Language Barriers and
Medical Interpretation
Academy Health
June 27, 2005, Boston, MA
“National Standards For Culturally & Linguistically
Appropriate Behavioral Health Care: Are We Kidding
Eric J. Hardt MD
Clinical Director, Geriatrics Section, Boston Medical Center
Medical Consultant to Interpreter Services
Linguistic Minorities in the USA
Definition of NES, LEP
13.8 %
6.0 %
17.9 %
8.1 %
18.7 %
8.1 %
Boston 2000
33.4 %
16.3 %
47.6 %
23.7 %
57.8 %
32.6 %
Language and Access Mandates
4. Offer and provide language timely
assistance services without charge
5. Inform patients of their right to
receive language assistance services
6. Interpreters and bilingual staff
7. Patient-related materials and signage
Interactions between Culture and
• Scenarios that include language
barriers are very likely to present
“cross-cultural” issues:
– Bias/discrimination/stereotypes/racism
[both personally-mediated and institutionalized]
– Culturally-mediated diversity in health-related
– Power differentials
E Hardt 2005
Roles for Medical Interpreters
in Relation to “Cultural” Issues
• Interpreter is the conduit via which the
culturally competent provider may explore
differences in health related beliefs and
• Interpreter adopts an expanded role that
includes explanation of features of medical and
of patient culture and brokerage of
relationships between patient and provider
E Hardt 2005
Clinical Issues I
• Outreach and marketing, signage,
telephone access
• “Taking a history” and doing the PE
• Clinical evaluation [ e.g. CAGE, Folstein
MMSE, peak flow]
• Ordering, interpretation, and performance
of tests [e.g. CAT, MRI, ETT, PFTs]
• Procedures [ e.g. colonoscopy, conscious
sedation, labor and delivery ]
• Patient education, counseling, discharge
instructions, preps, written materials
E Hardt 1988
Clinical Issues II
• Consent for treatment/procedures/studies
• Follow-up of test results, appointment
• Medication compliance, adverse drug
reactions, allergies
• Cost containment, managed care
• Risk management, medical errors,
standards of care
• Doctor-patient relationships, patient
E Hardt 1988
Research Data and
Needed to change attitudes of
law and policy makers,
remodel provider behavior
and clinical systems, establish
credibility for interpreters.
Do We Have Health Care
Disparities related to
Language Barriers?
How big are they? For what
groups? In what areas? How do
we document them? What are
the costs? What can be done?
Who should be doing it?
Selected Research Issues
• Inclusion of potential LEP subjects
• Translation and validation of instruments
• Research infrastructure and personnel,
information systems
• Definitions [ PLINE, NES, LEP; “interpreters”
and translators ] and data collection methods
• Role of IRBs
• Research agenda
• Budgets and funding; involvement of
Interpreter Services
E Hardt 2005
The Exclusion of Non-EnglishSpeaking Persons from Research
• Survey of 172 responding researchers on
provider-pt relations
• Only 22% included LEPs who were potential
• Reasons for exclusions:
didn’t think of it
translation issues
staffing issues
no potential LEP subjects
Frayne SM et al J Gen Intern Med 1996
• Provide service to LEP clients that are more
limited in the scope or that are lower in quality
than those provided to other persons
• Subject a LEP client to unreasonable delays in
the provision of services
• Limit participation in program or activity on
the basis of English proficiency
• Provide services to LEP persons that are not
as effective as those provided to those who
are proficient in English
• Require a LEP client to provide and interpreter
or to pay for the services of an interpreter
VNS of Western MA:
OCR Action I
• July 1998 intake RN and supervisor refused
to accept referral of Spanish-speaking
diabetic because “she didn’t speak English
and had no one to interpreter for her at
home…” They claimed that this was “the new
policy caused by budget cuts…”
• Patient was a recipient of Medicare/Medicaid
• Case reported to OCR by RN on behalf of pt.
