A pilot study of behavioural and psychological signs and symptoms of dementia in patients of Indian sub-continent origin admitted to a dementia day hospital in the United Kingdom

INTERNATIONAL JOURNAL OF GERIATRIC PSYCHIATRY

Int J Geriatr Psychiatry 2004; 19: 1195–1204.

Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/gps.1245

A pilot study of behavioural and psychological signs andsymptoms of dementia in patients of indian sub-continentorigin admitted to a dementia day hospital in theUnited Kingdom

Imran Haider1 and Ajit Shah1,2*

1West London Mental Health NHS Trust, London, UK2Imperial College School of Medicine, London, UK

SUMMARY

Background There is a paucity of cross-cultural studies of behavioural and psychological symptoms of dementia (BPSD).Method BPSD were examined in a consecutive series of Indian sub-continent origin and white indigenous elders admittedto a dementia day hospital using the BEHAVE-AD. The correlates of individual BPSD in each of the two ethnic groups andthe differences between the two ethnic groups were examined.Results There were no differences between the two groups on most of the demographic and clinical variables examined,except that Indian sub-continent elders had a greater number of children. There were no differences between the two groupson the MMSE scores, BEHAVE-AD total scores and BEHAVE-AD subscale scores (with one exception). Indian sub-continent origin patients had lower scores on the anxiety and phobias subscale. Within the Indian sub-continent origin group,Alzheimer’s disease (AD) was associated with activity disturbance and vascular dementia with affective disturbance. Withinthe indigenous group, aggressivity was associated with males and prescription of neuroleptics, and affective disturbancewith prescription of antidepressants.Conclusion There is a need to develop and evaluate translated versions of instruments that measure BPSD. After devel-opment of these instruments there is a need for cross-cultural population-based epidemiological studies of BPSD. Copyright# 2004 John Wiley & Sons, Ltd.

key words— Behavioural disturbance; dementia; BPSD; cross-culture

INTRODUCTION

Behavioural and psychological symptoms of demen-tia (BPSD) include disorders of behaviour, percep-tion, thought content and mood (Finkel, 1996a,b).Population-based cross-cultural epidemiological stu-dies of dementia have largely concentrated on cogni-tive impairment and neglected BPSD (Jitapunkalet al., 1996). Moreover, BPSD has been poorly stu-died in developing countries and in ethnic minoritygroups in developed countries.

However, studies of BPSD are emerging fromdeveloping countries including Hong Kong (Lamet al., 1997; Choy et al., 2001; Lam et al., 2001;Leung et al., 2001), Taiwan (Hwang et al., 1997;Fuh et al., 1999, 2001), Japan (Schreiner et al.,2000; Schreiner, 2001), India (Kar and Sharma,2001; Shaji et al., 2003), Turkey (Eker and Ertan,2000), Poland (Kloszewska, 1998) and Korea (Shahet al., 2004). Also, there are only a few cross-nationalstudies of BPSD: between Italy and USA (Binettiet al., 1998); between Nigeria, Jamaica and AfricanAmericans (Hendrie et al., 1996, 2000); betweenKorea and the UK (Shah et al., 2004); and, betweenHong Kong, Taiwan and the USA (Chow et al.,2002). There are few US studies (Cohen and Carlin,1993; Mintzer et al., 1996; Cohen et al., 1998a,b;

Received 25 May 2004Copyright # 2004 John Wiley & Sons, Ltd. Accepted 2 September 2004

*Correspondence to: Dr A. Shah, John Connolly Unit, West LondonMental Health NHS Trust, Uxbridge Road, Southall, MiddlesexUB1 3EU, UK. Tel: 0208 354 8140. Fax: 0208 354 8898.E-mail: [email protected]

Akpaffiong et al., 1999; Cohen and Magai, 1999;Chen et al., 2000; Harwood et al., 2001; Bassionyet al., 2002) and one Argentine study (Mangone,1996) of specific ethnic minority groups. These

studies have largely been of convenience samples(Table 1) of either Alzheimer’s disease (AD) or undif-ferentiated dementias. Instruments measuring BPSDin languages other than English, as illustrated inTable 2, have been developed with conventional trans-lation and back-translation techniques and subsequentevaluation of psychometric properties.

Table 3 illustrates the prevalence of depression insome American ethnic groups. Depression in demen-tia was less severe in African Americans than whiteAmericans in nursing homes, outpatient clinics andresearch center settings (Fabrega et al., 1988; Walkeret al., 1995; Cohen et al., 1998a,b; Cohen and Magai,1999). Table 4 illustrates the prevalence of disordersof mood in dementia in various settings in HongKong, Taiwan and Korea. The prevalence of depres-sion among community-dwelling African Americansin Indianapolis and Nigerians in Ibadan, and a conve-nience sample of Jamaicans was 14%, 4% and 6%respectively (Hendrie et al., 1996). The prevalenceof depression was similar in a convenience sampleof Alzheimer’s subjects in rural and Urban Taiwan,Hong Kong and America, ranging from 43%–49%(Chow et al., 2002).

