Dr B Wood
IIBCh ll, N{Prirri\letl
Dept of F;rmilv Mcciicine : Mccluns;r P O l i r x 2 2 2
trIeclunsa 0204
Curriculum vitae
Briirn \\'ils born in Capc To$'n, irncl obtlincrl thc NIBChB rtt UC-l- in I982. He cornpletccl l.ris intcrnship ilt Victori;1 Hospital, lVvnbcrg.
F r o m l 9 8 5 t o 1 9 9 0 h e n ' o r l t e c l i n m r a l I(s;rZulu, nrainh' rrt Oharles ]ohnsou Memoriirl Hospital in Nc1utr.r, thcn at Hlabisl Hospitirl.
In mitl 1990 he joinccl thc Fan.rih,Metlicinc cleprrrtnre nt irt N'[ec'lunsl. Hc completccl the M l r . t r A I e r l i r r I 9 9 l , l n t l . r t 1.r'c:crrr i s ippointecl as a F.-u.nih' Practitioncr irncl Lecture r rrt N'lcrluns;r. He is rcsponsible tbr thc clepirrtnrcr.tt's progrrlrme fbr fin;rl verrr mcclical stuclcuts. Hc is r.narrictl to Iudv ar.rcl thci' hai'c tso chilciren.
A Literature Review - Dr Brian Wood
Suntnta4,
Interpreted ntedi cal consu ltations are cornnlon in South Ah'ic;t, particularly in the pttblic sector. A contnton pc t'L'cpti o tl t'.tlst.s fhat i r t tc'4> re I t'd consttltatic>ns are inherentlv Lt rrsJ I i stitL' t o n', b t r t t I t c rcscart'h literature provides little et,idence lbr this viev. Conceptual moclels r>t'the i n te rp rc te d co n su I t:tti ott it rc
rli.scusscJ. r| t c e r i d c rt c'c co tl L'e nr i rr g v'ho shoLtld interyret is rcviev'cd, rcsearch tinclings ;trc discussed, ;tnd p'totential arctts lbr Iitrther research are highlightecl.
S Ali'F;tnt Pract 1993; l4: 347-53
I(EYWORDS:
Physici.rr, Famil,v; Interviews;
Interpreters; Ph1'sician -patient Relations; Language.
"Nurse, plerse lr.ill )rou ask the patier-rt hou' he uucicrstauds the cause of his cliseascl" A prolongcd,
incomprehensiblc interchar.rge then fi>llons. Eventualh, the nurse tllrns to the doctor. "The patient says 'Yes, he is'w,illing', doctor!"
Mant' interactior-rs betn,een doctors irnd pirtients in South Afiica are conductcd through an interpreter. I hlve u'<lrked deilv rr ith intcrpretcrs fbr 9 ycars, and had numcroLls exoeriences sir.r.rilar to that recorcled i r b o l e . I l e l t u n p r e p a r e d f i r r s r r c h consultirtions. Tl-re u,hole area of i rrtcrpcrsou:.rl com rn u rr i c;rtiorr receivecl little attention in the cnrriculum u'hen I \\'as all tur-rdergradr-rirte, let alone a sub- comporlent like cross-cr-rltural iutcrr icr'r's. The doctor-pirticur relationship has become ir core
concern fbr Famih, Practitioncrs. For gcneralists u.'orking ir-r a cross-cultural sctting, liil'lguagc arrd iuterPrctaticlrr is of cer.rtrirl irnoclrtance to the
developr-ner.ri of this relationship, and is receiving increasing attention. The Iiteratr.rre revierved u,;,rs sourced prcclor.ninandv through the r-rse of the Meclline datab;rse.
The litera'rtr-rre cleirling n'ith the usc of interpreters is largelv of reccnt origin.
