10.1192/bjp.161.6.802 Access the most recent version at doi:
1992 161: 802-808 The British Journal of Psychiatry
E Moore, RA Ball and L Kuipers
Expressed emotion in staff working with the long-term adult
Article cited in:
To obtain reprints or permission to reproduce material from this paper, please write
to this article at
You can respond
top right corner of the article or
Receive free email alerts when new articles cite this article - sign up in the box at the
The Royal College of PsychiatristsPublished by
on July 10, 2011 bjp.rcpsych.org
go to: The British Journal of PsychiatryTo subscribe to
British Journal of Psychiatry(1992),161,802â€”808
ExpressedEmotion in Staff Working
with the Long-Term
E. MOORE, A. A. BALL and L. KUIPERS
Staffâ€”patient relationships in long-term settings were examined in 35 staff and 61 patients.
Measures were also taken of the staff's general health, their coping style in relation to work
events, andjob satisfaction. A rangeof ratingsof EEwas evident instaff descriptionsof patients
undertheircare.Strainandcriticismin the relationshipwerenot associatedwith identified
stressorsinthe workplace,orthe generalhealthof the carer.Whenpatientsweregrouped
according tohigh-EE and low-EEinterviews, there were no significant
and attention-seeking behaviour, underactivity, and limited social interaction. The findings
haveimplications for staff training andfor themaintenance ofoptimal staff-patient
inservices supporting severely disabled patients.
Staff who work with chronically disabled psychiatric
patients have to deal with a widevariety of behaviour,
from high levels of physical or verbal abuse to low
motivation and self-neglect.
careoffered, anddifficulties maybeslowto change
(Bennett & Morris, 1983; Woods & Cullen, 1983;
Hall, 1989), while re-emergence of florid symptoms
is a constant risk (Isaacs & Bebbington, 1991). The
number of patients with severe, chronic disabilities
remains high (Mann & Cree, 1976), despite active
Patients inthis long-term group arelikely toneed
high levels of careand supportfor many years
Hall (1990) emphasisesthat care is a mutual
process, with the same importance attributed to the
â€œ¿?beliefs,goals, practices and emotions of the carer
as to the cared-forâ€•. It is known that empathy
and support require a considerable expenditure
of psychological energy from both professional
Bebbington, 1990), especially in those cases where
the patient's problems are many, various, and
resistantto help(Cherniss,1980).Typically, chronic
patientstakealongtime to respondto programmes,
so that stability and consistencyof approach are
Professional carers, whoserole is to form long
standing relationshipswith long-term
patients, would seem to have an advantage over
family carers: they have specific training
less time with patients. However, they may also be
vulnerable to those consequences of employment
with this client group, such asemotional exhaustion
and low personal accomplishment, that predict a
change of job (Jackson et a!, 1986). High staff
Patients may reject the
turnover and low staffing levels are not uncommon
in psychiatric settings (Sutton, 1981;Lavender, 1985)
and can bedisruptive (Jameseta!, 1990;Thorrncroft,
1991).Thus it was decided to investigate some of the
attributesof long-termwork with psychiatricpatients
that enable staff to stay in post and feel positive
about the work they do, and which may alsoaffect
the type of care that patients receive.
An index of the relationship betweenclient and
carer thathas been of value in predicting
hasbeenexpressed emotion(EE). EE isnowa well
established and replicated measure, although it
has traditionally been applied to relatives (Leff
& Vaughn, 1985; Kuipers & Bebbington, 1988).
However, there is considerable overlap between
behaviour and problems that staff and relatives
haveto dealwith in caring for patientswith chronic
illnesses (Creer et a!, 1982). Both Watts (1988) and
Herzog (1992)reported finding a full rangeof EE
in relationships between psychiatric patients and
staff. Herzog also suggested that the valuable
contribution of nursing and other direct-carestaff
had been overlooked. Indeed, one possible dis
incentive to offering long-term support aspart of
a multifaceted programme is that progress may
be difficult to attribute to any one intervention
(Cherniss, 1980). Staff may thus be deprived of the
vital sense of personalaccomplishmentthat helpsto
maintain enthusiasmand commitment.
The aim of the presentstudy wasto examinethe
staffâ€”patientrelationship using the standard EE
procedure, and to relate EE to stress, coping,
and job satisfaction in staff. It waspredicted that
staffâ€”patient EE ratings would be comparable to
those in relativeâ€”patientrelationships. Burden
associatedwith the careof disturbed relativesis a
well documented finding (Creer eta!, 1982; Fadden
eta!, 1987;Bledineta!, 1990;Jacksoneta!, 1990).It
wasthusfurther predictedthat high-EErelationships
with higher levels of stress, less adaptive coping
styles, and less job satisfaction in the carer.