Table 3 illustrates the prevalence of delusions andhallucinations in some American ethnic groups. TheBEHAVE-AD psychosis score was higher amongAfrican Americans than white Americans in outpati-ent settings (Cohen and Magai, 1999). Table 4 illus-trates the prevalence of delusions and hallucinations

Table 1. Samples in BPSD studies

Sample References

Out-patients Binnette et al., 1998; Cohenand Magai, 1999;Fabrega et al., 1988;Hendrie et al., 1996;Kloszewska, 1998;Harwood et al., 2000;Eker and Ertan, 2000;Fuh et al., 2001;Leung et al., 2001;Bassiony et al., 2002.

Tertiary Care Centre Chow et al., 2002.Dementia Research Cohen and Carlin, 1993;Centre Mangone, 1995; Chen et al., 2000.Psychogeriatric Service Shah et al., 2003ReferralsNursing Homes Cohen et al., 1998a,b;

Mintzer et al., 1996Lam et al., 1997;Schreiner et al., 2000;Schrenier, 2001.

Psychogeriatric inpatients Akpaffiong et al., 1999;Mintzer, 1996; Hwang et al., 1997;Kar and Sharma, 2001.

Population-based sample Hendries et al., 1996, 2000.Case-reports Ogunniyi et al., 2002

Table 2. BPSD insturments translated into languages other than English

Scale Language or Ethnic Group Reference

Rating Scale for Aggression in the Elderly Chinese Lam et al., 1997(RAGE) (Patel and Hope, 1992)Cohen-Mansfield Agitation Inventory (CMAI) Chinese Choy et al., 2001(Cohen-Mansfield, 1996) Japanese Schreiner, 2001

Black American Cohen et al., 1998bBlack American Akpaffiong et al., 1998

Cornell Scale for Depression in dementia Korean Shah et al., 2003(Alexopoulos et al., 1988) Japanese Schreiner and Morimoto, 2002Geriatric Depression Scale Black American Cohen et al., 1998b(Yesavage et al., 1983)Gestault Depression Scale Black American Cohen et al., 1998b(Abrahams and Alexopoulos, 1994)The Behavioural Pathology in Alzheimer’s Chinese Lam et al., 2001Disease Rating Scale (BEHAVE-AD) Korean Suh et al., 2001; Suh and Park, 2001(Reisberg et al., 1987)

Malayalam Shaji et al., 2003The Neuropsychiatric Inventory (NPI) Chinese Fuh et al., 2001; Leung et al., 2001(Cummings et al., 1994) Italian Binnetti et al., 1998

Japanese Hirono et al., 1997US Cubans Harwood et al., 2001

The Revised Memory and Behaviour Checklist Chinese Fuh et al., 1999(Teri et al., 1992) Spanish Harwood et al., 2001

1196 i. haider and a. shah

Copyright # 2004 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2004; 19: 1195–1204.

in dementia in various settings in Hong Kong, Taiwanand Korea. The prevalence of delusions in Nigerians,African Americans in Indianapolis and a conveniencesample of Jamaicans was 4%, 21% and 18% respec-tively (Hendrie et al., 1996); the prevalence of hallu-cinations in the same three groups was 4%, 4%, 22%respectively. The prevalence of delusions in AD suf-ferers in rural Taiwan, urban Taiwan, Hong Kong andAmerica was 40%, 31%, 58% and 40% respectively(Chow et al., 2002); the prevalence of hallucinationsin this group ranged from 14%–25%.

Apathy, irritability and agitation were more com-mon among Italian outpatients with Alzheimer’s dis-ease compared to American equivalents (Binnettiet al., 1998). Table 3 illustrates the prevalence of agi-tation in African and white Americans attending a

dementia assessment centre. Levels of agitationand aggression were higher among AfricanAmericans than white Americans in nursing homes(Mintzer et al., 1996). Table 4 illustrates the preva-lence of agitation and aggression in various settingsin Hong Kong, Taiwan, Korea and Japan. The preva-lence of agitation in rural Taiwan, urban Taiwan, HongKong and America among AD sufferers was 40%,46%, 61%, 49% respectively (Chow et al., 2002).

These studies illustrate that BPSD are common andan important feature of dementia in developing coun-tries and in ethnic minority groups in a given country.A study examining BPSD among Indian sub-continentelders in the UK was conducted because: (i) the pro-portion of ethnic elders as a function of all ethnicminority individuals has increased from 1% to 3%

Table 3. Prevalence of BPSD in some American ethnic groups

BPSD Dementia research centre Convenience Samples

African White African Asia Pacific Islanders Hispanic% % % % %

Depression 46a 35a 18b 60b 80b

Agitation 48a 47a

Hallucinations 16a 7a 29b 35b 53b

Visual hallucinations 7a 4a

Auditory hallucinations 13a 9a

Delusions 22a 10a 66b 75b 88b

Delusions of theft 15a 6a

Organised delusions 8a 6a

Data in nursing home residents were only available for depression.aCohen and Carlin, 1993.bChen et al., 2000.

