41 71.84 80.07 #,+
*c2=27.344
(SD) (8.16) (9.25) (8.84) p<0.001
Education Mean 643 3.71** 3.36 #
*c2=22.583
(SD) (5.01) (4.74) (4.49) p<0.001
Sex Male 24 9 14 c2=1.757
(%) (30.45%) (36.0%) (23.0%) p>0.05
SD: Standard Deviation; *Kruskal-Wallis test among three groups; **Mann-Whitney – controls versus CIND: U=522.500, p: 0.002; #Mann-
Whitney – controls versus AD: age (U=1254,000; p<0,001) and education (U=1140.500; p<0.001); +Mann-Whitney – CIND versus AD:
U=375.500, p<0.001

ers of patients with CIND, and 79 caregivers of normal NPI of 34.97, scores higher than those observed in our
elderly. study. Godinho et al.,23 studying a sample of 64 clinical
Table 1 shows the demographic characteristics of outpatients with Alzheimer’s disease, observed a mean
subjects and comparisons between the three groups. NPI score of 35. One possible explanation for this dif-
Table 2 gives the number of subjects who had at least ference is that our sample was community-based, while
one symptom in each subsection and the mean Zarit Moscoso et al.22 evaluated a clinical sample. Considering
scores for each NPI domain. the CIND subjects’ neuropsychiatric symptoms, popula-
Comparing the subjects by diagnostic group, there
was a statistically significant difference, as evaluated
using the Kruskal-Wallis test, for both total NPI score Table 2. Zarit scores according to each NPI domain.
(c2=65.848; p<0.001) and Zarit score (c2=79.266;
Number of Subjects Zarit
p<0.001), which were higher in the AD group. Mean
Domains (Score: Item ≥ 1) Mean (SD)
scores on the NPI and Zarit by diagnostic group are de-
Delusions 7 39.86 (25.73)
picted in Table 3.
Of the 61 AD patients included in the present study, Hallucinations 5 34.00 (18.00)
25 (41%) were classified as mildly demented (CDR=1), Agitation 17 27.24 (25.06)
29 (47.5%) moderately demented, and 6 (9.8%) as se- Depression 32 22.13 (19.06)
verely demented. To compare the groups, Mild AD was Anxiety 26 17.62 (17.45)
considered CDR 1, and Moderate and Severe AD as CDR Euphoria 3 41.33 (26.63)
2 and CDR 3,respectively, but no statistically significant
Apathy 35 20.71 (16.19)
differences were observed among the groups.
When evaluating groups using the Spearman rank Disinhibition 10 37.30 (24.66)
correlation coefficient, a significant positive correlation Irritability 18 25.22 (22.53)
was observed between the NPI and Zarit scores (Spear- Aberrant motor behavior 6 38.33 (21.23)
man=0.684; p<0.001). Stratifying subjects by diagnosis, Night-time disturbance 32 21.91 (21.12)
the positive correlation between NPI and Zarit remained Eating changes 15 24.33 (20.81)
strong for both CIND (Spearman=0.606; p=0.002) and
AD (Spearman=0.589; p<0.001) groups, but was weak
for controls (Spearman=0.300; p=0.008).
Table 3. NPI and Zarit scores by diagnostic group.
Controls CIND AD
DISCUSSION
Mean (SD) Mean (SD) Mean (SD)
In the present study, subjects with AD had mean scores
on the Zarit and NPI of 20.11 and 12.28 points, respec- NPI score 0.87 (2.94) 3.08* (4.69) 12.28**,# (13.29)
tively, scoring higher than controls. Moscoso et al.,22 in Zarit score 2.32 (3.96) 3.92 (6.05) 20.11**,# (17.38)
thirty-one elderly patients with AD from the outpatient SD: Standard Deviation; *Mann-Whitney test-controls versus CIND: U=693.500, p=0.002;
**Mann-Whitney test – controls versus AD: NPI (U=690.000; p<0.001) and Zarit (U=405.500;
unit of CEREDIC (Clinics Hospital – Cognitive Refer- p<0.001); #Mann-Whitney – CIND versus AD: NPI (U=378.500; p<0.001) and Zarit (U=198.000;
ence Center), observed a mean Zarit score of 31.77 and p<0.001)