The patients selectedhad all been in contact with one
of four servicesin the Camberwellareafor at leasttwo
years.Theywerelivingin oneof threeresidential hostel
units, or receiving day or residential care from a specialised
centre designed to meet the needs of the long-term adult
mentally ill. Thefacilities in Camberwell,whencompared
nationally, are considered to be of a high standard
(Salokangaset al, 1985).The serviceis comprehensive,
including daytime activities, a drop-in centre, in-patient
residential care, and associatedhostel care, so that a
continuous programme of support can be provided for
Initially, patients' casenotes werescreenedto ensure
that the primary diagnosiswaseither major psychosisor
neurotic disorder. Three patients with unusual primary
diagnoses,suchasepilepsyor personality disorder, were
excluded at this stage.From this samplepatients were
randomlyselected. Patientswerethuseligiblefor inclusion
if they suffered from a non-organic
had beenin hospital continuously for a minimum of one
year before transfer to oneof the facilities, and lived in
ahostelor attendedthedaycentrefor morethan2Â½ days
Thestaffworkedin multidisciplinary teams attheday
hospitalandin two of thethreehostels.Theywereincluded
iftheyhad beenintheir jobforatleast three monthsbefore
interview, and were key workers to the patients about
whom they actedasprimary informants. Most members
of staff (26 of 35) were interviewed on more than one
occasion. The patients about whom they talked were
selected at randomfrom their list of keypatientswho met
the inclusioncriteria. Wherethestaff hadalargenumber
of keypatients(a meanof 17in the daycentrecompared
with a meanof 4 in the hostels),the list wasreducedto
thosewith whom they had more regular contact.
The patients' clinical status was assessed using the
Present State Examination (PSE; Wing et al, 1974).
Interviews were conducted, before the rating of other
interview data, by RAB and EM, who wereboth trained
to usethe PSE. Information concerningpastpsychiatric
historywasrecorded alsofrom casenotes.
The key workers' interview (see below) included a
version of the Social Behaviour Schedule(SBS; Wykes
& Sturt, 1986),and provided additional information on
patients' symptomsandbehaviourproblems.This check
list rated items on a three-point scale,where0 indicated
absenceof the symptomor behaviour, that is, behaviour
which is acceptable,and a scoreof 2 is givenif the item
in question is present, and with some intensity, more
than half the time. The reliability of this typeof measure
hasbeenestablished (Wykes& Sturt, 1986).TheSBSscores
were added to provide an overall indication of behavioural
disturbance,and certain items werecombined to assess
patientdifficulty in four relatedareas:retardation; social
interaction; neuroticproblems; andbehavioural problems
(Brewin et a!, 1990).
Staff memberswere interviewed during work hours
usingaversionof theCamberwellFamily Interview (CFI;
Vaughn& Leff, 1976).This wasadaptedfor usewith a
non-relativeinformant andachronically ill population by
removing obviously inappropriate items, for example
â€œ¿?have you ever stayedaway overnight?â€•, but mirrored
the original ascloselyaspossibleto retain the reliability
and validity previously achievedwith the CFI (Leff &
Vaughn,1985).The interviews(conducted by EM and
taped with the participant's consent, and subsequently
rated for indices of EE: critical comments, hostility,
time budgets. The percentage of time that the staff member
spentworking with thepatientin question,asopposedto
other duties(e.g. administration) and contact with other
patients, wasalso calculated.
The staff interview was semistructured,and involved
questionsconcerning what the informants knew of the
history of the patients' problems and current difficulties,
thenatureof their presentrelationship,andtheemotional
strain(if any)that theworkersfelt in theirdailycaring
role with the patient. Answers to thesequestionswere
recorded by the interviewer using a three-point coding
system:0 indicatedno difficulty; 1, difficulty on some
occasions; and 2 represented significant
referenceeither to the staff members'feelingsor to the
patients'ability, depending on thequestion).
Staff werethen askedto completethe 60-itemGeneral
Health Questionnaire(GHQ; Goldberg, 1978).A 32-item
questionnaireabout coping style wasalso administered.