Table 4. The prevalence of BPSD in various settings in Hong Kong, Taiwan and Japan

BPSD Hong Kong Hong Kong Taiwan Japan KoreaNursing Home Clinical Sample Out-patient sample Nursing Home New Referrals

Affective disturbance and Anxiety (amalgamated) 24a *54f

‘Pure’ Anxiety 54b

Phobia 19a **61f

Dysphoria 38b

Euphoria 5b

Delusions 32a 45b 75f

Hallucinations 15a 21b 57f

Aberrant motor behaviours 57b

Agitation 85c 45b

Aggression 64d 45e 61f

Activity-distrubance 54d 76f

Data on all the listed BPSD were not available in all settings in all three countries.aLam et al., 2001.bFuh et al., 2001.cChoy et al., 2001.dLam et al., 1997.eSchreiner, 2001.fShah et al., 2004; *pure depression; **combined anxiety and phobias.

bpsd and ethnicity 1197


between 1981 and 1991 (OPCS, 1983, 1993) and islikely to be higher in the 2001 census; (ii) there areno studies of BPSD in this UK ethnic group; and(iii) BPSD cause distress to patients, relatives andcarers (Everitt et al., 1991), result in institutional care(Eastley and Mian, 1993) or long-term hospitalisation(Shah, 1992), physical restraint (Werner et al., 1989)and over-medication (Everitt et al., 1991). Thus, apilot cross-cultural study of Indian subcontient originelders and white indigenous elders attending a de-mentia day hospital in the UK was undertaken withthe following aims: (i) to ascertain the prevalenceand correlates of various types of BPSD; (ii) to ascer-tain any differences in the prevalence and correlatesof BPSD between the two ethnic groups; and (iii) toascertain any similarities or differences which wouldbe worthy of further cross-cultural population-basedepidemiological study.

METHOD

Sample

This study was based in a geriatric psychiatry servicein West London. The overall service included two16-bedded acute admission wards, a 16-beddedrespite and rehabilitation ward, two 15-place demen-tia day hospitals, a 15-place functional day hospital,and four community multidisciplinary teams servinga catchment area of 40 000 elderly over the age of65 years. This study was confined to the catchmentarea served by one of the four community multidisci-plinary because patients of Indian sub-continent ori-gin were concentrated in this area (total elderlypopulation of 10 000). This team had access to onlyone of the two dementia day hospitals and the currentstudy was confined to that day hospital. Patientswould be admitted to the day hospital for assessmentand treatment of dementia or suspected dementia afteran initial assessment at their home by the consultant,specialist registrar or the staff grade doctor followinga referral from the general practitioner; not everyreferred patient with dementia was admitted to theday hospital as ‘uncomplicated’ dementias were in-vestigated as out-patients and more severe cases wereadmitted into hospital. At the day hospital, patientsand their carers would receive a multidisciplinaryassessment from psychiatrists, nurses, occupationaltherapists, physiotherapists, social workers and thevoluntary sector agencies.

Ethnic minority individuals were defined as thosewith a cultural heritage distinct from the majority whitepopulation (Manthorpe and Herriarachy, 1993). Indi-viduals, irrespective of the country of birth, of Indian,

Pakistani, Bangladeshi and Sri Lankan origin wereincluded in the group defined as ‘Indian sub-continentorigin’. Individuals from the majority Caucasianpopulation (white, British and English speakers) wereclassified as ‘indigenous’ and were utilised for com-parison. Patients from other ethnic minorities wereexcluded.

A consecutive series of admissions of Indian sub-continent origin elders to the dementia day hospitalwere examined. Thirty-one subjects of Indian sub-continent origin were identified. The comparisongroup of 31 patients comprised of white indigenouselders admitted to the day hospital.

Collection of other data

Data were extracted from the detailed case-notesusing a pre-designed structured schedule similar tothat used in previous studies in the same ethnic group(Redelinghuys and Shah, 1997; Odutoye and Shah,1999). Data on age, sex, marital status, ethnicity,country of birth, religion, first language, number ofchildren, duration in the UK, duration of the present-ing illness, length of stay in the day hospital, psycho-tropic and non-psychotropic medication and the MiniMental State Examination (Folstein et al., 1975) werecollected. The ICD-10 (WHO, 1992) diagnosis wereavailable for each subject following their assessmentin the day hospital, and these were collapsed intobroader groups.

Measurement of BPSD

Several features of BPSD, including aggressive beha-viour, depression and psychotic symptoms were mea-sured using the BEHAVE-AD (Reisberg et al., 1987).It measures disorders of behaviour, perception, moodand thought content (Reisberg et al., 1987). This 25-item instrument, with seven subscales, scores eachitem on a four-point (0–3) rating and gives an overallseverity score for the whole instrument and each sub-scale. It has good reliability and validity on a numberof parameters (Sclan et al., 1996; Reisberg et al.,1996). The BEHAVE-AD was originally used byReisberg et al. (1987) to retrospectively extract infor-mation on BPSD from case-notes. In the current studythe BEHAVE-AD was used to extract information onBPSD from case-notes using the same approach asused by Reisberg et al. (1987) in their original pio-neering study. Information pertaining to BPSDwhether occurring during the actual day hospitalattendance or that occurring at home and reportedby other informants like carers would have been



carefully recorded by the medical, occupational ther-apy and nursing staff in the case-notes.

Data analysis

Simple descriptive statistics, chi-square test (continu-ity corrected), Mann–Whitney U test and Spearman’srank correlation coefficient were used for analysis.