Much rescarch originirtes in thosc d c r e l o p c c l r n d i r l l l r r e n t c o r r n t l i e s u4'rich are expcriencing an influx of P e < D l e l l ' < l r u l e s s r f l l r r e u t c o u n t r i e s . and are cncountering ne"'r, ciifliculties l'ith comr.nur-ricirtion rclatecl to language and culture.
Mr-rch of this literaturc;rrises out of i'r p s v c h i a t l i c s e t t i u g . i n d i c r t i n g recognition of the in-rportance of communication in the evaluation ernd manageurcnt of such illness fbrms.
Some earlv rvork u'as condr-rcted by
\\'estern trained doctors u'orking in nol.r-westcm settings. Fror]] Nortl]
America tl-rere is research dcaling u'itl-r commur.ricntion oroblcms
experier-rced b), tire ir-rdrgenous populatior-rs. Research initially fbcr-rssed cln the quality of the infbrrnation providecl to the doctor.
T h c t l t r . r l i n o f t h c i n t c r a c t i o u , particularh' the perceptions of the patient, have onl,v recellthr been given attenti()n.
Conceptual Models of Medical
I ranslailor1
Tl.rcoretical moclels hirve becr-r der,eloped to clescribe tl-re interaction rvl-rich bccurs. It has been acccotecl as r strrti|lg poirrt tlrlr, in corrtr,rit ttr the dvaclic clinician-pirtient n.rodel u,ith its single relationship, tl-re t r i i r u g r r l l r d o c t o r i n t e r p r e t c r - p a t i c n r model ir-n'olr,es 3 relationsl'rios.
347 SA Fan.rilv Practicc August 199.3 SA Huisrrrtspraktvk Augustus 1993
Within this structure there are a variety of possible ways for the interaction to be handled.'
I. [nterpreter as inrerviewer In this model, the interpreter conducts the interview along general guidelines provided by the doctor, who has no direct
interaction with the patient.
Interpreted interviews are certain to be with us for the forseeable future
2. Interpreter as instrument of the provider (Black box model) Here the doctor wishes to retain control of the interaction, and requires as direct a translation as possible of his or her questions and the responses ofthe patient.
This model may well be
appropriate in certain settings like international meetings, but it is Iimiting in a medical context.
Many basic concepts in medical thought are not directly
translatable into another Ianguage, and the same applies to the
patient's conceptualization of the disease process. However, it is clear from the research articles discussed below, that many of the authors consider this to be the applopriate way for an interpreter to function.
3. Interpreter as patient advocate Here interpreters see themselves as protecting the patient from
excesses ofthe doctor or the medical context to which the doctor belongs. It is unusual for
Interpreters in Medical Consultations
doctors to ask their interoreter to assume such a role.
4. Intetpreter as cttlture broker Often there is a substantial cultural gap between doctor and patient of which the interpreter will often be aware, although the doctor and patient may both be less conscious of their own cultural determinants. Over- identification with either the doctor or the patient may put pressure on the interpreter.
5. Modelof pannership
This model is considered to be the ideal. Here it is recognized that both the doctor and the interpreter have specialized knowledge to offer the patier-rt, and that for the patient ro receive optimal care, they need to
function as a team.
Who Interpretsf
Interpreters vary substantially. The
literature describes the r,rse of
personnel dedicated specifically to the task of translating.z':' +' s 11't't
Language and interpretation is of central importance in the doctor-patient relationship
arrangement is highly rated. It is probably uncommon in South Africa in a medical context, although it has been the practic€ in our courts for years.
The use on a casual basis of a wide variety of other medical, nursing,
348 SA Family Practice August 1993 SA Huisartsprakwk Augustus 1993 paramedical or even non-medical ancillary staffas interpreters, is also described.6,7,8 Such personnel are not paid to interpret, and receive no formal training or recognition for their services. This is probably the common situation in South Africa.
Very few articles on original research available
This arrangement has been criticized on two counts. Firstly, because interpretation is not formally part of their job description, such
interpreters receive no training and no recognition, and often experience tension between interpreting, and the requirements of their formal duties.