This was adapted for a work situation from the Adult
Health and Daily Living Form devised by Billings &
Moos(1981),byaskingthekeyworkerto state(in writing)
a particular event or aspectof their work which they
foundparticularlystressful andto relatetheirresponses
to that statement.Subjectsthenratedtheiruseof particular
coping mechanismson a four-point scale.Three coping
styleswereelicited from thequestionnaireto yield a total
score for avoidance, cognitive and behavioural strategies,
in relation to work events.
Finally, staff wereaskeddirectquestionswith respectto
their intention to remainin their job, the mostimportant
sourceof satisfactionthey experiencedat work, and the
aspect of the job they found least satisfying. This method
hasyieldedvaluableinformationin comparable studies
(Firth, 1986).Answersto thesequestionswerecategorised
accordingto a simplecoding system.
Expressed emotionwasratedby anexperienced rater
(LK), whowasbothblindtopatientdiagnosis andtostaff
was assessed by computingan intraclasscorrelation
coefficient (ICC) for eachof thecomponentscalesof EE
on eightrandomlyselected interviews.Completeagreement
(s.d.) score forpositiveremarks2.3(2.5)1.4
(s.d.) score forwarmth2.4
MOORE ET AL
was achieved on EE category (high or low) and the
presence andamount of hostility. ICCs for agreementon
There were 61 patients, whoseagesranged from 24 to
76years(mean45.5,s.d. 13.5).Thenumberof yearssince
their contact with psychiatric services ranged from 3 to 45
(mean 19.8, s.d. 11.1), and the number of years with
the unit rangedfrom 1to 20 (mean5.7, s.d. 4.4). At the
time of interview, 21weredaypatients,andtheremaining
40werehostelor in-patient residents.Therewere32men
and29women.Fiveethnicoriginswererepresented in the
sample: white British (n= 41), Afro-Caribbean (n= 15),
Irish (n= 3), Greek Cypriot (n= 1), and Asian (n= 1).
fewpatientshadaconsistentrecordof work achievement.
Severalpatientshadbeenmarried, but the majority were
without parental responsibility.
Only two patients were not taking any form of
medication; 57 were prescribedneuroleptic medication.
Using the CATEGO program, patients were classified as
follows: schizophrenia,26; paranoid psychosis,5; manic
and mixed affective psychosis,5; other psychoses,12;
uncertain, 13.In this sample,83.6Â°lo of thoseinterviewed
on thePSEhadIndexof Definition(ID) scores greater
than or equal to 5. The 10 subjectswhosediagnosisis
category.Wing et a! (1978)reportedthat the percentages
of subjects with scores above a threshold level of 5 in their
l00Â°lo, 77% and 20%, respectively.CATEGO diagnosesof
schizophreniaor other psychoses accountedfor 78.7% of
Of the 35 members of staff,
men. Their agesrangedfrom 22 to 59years(mean36.7,
s.d. 10.5). The length of time staff had been in their present
post rangedfrom 5 months to 27 years(mean7.1 years,
s.d. 6.6). Their experience in workingwith psychiatric
patients ranged from approximately 1 year to 27 years
(mean7.1, s.d. 6.5). At the time of interview,6 staff
members were charge nurses, 14 staff
enrolled nurses,9 nursing assistants,and 4 occupational
28 were women,seven
nurses, 2 state
Thelevelof criticismperinterviewrangedfrom nocriticism
to 20 criticalcomments (mean 3.3, s.d. 4.7), and the
numberof positiveremarksrangedfrom 0to 12(mean2.1,
s.d.2.3).Hostilitywaspresent in 11interviews, andthe
meanrating for warmth was2.1 (s.d. 1.1,range0â€”4). In
contrast to relativesof schizophrenicpatients,emotional
overinvolvementwasnot apparentin thestaff sample.The
profileof theEEcomponent scores for theinterviews and
the averageamount of time spentin face-to-facecontact
are presentedin Table 1.
Ratingsof staff on EEcomponentscalesin interviews
relating to 61 patients
Thecriticalcomments scoreof 6or moreandthepresence
of hostility were used to distinguish the high-EE from
low-EE interviews(Leff & Vaughn, 1985).This affords a
direct comparison with EE in the relatives of patients
with schizophrenia.No hostility
members whomadelessthan6criticalcomments. Seven
staff members (20%) made no criticism at all of their key
patient. Fifteen staff members(42.9Â°lo) made6 or more
criticismsin at leastoneinterview. Twenty staff members
(57%) wererated low EE and four (11.4%) high EE in
all their interviews.Analysisof thetotal sampleindicated,
however, that scoresderived from the staff interviews
weredependenton the patient in question rather than a
constantapproachtaken by the staff member.This was
confirmed by a low ICC (0.2) derived from a one-way
analysisof varianceof criticalcommentsperstaff member.