RESULTS

Sample

Thirty-one subjects of Indian sub-continent originwere admitted to the day hospital during the studyperiod, but only 28 case-notes were available datacollection. All 31 case-notes for indigenous com-parison group were available. The demographic andclinical characteristics of Indian sub-continent originand indigenous elders are illustrated in Table 5. Themedian scores on the BEHAVE-AD and it’s subscalesare illustrated in Table 6.

Comparison of the characteristics of the Indiansub-continent origin and the indigenousgroups (Tables 5–7)

Indian sub-continent origin elders had greater numberof children (Mann–Whitney U Test, Z¼�3.19,p¼ 0.001), lower severity score on the anxiety andphobia subscale of the BEHAVE-AD (Mann–WhitneyU Test, Z¼�2.6, p¼ 0.008) and a lower prevalence

of anxiety and phobias (X2¼ 5.09, 1d.f., p¼ 0.024).There were no differences between the two groupson any of the other measured variables.

Correlates of BEHAVE-AD in the Indiansub-continent origin sample

There was no significant association between totalBEHAVE-AD scores and age, sex, marital status,number of children, duration in the UK, duration ofthe presenting illness, length of stay in the day hospi-tal, precise diagnosis of dementia, MMSE scores, andprescription of neuroleptics, antidepressants, benzo-diazepines, mood stabilisers and drugs for physicalillness. This also applied to all the BEHAVE-AD sub-scales with two exceptions. BEHAVE-AD activitydisturbance subscale scores were higher in thosewith Alzheimer’s disease (Mann–Whitney U Test,Z¼�2.35, p¼ 0.036). BEHAVE-AD affective distur-bance subscale scores were higher in vascular demen-tia (Mann–Whitney U Test, Z¼�2.55, p¼ 0.036).

Correlates of BEHAVE-AD in the indigenous sample

There was no significant association between totalBEHAVE-AD scores and age, sex, marital status,number of children, duration of the presenting illness,length of stay in the day hospital, MMSE scores,precise diagnosis of dementia, and prescription ofneuroleptics, antidepressants, mood stabilisers, ben-zodiazepines and drugs for physical illnesses. Thisalso applied to all the BEHAVE-AD subscales with

Table 5. The demographic characteristics of Indians subcontinent origin and indigeous elders

Indian Indigeous Statistics

n % n %

Median Age (range) 78 (65–96) 79 (65–90) NSSex

Men 14 50 9 29 NSFemale 14 50 22 71

Marital StatusMarried 23 82 16 52 NSWidowed 5 18 15 48 NS

DiagnosisAlzheimer’s Disease 17 61 17 55 NSVascular dementia 6 21 8 26 NSOther dementias 5 18 6 19 NS

Duration of symptoms months, median (range) 36 (10–72) 48 (3–300) NSDuration in the UK years, median (range) 30 (0–45) NALength of stay in day hospital days, median (range) 113.5 (11–208) 111 (7–252) NSNumber of children Median (range) 3 (0–11) 2 (0–9) Mann–Whitney U Test,

Z¼�3.19 p¼ 0.001MMSE scores Median (range) 16.5 (5–24) 17 (4–25) NS

NS¼Not significant, NA¼Not Applicable.



a few exceptions. There was a positive correlationbetween age and the hallucinations subscale (rho¼0.42, p¼ 0.019). There was a negative correlationbetween diurnal rhythm disturbance subscale scoreand duration of the illness (rho¼�0.39, p¼ 0.032).There was a negative correlation between affectivedisturbance subscale score and length of stay in theday hospital (rho¼�0.37, p¼ 0.043), but thosereceiving antidepressants were likely to have higherscores on the affective disturbance subscale (Mann–Whitney U Test, Z¼�3.09, p¼ 0.02). Aggressivitysubscale scores were likely to be higher in men(Mann–Whitney U Test, Z¼�2.04, p¼ 0.041) andin those receiving neuroleptics (Mann–Whitney UTest, Z¼�2.26, p¼ 0.02).

DISCUSSION

This is the first study of BPSD of an ethnic minoritygroup in the UK. The main findings were: (i) therewere no significant differences between the twosamples on most demographic variables, prescriptionof various psychotropic drugs and drugs for physicalillness, precise diagnosis of dementia, duration of thepresenting illness, length of stay in the day hospital,MMSE scores, BEHAVE-AD total scores and on

all subscales scores (except the anxiety and phobiassubscale); (ii) Indian sub-continent elders had lowerscores on the anxiety and phobia subscale and a lowerprevalence of anxiety and phobias; (iii) within theIndian sub-continent origin group diagnosis of ADwas associated with higher scores on the activity dis-turbance subscale, and vascular dementia was associ-ated with higher scores on the affective disturbancesubscale; and (iv) within the indigenous group, malesand those receiving neuroletics had higher aggressiv-ity subscale scores, and those receiving antidepres-sants had higher affective disturbance subscale scores.