VNS of Western MA:
OCR Action II
• By November 1998 the VNS had entered into a
compliance agreement with the OCR:
– Services for LEP patients were restored
– VNS contracted with a telephone interpretation
agency and instructed staff re its utilization
– Bilingual staff were recruited and hired and
matched to patients when possible
– The VNS was deemed by the OCR to be once
again in compliance with Title VI of the Civil
Rights Act of 1964 and eligible for federal money
Studies on Language
• Satisfaction
• Access
Utilization of Health Care
Quality of Care
Impact of Language Barriers on
Patient Satisfaction in an
Emergency Department
• Survey of 2333 pts in 5 urban academic EDs
• 15% NES (? LEP status)
• Overall satisfaction: 52% for NES vs. 71%
for ES
• Willingness to return: 86% for NES vs. 91.5%
for ES
• NES pts more likely to report overall
problems with care, communication and
Carrasquillo O et al JGIM 1999
Effect of Spanish Interpretation
Method on Patient Satisfaction
• 233 Eng-speaking [ES] and 303 Span-speaking
[SS] pts in CO urban walk-in clinic, mean age 32
• 128 of SS seen by language concordant MD [LC]
• 59 SS used AT&T, 69 SS used family members,
47 SS used ad hoc interpreters
• Overall satisfaction was identical for ES, LC, and
AT&T at 77 % Vs 54 % for those using family
and 49% for those using ad hoc interpreters
Linda Lee et al, JGIM 2002
Patient Assessment of
Medicaid Managed Care
• Consumer Assessment of Health Plans
Study [49,327 PTs/14 states, 1999-2000]
• Linear regression model within/between
plans; telephone/mail survey in Eng & Span
• NES reported lower ratings of care [access,
timeliness, provider communication, staff
helpfulness, & composite]
• White NES and Hispanic Spanish-speakers
clustered in worse plans
• Most observed racial/ethnic difference in
ratings attributable to within plan variation
including those for NES Asians
Weech-Maldonado et al, JGIM 2004
Importance of MD Training in
Use of Interpreters in the OPD
• 158 MD questionnaires about last clinic visit
involving an interpreter [?type] at SFGH
• 85 % satisfied with ability to Dx and Rx; but
only 45 % satisfied with ability to educate
and empower the PTs about Dx, Rx, meds
• Previous training in interpreter collaboration
was associated with higher IS use and
better satisfaction with medical care
Karliner L et al, JGIM 2004
Studies on Language
• Satisfaction
• Access
Utilization of Health Care
Quality of Care
One in Five Have Gone Without Care When
Needed Due to Language Obstacles
19% Have Not sought care when needed due to language barrier
HQ11: In the course of the past year, how many times were you sick, but decided not
to visit a doctor because the doctor didn’t speak Spanish or have an interpreter?
Source: Wirthlin Worldwide 2002 RWJF Survey
Racial/Ethnic Differences in
Children’s Access to Care
• Data from 1996 Medical Expenditure Panel Survey
• 6900 US children, 9% lacking usual source of care
• 6.0% of Whites, 12.5% of AAs, 17.2% of Hispanics
• For Hispanics, 40.7% were interviewed in Spanish,
59.3% were interviewed in English
• Hispanic LEPs 27% as likely as Whites to have
regular source of primary care
• No difference between English-speaking Hispanics
and Whites
Weinick RM et al Am J Public Health 2000
Slide 7
Smoking Cessation Counseling
Percent of current smokers counseled by physician to quit
Source: The Commonwealth Fund 2001 Health Care Quality Survey.
Studies on Language
• Satisfaction
• Access
• Utilization of Health Care
• Quality of Care
• Costs
• Interventions
Does a Physician-Patient Language
Difference Increase the Probability of
Hospital Admission?
• Prospective observational study of 653 adult [AP] and 79 pediatric
[PP] pts in the ED at NYU Med Center Queens
• 14.7% of APs and 12.7% of PPs preferred non-English [NES]
• 52% of NES APs and 17% of NES PPs used “interpreters”
• No trained or professional interpreters were used
• NES APs were more likely to be admitted than ES controls, [35%
vs. 21%, RR 1.70 {1.14-2.53}]. No difference for PPs.