Thus staff members might be rated as low-EE when
discussing onepatientandhigh-EEwhendiscussing another
(n= 11,31.4Â°lo), suggesting anindependence ofinterviews.
was expressed by staff
The independence of interviewsmeantthat it wasnot
necessary to accountfor the nestedsamplingdesign(i.e.
somestaff membersdiscussing morethan onekeypatient)
in thefollowing analyses. Thehigh-EEandlow-EEgroups
wereclassifiedin two different ways.The first (method1)
involved averaging the number of critical comments
per staff member across the interviews in which they
participated; the second(method 2) by using the staff
member's highest scorefor critical comments.The two
groups werethen comparedusing t-tests.
Thegroupsdid not differ significantly on anysocialor
demographic variables, including
they had spentin their job, and the position they held,
irrespectiveof the method of classification used.
Staff members classified as high EE by method 1
reportedspendingsignificantly moretime than their low
(t= â€”¿?2.75,d.f. = 59,p<0.Ol), and proportionallymoretime
with thesepatientsthanin otherdutiesor with otherpatients
age, the length of time
(1= â€”¿?2.56,d.f. =59,
participants could not be distinguished by their GHQ
score,total scoreson thecopingstrategiestheyemployed
at work, or theirlevel of satisfaction
That contact time was reported as being greater in
high-EE relationships was evident (t= â€”¿?
P.czo.05) using method2. High-EE
behavioural (t= â€”¿? 3.28, d.f. = 50, P<0.Ol), emotionally
discharging (1= â€”¿?2.50, d.f. = 50, P<0.05),
seeking (1= â€”¿? 2.77, d.f. = 50, P<0.0l),
solving(t= â€”¿? 2.19,d.f. = 50,P<0.05) copingstrategies for
work-related stressfulevents.(It shouldbebornein mind
that the a coefficients for internal reliability on the
coping-style questionnaire scales were not high when
To investigatethe issueof whetherrelationshipsrated
high EE were associated with higher contact time because
of greaterperceivedproblemsof the patient, an analysis
of covariance was conducted between EE groups, with
contact time asa dependentvariable, for eachbehaviour
problem as a covariate. The significant relationship
betweencontact time and EE was not affected by the
inclusion of patient behaviour problems (SBS scores)
ascovariates.Hence,the greatercontact time associated
with thehigh-EEgroupcouldnot beattributedto problem
behaviour. Face-to-facecontact time wasthencorrelated
with the behaviourand role performanceitems from the
staff semistructuredinterview. Only poor self-careability
(i.e. requiring regular prompting with personalhygiene)
Fewstaff wereratedasusingavoidant strategies(mean
total 3.7, s.d. 2.9) in dealingwith work-relatedproblems
ratherthan to activebehaviouralcoping(meantotal 19.1,
s.d. 5.0)and cognitivecoping(meantotal 17.4,s.d. 5.9).
The mean GHQ score (assessed by the 1-0 â€˜¿?GHQ
scoring' method) was 6.9 (s.d. 9.6). The commonly
cut-off of 11/12(Goldberg, 1972)wasusedto distinguish
casesfrom non-cases,which placedsevenstaff members
(20%) in thehigh-scoringgroup.Thisprevalence of minor
psychiatric morbidity is similar to that reported for this
age group in community samples (e.g. Finlay-Jones
& Burvill, 1977).Some staff in the day hospital were
experiencing considerably poorer general health than
others (mean total score 49.6, s.d. 22.6). A modified
1-test, using the separate variance estimate (SPSS, 1990),
revealedthat day-hospitalstaff still scoredmorehighly on
theGHQ thanstaff workingelsewhere (1=3.17,d.f. =24.7,
Only 13of the staff (37%)saidthat theyhopedto
remainin theirjobs, at leastfor thetimebeing;11(31.4%)
were uncertain about their intentions, and 11definitely
intended leaving their job in the near future. A similar
number (12; 34.3%) felt dissatisfiedwith more than one
aspectof their work. Thesources of stressandsatisfaction
identified by respondentsareillustrated in Table 2. Role
overload (i.e. limited resourcesand staff shortage)was
more frequently cited as a problem by staff in the day
hospital, where the staff:patient
Two membersof staff found abuseby patientsto bethe
most distressingaspectof their work. Lack of patient
with the job.
2.51, d.f. = 59,
staff were using more
ratios were lower. Hostel
Sourcesof stressandsatisfaction intheir work identified
by day-careand hostel staff
progress was also seen as a significantsource of stress.