Before discussing the results, some methodologicallimitations are considered. The BEHAVE-AD wascompleted by reviewing the case-notes and datamay be of variable quality; this was minimised byone of the authors (IH) carefully and systematicallyreviewing the case-notes by applying the definitionof each BEHAVE-AD item as described by Reisbergand colleagues (1987). This strategy is similar tothat used by Reisberg et al. (1987) in their originalpioneering study using the BEHAVE-AD. TheBEHAVE-AD has emerged as a semi-structured inter-view subsequent to its original development, whereit was used to extract data from case-notes. Dataon other parameters were carefully defined and

Table 6. Comparison of BEHAVE-AD and subscale scores between the two groups

Indian Median (range) Indigenous Median (range) Statistics (Mann–Whitney U Test)

Total BEHAVE-AD score 5 (0–15) 4 (0–28) NSBEHAVE-AD subscale scores

Delusions 0 (0–6) 0 (0–8) NSHallucinations 0 (0–4) 0 (0–4) NSActivity disturbance 0 (0–3) 0 (0–6) NSAggressivity 0 (0–7) 0 (0–8) NSAffective disturbance 0 (0–3) 0 (0–3) NSDiurnal rythm disturbance 0 (0–2) 0 (0–3) NSAnxiety & phobia 0 (0–1) 0 (0–3) Z¼�2.6, p¼ 0.008

NS¼Not significant.

Table 7. Prevalence of BPSD

Indians Indigenous StatisticsChi Square

N % N %(28) (31)

Delusions 7 25 9 29 NSHallucinations 7 25 5 16 NSActivity disturbance 11 39 10 32 NSAggressivity 13 56 12 38 NSDiurnal rythm disturbance 4 14 4 13 NSAffective disturbance 7 25 9 27 NSAnxiety and phobia 1 4 9 29 X2¼ 5.09, p¼ 0.024

All calculations 1 degree of freedom, NS¼Not significant.



systematically collected from the case-notes using apre-designed structured schedule similar to that usedin previous studies (Redelinghuys and Shah, 1997;Odutoye and Shah, 1999). The researcher was unableto be blind to the ethnicity status of the patients. Thesample size was small and it is possible that the studymay not have been adequately powered for evaluatingdifferences between the two ethnic groups for theBEHAVE-AD items. This study was of a conveniencesample of patients from one ethnic minority groupadmitted to one dementia day hospital, and the find-ings are likely to be biased by clinical factors thatdetermine admission to this day hospital. Thus, thefindings need to be viewed with caution when con-sidering other ethnic groups and patients in othertreatment settings.

The definition and identification of ethnic minoritypatients can pose difficulties (McKenzie and Crowcroft,1996). A widely accepted contemporary definition ofethnicity was used (Manthorpe and Hettiaratchy,1993). The ethnicity of patients was identified by anextensive review of the case-notes, where data oncountry of birth, country of origin, language and re-ligion were recorded. Patients from several Indiansubgroups were amalgamated; such amalgamationhas been criticized because individual ethnic subgro-ups can and do differ. However, this argument is morepowerful when allocation of resources is consideredrather than in a preliminary pilot study such as this.

Indian sub-continent origin patients had greaternumber of children and this simply reflects populationdemography consistent with previous literature onclinical samples of this ethnic group in this area(Redelinghuys and Shah, 1997; Odutoye and Shah,1999). There were no differences between the twogroups on MMSE scores, and BEHAVE-AD totaland subscale scores (except for the anxiety and pho-bias subscale). There may be several explanations forthese findings including the methodological issuesdiscussed above. First, it may be an artifact due toType 2 statistical error. Second, documentation ofBPSD in the case-notes may have been poor or themethod of data collection was unreliable. Neitherexplanation is likely as anecdotally the day hospitalstaff appeared to be very meticulous in their notekeeping and the method of data extraction from thecase-notes was similar to that used by Reisberg et al.(1987) in their original pioneering work; also, case-notes from this day hospital were used to extract datain an earlier study with evidence of face validity(Redelinghuys and Shah, 1997). It is possible thatthe BEHAVE-AD, as used in this study, does not ade-quately measure BPSD in some ethnic groups.

However, findings from correlates of BPSD withinthe two ethnic groups further support the validity ofthe BEHAVE-AD as used in the current study. Withinthe Indian sub-continent group diagnosis of AD wasassociated with increased activity disturbance andthat of vascular dementia with increased affective dis-turbance, and both these findings are consistent withthe literature (Shah, 1999; Brodaty and Luscombe,1996). Within the indigenous group aggressivityscores were associated with males and prescriptionof neuroleptics, both of which are consistent withthe literature (De Deyn et al., 1999; Katz et al.,1999; Shah, 1999). Similarly, the association betweenaffective disturbance and prescription of antidepres-sants in the indigenous group is consistent with thegrowing literature on the treatment of depression indementia with antidepressants (Katona et al., 1998).Third, the consultant psychiatrist or the junior doctorsfirst assessed the patient at home, after referral fromthe general practitioner, before admitting them tothe day hospital for further assessment. These clini-cians are likely to apply similar clinical criteria toboth groups whilst selecting patients for the day hos-pital and this may explain any absence of differencesbetween the two groups on BPSD. Finally, the find-ings may be genuine and independent of any of themethodological flaws.

Indian sub-continent origin patients had a lowerprevalence and lower severity scores for anxiety andphobias subscale on the BEHAVE-AD. The 4% pre-valence of anxiety and phobias among the Indiansub-continent origin patients is also lower than the18% and 13% prevalence of anxiety and fearfulnessrespectively among psychiatric inpatients with undif-ferentiated dementia in India (Kar and Sharma, 2001);however, comparison of our sample with that fromIndia is problematic as it does not take into accountmigration, acculturation and changes in lifestyleand cultural values. This low prevalence is difficultto explain and may in part be due to the methodologi-cal issues discussed earlier. However, health seekingbehaviour of patients and carers, and cultural factorsmay also be important explanations (Shah et al.,2004).