Secondly, issues of confidentiality may arise, if for instance, the doctor is unaware that the Datient and the interpreter already know each other.e'ro
All authors who have considered the subject agree that the most
unsatisfactory interpreters are relatives or family members of the patient.
Their relationship with the ill person introduces interactions into the interview which are complex, and greatly influence the information obtained. This is particularly so when the interpreter is the child of the ill
p e r s o n . r 1 . 1 2 , 1 3
Research Findings
Price compared the proficiency of three interpreters, one of whom was a psychiatric patient described as being in full remission.6 It was found that the completely inexperienced patient was by far the most accurate
interpreter, with fewest omissions and
n-ristranslations. When thc language proficiency ofthc three rvas
compared, it rvas found that the patient had substantially better command of English than the other two, both of whom were medical orderlies.
Lang ar-rdiotaped consultations where mcdicai orderlies r'vere also used as interpreters, in Papua New Gr-rinea.7 With the cn.rphasis on identilying mistranslations, he noted that most errors could be described as either additions or omissions. The interpreters werc prone to initirte their ou'n questions, either to clari$' an ans\'ver grven by a patlent, or to demonstratc their own medical acllmen. Interpreters were perceived a s d i s p l a f i n g l i t t l e s e n s i t i v i r y , interrupting the flow of patients' responses indiscriminately, when the amount to be translated became s u b s t a n t i a l . N o i n d i c a t i o n i s contained in this study of the fi'equeno, of mistranslations. In a
The experience and
perceptiolls of the patient arc seldom given
similar study in Nigeria, Launer found that interpreters often did not provide a word-for-u'ord translation of what the doctor or paticnt said.8 Some deviations rvere thought to have been helpful, some confbsing or incorrect.
In a psychiatric context in New York, Marcos looked at aspects of the use of interpreters for patients whose first language was Spanish.'t Starting from the assumption that "faithfulness is the fundamental obiect of
interpretation". he identifi ed three maror sources ol lnterpretrve distortions: ( I ) Omissions, substitutions. additions and
condensations arising from deficient language or trar-rslational skills of the interpreter. (2) Interpreters' lack of
No longer only doctor-patient relationship. It becomes the doctor- interpreter-patient relationship
psychiatric sophistication, resulting for instance , in the r-rormalization of thought disorders. (Pricc made a sirnilar obseruation.') (3) Interpreters' attitucles towards either the Datient or t l r e c l i r r i c i a n , s o m e t i m e s l e a d i n g t o deliberate distortions.
Sabin reported the unanticipated suicidc of two Spanish speaking patients after interpreted
consultations.2 He found that the severiq' of the illness was accurately, identified, but that psychotic features were given precedence over
recognition of the depth of the patients' despair. This was despite the use of a translator fluent in both l a n g u a g e s , a n d w i t h t r r i n i n g i n school counselling. Sabin concluded that at'Ibctive components of mer-rtal illness are prone to being
mistranslated.
Marcos has provided evidence that schizophrenic patients whose native language was Spanish, when
intenrier.ved in English without an interpreter, lvere scored as showing more evidence of psychopathy than when they rvere intewiewed in Spanish. The patients gcnerally
349 SA F;rmil,v Practice August 1993 SA Hr-risartsprakn,k Augustus I 993
appeared to be more proficient in Spanish than in English.rs
A study conducted by Kline et al had a diflbrent focus al.rd produced somelvhat surprising results.'6 This study was also conducted in the c o n t e x t o f a p s 1 c h i a t r i c o u t p a t i e u t clinic. Qucstionnaires u,ere administered to both patients and psychiatric registrars after interpreted consultations. The registrars on11, occasionally made use of an
interoreter. A wide discordance rvas found between the psychiatric registrars who had little faith in the potcutial for a thcrapeutic
relationship to dcvelop when an interpreter lvas used, and their patients, rvho felt very positive about the interaction, and were kcen to rcturn under the same circurnstances.