Conversely, if staff noted some improvement in the
patient's condition,this was considered
rewarding,particularly by hostelworkers.Teammember
ship wascited asa satisfying feature of the work more
often in the day hospital. Other aspectsof the job (e.g.
hoursworked) werementioned,while five staff members
were not satisfied with any aspectof their job at the
time of interview. In the six months that have elapsed
since the data collection, eight staff
have left their jobs in the day hospital, and two hostel
and low-EE groupsin termsof the certainty with which
their diagnosis was classified (ID scores), the syndrome
clusters which characterised their illness, or total impairment
asmeasuredby the SBStotal scoreor the subscores. The
average total SBSscorefor the groupwas8.1, which
suggests that this mixedsampleof dayhospital, hostel,or
residential patients hadhighlevels of impairment (Wykes
& Sturt, 1986).Inappropriate or difficult behaviour (5
items)contributed most to this overall score(mean2.34,
s.d. 1.82). Retardation (2 items) was lessmarked (mean 0.59,
s.d. 1.05)than limited social interaction (3 items) (mean 1.39,
s.d. 1.49). The mean score for neurotic behaviour was 1.61
Brewin et a!(1990) found that interviews
staff andpatients'relativesyieldedreportsof higherlevels
of symptomsandbehaviourproblemsin patientsthan the
equivalent interviews withthy staff.Totestforsuchabias
in thisdataset,totalandsubscale scores werecompared
by thesettingin whichtheinformants worked(residential
versusday care).However,in this sample,differencesof
setting seemto have had a reverseinfluence: neurotic
behaviour in patients was significantly
bereportedby daystaff thanby hostelstaff (1=2.47,
d.f. =49,P<0.05). Noneof theothersubscores differed
more likely to
806 MOORE ET AL
would seem more accurate to view the indices
of EE as a â€˜¿?snapshot'of the state of ongoing
transactions between two individuals (Hahiweg et a!,
1989;Miklowitz et a!, 1989;Birchwood & Preston,
1991), which is likely to be subject to variation
thepatient hasbecomeill, thepresentingcharacter
istics of the patient are likely to influence the
subsequentdevelopmentof the relationship, and in
general, â€˜¿?late' relationships are more dependent
on the attributes of the patient. This situation is
paralleled in spouse carers, who are less likely to
carers, but may still be critical of their partners
(Hooley, 1987).That staff membersmay becritical
of their key patient is clear, and an analysisof the
frequency and content of criticism in the staff group
is currently underway.
Self-report measures indicate that there are
genuineareasof dissatisfaction and stressin work
with thelong-term adult mentally ill. Interpretation
of thefindings without adequatefollow-up datacan
only be speculative. Previous research (Pines &
Maslach, 1978; Cherniss, 1980)would lend credence
to the associationbetweenlow staff:patient ratios
andgreaterstress(higher GHQ scores)in the carer.
An analysis of the organisational structure of the
setting and its effects on the staff (e.g. Handy, 1991)
might also be indicated. Variability
service across settings (i.e. day versus residential
care) on objective measures like the GHQ cautions
againstthegeneralisationof findings from oneunit
to another, evenwhenthe client group is the same.
Schizophrenic patients are particularly vulnerable
to over- and understimulating environmental pressure
(Wing, 1989). A harmful milieu may be created
both by the â€˜¿?burned out' worker (Pines& Maslach,
1978), and the intolerance and criticism associated
with high EE (Leff et a!, 1985).Our resultssuggest
however that these two issuesmay not be related in
staff. This maybebecausestaff areableto separate
patient-related stressfrom other job-related stressin
these particular settings. High-EE relationships
betweenstaff membersand long-term patientscan
exist without poor generalhealth and lack of job
satisfaction in the carer. Workers who experience
significantstressat work mayhaveconsistentlygood
relationshipswith their patients,at leastin that they
areacceptingand tolerant (low EE) of the patient's
difficulties. Some patients are obviously more
difficult to make positive relationships with, for
examplethosewho lack interaction skills or behave
It may be that, in
than are parent
within the same
Patientcharacteristicssignificantly associatedwith staff
criticism and warmth
between criticismand warmth and patient
Sincecriticism is not exclusiveto high-EE relationships,
it was decidedthat some examination
irrespectiveof EE classification would be informative.