Caution should be exercised in generalising thefindings of this study to other psychogeriatric servicesand ethnic groups. The findings of this study cannotbe assumed to apply to all ethnic minority groupsbecause the clinical features and presentation ofBPSD are likely to be influenced by migration, degreeof acculturation, life style changes, language, educa-tion, cultural values, health seeking behaviour, stigmaand the ability of clinicians to recognise BPSD in



ethnic minority patients. Although all the findingscould be potentially explained by differing accuracyof the BEHAVE-AD in the two groups, health seekingbehaviour of patients and relatives, cultural factors,and knowledge, expectation and recognition of BPSDamong professionals, the possibility that the findingsmay be genuine remains. This ideally requires furtherstudy in a cross-cultural population-based epidemio-logical studies. Such a study would allow elucidationof genetic and environmental factors and the effect ofgene-environment interaction on the developmentof BPSD. The well established cross-cultural studiesof cognitive impairment in dementia elegantly illus-trate this. However, in order to avoid possible biasintroduced by retrospecive used of BEHAVE-ADand the influence of cultural factors, there is a clearneed to formally develop and evaluated translated ver-sions of instruments measuring BPSD, such as theBEHAVE-AD, in this and other ethnic groups beforeconducting population-based studies. Development ofinstruments measuring BPSD in different ethnic min-ority groups should follow the well established princi-ples of cross-cultural development of measurementinstruments (Shah and Lindesay, 2000).

ACKNOWLEDGEMENTS

We wish to thank the Penny Sangam Day Hospitalstaff for their generous help and Dr Geetha Oommenfor allowing us to study case-notes of her patients.Comments of the two referees were greatly appreciated.

REFERENCES

Abrahams RC, Alexopoulos GS. 1994. Assessment of depression indementia. Alz Dis Rel Disord 8: S227–S229.

Akpaffiong M, Kunik ME, Hale D, Molinari V, Orengo C. 1999.Cross-cultural differences in demented geropsychiatric inpati-ents with behaviour disturbances. Int J Geriatr Psychiatry 14:845–850.

Alexopoulous GS, Abrahms RC, Young RC et al. 1988. CornellScale for Depression in Dementia. Biol Psychiatry 23: 271–284.

Bassiony MM, Warren A, Rosenblatt A, et al. 2002. Isolated hallu-cinosis in Alzheimer’s disease is associated with African Amer-icans race. Int J Geriatr Psychiatry 17: 205–210.

Binette G, Mega MS, Magni E, et al. 1998. Behavioural disorders inAlzheimer’s disease: a transcultural perspective. Arch Neurol 55:539–544.

Brodaty H, Luscombe G. 1996. Studies on affective symptoms anddisorders: depression in persons with dementia. Int Psychogeria-tris 8: 609–622.

Chen JC, Borson S, Scanlan JM. 2000. Stage-specific prevalence ofbehavioural symptoms in Alzheimer’s disease in a multi-ethniccommunity sample. Am J Geriatr Psychiatry 8: 123–133.

Chow TW, Liu CK, Fuh JL, et al. 2002. Neuropsychiatric symp-toms of Alzheimer’s disease differ in Chinese and Americanpatients. Int J Geriatr Psychiatry 17: 22–28.

Choy CNP, Lam LCW, Chan WC, Li SW, Chiu HFK. 2001. Agita-tion in Chinese Elderly: validation of the Chinese version of theCohen-Mansfield Agitation Inventory. Int Psychogeriatr 13:325–335.

Cohen CI, Carlin L. 1993. Racial differences in clinical and socialvariables among patients evaluated in a dementia assessmentcentre. J Nat Med Assoc 85: 379–384.

Cohen CI, Hyland K, Magai C. 1998a. Interracial and intraracialdifferences in neuropsychiatric symptoms, sociodemographyand treatment among nursing home patients with dementia.Geontologist 38: 353–361.

Cohen CI, Hyland K, Magai C. 1998b. Depression among AfricanAmerican Nursing Home patients with dementia. Am J GeriatrPsychiatry 6: 162–175.

Cohen CI, Magai C. 1999. Racial differences in neuropsychiatricsymptoms among dementia outpatients. Am J Geriatr Psychiatry7: 57–63.

Cohen-Mansfield J. 1986. Agitated behaviour in the elderly II:preliminary results in cognitively deteriorated. J Am GeriatrSociety 34: 722–727.

Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, CarusiDA, Gorbein J. 1994. The neuropsychiatric inventory: compre-hensive assessment of psychopathology in dementia. Neurology44: 2308–2314.

De Deyn PP, Rabheru K, Rasmussen A, et al. 1999. A randomisedtrial of risperidone, placebo, and haloperidol for behaviouralsymptoms of dementia. Neurology 53: 946–955.

Eastley RJ, Mian IH. 1993. Physical assaults by psychogeriatricpatients: patient characteristics and implications for placement.Int J Geriatr Psychiatry 8: 515–520.