• Difference persisted after multivariate analysis for age,
gender, acuity level, and presence of an “interpreter”.
Lee ED et al Acad Emerg Med 1998.
Effect of English Language
Proficiency on Length of Stay I
• Retrospective review of administrative data
on consecutive admissions to 3 major
Toronto teaching hospitals 1993-1999
• LOS differences analyzed for 23 medical
and surgical conditions [59,547 records]
and then meta-analysis of 220 case mix
groups [189,119 records]
• Similar analysis for in-hospital mortality
John-Baptiste A et al, JGIM 2004
Effect of English Language
Proficiency on Length of Stay II
• LOS for LEP patients longer for 7 of 23
conditions [unstable coronary syndromes and
chest pain, CABG, stroke, craniotomy,
diabetes, hip replacement, GI procedures]
• Differences range from 0.7 to 4.3 days
• Overall LEP LOS 6% longer [ approx 0.5 days ]
• No increased risk of in-hospital death
John-Baptiste A et al, JGIM 2004
Studies on Language
• Satisfaction
• Access
• Utilization of Health Care
• Quality of Care
• Costs
• Interventions
Ethnicity as a Risk Factor
for Inadequate Emergency
Department Analgesia
• 139 pts with long bone fracture in UCLA ED
• 108 NHWs, 31 Hispanic (42% NES, ?LEP)
• Hispanics twice as likely to get no ED pain
Rx [OR 7.46; 95% CI, 2.22-25.02; p<0.01]
• NES status was borderline significant
predictor [OR 3.12; 95% CI, 0.98-9.83;
Todd KH et al JAMA 1993
Understanding Instructions for
Prescription Drugs Those Prescribed
No Interpreter Needed
Understood Instructions
Did Not Understand
Interpreter Needed
Not Available
No instructions Given
Source: Andrulis D, et al. What a Difference an Interpreter Can Make:
Health Care Experiences of Uninsured with Limited English Proficiency, March 2002
Quality of Diabetes Care for NonEnglish-Speaking patients: A
Comparative Study
• Retrospective cohort study of 622 diabetics, 93
• Academic medical center and county hospital
• Virtually all LEPs (24 languages) arrived with
professional interpreters
• LEPs more likely to get
– 2 or more Hgb AlC per year
– 2 or more clinic visits per year
– 1 or more dietary consults
• No differences in other labs, complications, use of
other services, and total changes.
Tocher TM et al West J Med 1998
Studies on Language
Utilization of Health Care
Quality of Care
• Costs
• Interventions
Language Barriers and Resource
Utilization in a Pediatric ED
• 2467 patients in an urban, academic
pediatric ED
• 12% LEP, 8.5% with LB with MD
• For cases with LB:
– higher test ($145 vs. $104)
– Longer ED stay (165 vs. 137 minutes)
• Analysis of covariance:
– LB accounted for $38 and 20 minutes
Hampers Pediatrics 1999 LC et al
Does the Use of Trained Medical Interpreters
Affect ED Services, Charges, and Follow-up?
• Retrospective chart reviews of 503 pts in Boston Med Ctr ED
• CC: CP/SOB, HA, ABD pain, pelvic pain/vag bleeding
• 66 Eng-speakers [ESPs], 63 Spanish, Haitian, Cape Verdean
pts using hospital interpreters [IPs], 374 LEP pts not using
interpreters [NIPs]
• NIPs had shortest ED stay [p .001] and fewest tests [p .04]
and prescriptions [p .03]
• IPs were more likely to make clinic follow-up and less likely
to return to the ED than NIPs [p .03]
• Among non-admitted pts, return visit ED charges and total
subsequent 30 day charges were reduced for IPs compared
to NIPs and ESPs.
Bernstein J et al. Journal of Immigrant Health 2002; 4: 171-176.
