Access to Language Interpretation Services and its Impact ......the following: if language barriers are addressed, if the intervention involved a trained interpreter, and if clinical

April 2018

Access to Language Interpretation Services and its Impact on Clinical and Patient Outcomes: A Scoping ReviewNazeefah Laher, Anjum Sultana, Anjana Aery, & Nishi Kumar

Wellesley Institute works in research and policy to improve health and health equity in the GTA through action on the social determinants of health.

By: Nazeefah Laher, Anjum Sultana, Anjana Aery, & Nishi Kumar

Report© Wellesley Institute 2018

Copies of this report can be downloaded from www.wellesleyinstitute.com.

Statement on Acknowledgement of Traditional LandWe would like to acknowledge this sacred land on which the Wellesley Institute operates. It has been a site of human activity for 15,000 years. This land is the territory of the Huron-Wendat and Petun First Nations, the Seneca, and most recently, the Mississaugas of the Credit River. The territory was the subject of the Dish With One Spoon Wampum Belt Covenant, an agreement between the Iroquois Confederacy and Confederacy of the Ojibwe and allied nations to peaceably share and care for the resources around the Great Lakes.

Today, the meeting place of Toronto is still the home to many Indigenous people from across Turtle Island and we are grateful to have the opportunity to work in the community, on this territory.

Revised by the Elders Circle (Council of Aboriginal Initiatives) on November 6, 2014

10 Alcorn Ave, Suite 300Toronto, ON, Canada M4V [email protected]

mailto:[email protected]

TABLE OF CONTENTS

Introduction .................................................................................................................................................1

METHODS ....................................................................................................................................................2

Search Strategy and Study Selection .....................................................................................................2

Data Extraction and Synthesis ..............................................................................................................3

Terms and Definitions ...........................................................................................................................4

RESULTS .......................................................................................................................................................5

Quality Appraisal ...................................................................................................................................5

Description of Studies ...........................................................................................................................5

Theme 1: Access to and Mode of Interpretation ..................................................................................6

Theme 2: Clinical Outcomes of Interest ...............................................................................................8

Theme 3: Patient Outcomes ................................................................................................................10

Discussion ..................................................................................................................................................11

Interpretation of Findings...................................................................................................................11

Conclusion .................................................................................................................................................15

Appendices .................................................................................................................................................16

Appendix A ...........................................................................................................................................16

Appendix B ...........................................................................................................................................17

Endnotes ....................................................................................................................................................31

LANGUAGE INTERPRETATION SERVICES SCOPING REVIEW - WELLESLEY INSTITUTE 1

IntroductionLanguage barriers are a significant challenge in ensuring equitable access to health care. 1,2,3,4

Within the Canadian context, language is often cited as an obstacle to health care services for

immigrant populations, in particular newcomers.5 Patients and families who do not speak an

official language face obstacles accessing health services,6,7 receiving physician and hospital

care, as well as participating in health promotion and prevention programs.8 Language

barriers can result in inappropriate medical testing,9 increase the risk of adverse medication

reactions,10 and pose a significant barrier to medical comprehension.10 It has also been

associated with increased risk of hospital admission,11 medication adherence,12,13 less optimal

palliative care,14 and disparities in diagnostic testing.15 Also language barriers appear to have

measurable impacts on patient satisfaction, specifically on communication with health

care providers16 and on the care received.17 This carries over to health care providers who

also experience dissatisfaction in their interactions with patients when there is a language

barrier.18

In Ontario, data suggests that the issue of language barriers may impact more and more

residents. Census data from 2016 shows an increasingly diverse linguistic landscape.19 Nearly

half of all Canadians with a mother tongue other than English or French live in Ontario20

and almost 15 percent of Ontarians and 25 percent of Torontonians speak a non-official

language at home.20 Furthermore, approximately 80 percent of Ontarians who have no

knowledge of English or French are concentrated in Toronto.21 This speaks to the growing

need for linguistic accommodations in Toronto and Ontario. While Census data22 says that

vast majority of Canadians speak one of the official languages, it is not clear exactly how many

of Canadians are fully fluent or how many have limited official language proficiency. A 2001

review23 of the research literature related to language access suggested that 1 in 10 Canadians

may need an interpreter in a health care setting. It’s not just new immigrants that experience

language barriers in health care settings. Some long-term immigrants who have been fluent

in English or French for most of their lives are losing their English skills due to the impact

of dementia.24 In the era of Patients First25, which provides an opportunity to ensure that

patients have the right care, in the right place, at the right time, we also need the necessary

supports in place so patients can make informed decisions about their health care.

Ontarians who are not proficient in, or unable to speak an official language (English or

French), face inequities in health care settings. Health Quality Ontario defines health

equity as the ability of all people “to reach their full potential [and] receive quality care that

is fair and appropriate for them – regardless of where they live, what their economic and

social status, language, culture, gender or religion [is].”26 It is important to consider the

health equity impacts of language barriers in health care in the context of an increasingly

multicultural, multiracial, and multilingual province such as Ontario.27, 28


Despite the growing body of work on the impact of language barriers on access to health care

services, and the growing linguistic diversity in Ontario, there has been limited uptake of

formalized language supports across Ontario’s health care system. In a review of strategies to

increase language accessibility for patients who have limited English proficiency, a number

of supports are identified such as bilingual providers, non-professional interpreters such as

family members and volunteers, translated written materials, computer-assisted translation,

and most common, language interpretation services.29 Some health care institutions in

Ontario offer in-person professional language interpretation services to patients, but in

practice, many use informal methods such as communicating through bilingual hospital staff

or the patients’ family members.30

The purpose of this scoping review is to examine the impact of language interpretation

services on patients’ health and clinical outcomes to better understand its effectiveness as an

intervention. We chose language interpretation services compared to other interventions to

address language accessibility because of the formalized nature of the intervention as well as

to provide relevancy to similar interventions within the Canadian context.31

Systematic reviews that have examined the efficacy of language interpretation services have

focused on specific settings such as in psychiatric services32, or excluded outcomes of interest,

for example relating to the clinical experience of accessing health care services such as length

of stay, and re-admission rates.33 In general, these systematic reviews found that trained

interpreters were more effective than untrained interpreters in producing positive patient

outcomes when it came to psychiatric assessment and diagnoses, patient understanding of

treatment plans, patient satisfaction, and errors of clinical significance. However, many of the

studies in these reviews were conducted nearly 10 years ago and do not include more recent

research studies. Moreover, there has been limited exploration of the effectiveness of these

services across all potential modes of interpretation services such as in-person, phone, video

among others.

METHODS

Search Strategy and Study Selection

This scoping review was conducted using the Arksey & O’Malley methodology.34 We focused

on peer-reviewed academic literature and followed these steps: 1) identified our research

question; 2) selected our databases; 3) developed our search strategy and searched for

relevant studies; 4) used inclusion and exclusion criteria to select relevant studies; 5) charted

the data; 6) collated and summarized the results.

After identifying our research question (“What is the link between access to interpretation

services in health care and patient outcomes?”), we developed a search strategy with support


from a librarian at the University of Toronto. We adapted this search strategy to effectively

search our five selected scholarly databases (Medline, Scopus, PubMed, PsychInfo, and

Embase) (Appendix B). We selected these databases because of their focus on health and

social sciences. We selected studies based on the following inclusion criteria: articles had

to be peer-reviewed, in English, published in the last 17 years (2000 onwards), focused on

the provision of interpretation services in health care settings, and have a patient or clinical

outcome. This project focused on the needs of people who had barriers with spoken language

as well as the use of formal interpretation. As such, interventions that exclusively focused

on informal interpretation or those that did not address spoken language barriers were

out of scope for our project. Subsequently, we excluded research that exclusively focused

on non-professional or informal interpretation (such as by volunteers, family members

or by healthcare providers), and studies that focused on interpretation for American Sign

Language, or assessments of aides and tools for addressing communication disorders.

Articles that met the inclusion criteria were extracted and we collated key information on

relevant factors including: study design, country/region, type of service provided, how service

need was identified, type of interpretation provided, interventions, outcomes of interest,

study population, clinical care setting, and key findings.

Data Extraction and Synthesis

Our initial search of the five selected databases yielded 9,961 articles. A total of 3,691 articles

were excluded due to duplication. The remaining 6,270 articles were screened by titles, then

abstracts, and then full text screening, leaving a total of 30 articles that met all inclusion

criteria.