260 Neuropsychiatric symptoms and caregiver burden     Folquitto JC, et al.

 Dement Neuropsychol 2013 September;7(3):258-262

tion-based studies have shown that these symptoms are of studies from different regions of the world (North
more frequent among this patient group compared with America, Europe/Australia and Asia) a significant asso-
healthy elderly.24,25 ciation between neuropsychiatric symptoms and care-
Regarding the severity of dementia, there were no giver burden was reported.26 Moreover, the incidence of
significant differences in caregiver burden scores by de- behavioral problems seems to have a higher potential to
mentia group. According to Moscoso et al.,22 there is no cause stress than persistence of these symptoms,28 but
consensus on the influence of the severity of dementia constant caregiving may significantly increase the risk
on caregiver stress. of caregiver stress and burden.29
Considering the mean Zarit scores for each NPI do- In conclusion, several factors influence the presence
main in Table 2, we found that Delusions, Hallucina- of caregiver burden, with neuropsychiatric symptoms
tions, Euphoria, Disinhibitions and Aberrant Motor showing a significant association with higher rates of
Behavior were the domains with higher Zarit scores. caregiver stress, mainly related to symptoms such as
Additionally, we observed a good correlation between aggression and delusions. Our study found a significant
the scores on the NPI and Zarit inventories. Moscoso association between Zarit and NPI scores while subjects
et al.22 also observed a significant association between who presented one or more symptoms, such as delu-
caregiver burden and total NPI scores. Several studies sions, hallucinations, euphoria, disinhibition and aber-
have observed a relationship between neuropsychiatric rant motor behavior, had the highest scores on the Zarit
symptoms and caregiver stress. Fialho et al.26 evaluated inventory. The adequate treatment and management of
83 caregivers of patients diagnosed with dementia in these neuropsychiatric symptoms in patients with de-
the state of Minas Gerais, Brazil, and found a positive mentia can have a significant impact on the quality of
correlation between NPI and Zarit scores. In a review life of patients and their caregivers.