Language Barriers in Health
Care: Costs and Benefits of IS
• Follow up analysis of intervention study at
major HMO as it increased interpreter
services [IS]
• Average cost of IS per LEP member $234/yr
• For HMO overall, total costs averaged $0.20
per member per month
• Average cost of IS encounter $79 at the
time which can be expected to decline with
increasing efficiency
Jacobs E, et al. AJPH 2004; 94:366-369
Studies on Language
Utilization of Health Care
Quality of Care
• Interventions
Effects of Interpreters on the
Evaluation of Psychopathology in
Non-English-Speaking Patients
• 2 Public hospitals in NYC with no official
• 30 psychiatric interpreter-interviews daily
• Interpreters were other pts, friends, family, staff
• Open discussions with providers and bilingual
• Content analysis of 8 audio-taped interviews
• Distortions resulted from interpreters’ poor
language skills, lack of psychiatric knowledge,
and attitudinal issues
Marcos LR Am J Psychiatry 1979
When Nurses Double as Interpreters:
Spanish-speakers [SS] in Primary Care
• 21 SS pts with first walk-in visit to primary care clinic
with untrained nurses used to interpret
• Transcripts revealed serious miscommunication that
affected understanding or credibility in 1/2 cases
• MDs resisted reconceptualization in face of
• Nurse provided data expected clinically vs. actual
• Nurse interpretation reflected unfavorably on pts
• Pts used cultural metaphors incompatible with
Western clinical nosology not always interpreted
Elderkin-Thompson et al, Soc Sci Med 2001
Impact of Interpretation
Method on Clinic visit Length
• Time motion study of 613 visits to PCU in
RI with 28% LEP pts [90% Span-speakers]
• Interpreted pts spent longer in clinic [93.6
vs. 82.4] and w/ provider [32.4 vs. 28.o]
• Patients using telephone and patientprovided interpreters took longer; those
using hospital interpreters did not
• Authors calculated potential cost savings
of reduced telephone usage and more
efficient MD utilization in terms of potential
hospital interpreters hired
Fagan MJ et al JGIM 2003; 18: 634-638
Medical Interpreters Have
Feelings Too I
• Anonymous questionnaire of all 22
members of interpreter service of GRC
• 5 had exposure to severe trauma [war,
torture, detention, beatings]
• 7 reported more than 50 % of sessions
involved patients with exposure to
• 5 frequently experienced difficult feelings
during interpreting sessions
Medical Interpreters Have
Feelings Too II
• 66 % had frequently painful memories
• 83 % reported seeing patients outside
of the consultation setting
• Interpreters expressed the need to talk
and share feelings after the session
with the medical doctor [83 %] or with
relatives or spouse [44 %]
Louton L et al Soz Praventivmed 1999
Mandates for Medical
Interpreter Services
CLAS Standards
Office of Civil Rights [ORC] position
State laws [26 states and increasing]
Regulatory and review organizations
Risk management
Possible cost savings, market
Outcomes, quality
Massachusetts ED Interpreter Bill
[Effective July 1, 2001]
• Section 25J. Every acute-care hospital shall
provide competent professional interpreter
services in connection with all emergency room
services and acute inpatient psychiatric services
provided to a non-English- speaker or person
who has difficulty in speaking or understanding
the English language.
• Section 3c. Any non-English- speaker who is
denied effective health care services by a health
care provider by reason of the provider’s not
providing competent professional interpreter
services should have a right of action in a
superior court.
• Governmental units are to reimburse the cost of
interpreters for any mandated provider.
Selected Issues re
• Documentation: language status of patient
in IS; interpreter utilization by site, shift,
language, etc.
Risk Management: informed consent, staff
education re expectations and availability
Clinical outcome measures including
satisfaction, utilization, and quality
Research inclusion and activity, related
Training activity for staff and interpreters;
notification of rights for patients
E Hardt 2005
Might Language Competence
Facilitate Cultural Competence?
• Skills training viz language may invite and
synergize with efforts to learn content and
change attitudes while starting with a less
threatening set of goals
• Interpreter Services Department often
catalyze/lead organizational efforts at CC
• Methodology of organization’s approach to
language-based disparities can model
approach to other areas of disparities and
growth potential
E Hardt 2005
References and Bibliography
• See NCIHC website [ National Council
on Interpreting in Health Care],
• for annotated
bibliography August 2003
• email me at:

Slide 1