Medline: 2,249 Embase: 2,034CINAHL: 2,835

PsychInfo: 1,620Scopus: 1,223

Title Screened: n = 6,270

Abstract Screening: n = 1,845

Full Text Screening: n = 114

Full Text Screening: n = 114

Duplicates Removed: n = 3,691

Excluded: n = 4,425

Excluded: n = 1,731

Excluded: n = 84


At every stage of the screening process, each article was screened by two of the four

researchers and all four researchers resolved conflicts during the screening together.

Abstract titles were screened to identify if the study was relevant to language interpretation.

We then created a data extraction form to screen abstracts. The screening form identified

the following: if language barriers are addressed, if the intervention involved a trained

interpreter, and if clinical or patient outcomes are assessed. We used this extraction form

again for the full text screening. During this screening process we chose to exclude qualitative

studies, as we were primarily interested in measurable health or clinical outcomes. Data

extraction was divided amongst all four researchers. Throughout the screening and extraction

process, researchers met frequently to ensure consistency was maintained throughout the

process. A total of 30 articles were included for review.

Research team members then independently reviewed the results and identified themes

based on our outcomes of interest. These included but are not limited to: number of

clinically significant errors, admission and readmission rates, length of stay and length of

appointment, preventive care, chronic disease management, mental health assessments,

quality of care, medical adherence, service utilization and health care usage. The research

team then discussed and built consensus on key themes which were summarized and

included in the table in Appendix B.

As a point of reference, the table below includes frequently used terminology with an

accompanying definition.

Terms and Definitions

Term Definition

Language Interpretation Services An intervention that can increase language accessibility of the health care system.

Trained Interpreters Trained interpreters have received some form of training, whether formal in a classroom or on the job from the health care institution or organization that they are working at.

Untrained Interpreters Untrained interpreters are a broad term to refer to any person who engages in interpretation who has not been formally trained, whether in a classroom setting or on the job. In this scoping review, we refer to ad-hoc interpreters (individuals who conduct interpretation services without any clear training) as untrained interpreters.

In-person interpreters In-person interpreters are interpreters who are physically present during the health encounter a patient has with their health care provider.

Remote interpretation A modality of interpreting that is conducted at a distance, usually by telephone or video. Remote interpretation can either be consecutive or simultaneous.

Consecutive Interpretation In consecutive interpretation the interpreter listens to a unit of speech in the source language and then conveys that message into the target language. This mode is more often used in health care


Simultaenous Interpretation Simultaneous interpretation is not necessarily interpreting word to word, it is the mode in which the interpreter interprets the message into the target language at almost the same time as the original message is being said.There are studies included in this review that use “remote simultaneous medical interpretation,” referring to simultaneous interpretation using a telephone or video conferencing.

Language Concordant Providers Health care providers who are able to speak the same language as their patients.

LEP (Limited English Proficiency) Patients who do not speak English or are not proficient in English.

ESP (English Speaking Patients) Patients comfortable speaking English/choose to speak English in health care settings

Length of Stay Time spent in the hospital or in health care settings from intake to discharge

Length of Appointment Time spent during an appointment with a health care provider from the beginning to the end of the appointment

Clinical Outcomes For the purposes of this scoping review, we define clinical outcomes as issues related to institutional level factors impacting a patient’s experience within the health care system such as length of stay or medical errors for example.

Patient Outcomes For the purposes of this scoping review, we define patient outcomes as issues related to a patient’s uptake of services such as preventive services or the impact of LIS on access and quality of treatment services.

RESULTS

Quality Appraisal

After data extraction was completed, the researchers conducted a quality appraisal of the 30

articles included in the review using the Quality Assessment Tool for Quantitative Studies,

developed by the Effective Public Health Practice Project (EPHPP).35 One reviewer completed

the quality appraisal per article. There were six strong articles, 16 articles that were of

moderate strength, and eight weak articles (Appendix B).

Description of Studies

Thirty articles were included in our scoping review (Appendix B). Of these, the majority of

the studies were cohort studies (14) and cross-sectional studies (9). The review included two

randomized control trials, one randomized clinical trial, one case control study, and three

pre-post observational studies. The majority of the studies (25) were from the United States.

There were also studies from Switzerland (1), the United Kingdom (1), and Australia (3).

Through a process of iterative thematic coding, we categorized the outcome data extracted

from the finalized list of screened articles into three main themes: Access to and Mode of

Interpretation; Clinical Outcomes; and Patient Outcomes.


For the first theme: Access to and Mode of Interpretation, we explored the effectiveness of

professionally trained interpreters compared to untrained interpreters, such as volunteers

and family members. We also considered differences between five different modes of

interpretation (in-person, telephone, video, bilingual providers, and remote simultaneous

interpretation services) as well as how frequency and consistency of interpretation services

affected outcomes.

The second theme focused on Clinical Outcomes, which in our scoping review refers to how

language interpretation services impacted institutional and organizational level outcomes.

This can include various outcomes associated with a health care encounter such as duration

of a clinic visit and re-admission rates.

Finally, the last theme focused on the impact of interpretation services on various types of

Patient Outcomes such as changes in diagnostic, treatment and management outcomes, as

well as uptake of preventive services.

Theme 1: Access to and Mode of Interpretation

Comparing Trained Interpreters vs. Untrained or No Interpreters

In our scoping review, 14 studies examined the effectiveness of trained interpreters compared

to untrained or no interpreters. Of the 14 articles, six36, 37, 38, 39, 40, 41 showed that trained

interpreters led to better clinical outcomes, three34, 35, 42 showed that trained interpreters led to

better patient outcomes, and five43,44, 45, 46, 47 showed a decrease in clinical errors with trained

interpreters.

One of the most substantial differences when comparing trained interpreters to untrained or

no interpreters was seen in differences in errors of clinical significance. Five studies,40,41,42,43,44

were able to demonstrate that the rates of errors were less frequent and less severe for trained

interpreters compared to untrained interpreters or no interpreters. Flores and colleagues

found that the proportion of errors was lower for professionally trained interpreters (12%)

compared to untrained interpreters (22%) and no interpreters (20%).42 There were higher rates

of omissions and incorrect interpretation errors for untrained interpreters (46.3%; 31.6%) and

no interpreters (54.2%; 35.9%) compared to trained interpreters (41.9%; 13.6%), respectively.42

This was also seen by Napoles’ research team who found the odds of a moderately or highly

clinically significant error was lower for in-person trained interpreters compared to untrained

interpreters (OR = 0.25, p = 0.05).40 There was also a reduced likelihood of understanding of

health issues such as heart disease risk when family and friends were used as interpreters

compared to professional interpreters.41

The evidence from our scoping review indicates that having access to trained interpreters

improves clinical and patient outcomes in comparison to ad-hoc, informal, or no interpreters.


Comparing Different Modes of Interpretation

Seven studies36, 40, 48, 49, 50, 51, 52 evaluated different modes of interpretation. This included

both simultaneous interpretation and interpretation (table 1), delivered in-person, through

telephone, and through video conferencing. Some studies also used bilingual providers who

provided services in the patient’s language. The evidence on which mode was best differed

across studies and was dependent on multiple factors including the health care setting

context and type of outcome.

In another study,47 patients that had in-person interpreters were more likely to be seen

faster (65%) than those who had telephone interpreters (52%) or bilingual providers (42%).

Patients with in-person interpreters (8%) also had a lower frequency of complex medical

administration compared to their counterparts who received alternative support from

telephone interpreters (15%) or bilingual providers (15%).47

Length of encounter also varied depending on mode of interpretation. For example,

one study that looked at errors of clinical significance found that remote simultaneous

medical interpretation encounters were on average shorter and non-remote-simultaneous

medical encounters (which includes remote consecutive as well as in person simultaneous

and consecutive interpretation) were 12 times more likely to result in errors of clinical

significance.44 Remote simultaneous medical interpretation encounters were also associated

with greater rates of detection and treatment of depression, however it was not seen to be

statistically significant.45 The mode of interpretation may also have implications for patients’

understanding of diagnoses. Lion and colleagues found that parents were more likely to

correctly identify their child’s diagnosis if they were exposed to video interpretation (74.5%)

compared to telephone interpretation (59.8%).46

Impact of Frequency and Consistency of Interpreter Use

There were four studies 37,53,56,60 that commented on the impact of increased frequency of

interpreter use on clinical and patient outcomes.