REFERENCES
1. Finkel SI, Silva Jc, Cohen G, et al. Behavioral and psychological signs nados na avaliação de demência em idosos. Arq Neuropsiquiatr 2003;
and symptoms of dementia: a consensus statement on current knowl- 61:601-606.
edge and implications for research and treatment. Int Psychogeriatr 14. Roth M, Tym E, Mountjoy CQ, et al. CAMDEX: a standardised instru-
1996;8:497-512. ment for the diagnosis of mental disorder in the elderly with special
2. Tatsch MF, Bottino CMC, Azevedo D, et al. Neuropsychiatric Symptoms reference to the early detection of dementia. Br J Psychiatry 1986;149:
in Alzheimer Disease and Cognitively Impaired, Nondemented Elderly 698-709
From a Community-Based Sample in Brazil: Prevalence and Relation- 15. Bottino CMC, Almeida OP, Tamai S, et al. Entrevista estruturada para
ship With Dementia Severity. Am J Geriatr Psychiatry 2006;14:438-445. diagnóstico de transtornos mentais em idosos - CAMDEX – The Cam-
3. Lyketsos CG, Corrillo MC, Ryan JM, et al. Neuropsychiatric symptoms bridge examination for mental disorders of the elderly. Brazilian version
in Alzheimer’s disease. Alzheimer’s Dement 2011;7:532-539. (translated and adapted on behalf of the editors, Cambridge University
4. Vega UM, Marinho V, Engelhardt E, et al. Neuropsychiatric symptoms Press); 1999.
in dementias: preliminar report of a prospective outpatient evaluation in 16. Chaves ML, Camozzato AL, Godinho C, et al. Validity of the clinical de-
Brazil. Arq Neuropsiquiatr 2007;65:498-502. mentia rating scale for the detection and staging of dementia in Brazilian
5. Badia LX, Surinach LN, Gamisans RM. Calidad de vida, tiempo de dedi- patients. Alzheimer Dis Assoc Disord 2007;21:210-217.
cation y carga percebida por el cuidador principal informal del enfermo 17. Cummings JL, Mega M, Gray K, et al. The Neuropsychiatric Inventory:
de Alzheimer. Aten Primaria 2004;34:170-177. comprehensive assessment of psychopathology in dementia. Neurol-
6. George LK, Gwyther LP. Caregiver well-being a multidimensional exam- ogy 1994;44:2308-2314
ination of family caregivers of demented adults. Gerontology 1986;26: 18. Camozzato AL, Kochhann R, Simeoni C, et al. Reliability of the Brazilian
253-259. Portuguese version of the Neuropsychiatric Inventory (NPI) for patients
7. Ryan KA, Weldon A, Persad C, et al. Neuropsychiatric symptoms and with Alzheimer’s disease and their caregivers. Int Psychogeriatr 2008;
executive functioning in patients with mild cognitive impairment: rela- 20:383-393.
tionship to caregiver burden. Dement Geriatr Cogn Disord 2012;34: 19. Scazufca M. Versão Brasileira da escala Burden Interview para avalia-
206-215. ção de sobrecarga em cuidadores de indivíduos com doenças mentais.
8. Allegri RF, Sarasola D, Serrano CM. Neuropsychiatric symptoms as a Rev Bras Psiq 2002;24:12-17.
predictor of caregiver burden in Alzheimer’s disease. Neuropsychiatr Dis 20. American Psychiatry Association. Diagnostic ans statistical manual of
Treat 2006;2:105-110. mental disorders, 4th Edition. Washington, DC. American Psychiatry As-
9. Folstein MF, Folstein SE, McHugh PR. Mini Mental State: a practical sociation; 1994.
method for grading the cognitive state of patients for the clinician. J 21. Ebly EM, Hogan DB, Parhad IM. Cognitive impairment in the nonde-
Psychiatr Research 1975;12:189-198. mented elderly. Results from the Canadian Study of Health and Aging.
10. Fuld PA, Muramoto O, Blau AD, et al. Cross-cultural and multi-ethnic Arch Neurol 1995;52:612-619
dementia evaluation by mental status and memory testing. Cortex 22. Moscoso MA, Marques RCG, Ribeiz SRI, et al. Profile of caregivers of
1988;24:520-571. Alzheimer’s disease patients attended at a reference center for cognitive
11. Jorm AF, Korten AE. Assessment of cognitive decline in the elderly by disorders. Dement Neuropsychol 2007;1:412-417.
informant interview. Br J Psychiatry 1988;152:209-213. 23. Godinho C, Gorczevski I, Heisler A, et al. Clinical and demographic
12. Lehfeld H, Reisberg B, Finkel S, et al. Informant-rated activities-of-daily- characteristics of elderly patients with dementia assisted at an outpa-
living (ADL): results of a study of 141 items in the USA, Germany, Rus- tient clinic in Southern Brazil. Dement Neuropsychol 2010;4:42-46.
sia, and Greece from the International ADL Scale Development Project. 24. Lyketsos CG, Lopez O, Jones B, et al. Prevalence of neuropsychiatric
Alzheimer Dis Assoc Disord 1997;11:39-44 symptoms in dementia and mild cognitive impairment: results from the
13. Bustamante SEZ, Bottino CMC, Lopes MA, et al. Instrumentos combi- cardiovascular health study. JAMA 2002;288:1475-1483.

Folquitto JC, et al.     Neuropsychiatric symptoms and caregiver burden 261

 Dement Neuropsychol 2013 September;7(3):258-262

25. Geda YE, Roberts RO, Knopman DS, et al. Prevalence of neuropsy- trends and reports in dementia caregiver burden. Alzheimer Dis Assoc
chiatric symptoms in mild cognitive impairment and normal cognitive Disord 2004;18:99-109.
aging: population-based study. Arch Gen Psychiatry 2008;65:1193- 28. Gaugler JE, Wall MM, Kane RL, et al. The effects of incident and per-
1198. sistent behavioral problems on change in caregiver burden and nursing
26. Fialho PPA, Koenig AM, Santos EL, et al. Dementia caregiver burden in home admission of persons with dementia. Med Care 2010;48:875-883.
a Brazilian sample: Association to neuropsychiatric symptoms. Dement 29. Bandeira DR, Pawlowski J, Gonçalves TR, et al. Psychological distress
Neuropsychol 2009;3:132-135. in Brazilian caregivers of relatives with dementia. Aging Ment Health
27. Torti FM, Gwyther LP, Reed SD, et al. A multinational review of recent 2007;11:14-19.

262 Neuropsychiatric symptoms and caregiver burden     Folquitto JC, et al.