Pearson product-moment correlations showed thatstaff
criticism and warmth were related to certain aspects of the
staffâ€”patient relationship (Table 3), but not to any other
staff variables.Themoretimespentin face-to-facecontact
with patients,andthegreaterthestraindescribed bythe
staff member,thegreaterwasthe likelihood of criticism.
Perceivedpatient deficits in warmth, social interaction,
and generalactivity levelswereassociatedwith criticism,
as were aggressive and attention-seeking tendencies.
Warmth expressed by staff memberswas negatively
associated withalackof warmthshown bythepatientand
of the data
This study has shown that EE attitudes in staff
overlap with those of other carers of long-term
psychiatric patients.The only difference
staff in this sample were lessvulnerable to over
involvement in their relationships with patients than
are familycarers. Staff would seem to have less
cause to become overconcerned about patients'
welfare, althoughthepossibility of overinvolvement
should not be excluded.
The proportion of staff members exhibiting
high EE in at least one interview was just under
half (n= 15, or 43%), of the same order as the
distributionof critical comments
relatives of schizophrenic patients in a variety of
cross-culturalstudies (Leff & Vaughn, 1985). The
fact that EE in this sample was more dependent
on the attributes of the patient than of the staff
member suggeststhat EE does not represent a
stable, enduring, trait-like quality in the carer. It
made by the
The optimal style of interaction associatedwith
different patient groups is still uncertain (Smith
& Birchwood, 1990), although the EE research
paradigmhasgenerated muchinformation of clinical
value (Kuipers & Bebbington, 1988). Staff who
are most supportive of the long-term patient with
schizophrenia may be those who strike the right
balance between tolerance and limit setting on
unacceptablebehaviour; and, in thosewith negative
of personal care and subjecting the patient to too
& Heinl, 1984;Garety& Morris, 1984;Wing, 1989)
have suggestedthat this combination of qualities
in the trained carer is beneficial, but such an
equilibrium may not be easy to maintain. What
starts out, for example,asencouragementto over
come a motivational deficit can easily becomean
interchange of nagging by the staff member and
obstructive behaviour by the patient, as described
in somehigh-EE interviews. Potentially destructive
relationships might be less likely to develop if
these issues were more generally recognised and
incorporated into training and supervisionpractices
with long-term staff.
It is possiblethat the amount of contact between
direct-care staff and patients, although lessthan
that found with somerelatives,would affect patient
outcome. Fairly limited contact (on averageless
than eight hours a week) was certainly found to
be sufficient to generatecritical and even hostile
attitudes in staff. The results do not support the
hypothesis that the greater contact time that was
reported in high-EE interviews could be accounted
for by increasedprofessional input with patients
whose difficultiespose particular
rehabilitation. Patient difficulty with self-carewas
associatedwith higher contact time in all relation
ships (high and low EE).
The relationship between EE in staff and patient
outcomeisdifficult to establishfor patientsengaged
in a variety of complementary services(e.g. day care,
drop-in, residential hostel) and subject to various
issueis being examinedby a follow-up study of a
subsample of patients residing in a hostel, whose
contact time with their carers is comparable to
that of relatives, and by a further study of direct
interaction betweenpatients and staff.
Ratings of high and low EE are not confined to
relatives caring for patients, but can also be made
on staff who form key-worker relationshipswith
long-term patients. However, the number and
intensity of reported stressors in the workplace
(e.g. staff shortageor a poorly defined role) seems
to beunrelatedto high-EEstaff-patient relationships
in this sample.Staff who work with this patient
group may face similar
asdo relatives, but the quality of the staffâ€”patient
relationship depends, for the most part, on the
individual contribution of the two agents, and is
affected by the warmth and sociability of the patient.
The majorityof staff
had extensiveexperienceand training in dealing
with the special difficulties associated with the
long-term patient group. Certain characteristics
of someof thepatients,suchaslack of spontaneous
interaction, underactivity, and aggression, were
associated with criticism in both high-EE and
low-EE relationships. It would be beneficial to
recognise these potential areas of dissatisfaction
in staff training programmes.
It is likely that other variablesinfluence the staff
member's approach to care. These may include
the â€˜¿?guidingphilosophy' of the service, how well
informed thecareris, andtheexpectationstheyhave
of patient progress. Expressedemotion seemsto
be a useful starting point in exploring the factors
that areof importance in creating and maintaining
optimal staffâ€”patient relationships.
difficulties in the client
members in this study
We are indebted to all the patients and staff who so kindly
cooperated withtheinterview procedures. Wethankconsultants
Dr lsaacs, Dr Birley, and Dr Bebbington who allowed access
to patients in their care. We also thank Dr Graham Dunn and
Mr PaulStanworth for adviceon statistics. Thefirst authoris
financed byagrant awarded bythe Economic andSocial Research
Council, whose assistance is gratefullyacknowledged.