I(ine concluded that mental health profbssionals rvorking u'ith an
interpreter in a cross-cultural context, may not understand what patients want or fcel and may instead project t h e i r d i s c o m t o r t o n t o p a t i e n t s . Bucci and Baxter looked at linguistic insecurity in a multicr"rltural speech context.rT Their patients were speaking a second language or a non-
Family members or friends are the most unsatisfactory
interpreters
standard form of English to Standard American speaking therapists. No interpreter was readily available. They point out that even speakers r.vho are apparently competent in the language of the consultation may be fbeling insecure, and describe a method for detecting such insccurity.
Dodd obserr,,ed in the introduction to his study in Saudi Arabia, that the Iiierature has assumed that oatients are "disadvantaged" when ihel' do not share a common language with their physician.'8 Dodd cornpared the diagnoses reached by English speaking doctors using interpreters,
The depth of the patient's despair not interpreted
with those reached by Arabic doctors, and fbund that differences were related to the experience ofthe doctor and not to the use of interpretation. Dodd decided the instrument for measurement was too crude to find differences which must exist. Despite the lack of evidence which he provided, he concluded that
"there is no doubt that, with regard to mental diseases and ill-deflned conditions, the patierrt is
disadvantaged when the doctor and Datient do not share a common ianguage".
Kaufert and Koolage have published initial observations concerning role conflict expericnced by native Canadian medical interoreters.'' It was noted that: "Interoretcrs are often faced r'vith confliits between their role as health system employees and their role as culture-brokers and patient advocates". A number of examples are provided of the
interpreter being pressurised by either the clinician, with a biomedical agenda, or the patient, who looks to the interpreter for cultural support.
A postal survey of GPs seeing Asian patients in Newcastle was conducted by Wright." All interpreters used
Interpreters in Medical Consultations
rvere family members of patients. The doctors felt that members of this communiry werc prone to complain of trivial ailments, while infrequently presenting with psychosocial problems. One GP experienced a drop in trivial complaints on learning some Hindi, and Wright postulated a connection between inadeouate communication arrd unn.cissn,-r' complaining.
Also in an Asian context, in Leicester, Ebdcn et al, audiotaped and
transcribed consultations translated into Gujarati." Again the interpreters were provided by the patient, and were family members. 23-52% of questions asked by the doctor were mistranslated or not translated at all!
It is difficult to believe that these interactions appeared to be f-airly normal to the doctors involved.
Anatomical and technical terms were particularly prone to mistranslation.
Children were found to be
If patient and interpreter already know each other) sensitive issues of
confidentialiry may arise
embarrassed to translate questions about menstruation or bowel movements for their parents, and there was a general tendency for questions about bodily functions to be ignored. In a similar context in Bradfbrd, Ahmad et al, found a strong preference fbr doctors who could speak the patient's language, where the patient's proficiency in English was poor.20
A common recommendation made bv authors is that doctors should
350 SA Famil,v Practicc August 1993 SA Huisartsprakwk Augustus 1993 learn the language of their patients, but the results of attemots to do so have not received mrrch attention.
F,rzinger has provided evidence that Ianguage acquisition does not by i tself solve communication
difficulties." If the system for scoring patient centredness developed by Henbest is applied to the four consultations recorded by Erzinger,
Faithfulness is the fundamental object of i nterpretation
poor scores are obtained (l = closed responses only, in 2 consultations), unless the doctor's language
proficiency is ofa very high standard (2 = open responscs to patient oflers, also in 2 consultations)." No
interpreter was present during these consultations.
There is a vast literature dealine \'vith issues relcvant to the translatioi of a research questionnaire into another language. The relevance of this research to the usc ofintcrpreters is debatable, but it does give insight into the linguistic and social comolexities involved in translation.