Eker E, Ertan T. 2000. Behavioural and psychological symptoms ofdementia in Eastern and southeastern Europe and the middleeast. Int Psychogeriatrics 12(Suppl. 1): 409–413.

Everitt DE, Fields DR, Soumeraii SS, Avorn J. 1991. Residentbehaviour and staff distress in nursing homes. J Am GeriatrSociety 39: 792–798.

Fabrega H, Mezzich J, Ulrich RF. 1988. Black-white differencesin psychopathology in an urban psychiatric population. CompPsychiatry 29: 285–297.

Finkel SI. 1996a. New focus on behavioural and psychologicalsigns and symptoms of dementia. Int Psychogeriatrics8(Suppl. 3): 215–216.

KEY POINTS

� There is a paucity of cross-cultural studies of

behavioural and psychological symptoms of

dementia.

� This study did not reveal any major differences

in BPSd between patients of Indian subconti-

nent origin.

� There is a need to develop instruments measur-

ing BPSD in different ethnic minority groups.

� There is a case for cross-cultural population-

based epidemiological studies of BPSD.



Finkel SI. 1996b. Research methodological issues in evaluatingbehaviour disturbance in dementia. Int Psychogeriatrics8(Suppl. 2): 149–150.

Foli S, Shah AK. 2000. Measurement of behaviour disturbance,non-cognitive features and quality of life. In Dementia, O’BrienJ, Burns A, Ames D (eds). Arnold: London; 87–100.

Folstein MF, Folstein SE, McHugh PR. 1975. Mini Mental State:a practical method of grading the cognitive state of patients forthe clinician. J Psychiatr Res 12: 189–198.

Fuh JL, Liu CY, Mega MS, Wang SJ, Cummings J. 2001. Beha-vioural disorders and caregivers’ reaction in Taiwanese patientswith Alzheimer’s disease. Int Psychogeriatrics 13: 121–128.

Fuh JL, Liu CY, Wang SJ, Wang HC, Liu HC. 1999. Revised mem-ory and behaviour checklist in Taiwanese patients with Alzhei-mer’s Disease. Int Psychogeriatrics 11: 181–189.

Harwood DG, Baker WW, Ownby RL, Bravo M, Aguero H, DuaraR. 2001. The Behaviour Problems Checklist-Spanish: a preli-minary study of a new scale for the assessment of depressivesymptoms and disruptive behaviour in Hispanic patients withdementia. Int Psychogeriatrics 13: 23–35.

Hendrie HC, Baiyewu O, Eldermire D, Prince C. 1996. Cross-cultural perspectives: Caribbean, Native American and Yoruba.Int Psychogeriatrics 8(Suppl 3): 483–486.

Hendrie H, Gao S, Baiyewu O. 2000. A comparison of symptomsof behavioural disturbance in Yoruba and African Americansindividuals with dementia. Int Psychogeriatrics 12(Suppl. 1):403–408.

Hirono S, Mori E, Ikejiri Y, Imamura T, Shimomura T. 1997. AJapanese version of Neuropsychiatric Inventory: utility of theassessment for psychiatric symptoms in dementia. Brain &Nerve 49: 266–271.

Hwang JP, Yang CH, Tsai SJ, Liu KM. 1997. Behavioural distur-bances in psychiatric inpatients with dementia of Alzheimer’sType in Taiwan. Int J Psychiatry 12: 902–906.

Jitapunkal S, Lailert C, Worakul P, et al. 1996. Chula Mental Test:a screening test for elderly people in less developed countries.Int J Geriatr Psychiatry 11: 715–720.

Kar N, Sharma PSV. 2001. Behavioural and psychological symp-toms in dementia—clinical features in an Indian population.Int J Geriatr Psychiatry 16: 537–542.

Katona CLE, Hunter BN, Bray J. 1998. A double blind comparisonof the efficacy and safety of paroxetine and imipramine in thetreatment of depression in dementia. Int J Geriatr Psychiatry13: 100–108.

Katz IR, Jeste D, Mintzer J, Clyde C, Napolitano J, Brecher M.1999. Comparison of risperidone and placebo for psychosisand behavioural disturbances associated with dementia: a rando-mised, double blind trial. J Clin Psychiatry 60: 107–115.

Kloszewska I. 1998. Incidence and relationship between beha-vioural and psychological symptoms of Alzheimer’s disease.Int J Geriatr Psychiatry 13: 785–792.

Lam LCW, Chiu HFK, Ng J. 1997. Aggressive behaviour in theChinese elderly-validation of the Chinese version of the ratingscale for aggressive behaviour in the elderly (RAGE) in hospitaland nursing home settings. Int J Geriatr Psychiatry 12: 678–681.

Lam LCW, Tang WK, Leung V, Chiu HFK. 2001. Behaviouralprofile of Alzheimer’s disease in Chinese elderly—a validationstudy of the Chinese version of the Alzheimer’s disease beha-vioural pathology rating scale. Int J Geriatr Psychiatry 16:368–373.

Leung VPY, Lam LCW, Chiu HFK, Cummings JL, Chen QL. 2001.Validation study of the Chinese version of the neuropsychiatricinventory (CNPI). Int J Geriatr Psychiatry 16: 789–793.