One study50 found that when a patient had more frequent interpreted health care encounters,

there was a significant reduction in hospital admissions due to hyperkalemia, hypertension

and fluid overload. Specifically, the greatest reduction was seen for rates of hyperkalemia.

When a patient did not have any health care encounters interpreted, there was an average

of 10 hospital admissions for hyperkaliemia.50 However, when a patient had 5 interpreted

health care encounters, the number of hospital admissions for hyperkalemia dropped down

to 2.

In a study by Jimenez and colleagues, that measured the association between parental

language proficiency, interpreted care, and postsurgical pediatric pain management, it was

found that children of families who received less than 2 interpreted visits per day had higher


rates of post-analgesic pain score (1.6 ± 1.4 vs 0.7 ± 1.2; P = .004) compared to children with

more frequent interpretation.60

Researchers found that patient stroke rehabilitation patients who had more frequent

interpreter use had improved Functional Independence Scores.56 There was no evidence of

impact on other outcomes such as length of stay or discharge destination. The frequency

of interpreter use combined with the mode of interpreter use can also lead to a positive

impact. For example, one study found that patients who received 100% of their primary care

visits with language concordant providers (also known as bilingual providers) were the least

likely to have diabetes related emergency department visits compared to all other types of

interpretation which included in-person interpretation, and no interpretation.37

Theme 2: Clinical Outcomes of Interest

The impact of language interpretation services on clinical outcomes was observed in 17 of the

reviewed articles.

Length of Stay

For length of stay in hospital, evidence was mixed in hospital settings, and context specific,

with some studies finding that length of time increased when trained interpreters were

used while others finding it decreased the length of stay. Three studies54,55,56 reported that

interpretation increased length of stay, one study57 reported that interpretation decreased

length of stay, and one58, 59 reported no significant difference in length of stay.

Length of Appointment

Three studies36,47,57 provided evidence on the impact of interpretation on the length of an

appointment. In a study conducted by Fagan and colleagues, they showed that telephone

interpretation was associated with increased time spent with the provider (36.3 min vs.

28.0 min, P < 0.001), compared to patients not using an interpreter.36 In another study, 47 researchers looked specifically at the type of interpretation mode and the impact on

provider time and found patients who had in-person interpreters (116 min) had on average

significantly shorter throughput time compared to patients who had telephonic interpreters

(141 min) and bilingual providers (153 min). Researchers found that for every hour of

in-person interpretation (60 minutes), approximately 2 additional minutes were required

for an encounter with a bilingual provider and 9 additional minutes would be added for

an encounter using telephone interpretation. The third study57 found that limited English

proficiency (LEP) patients using a trained interpreter had longer appointments, such as

patients speaking Spanish (averaging 9.1 additional minutes of physician time) and Russian

(averaging 5.6 additional minutes of physician time) compared to English speaking patients

who did not use an interpreter.


Service Utilization

Clinical outcomes related to service utilization were also impacted by access to interpretation

services, demonstrated in four studies.33,35,38,60 Two of these studies studies33,57 found that

using professional interpreters led to a significant increase in referrals for primary, specialty,

and psychological care, and showed an uptake in the number of office visits. For general

medical referrals, access to both untrained interpreters and trained interpreters allowed

access to 48 percent and 42 percent of patients respectively, compared to 31 percent for

patients who did not use an interpreter.57 However, a study conducted by Sarver & Baker,

found no significant difference in appointment compliance rates for patients with and

without an interpreter.38 One study35 also found that English speaking patients had a

significantly greater volume of service use during an emergency department visit compared to

patients with limited English proficiency with and without an interpreter.35

Admission and Readmission Rates

In the four studies33,54,56,61 that examined admission and readmission rates in hospitals,

there was consistent evidence that showed that language interpretation services decreased

admission rates and re-admission rates. One5 found that upon giving patients access to

24-hour interpretation, 30-day readmission rates decreased by approximately 5% more during

the study period, compared to the 18 months before the intervention.9 Likewise, Lindholm

and colleagues56 found that patients who received interpretation at both admission and

discharge were less likely to be readmitted 30 days after discharge (14.9%) compared to those

only receiving interpretation at admission (16.9%), only at discharge (17.6%), or those who did

not receive interpretation at admission nor discharge (24.3%).

Medical Errors

There were three studies40,42,44 that evaluated how access to language interpretation services

decreased the likelihood of errors of clinical significance. Certain modes of interpretation

were less likely to produce errors in general as well as errors of clinical significance. For

example, one study44 found that remote simultaneous medical interpretation services were

less likely to produce errors of clinical significance. Non-remote simultaneous medical

interpretation was associated with a 12 times greater chance of potential medical errors

of moderate or greater clinical significance compared to remote simultaneous medical

interpretation. Non-remote in this case refers to in person simultaneous and consecutive

interpretation as well as remote consecutive interpretation. We also see that professional

interpreters (12%) are less likely to commit errors compared to ad hoc interpreters (22%) and

no interpreters (20%).42


Theme 3: Patient Outcomes

Our scoping review has found 11 studies that show the efficacy of language interpretation

services in the diagnosis, treatment, and management of various health conditions, as well as

uptake of preventive care.33, 34, 35, 37, 39, 41, 49, 55, 57, 59, 60

Diagnostic, Treatment and Management Outcomes

Four studies indicated the impact of interpretation services had on facilitating the diagnosis,

treatment and management of health conditions.39,55,57,60

Eytan and colleagues found that the use of interpreters was helpful in encouraging patients

to report traumatic events and psychological symptoms.57 The report of adverse events and

past exposure to situations like war, the violent death of a relative or a missing relative was

more frequently reported when trained interpreters (77%) were present compared to when

no interpreters (55%) or untrained interpreters (46%) were present.57 The use of trained

interpreters was also associated with increased referrals to medical care (42%) compared to

no interpreters (31%).57

The use of interpretation has also been seen to lead to higher rates of pain control, timely

response to patient pain, as well as help from staff compared to patients who did not always

receive interpretation.62 For stroke rehabilitation patients increased interpretation usage was

also associated with greater improvement in Functional Independence Measure efficiency

score, which is a significant measure for stroke rehabilitation.55 Patients with refugee status

who used interpreters (27%) were seen to have greater improvements in mental health

outcomes such as for scores on the Impact of Events Scale compared to refugees without

interpreters (14%).39

Preventive Health Services

Four studies in our scoping review examined the impact of language interpretation services

on the uptake of preventive health services.34,35,37,63

All studies that examined preventive health service use found that language interpretation

services increased uptake. Patients who had access to trained interpretation services were

more likely to have had rectal exams conducted compared to people that did not have access

to interpretation services (0.26 vs. 0.02; P =0.05; not shown).34 In another study60 it was found

that trained in-person medical interpreters increased the likelihood of mammograms (OR

= 1.85), clinical breast exams (OR = 3.03), and pap smears (OR = 2.34), compared to patients

that did not use any interpreters. Patients who had bilingual providers were also twice as

likely to have a clinical breast exam or a pap smear compared to patients who did not have

any interpreters.60 In another study, it was found that when patients received interpretation


services, they had higher rates of preventive services uptake for rectal exams and fecal occult

blood tests compared to their peers who did not receive any interpretation services.35

DiscussionThis is the only scoping review to our knowledge that has synthesized evidence on the

impact of language interpretation services on both the clinical and patient outcomes

across a range of health care settings. There is a significant body of evidence that trained

interpreters were consistently effective at producing better clinical and patient outcomes

for patients in comparison to patients who had access to only untrained interpreters or

no interpreters.33,34,40,41,42,49,57 This review also demonstrated that the issue of language

accessibility in health care was addressed through many different modes of interpretation. 40,45,46,47,48

Interpretation of Findings

Access to and Mode of Interpretation

Overall, the literature reports that trained interpreters are more effective than untrained

interpreters or no interpreters. The use of trained interpreters was associated with lower

rates of readmissions,56 fewer errors of clinical significance,42 and produced more favorable

outcomes for uptake of preventive services.35

While access to language interpretation appears to be beneficial with respect to clinical

outcomes and patient outcomes, it is not possible to definitively conclude which mode of

interpretation is most effective. The most effective mode of interpretation varies across

contexts and subgroups of patients depending on things such as the type of health problems

or acuity. There is no such thing as a one-size-fits-all approach in medical interpreting.