Drennan et al have orovided an interesting local account of the interactions which occurred in the preparation of a Xhosa version of the Beck Depression Inventory.'?3 They highlight the power relationships which occur in what is essentially an interpersonal process, leading to the potential for a struggle for control.
A component ofa study undertaken by Henbest to evaluate the patient centered approach in a non-western context, suggested that the use ofan interpreter does not inevitably lead to
poor patient-centredness scores.2a Interpreted consultations were found to be more patient centered overall than consultations where no
interpreter was used. The instrument used had previously been validated in a monolingual context.
Summary of Research Findings
I. The most common tvDes of errors identified were omissions, additions, condensations and substitutions.tn
2. Errors are reduced when the interpreter is proficient in both the languages used.u
3. Interpreters are inclined to take control of consultations (role exchange).7'8
4. If the interpreter has limited insight into the possible disease process) certain types of problems are likely to be normalized (eg thought disorders),u,'o 5. Interpreters are a third party in a
consultation, and influence the course of the consultation. Their attitudes towards the people concerned, ethical position and cultural background infl uence the interpretation provided. te 6. Patients may appreciate the
presence of an interpreter, rvhile the doctor may feel
uncomfortable about conducting an interview with an interoreter.
Both are likely to be unaware of this divergence.'6
7. No clear disadvantage has been demonstrated in a large group of patients when an interpreter was u s e d . r s
8. Anatomical and technical terms are prone to be mistranslated.rz 9. Language acquisition by the
doctor, who then stops usirrg an interpreter, does not rapidly solve comm unic ation diffi culties, unless the doctor is hiehlv proficient.2'
10. The use of an interoreter is not inevitably associatei with poor patient-centredness scores.2a I l. Emotional content may be the component of the interaction most easily lost when an interpreter is used.2 It may xl5e be easily lost when a second language is spoken poorly by the oatient.rT
The emotional content is the one most likely to get lost
The use of family members as interpreters is associatd with a large number of mistranslations, and is not the ideal.tt' t2' t3 Even the use of a trained, highly proficient interpreter can be associated with misunder- standings (which can certainly occur when both the doctor and patient speak the same language and no interpreter is present!)' Patient proficiency in their second language if it is being used in the consultation, leads to a perception ofa reduced need for a doctor who speaks the patient's first language .20
Commonly Repeated Recommendations Lacking Supporting Evidence
The evidence available suggests that the use of,casual interoreters is associated with frequent problems, particularly when they are family members. High levels of satisfaction have been reported when professional interpreters are used, for instance by Richter et al.a However, no studies have been done comparing trained interpreters with, for instance, nurses who regularly int€rpret but have no formal training.
In the light of identified limitations to the use ofthe interpreters, various recommendations have been made concerning appropriate behaviour for doctors utilizing interpreters. These include asking short, simple
questions, and rephrasing and repeating qLrestions, to check the quality of translations.T' 8' 25 These seem to make sense, but the resulting consultations have not been reported.
Such questions are likely to be closed, and not conducive to a oatient centred interview.l8' 2a
Reference is made in the literature to the importance of building an understanding with the interpreter before beginning the consultation with the patient.s This is usually said in the context of a brief interaction, and no attention has been given to the effects of a long-term relationship benveen a doctor and an interpreter.
The effects of exclusively or predominantly interacting with patients through an interpreter have not been studied. Manv doctors in the South African public service are in this position, and fhe fiterature does not provide information concerning the adjustments likely to occur.
t 2 .
r 3 .
t4.