Mangone CA. 1996. Cross perspectives: Argentina. Int Psychoger-iatrics 8(Suppl. 3): 473–478.

Manthorpe J, Hettiaratchy P. 1993. Ethnic minority elders inBritain. Int Rev Psychiatry 5: 173–180.

McKenzie K, Crowcroft NS. 1996. Describing race, ethnicity andculture in medical research: describing the group is better thantrying to find a catch all name. Br Med J 312: 1051.

Mintzer JE, Nietert P, Costa K, Waid LR. 1996. Cross-culturalperspectives: agitation in dementia patients in the United States.Int Psychogeriatrics 8(Suppl. 3): 487–490.

Odutoye K, Shah AK. 1999. The characteristics of Indian subcon-tinent origin elders newly referred to a psychogeriatric service.Int J Geriatr Psychiatry 14: 446–453.

Office of Population Census and Surveys. 1983. 1981 Census:Country of Birth. Great Britain. HMSO: London.

Office of Population Census and Surveys. 1993. 1991 Census:Ethnic Group and Country of Birth. Great Britain. OPCS:London.

Ogunniyi A, Akang EEU, Gureje O, et al. 2002. Dementia withLewy Bodies in A Nigerian: a case report. Int Psychogeriatrics14: 211–218.

Patel V, Hope T. 1992. A rating scale for aggressive behaviour in theelderly—the RAGE Scale. Psychol Med 22: 211–221.

Redlinghuys J, Shah AK. 1997. The characteristics of ethnic eldersfrom Indian subcontient using a geriatric psychiatry service inWest London. Ageing and Mental Health 1: 243–247.

Reisberg B, Borensteen J, Sabbs S, et al. 1987. Behavioural symp-toms in Alzheimer’s disease: phenomenology and treatment.J Clin Psychiatry 48(Suppl.): 9–15.

Reisberg B, Auer SA, Monteiro I, et al. 1996. Behavioural distur-bances of dementia: an overview of phenomenology and metho-dologic concerns. Int Psychogeriatrics 8(Suppl. 2): 169–180.

Schreiner A, Shiotain H, Yamamoto E. 2000. Agitated behaviourin elderly nursing home residents with dementia in Japan.J Gerontol 55B(3): P1–P7.

Schreiner AS. 2001. Aggressive behaviours among dementednursing home residents in Japan. Int J Geriatr Psychiatry 16:209–215.

Schreiner AS, Morimoto T. 2002. Factor structure of the CornellScale for Depression in Dementia among Japanese poststrokepatients. Int J Geriatr Psychiatry 17: 715–722.

Sclan SG, Saillon A, Franssen E, et al. 1996. The Behaviour Pathol-ogy in Alzheimer’s Disease Rating Scale (BEHAVE-AD): relia-bility and analysis of symptom category score. Int J GeriatrPsychiatry 11: 819–830.

Shah AK. 1999a. Aggressive behaviour in the elderly. Int J Psychia-try Clin Prac 3: 85–103.

Shah AK. 1992. Violence and psychogeriatric inpatients. Int J GeriatrPsychiatry 7: 39–44.

Shah AK, Dighe-Deo D. 1998. Elderly Gujaratis and psychogeria-trics in a London psychogeriatric service. Bull Int PsychogeriatrAssoc 14: 12–13.

Shah AK, Elanchenny N, Suh G. 2004. A comparative study ofbehavioural and psychological symptoms in patients withdementia referred to a psychogeriatric service in Korea and theUK. Int Psychogeriatrics 16: 219–236.

Shaji S, Bose S, Jacob Roy K. 2003. A study of behavioural andpsychological symptoms in Alzheimer’s Disease: Alzheimer’sand Related Disorders Society of India. Dementia News 3: 5.

Suh GH, Park JH. 2001. The Behaviour Pathology in Alzheimer’sDisease Rating Scale, Korean Version (BEHAVE-AD-K): factorstructure among Alzheimer’s disease inpatients. J Kor GeriatrSoc 5: 86–91.



Suh GH, Son HG, Shin H, et al. 2001. Reliability and analysis ofsymptom category scores of the Behaviour Pathology in Alzheimer’sDisease Rating Scale, Korean Version (BEHAVE-AD-K). J KorGeriatr Soc 5: 50–57.

Teri L, Traux P, Logsdon R, Uamoto J, Zarit S, et al. 1992. Assess-ment of behavioural problems in dementia: the revised memoryand behaviour checklist. Psychol Ageing 7: 627–631.

Walker S, Borson S, Katon W, et al. 1995. Differential characte-ristics of black and white nursing home residents. Am J GeriatPsychiatry 3: 229–238.

Werner P, Cohen-Mansfield J, Braun J, Marx MS. 1989. Physicalrestraint and agitation in nursing home residents. J Am GeriatrSoc 37: 1122–1126.

World Health Organisation. 1992. The ICD-10 Classification ofMental and Behavioural Disorders: Clinical Descriptions andDiagnostic Guidelines. WHO: Geneva.

Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, et al. 1983.Development and validation of a geriatric depression scale: apreliminary report. J Psychiat Res 17: 37–49.



Documents

A pilot study of behavioural and psychological signs and symptoms of dementia in patients of Indian sub-continent origin admitted to a dementia day hospital in the United Kingdom