In Theory and Practice of Psychiatric Rehabilitation (cdi
F. N. Watts& D. H. Bennett). Chichester: Wiley.
BEIucowITz, R. & HEINL, P. (1984) Themanagement of schizo
phrenic patients: thenurses' view.Journal ofAdvanced Nursing,
BILLINGS, A.G.& Moos,R.H.
responses and social resourcesin attenuating the stressof life
events. Journalof Behavioural Medicine, 4, 139-157.
BIRCHWOOD,M. & PRESTON, M.
Clinical Problems: A Cognitive-Behavioural Approach (eds
W. Dryden & R. Rentoul). London: Routledge.
BLEDIN, K. D., MACCARThY, B., KUIPERS, L., etal(1990)Daughters
ofpeople withdementia: expressed emotion, strainandcoping.
British Journal of Psychiatry, 157,221-227.
BItEwIN,C. R., VELTRO, F., Wn@io, J. K., ci al (1990) The assess
mentof psychiatricdisability in the community. A comparison
of clinical,staff and family interviews. BritishJournalof
D. & Moaais, 1. (1983) Supportand rehabilitation.
(1981)The roleof coping
(1991) Schizophrenia.In Adult
Job Stressin the Human
808 Download full-text
In Long TermCommunityCare.Psychological Medicine,
Monograph Supplement 2 (ed. J. K. Wing), pp. 29â€”39.
Cambridge: Cambridge University Press.
FADDEN, 0., BEBBINOTON, P. & KulpElts,
care:theimpactof functional psychiatricillnesson thepatient's
family. British Journal ofPsychiatry,
FINI@Y-Jo@s,R. A. & BURVILL, P. W.
minorpsychiatric morbidity inthecommunity. Psychological
Medicine, 7, 475â€”489.
FIRm, J. (1986) Levels and sources of stress in medical
GARETY, P.A.& MORRIS,1. (1984)
psychiatricpatients:organisation,attitudesandquality of care.
PsychologicalMedicine, 14, 183â€”192.
GOLDBERG,D. P. (1972) The Detection of Psychiatric Illness by
Questionnaire. Oxford: OxfordUniversityPress.
(1978) Manual of the General Health Questionnaire.
HALL, J. N. (1989)Chronic psychiatric handicaps.In Cognitive
Behaviour TherapyforPsychiatric Problems (edsK. E. Hawton
et at). Oxford: Oxford UniversityPress.
(1990)Towardsa psychologyof caring. British Journal of
ClinicalPsychology, 29, 129-144.
HAHLWEG,K., GOLDSTEIN,M. J.,
(1989) Expressed emotion and patientâ€”relative
families of recentonsetschizophrenics.Journal of Consulting
and Clinical Psychology,57, 11â€”18.
HANDY, J. (1991) Stress and contradiction
Human Relations,44, 39â€”53.
HERZOG,T. (1992) Nurses, patients and relatives: a study of
family patternson psychiatricwards.In Family Intervention in
Schizophrenia: Experiencesand Orientations in Europe (eds
C. L. Cazzullo & 0. Invernizzi). Milan: ARS (in press).
HOOLEY,J.M. (1987)The nature
emotion. In Understanding Major Mental Disorder: The
Contributionof FamilyInteractionResearch (edsK. Hahiweg
& M. J. Goldstein). NewYork:FamilyProcess Press.
IsAAcs,A. D. & BEBBINGTON,P. E. (1991) Strategiesfor the
management of severe psychiatric illnessin thecommunity.
InternationalReviewof Psychiatry,3, 71â€”82.
JACKSON, H.J., SMITh, N.& MCGORRY,
betweenexpressedemotion and family burden in psychiatric
JACKSON, S. E., SCHWAB, R. L. & SCHULER, R. 5. (1986)
an understanding of the burnoutphenomenon. Journalof
AppliedPsychology, 71, 630-640.
D. V., FINBERG, N. A., S@i,
increasein violence on an acute psychiatric ward. British Journal
of Psychiatry,156, 846-852.
KUIPERS, L. & BEBBINGTON, P. (1988)
schizophrenia: theoretical and clinical implications. Psychological
Medicine, 18, 893â€”909.
& (1990) Workingin Partnership:Cliniciansand
Carers in the Management of Longstanding Mental illness.