3 5 1 S A F a m i l y P r a c t i c e A u g u s t I 9 9 3 SA Huisartspraktyk Augustus 1993
Interpreters 1n Medical Consultations
Emotive statements have been made, like that of Andreyev, who asked, "Is it too radical a position to suggest that every individual deserves a doctor who can speak to him or her in his or her own languagel"2u However, language acquisition as an adult is not easy. Sustained, high levels of motivation are required, often for years, to reach adequate proficiency, unless there is no available alternative to learning the language concerned. Usually, as in western countries experiencing immigration, there are much simpler alternatives for the doctor, like expecting the patient to learn the
doctor's language.'3 Furthermore, the research literature does not orovide convincing evidence that interpreted interactions are always
unsatisfactory. I 8 Also, Erzinger's
A plea for trained interpreters
study suggests that high levels of language and cultural knowledge are required by the doctor for the interaction to be satisfactory." No studies have been found ofthe effects of the doctor understanding some of the patient's language, while
retaining an interpreter, although this is frequently recommended.2u
Local Relevance
It presently seems to be rare to have personnel dedicated to interpreting in South Africa. Usually nurses fulfill this role. and have been made to feel that it is their responsibility.
Experience suggests that the quality of the interpretation varies
substantially. The importance of the interpreter role is rarely emphasized, no formal recognition is given to nurses providing interpretation, and it is generally not remunerated.
gTH FAMILY
PRACTITION ERS CONG RESS
CAPE TOWN 1Oth - 14th April 1994
.,FAMILY MEDICINE FOR ALL"
* Take a break from your practice - enjoy a week in the glorious Cape
* Join your colleagues in workshops, discussions and update sessions all relevant to Family Practice/Primary Care
* Take advantage of special rates at the Cape Sun and other hotels
* Consider presenting a paper yourself
Contact Cape Town Academy Office for details Tel: (021) 531- 4561 Fax: (021) 531- 2982
"LEARN SOME M.ORE - lN APRIL '94n'
352 SA Family Practice August 1993 SA Huisartspraktyk Augustus 1993
At least some doctors make an effort to learn the language of their patients.ll Despite the
recommendations made in the l i t e r a t u r e t o c o n t i n u e u s i n g a n interpreter, interpreters who have other duties (usually "nursing related), are keen to let the doctor
"manage".'u Doctors in this position lack adequate information to enable them to insist that they still require the presence ofan interpreter.
In RSA nurses have been made to feel it is their resDonsibilitv
It is recognized in the literature that it is difTicult to provide an
interpretation service to doctors in private practice.4.26 It could be argued that those doctors who are frequentlv dealing with patients rvho speat a different language to their own, have a responsibility to provide
interpretation or learn the language.
This issue is bound to become mclre complex and pressing if in future we have a national health scheme, rvith registered patients, for whom doctors have a statutory respor-rsibility.
In conclusion, it is suggested that a m o n g o t h e r s , t h e f o l l o w i n g is s u e s are worth exploring further:
1. The effects ofthe use ofan interpre ter on Ihe patient- centredness of the consultation.
2. The effects of the develooment of a n o n g o i n g r e l a t i o n s h i p b e r * e e n the doctor and interpreter.
3. The impact on the interaction of the doctor's efforts to learn the patient's language, when an interpreter is used and when no interpreter is prese nt.
References
I . Wcstermever J. Working with an interpreter in psychiatric assessment and treatment. J Nervous Mental Dis 1990;
l 7 B ; 7 4 5 - 9 .
2. Sabin lE. Translating despair. Am J Psychiatry 1975; 132: 197 -9 3. Faust S, Drickey R. Working with
intcrpreters. I Family Pract 1986;22:
1 3 1 - B
4. Richter R, Daly S, Clarke J. Overcoming language difficulties rvith migrar-rt patients.
M c d I A u s t 1 9 7 9 ; l: 2 7 5 6
5. Cox fL. Aspects of transcultural psychiatry.
B r i t I Psychiat 1 9 7 7 ; 1 3 0 : 2 l l 2 ) , 6 . l r i c c l . F o r e i g n Ia n g u a g c i r r r e r p r e t i n g i n
psychiatric practice. Aust NZ J Psychiatry 1 9 7 5 ; 9 : 2 6 3 - 7
7. Lang R. Orderlies as intcrpreters in Papua Ncw Guinea. Papua New Guinea Mcd J 1 9 7 5 ; l 8 : 3
8. Launer J. Taking medical histories through interpreters: practise in a Nigerian outpatient department. B Med I I97B,2, ( 6 1 4 2 ) , 9 3 4 - 9 3 5
9. Westermeyer |. Clinical considerations in cross cultural diagr-rosis. Hosp Cornrnr,rnity P s l , c h i a t r y 1 9 8 7 ; 3 8 : 1 6 0 5 .