Telephone interpretation for example may not be the best approach to patients with dual

diagnosis, those who are hard on hearing clients, or children [new endnote]. This is because

the studies included in this review did not compare all available modes of interpretation

and it is difficult to evaluate the role of varying contexts. It was evident that context played an

important role in determining which mode to use, however demonstrating which mode is

best for each context is will require further research and exploration.44,64

Previous scoping and systematic reviews have not evaluated the impact of frequency and

consistency on language interpretation services. In our scoping review, we found evidence

suggesting that the frequency of interpreted health care encounters as well as the consistency

of access to language interpretation services seem to be a contributing factor to clinical and

patient outcomes.37 More research is required to better establish the relationship between

frequency of interpretation services and patient outcomes. However, the available evidence


suggests the value of having consistent access to interpretation services for patients with

limited proficiency.

Impact of Language Interpretation Services on Clinical Outcomes of Interest

This review found that the impact of language interpretation services on length of stay in

hospitals after length of stay was mixed. Length of stay is often used as a measure of health

care system efficiency and some studies suggest that trained interpreters can reduce length

of stay of patients by almost a day.54,56 While a shorter length of stay is often considered more

favorable, there is evidence to suggest more time is needed for appropriate high-quality

care.65,66,67 Therefore, even though studies showed both an increase and decrease in length

of stay with the use of an interpreter, more research is needed to understand whether an

increased length of stay is indicative of more appropriate and thorough care in health care

settings.

Studies suggest that length of appointment is longer when an interpreter is used, but how

much longer depends on the mode of interpretation.36,47,53 Increased length of appointment

can be due to many factors, such as more information being exchanged or extra time due

to the interpretation.36 Telephone interpretation appears to take more time than in-person

interpreters. This may be due to challenges when using the phone because there is an

inability to communicate through non-verbal means.47 In-person interpretation may be more

responsive to the needs of patients as there is a greater ability to convey and pick up on verbal

cues, body language and other interpersonal cues compared to telephone interpretation.

Additionally, a decrease in health care usage does not necessarily imply more appropriate

care.53 As Hampers and colleagues found, health care usage, or service utilization rates as

noted in their study, for patients receiving interpretation were similar to that of English

speaking patients. This study suggests that access to interpretation services removed a barrier

between the physician and the patient and brought patient service utilization closer to that

of English speaking patients.53 in many cases patients with language barriers are unable

to communicate with their health provider, and are thus are receiving fewer services than

English speaking patients.33 Professional interpreters open a line of communication between

patients and health care providers, thus allowing providers to have a more comprehensive

understanding of patients’ symptoms, pain, queries, and discomforts.57 In two studies33,57 this

led to increased referrals to psychological care and specialists, allowing patients to receive the

appropriate care.

Professional interpretation has been consistently shown to decrease hospital admission and

readmission rates. Patients with limited English proficiency have higher readmission rates

compared to English speaking patients, but one study showed that providing services from

trained interpreters significantly decreases the likelihood of being readmitted.68 The positive


impacts of interpretation throughout the care continuum contribute to this reduction in

admission and readmission rates. With improved health outcomes2, better communication

with health providers,57 and increased referrals to specialists57, patients are less likely to

experience adverse health implications that require admission and readmission to the

hospital.

Finally, the most definitive evidence that examined the impact of language interpretation

services on patient outcomes was related to the frequency of errors of clinical

significance.42,44,45 Trained interpreters were less likely to have clinically significant errors in

comparison to untrained interpreters or encounters with no interpreters. This is consistent

with previous systematic reviews that also found that patients with limited English

proficiency experience worse health care quality with untrained interpreters compared to

patients with trained interpreters.69,70 These studies also compared a range of interpreter

modes and two studies44,45 have found that remote simultaneous interpretation services were

the best at reducing the rate of medical errors of clinical significance

Impact of Language Interpretation Services on Health Outcomes of Interest

Language interpretation services were found to be effective in improving uptake of preventive

services.34,35,37,61

This was found for a range of services including mammograms, clinical breast exams, and

rectal exams. This is significant to note because there has been a substantial body of research

that has demonstrated disparities in breast and cervical cancer screening rates for racialized,

ethnic minority and immigrant women.69 Addressing language barriers has been noted to be

an important step towards addressing cancer-related inequities for racially and ethnically

diverse patients as well as immigrants and newcomers.70 Evidence suggesting that language

interpretation services can contribute to reducing inequities between LEP patients and

English speaking patients (ESP) indicates a potentially effective solution to addressing health

inequities.

Finally, the most definitive evidence that examined the impact of language interpretation

services on patient outcomes was related to the frequency of errors of clinical

significance.42,44,45

Trained interpreters were less likely to have clinically significant errors in comparison to

untrained interpreters or encounters with no interpreters. This is consistent with previous

systematic reviews that also found that patients with limited English proficiency experience

worse health care quality with untrained interpreters compared to patients with trained

interpreters.71,72 These studies also compared a range of interpreter modes and two studies44,45

have found that remote simultaneous interpretation services were the best at reducing the

rate of medical errors of clinical significance.


Limitations and Areas of Future Research

According to Grant and Booth73 , a scoping review is a “preliminary assessment of potential

size and scope of available research literature [and] aims to identify nature and extent of

research evidence.” A limitation to this methodology is the comprehensiveness of searching

is limited by time and scope. Due to the inclusion and exclusion criteria, this scoping review

was not exhaustive and did not include all studies related to language interpretation. Part

of the exclusion criteria was that it only included studies from 2000 onwards. This enabled

the study to focus on the most recent evidence when considering evaluations of language

interpretation services, making them as closely relevant to the current context as possible. It

is possible that there are studies prior to 2000 that may have value in understanding language

interpretation and its relationship to clinical and patient outcomes.

This scoping review also highlighted gaps in the literature and areas of future research. While

this review did include studies from Europe and Australia, a significant proportion of the

included studies were conducted in the United States and none of the studies included were

from Canada. There are aspects of the Canadian context that this study could not address

such as having a publicly financed health care system, the presence of two official languages,

and the high degree of linguistic diversity in certain regions such as the Greater Toronto Area

(GTA).74 The lack of Canadian academic literature on language interpretation points to a need

for more research on this topic in a Canadian context.

While we did not exclude research from non-hospital settings, nearly all studies were

from hospital settings. From our survey of the research literature, there is a gap in studies

investigating the use of language interpretation services in non-hospital settings. Language

interpretation services in hospital settings may not be generalizable to settings such as

smaller scale family practices which do not have the same level of resources. Family doctors

are often the first point of entry into a health care system and reducing language barriers in

such a setting would be warranted to ensure that a patient’s complete clinical history is able

to be taken and a positive relationship is able to be built.

Additionally, this review found limited data on the financial costs of providing interpretation

services. Some studies looked at costs of programs or service use, but none of the studies

conducted any sort of cost effectiveness analysis or cost benefit analysis. In the future,

as language interpretation services continue to expand, it may be helpful to understand

the resources required for implementation, particularly regarding different modes of

interpretation. However, this must be compared against arguably the most important

variable, clinical and patient outcomes and the ability of patients to achieve their full health

potential.


ConclusionEnsuring equitable access to health care is an important priority that benefits all Ontarians.

It improves the health of communities and addresses health disparities. Language barriers

have been noted to be a significant issue preventing access to health care services. This is

especially pronounced in a province as linguistically diverse as Ontario. Ontario’s health care

system should be accessible to its diverse communities. Therefore, it is important to better

understand the effectiveness of interventions that seek to address language barriers and

facilitate greater language accessibility of the health care system.

Access to language interpretation services can enhance the accessibility of Ontario’s health

care system. As this scoping review suggests, language interpretation services can have

a measurable impact on the clinical and patient outcomes. Ontario’s linguistic diversity

is likely to continue to grow, so it is important that health care institutions and providers

across the province consider how language needs in health care settings are addressed and

managed.

Language accessibility will also become increasingly relevant to the operations of our

health care systems. As such, it is important that resources and energy be directed towards

evaluating interventions that aim to address these concerns. Future work should look at

further evaluating the impact of language interpretation services and other interventions that

improve language accessibility in the health care system within the Canadian context and

within Ontario specifically.