LAVENDER, A. (1985)Qualityof care
staysettings. In New Developments inClinical
(ed.F. N. Watts).BritishPsychological Society,Leicester.
C., STURT, E. & WYKES, T.(1982)The role of relatives.
L. (1987)The burdenof
(1977) The prevalenceof
A new unitfor long-stay
NUECHTERLEIN,K. H., ci al
and originsof expressed
@ A. K., ci al (1990) An
and staff practicesin long
L@sr,J. P., KUIPERS, L., BEREowrrz, R., ci al (1985)A controlled
trial of social intervention in the families of schizophrenic
patients: 2 year follow up. British Journal of Psychiatry, 146,
â€”¿?& VAUGHN,C. (1985) ExpressedEmotionin Families.
London: Guilford Press.
MANN, S. A. & Cam,
anational sample survey offifteen mental hospitals inEngland
andWales,1972â€”1973.Psychological Medicine, 6, 603â€”616.
Mnu.owrrz, D.J.,GOLDSTEIN, M. J.,DuANE, J.A., cial(1989) Is
expressed emotionan indexof a transactional process? I.
Parent's affectivestyle.FamilyProcess, 28,153â€”167.
Moasus, I. (1991) Residential care. In Community Psychiatry: The
Principles (eds D. H. Bennett & H. L. Freeman). London:
NICHOLS,K. A. (1985) Psychological care by nurses, paramedical
and medicalstaff: essential developments for the general
hospitals.British Journal of Medical Psychology,58, 231â€”240.
O'DElscoLL, C., MARSHALL,J. & REED,J. (1990) Chronically
patientsin a district generalhospital unit. A surveyand two
yearfollow-up inaninner-London health district. British Journal
PINES,A. & MASLACH,C. (1978) Characteristics of staff burnout
SALoIcANc@S, R. K. R., DER,0. & WING,J. K. (1985)Community
psychiatricservicesin Englandand Finland. SocialPsychiatry,
S,.irm, J. & Bmcuwoon, M. (1990)Relativesand patients as partners
in themanagement of schizophrenia: thedevelopment of a
service model.BritishJournalof Psychiatry, 156,654-600.
SPSS/PC (1990) Statistics Version 4.0. Chicago: SPSS Inc.
SI.rrroN, 0. (1981) The role ofthenurse. The organisation
administration of nursing servicesfor long stay patients. In
Handbookof PsychiatricRehabilitationPractice(edsJ. K. Wing
& B. Morris). Oxford: Oxford University Press.
THORNICROFr,0. (1991)Theconceptof casemanagement for long
term mental illness. International Review of Psychiatry, 3,
VAUGHN, C. & LEFF,J.(1976) The
emotion inthefamilies of psychiatric patients. BritishJournal
of Socialand ClinicalPsychology, 15, 157-165.
WATIS, S. (1988) A descriptive investigation of the incidence of high
expressedemotion instaffworking withschizophrenic
a hospital setting. Unpublished dissertation for the Diploma
in Clinical Psychology, the British Psychological Society,
WING, J. K. (1989) The concept of negative symptoms. British
Journalof Psychiatry,155(suppl.7), 10â€”14.
W. (1976)New long-staypsychiatricpatients:
â€”¿?,COOPER,J. E.& SART0RIU5, N.(1974) TheMeasurement
and Classification of Psychiatric Symptoms. Cambridge:
Cambridge University Press.
S. A., LEFr,J. P., ci al (1978)The conceptof a
â€˜¿?case'inpsychiatric population surveys. Psychological
WooDs,P. A. & CULLEN, C. (1983)Determinants of staff behaviour
in long-term care.Behavioural Psychotherapy, 11,4â€”17.
WYKES,T. & STURT,E. (1986) The measurement of social behaviour
in psychiatric patients: an assessmentof the reliability and
validity of the SBS.British Journal of Psychiatry,148,
5Correspondence: Department of Psychology, Instituteof Psychiatry,Dc Crespigny Park,LondonSE58AF
MOORE ET AL
E. Moore, BSc,PostgraduateStudent, Institute of Psychiatry, London SE5 8AF; R. A. Ball, MBChB, MPhil,
MRCPsych,ConsultantPsychiatrist, Queen Mary'sUniversity Hospital,
Registrar,MaudsleyHospital, London SES8AF; *L. Kuipers,BSc, MSc, Phi),SeniorLecturer in Psychology,
Institute of Psychiatry,andHonorary TopGradeClinicalPsychologist,MaudsleyHospital, London SE58AF
Roehampton, formerly Senior