10. Chesher T. Communicating tvith paticnts.
Do your paticnts understand youf Are you speaking the same languagel Med J Aust 1 9 8 5 ; 1 4 2 : 5 8 4 - 5
ll. Wright CM. Language and
communication problems in an Asian community. ] R Coll Gcn Pract l9B3; 33:
1 0 1 4
12. Ebden P, Carey Ol, Bhatt A, Harrison B.
Thc bilingual consultatior"r. Lancet 1988;
I : 3 4 7
13. Festenstcin F. Bilingual consultation.
L a n c e t l9 B 8 ; 1 : 8 8 8 9
14. Marcos LR. Ellbcts of interpreters on the evaluatior.r of psychopathology in non- English speaking patients. Am J Psvchiat 1 9 7 9 ; 1 3 6 : \ 7 1 4
15. Marcos L\ A-lpert M, Urcuyo L, I(cssclman M. The effect of ir-rtenierv language on the evaluation of ps),chopathologv in Spanish-American scl-rizophrenic patients. Am J Psychiatry ) . 9 7 3 ; 1 3 0 : 5 4 9 - 5 3
16. I(ir-re F, Acosta FX, Austir-r W, lohnsorr RG. The misunderstood Spanish speaking patient. Am J Psychiatry 1980,I37,(12),
r 5 3 0 r 5 3 3
17. Bucci W, Baxter M. Proble ms of linguistic ir.rsecurity ir-r multicultural speech contexts.
A n n N Y A c a d S c i 1 9 8 4 ; 4 3 3 : 1 8 5 2 0 0 18. Dodrl W. Do interpreters affect
consultationsl Fam Practice 1983; l:42-7 19. IGufert lM, i(oolage WW. Rolc conflict
among "culture brokers": The experience of native Canadian medical interpreters.
S o c S c i M e d 1 9 B 4 ; l 8 : 2 8 3 - 6
20. Ahmad WIU, I(ernohar"r EEM, Bakcr MR.
Patients' choice of general practitioner:
influence ofpatients' flucncv in English and the etl-rnicity and sex ofthe doctor. J R C o l l G e n P r a c t 1 9 8 9 , 3 9 , f 5 3 - 5 21. Erzinger S. Communication betu'een
Spanish-speaking patients and their doctors in medical encounters. Cult Med P s y c h i a t r y l 9 9 l ; l 5 : 9 l - 1 1 0
22. Hcnbest RJ, Stervart MA. Patient- centredness in the consultation. l. A mcthod fbr measurement. Fam Pract 1 9 8 9 : . 6 : 2 4 9 - 5 4
23. Drennan G, Levett A, Sn'artz L. Hidden dimensions of poner and resistance in the translation process: a South Afiican study.
C u l t M e d P s y c h i a t r y 1 9 9 1 ; i 5 ; 3 6 1 8 I 24. Henbest Rl, Fehrsen GS. Patient
centeredness: is it a;rplicable outside thc Wcstl Its rneasure mcnt and eflect on o u t c o m e s . F a m P r a c t f9 9 2 ; 9 : 3 l l - 7 25. Bal P. Dealing rvith the disadvant:rged:
communicating lvith non-Englisl-r speakir-rg patients. Br Mecl J t98l; 283;
268
26. Andreyer. HlN. Getting the messagc a c r o s s . B r M c d I 1987; 294:1534 5.
353 SA Famill, Practice August 1993 S A H t : i s a r t s p r a l < q k A u g u s r r r s I 9 9 3