This review has demonstrated the value of providing patients with services from trained

interpreters. As Ontario becomes increasingly linguistically diverse, language barriers will

continue to pose a challenge for health and health equity. Ontarians deserve the opportunity

to ensure not only that they have the right care at the right time, but also that they have

the necessary supports to make informed decisions about their health care. Investing in

interpretation services, and continuing to evaluate best practices, are important steps for

Ontario to improve patient care and health equity.


Appendices

Appendix A

Search Strategies

Database Records Search Strategy Limits

Medline 2,249 [exp Patients/ OR (patient or patients).mp.] AND [exp Translating/ OR medical interpret*.mp. OR health interpret*.mp. OR cultural interpret*.mp. OR (translat* adj3 service*).mp. OR (interpret* adj3 service*).mp. OR (language* adj3 service*).mp. OR (interpreter or translator).mp. ]

Limit to [English language AND yr="2000 -Current" AND journal article]

Embase 2,034 [exp Patients/ OR (patients or patients).mp.] AND [exp Translating/ OR medical interpret*.mp. OR health interpret*.mp. OR cultural interpret*.mp. OR (translat* adj3 service*).mp. OR (interpret* adj3 service*).mp. OR (language* adj3 service*).mp. OR (interpreter or translator).mp.]

Limit to [English AND 13 to yr="2000 -Current"

CINAHL 2,835 [(MH "Patients") OR (TX patient OR patients)] AND [(MH "Translations") OR (MH "Interpreter Services") OR TX medical interpret* OR (TX health interpret* OR TX cultur* interpret*) OR (TX translat* N3 service*) OR (TX interpret* N3 service*) OR (TX language N3 service*) OR (TX interpreter OR translator)

Limit to [(Published Date: 20000101-20171231) AND English AND Journal Articles)

PsychInfo 1,620 [exp Patients/ OR (patient or patients).mp.] AND [medical interpret*.mp. OR health interpret*.mp. OR cultural interpret*.mp. OR (translat* adj3 service*).mp. OR (interpret* adj3 service*).mp. OR (language* adj3 service*).mp. OR (interpreter or translator).mp. OR (exp foreign language translation/ or exp foreign languages/ or exp interpreters/)]

Limit to [English language and journal article and yr="2000 -Current"]

Scopus 1,223 TITLE-ABS-KEY("Medical interpret*" OR "health interpret*" OR "language Pre/3 service" OR "translator" OR "interpreter" OR "interpret* Pre/3 service" OR "translat* Pre/3 service" OR "cultural interpret*") AND TITLE-ABS-KEY("Patient" OR "patients") AND TITLE-ABS-KEY("health" OR "health care" OR "health service" OR "healthcare")

Limit (PUBYEAR > 1999)


App

endi

x B

Sum

mar

y of

Fin

ding

s Ta

ble

Art

icle

Id

entif

icat

ion

Num

ber

Art

icle

Titl

eA

utho

r and

Ye

arSa

mpl

e In

form

atio

nTh

eme

1: A

cces

s to

and

Mod

es o

f In

terp

reta

tion

Com

pari

ng T

rain

ed In

terp

rete

rs

vs. U

ntra

ined

Inte

rpre

ters

vs.

No

Inte

rpre

ters

Freq

uenc

y of

Inte

rpre

tatio

n

Com

pari

ng D

iffer

ent M

odes

of

Inte

rpre

tatio

n

Them

e 2:

Clin

ical

Out

com

es

of In

tere

st

Leng

th o

f Sta

y

Leng

th o

f app

oint

men

t

Adm

issi

on/R

eadm

issi

on ra

tes

Serv

ice

utili

zatio

n

Med

ical

Err

ors

Them

e 3:

Hea

lth

Out

com

es

Prev

entio

n

Dia

gnos

tic

Qua

lity

App

rais

al

1D

etec

tion

of

Dep

ress

ion

with

Diff

eren

t In

terp

retin

g M

etho

ds

Am

ong

Chi

nese

an

d La

tino

Prim

ary

Car

e Pa

tient

s:

A R

ando

miz

ed

Con

trol

led

Tria

l

Leng

, C.F

., C

hang

rani

, J.,

Tsen

g, C

., &

G

any

F. (2

010)

.

Gro

up 1:

Gen

eral

LEP

ra

ndom

ized

to re

ceiv

e Re

mot

e Si

mul

tane

ous

Med

ical

Inte

rpre

ting

(RSM

I)

Gro

up 2

: Gen

eral

LEP

ra

ndom

ized

to re

ceiv

e us

ual a

nd c

usto

mar

y (U

&C

) in

terp

retin

g

Gro

up 3

: Lan

guag

e co

ncor

dant

pat

ient

s

Sett

ing:

New

Yor

k C

ity, U

SA

Stud

y D

esig

n: R

ando

miz

ed

Con

trol

Tria

l

Com

parin

g D

iffer

ent M

odes

of

Inte

rpre

tatio

n

RSM

I was

ass

ocia

ted

with

gre

ater

ra

tes

of d

etec

tion

and

trea

tmen

t of

dep

ress

ion,

how

ever

, the

resu

lts

wer

e no

t sta

tistic

ally

sig

nific

ant.

The

appa

rent

sup

erio

rity

of R

SMI m

ay

have

bee

n du

e to

pra

ctic

al is

sues

su

ch a

s tim

e an

d eff

icie

ncy.

RSM

I is

mor

e pr

ivat

e an

d si

mul

ates

a n

atur

al

conv

ersa

tion

betw

een

lang

uage

di

scor

dant

par

ticip

ants

; non

-ver

bal

cues

may

als

o be

mor

e ob

viou

s w

ithou

t a d

istr

actio

n by

third

per

son

in th

e ro

om

Stro

ng

2Eff

ect o

f Te

leph

one

vs V

ideo

In

terp

reta

tion

on

Pare

nt

Com

preh

ensi

on,

Com

mun

icat

ion,

an

d U

tiliz

atio

n

in th

e Pe

diat

ric

Emer

genc

y D

epar

tmen

t

A R

ando

miz

ed

Clin

ical

Tria

l

Lion

, K. C

, et a

l. (2

015)

.LE

P Yo

uth.

LEP

with

ano

ther

in

terp

rete

r mod

e. T

estin

g th

e eff

ect o

f tel

epho

ne

vs. v

ideo

inte

rpre

tatio

n on

com

mun

icat

ion

durin

g pe

diat

ric e

mer

genc

y ca

re.

Sett

ing:

Sea

ttle

, WA

, USA

Stud

y D

esig

n: R

ando

miz

ed

Clin

ical

Tria

l

Com

parin

g D

iffer

ent M

odes

of

Inte

rpre

tatio

n

Am

ong

208

pare

nts

who

com

plet

ed

the

surv

ey, t

hose

in th

e vi

deo

arm

w

ere

mor

e lik

ely

to n

ame

the

child

’s di

agno

sis

corr

ectly

than

thos

e in

th

e te

leph

one

arm

(P =

0.03

) and

le

ss li

kely

to re

port

freq

uent

laps

es

in in

terp

rete

r use

(P =

.04)

. No

diffe

renc

es w

ere

foun

d be

twee

n th

e vi

deo

and

tele

phon

e ar

ms

in p

aren

t-re

port

ed q

ualit

y of

com

mun

icat

ion

(P =

.43)

or i

nter

pret

atio

n (P

= .6

9).

Pare

nt-r

epor

ted

adhe

renc

e to

the

assi

gned

mod

ality

was

hig

her f

or

the

vide

o ar

m (P

= .0

04).

Use

of

vide

o in

terp

reta

tion

show

s pr

omis

e fo

r im

prov

ing

com

mun

icat

ion

and

patie

nt c

are

in th

is p

opul

atio

n.

Stro

ng


3Pr

ofes

sion

al

Lang

uage

In

terp

reta

tion

and

Inpa

tient

Len

gth

of S

tay

and

Read

mis

sion

Ra

tes

Lind

holm

, M.,

Har

grav

es J.

L.,

Ferg

uson

, W.

J., &

Ree

d, G

. (2

012)

.

Gen

eral

LEP

who

rece

ive

inte

rpre

tatio

n at

adm

issi

on/

disc

harg

e vs

. LEP

who

did

no

t

Sett

ing:

Mas

sach

uset

ts,

USA

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ohor

t

Leng

th o

f Sta

y

Patie

nts

who

did

not

rece

ive

inte

rpre

tatio

n du

ring

adm

issi

on +

dis

char

ge h

ad

a lo

nger

leng

th o

f sta

y (+

1.5

days

), co

mpa

red

to th

ose

who

had

inte

rpre

ters

dur

ing

adm

issi

on +

dis

char

ge

(p<0

.001

). In

terp

reta

tion

at d

isch

arge

onl

y w

as n

ot

sign

ifica

ntly

ass

ocia

ted

with

leng

th o

f sta

y, th

ough

in

terp

reta

tion

at a

dmis

sion

w

as.

Read

mis

sion

s

Patie

nts

rece

ivin

g in

terp

reta

tion

durin

g bo

th

adm

issi

on A

ND

dis

char

ge w

ere

less

like

ly to

be

read

mitt

ed

with

in 3

0 da

ys (1

4.9%

) co

mpa

red

to th

ose

rece

ivin

g it

at a

dmis

sion

(16.

9%),

disc

harg

e (1

7.6%

) or n

eith

er (2

4.3%

)(p

<0.0

01).

Patie

nts

who

got

in

terp

reta

tion

at a

dmis

sion

an

d/or

dis

char

ge w

ere

still

less

lik

ely

to b

e re

adm

itted

.

Stro

ng

4In

accu

rate

La

ngua

ge

Inte

rpre

tatio

n an

d Its

Clin

ical

Si

gnifi

canc

e in

the

Med

ical

En

coun

ters

of

Span

ish-

spea

king

La

tinos

Náp

oles

, A.

M.,

Sant

oyo-

Ols

son,

J.,

Karli

ner,

L.S.

, G

rego

rich,

S.

E., &

Per

ez-

Stab

le, E

. J.

(201

5).

Gen

eral

LEP

(n=3

2; fa

ce-

to-f

ace=

5; v

ideo

=27,

fa

mily

=5) v

s. L

EP w

ith

othe

r int

erpr

eter

mod

e (T

o co

mpa

re a

ccur

acy

of in

terp

reta

tion

usin

g in

-per

son,

vid

eo, o

r ad-

hoc

inte

rpre

ters

)

Sett

ing:

Cal

iforn

ia, U

SA

Stud

y D

esig

n: C

ross

-se

ctio

nal

Com

parin

g D

iffer

ent M

odes

of

Inte

rpre

tatio

n

The

odds

of a

mod

erat

ely

or h

ighl

y cl

inic

ally

sig

nific

ant e

rror

wer

e lo

wer

for f

ace-

to-f

ace

prof

essi

onal

in

terp

rete

rs th

an fo

r ad-

hoc

inte

rpre

ters

(OR=

0.25

, p=0

.05)

.

Stro

ng

5Kn

owle

dge

of

Hea

rt D

isea

se R

isk

Am

ong

Span

ish

Spea

kers

with

D

iabe

tes:

The

Role

of

Inte

rpre

ters

in

The

Med

ical

En

coun

ter

Wag

ner,

J.,

Abb

ott,

G.,

& L

acey

, K.

(200

5).

To in

vest

igat

e he

art d

isea

se

risk

know

ledg

e am

ong

Span

ish

spea

kers

with

di

abet

es a

nd e

xam

ine

whe

ther

use

of a

d-ho

c in

terp

rete

rs is

ass

ocia

ted

with

hea

rt d

isea

se

know

ledg

e. L

EP p

atie

nts

wer

e as

ked

to fi

ll ou

t a

surv

ey in

dica

ting

thei

r he

art d

isea

se k

now

ledg

e an

d w

heth

er th

ey u

se

ad-h

oc in

terp

reta

tion.

G

ener

al L

EP p

artic

ipan

ts

used

frie

nds/

fam

ily d

urin

g ap

poin

tmen

ts

Sett

ing:

Con

nect

icut

, USA

Stud

y D

esig

n: C

ross

-se

ctio

nal

Com

parin

g Tr

aine

d In

terp

rete

rs

vs. U

ntra

ined

Inte

rpre

ters

vs.

No

Inte

rpre

ters

Know

ledg

e of

hea

rt d

isea

se o

vera

ll w

as lo

w. T

hose

who

use

d fa

mily

or

frie

nds

had

low

er s

core

s (p

<001

) th

an th

ose

who

did

not

use

fam

ily

or fr

iend

s (1

8.8,

SD

=4.1,

p<0

.01)

. U

sing

frie

nds/

fam

ily in

stea

d of

a

prof

essi

onal

inte

rpre

ter i

s as

soci

ated

w

ith lo

wer

und

erst

andi

ng o

f hea

rt

dise

ase

risk.

Dia

gnos

tic

Know

ledg

e of

hea

rt

dise

ase

over

all w

as lo

w.

Regr

essi

on a

naly

sis

indi

cate

that

thos

e w

ho

wan

ted

but d

id n

ot h

ave

a pr

ofes

sion

al m

edic

al

inte

rpre

ter s

core

d lo

wer

(p

<0.0

1) th

an th

ose

who

did

not

wan

t an

inte

rpre

ter (

p<0.

01).

Stro

ng


6C

ompa

rison

of

Thr

ough

put

Tim

es fo

r Li

mite

d En

glis

h Pr

ofic

ienc

y

Patie

nt V

isits

in

the

Emer

genc

y D

epar

tmen

t Be

twee

n D

iffer

ent

Inte

rpre

ter

Mod

aliti

es

Gro

ver,

A.,

Dea

kyne

, S.

, Baj

aj, L

., Ro

osev

elt,

G.E

., (2

012)

.

Dat

a fr

om 11

96 fa

mili

es

wer

e an

alyz

ed. 1

st c

ohor

t co

nsis

ted

of L

EP fa

mili

es

pres

entin

g on

stu

dy

reen

rollm

ent d

ays

that

w

ere

seen

and

man

aged

by

a v

erifi

ed b

iling

ual

prov

ider

in th

e st

anda

rd

cour

se o

f ED

wor

kflo

w. I

f th

ey c

ould

not

be

seen

by

bilin

gual

pro

vide

r, th

ey

wer

e ra

ndom

ized

to e

ither

in

-per

son

or te

leph

onic

in

terp

rete

r.

Vario

us m

odal

ities

use

d –

in-p

erso

n in

terp

reta

tion,

re

mot

e te

leph

onic

in

terp

reta

tion,

and

bili

ngua

l pr

ovid

ers.

Sett

ing:

USA

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ohor

t

Com

parin

g D

iffer

ent M

odes

of

Inte

rpre

tatio

n M

ore

patie

nts

in th

e in

-per

son

coho

rt w

ere

seen

qui

ckly

co

mpa

red

to b

oth

tele

phon

ic a

nd

bilin

gual

coh

orts

.

Stat

istic

ally

sig

nific

ant d

ecre

ase

in

com

plex

med

icat

ion

adm

inis

trat

ion

in th

e in

-per

son

inte

rpre

ter g

roup

as

wel

l as

a lo

wer

adm

issi

on ra

te w

hen

com

pare

d to

the

tele

phon

ic a

nd

bilin

gual

inte

rpre

ter g

roup

.

Com

parin

g D

iffer

ent M

odes

of

Inte

rpre

tatio

n

The

in-p

erso

n co

hort

(116

min

) had

a

sign

ifica

ntly

sho

rter

tota

l thr

ough

put

time

than

tele

phon

ic (1

41 m

in) a

nd

bilin

gual

pro

vide

r (15

3 m

in) c

ohor

ts

(P <

0.00

1).

For e

very

hou

r of i

n-pe

rson

in

terp

reta

tion,

app

roxi

mat

ely

2 ad

ditio

nal m

inut

es w

ere

requ

ired

for a

bili

ngua

l vis

it an

d 9

addi

tiona

l m

inut

es w

ould

be

adde

d to

a

tele

phon

ic v

isit,

con

trol

ling

for

adm

issi

on ra

te, c

ompl

ex m

edic

atio

n ad

min

istr

atio

n, fa

st tr

ack

stat

us a

nd

invo

lvem

ent o

f tra

inee

s.

Leng

th o

f App

oint

men

t

The

mod

e of

inte

rpre

tatio

n im

pact

ed th

e le

ngth

of t

he

appo

intm

ent.

In-p

erso

n in

terp

reta

tion

had

a sh

orte

d le

ngth

of a

ppoi

ntm

ent

com

pare

d to

tele

phon

ic

inte

rpre

tatio

n an

d a

visi

t with

a

bilin

gual

pro

vide

r.

Stro

ng


7Tr

aine

d M

edic

al

Inte

rpre

ters

in

the

Emer

genc

y D

epar

tmen

t: Eff

ects

on

Serv

ices

, Su

bseq

uent

C

harg

es, a

nd

Follo

w-u

p

Bern

stei

n, J.

, Be

rnst

ein,

E.,

Dav

e, A

., H

ardt

, E.

, Jam

ess,

T.

, Lin

den,

J.,

Mitc

hell,

P.,

Ois

hi, T

., &

Saf

i, C

. (20

02).

The

stud

y w

as d

one

to

inve

stig

ate

the

impa

ct o

f an

Inte

rpre

ter S

ervi

ce o

n th

e in

tens

ity o

f Em

erge

ncy

Dep

artm

ent s

ervi

ces,

ut

iliza

tion,

and

cha

rges

.

In a

ll 26

573

ED

reco

rds

from

Jul

y to

Nov

embe

r 19

99, r

esul

ted

in a

dat

a se

t of 5

00 p

atie

nts

with

si

mila

r dem

ogra

phic

ch

arac

teris

tics,

chi

ef

com

plai

nt, a

cuity

, and

ad

mis

sion

rate

.

Thre

e gr

oups

– In

terp

rete

d Pa

tient

s (IP

s) n

= 6

3; N

on-

inte

rpre

ted

patie

nts

(NIP

)s

n =

374;

and

ESP

s n

= 63

)

Sett

ing:

Mas

sach

uset

ts,

USA

Stud

y D

esig

n: M

atch

ed

Coh

ort S

tudy

Com

parin

g Tr

aine

d In

terp

rete

rs

vs. U

ntra

ined

Inte

rpre

ters

vs.

No

Inte

rpre

ters

Non

-inte

rpre

ted

(NIP

s) p

atie

nts

who

did

not

spe

ak E

nglis

h ha

d th

e sh

orte

st E

D s

tay

(leng

th o

f sta

y).

Engl

ish

spea

king

pat

ient

s ha

d th

e lo

nges

t len

gth

of s

tay.

ESP

s st

ayed

in

ED a

n av

erag

e of

3 h

ours

long

er th

an

did

NIP

s

NIP

s ha

d lo

wes

t sub

sequ

ent c

linic

ut

iliza

tion.

The

thre

e gr

oups

diff

ered

in

the

30-d

ay p

erio

d fo

llow

ing

the

inde

x vi

sit.

IPs

rece

ived

mor

e pr

imar

y ca

re a

nd s

peci

alty

clin

ic re

ferr

als

than

did

eith

er N

IPs

or E

SPs.

ESPs

rece

ived

a s

igni

fican

tly g

reat

er

inte

nsity

and

vol

ume

of s

ervi

ces

than

di

d ei

ther

IPs

or N

IPs

durin

g th

eir

inde

x ED

vis

it.

Inte

rpre

ted

patie

nts

(IPs)

had

low

est

retu

rn E

D v

isits

. IPs

wer

e m

ore

likel

y th

an N

IPs

to d

emon

stra

te fo

llow

up

clin

ic v

isits

and

less

like

ly th

an N

IPs

to re

turn

to th

e ED

.

NIP

s w

hose

acu

ity le

vel (

inte

nsity

of

illne

ss) d

id n

ot d

iffer

sta

tistic

ally

at

base

line

from

IPs

or A

sp’s,

rece

ived

th

e fe

wes

t tes

ts a

nd p

roce

dure

s,

wer

e le

ast l

ikel

y to

hav

e an

IV s

tart

ed

and

rece

ived

the

few

est m

edic

atio

ns

durin

g th

eir E

D s

tay.

Leng

th o

f Sta

y

Non

-inte

rpre

ted

(NIP

s) p

atie

nts

who

did

not

spe

ak E

nglis

h ha

d th

e sh

orte

st E

D s

tay

(leng

th o

f st

ay).

Engl

ish

spea

king

pat

ient

s ha

d th

e lo

nges

t len

gth

of s

tay.

ES

Ps s

taye

d in

ED

an

aver

age

of 3

hou

rs lo

nger

than

did

NIP

s

Serv

ice

Util

izat

ion

NIP

s ha

d lo

wes

t sub

sequ

ent

clin

ic u

tiliz

atio

n. T

he th

ree

grou

ps d

iffer

ed in

the

30-d

ay

perio

d fo

llow

ing

the

inde

x vi

sit.

IPs

rece

ived

mor

e pr

imar

y ca

re

and

spec

ialty

clin

ic re

ferr

als

than

did

eith

er N

IPs

or E

SPs.

ESPs

rece

ived

a s

igni

fican

tly

grea

ter i

nten

sity

and

vol

ume

of

serv

ices

than

did

eith

er IP

s or

N

IPs

durin

g th

eir i

ndex

ED

vis

it.

Read

mis

sion

s

Inte

rpre

ted

patie

nts

(IPs)

had

lo

wes

t ret

urn

ED v

isits

. IPs

w

ere

mor

e lik

ely

than

NIP

s to

de

mon

stra

te fo

llow

up

clin

ic

visi

ts a

nd le

ss li

kely

than

NIP

s to

retu

rn to

the

ED.

Dia

gnos

tic

Non

-inte

rpre

ted

(NIP

s)

patie

nts

who

did

not

sp

eak

Engl

ish

had

the

few

est t

ests

, IVs

and

m

edic

atio

ns

NIP

s w

hose

acu

ity le

vel

(inte

nsity

of i

llnes

s) d

id

not d

iffer

sta

tistic

ally

at

base

line

from

IPs

or A

sp’s,

re

ceiv

ed th

e fe

wes

t tes

ts

and

proc

edur

es, w

ere

leas

t lik

ely

to h

ave

an

IV s

tart

ed a

nd re

ceiv

ed

the

few

est m

edic

atio

ns

durin

g th

eir E

D s

tay.

Mod

erat

e

8D

oes

Inte

rpre

ter-

Med

iate

d C

BT

with

Tra

umat

ized

Refu

gee

Peop

le W

ork?

A

Com

paris

on o

f Pa

tient

Out

com

es

in E

ast L

ondo

n

D’A

rden

ne,

P.D.

, Rua

ro,

L., C

esta

ri, L

., Fa

khou

ry, W

., &

Prie

be, S

. (2

007)

.

3 gr

oups

of p

atie

nts

– re

fuge

es w

ho re

quire

d in

terp

rete

rs, r

efug

ees

who

di

d no

t req

uire

inte

rpre

ters

, En

glis

h sp

eaki

ng n

on-

refu

gees

. Pat

ient

s re

ferr

ed

to a

n Ea

st L

ondo

n m

enta

l he

alth

ser

vice

spe

cial

izin

g in

the

psyc

holo

gica

l tr

eatm

ent o

f pos

t-tr

aum

atic

st

ress

dis

orde

r bet

wee

n 20

00 to

200

4.

44 re

fuge

e pa

tient

s re

quiri

ng in

terp

rete

rs,

36 re

fuge

e pa

tient

s no

t re

quiri

ng in

terp

rete

rs, a

nd

48 n

on-r

efug

ee p

atie

nts.

Sett

ing:

Eas

t Lon

don,

UK

Stud

y D

esig

n:

Retr

ospe

ctiv

e St

udy

Com

parin

g Tr

aine

d In

terp

rete

rs

vs. U

ntra

ined

Inte

rpre

ters

vs.

No

Inte

rpre

ters

Refu

gees

with

inte

rpre

ters

hav

e pr

opor

tiona

lly m

ore

impr

ovem

ent

than

refu

gees

with

inte

rpre

ters

.

Dia

gnos

tic

All

grou

ps s

how

ed

sign

ifica

nt im

prov

emen

t in

the

Impa

ct o

f Eve

nts

Scal

e (IE

S) w

hich

as

sess

ed s

ever

ity o

f PT

SD s

ympt

oms,

as

wel

l as

the

Beck

Dep

ress

ion

Inve

ntor

y (B

DI).

Refu

gees

with

inte

rpre

ters

ha

ve p

ropo

rtio

nally

m

ore

impr

ovem

ent t

han

refu

gees

with

inte

rpre

ters

.

Mod

erat

e


9D

oes

Engl

ish

prof

icie

ncy

impa

ct o

n he

alth

outc

omes

fo

r inp

atie

nts

unde

rgoi

ng s

trok

e

reha

bilit

atio

n?

Dav

ies,

S.E

., D

odd,

K.J

., Tu

, A

., Zu

cchi

, E.,

Zen,

S.,

Hill

, K.

D. (2

016)

.

To d

eter

min

e w

heth

er

Engl

ish

prof

icie

ncy

and/

or fr

eque

ncy

of in

terp

rete

r us

e im

pact

s on

hea

lth

outc

omes

for i

n pa

tient

st

roke

reha

bilit

atio

n

Peop

le a

dmitt

ed

for i

npat

ient

str

oke

reha

bilit

atio

n

Hig

h En

glis

h pr

ofic

ienc

y gr

oup

of n

ativ

e or

nea

r na

tive

Engl

ish

prof

icie

ncy

(n=8

0); a

nd a

low

Eng

lish

prof

icie

ncy

grou

p co

mpr

ised

peo

ple

who

pr

efer

red

a la

ngua

ge o

ther

th

an E

nglis

h (n

=80)

.

Sett

ing:

Aus

tral

ia

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ase

Con

trol

St

udy

Freq

uenc

y of

Inte

rpre

tatio

n

Incr

ease

d in

terp

rete

r usa

ge

impr

oved

Fun

ctio

nal I

ndep

ende

nce

Mea

sure

(FIM

) effi

cien

cy s

core

but

di

d no

t sig

nific

antly

alte

r oth

er

outc

omes

Leng

th o

f Sta

y

No

sign

ifica

nt d

iffer

ence

s w

ere

foun

d be

twee

n th

e gr

oups

in

leng

th o

f sta

y.

Dia

gnos

tic

The

low

Eng

lish

prof

icie

ncy

grou

ps

show

ed g

reat

er

impr

ovem

ent i

n FI

M fr

om

adm

issi

on to

dis

char

ge

Incr

ease

d in

terp

rete

r us

age

impr

oved

FIM

eff

icie

ncy

but d

id n

ot

sign

ifica

ntly

alte

r oth

er

outc

omes

Sign

ifica

nt d

iffer

ence

s w

ere

foun

d fo

r FIM

eff

icie

ncy

and

FIM

m

otor

effi

cien

cy w

ith

low

est i

nter

pret

er u

sage

re

sulti

ng in

low

est

effic

ienc

y; h

owev

er n

o st

atis

tical

ly s

igni

fican

t di

ffere

nces

wer

e fo

und

betw

een

the

quar

tiles

of

the

perc

enta

ge o

f th

erap

y en

coun

ters

with

an

inte

rpre

ter f

or F

IM a

nd

leng

th o

f sta

y.

Mod

erat

e

10Im

pact

of

Inte

rpre

ter

Serv

ices

on

Del

iver

y of

Hea

lth

Car

e to

Lim

ited-

Engl

ish-

prof

icie

nt

Patie

nts

Jaco

bs, E

.A.,

Laud

erda

le,

D.S.

, Mel

tzer

, D.

, Sho

rey,

J.M

., Le

vins

on, W

., &

Thi

sted

, R.A

. (2

001)

.

4380

adu

lts c

ontin

uous

ly

enro

lled

in a

sta

ff m

odel

H

MO

for t

wo

year

s of

st

udy

who

eith

er u

sed

the

com

preh

ensi

ve in

terp

rete

r se

rvic

es (I

SG, N

=327

) or

wer

e ra

ndom

ly s

elec

ted

into

a 10

% c

ompa

rison

gr

oup

of a

ll ot

her e

ligib

le

adul

ts (C

G, N

= 4

053)

.

Maj

ority

of p

atie

nts

in th

e IS

G s

poke

Spa

nish

(n=2

57)

ISG

was

mor

e fe

mal

e (6

3%

vs. 5

7%, P

<0.

05),

olde

r (m

ean

age

– 46

+/-

14 y

ears

, P<

0.0

1) a

nd li

ved

in a

ZIP

co

de w

ith a

low

er m

edia

n in

com

e co

mpa

red

to C

G.

Sett

ing:

Mas

sach

uset

ts,

USA

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ohor

t Stu

dy

Com

parin

g Tr

aine

d In

terp

rete

rs

vs. U

ntra

ined

Inte

rpre

ters

vs.

No

Inte

rpre

ters

Clin

ical

ser

vice

use

incr

ease

d si

gnifi

cant

ly in

the

ISG

com

pare

d to

th

e C

G fo

r offi

ce v

isits

(1.8

vs.

0.7

0,

P<.0

1); p

resc

riptio

ns w

ritte

n 1.7

6 vs

. 0.

53, P

< 0

.01)

, and

pre

scrip

tions

fil

led

(2.3

3 vs

. 0.8

6; P

< 0

.01)

.

Serv

ice

Util

izat

ion

Ove

rall

ther

e w

as a

n in

crea

se in

clin

ical

ser

vice

fo

r bot

h gr

oups

in a

ll bu

t of

the

mea

sure

s, w

hich

w

as u

rgen

t car

e ca

lls. I

t in

crea

sed

in C

G b

ut n

ot in

IS

G. A

ll ot

hers

incr

ease

d fo

r bo

th g

roup

s: o

ffice

vis

its,

phon

e ca

lls, u

rgen

t car

e vi

sits

, pre

scrip

tions

writ

ten,

pr

escr

iptio

ns fi

lled.

For t

hree

mea

sure

s, th

ere

was

a s

igni

fican

t inc

reas

e fo

r pat

ient

s in

the

ISG

gro

up

com

pare

d to

the

CG

gro

up:

offic

e vi

sits

ISG

1.8

vs. C

G 0

.7; #

of

pre

scrip

tions

writ

ten

for t

he

ISG

was

1.76

vs.

0.5

3 fo

r CG

; pr

escr

iptio

ns fi

lled

for I

SG w

as

2.33

pre

scrip

tions

per

per

son

per y

ear o

ver t

he s

tudy

per

iod

com

pare

d to

0.8

6 fo

r CG

. Thi

s st

ayed

eve

n af

ter a

djus

ting

for

age,

gen

der,

inco

me.

Prev

entio

n

Ove

rall,

ther

e w

as

an in

crea

se in

bot

h gr

oups

for a

ll pr

even

tive

mea

sure

s.

The

incr

ease

of r

ecta

l ex

ams

of 0

.26

per

man

ove

r age

40

was

si

gnifi

cant

. Rec

tal e

xam

s in

crea

sed

sign

ifica

ntly

m

ore

for I

SG c

ompa

red

to C

G b

ut th

is w

as n

ot

sign

ifica

nt a

fter

adj

ustin

g fo

r age

, gen

der a

nd

med

ian

inco

me.

Mod

erat

e


11O

verc

omin

g La

ngua

ge B

arrie

rs

in H

ealth

Car

e:

Cos

ts a

nd

Bene

fits

of

Inte

rpre

ter

Serv

ices

Jaco

bs, E

.A.,

Shep

ard,

D.S

., Su

aya,

J.A

., &

Sto

ne, E

.L.

(200

4).

They

ass

esse

d th

e im

pact

of

inte

rpre

ter s

ervi

ces

on

the

cost

and

util

izat

ion

of h

ealth

car

e se

rvic

es

amon

g pa

tient

s w

ith li

mite

d En

glis

h pr

ofic

ienc

y. T

hey

mea

sure

d th

e ch

ange

in

del

iver

y co

st o

f car

e pr

ovid

ed to

pat

ient

s en

rolle

d in

a h

ealth

m

aint

enan

ce o

rgan

izat

ion

befo

re a

nd a

fter

inte

rpre

ter

serv

ices

wer

e ad

ded.

The

time

perio

d of

Jun

e 1,

1995

thro

ugh

to M

ay 3

1, 19

97. 3

80 p

atie

nts

in th

e in

th

e in

terp

rete

r gro

up a

nd

4119

in th

e co

mpa

rison

gr

oup.

Sett

ing:

Mas

sach

uset

ts,

USA

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ohor

t

Com

parin

g Tr

aine

d In

terp

rete

rs

vs. U

ntra

ined

Inte

rpre

ters

vs.

No

Inte

rpre

ters

Rela

tive

to th

e co

mpa

rison

gro

up,

the

inte

rpre

ter s

ervi

ce g

roup

sho

wed

si

gnifi

cant

ly g

reat

er in

crea

ses

per

pers

on p

er y

ear i

n th

e fo

llow

ing

serv

ices

: per

cent

age

of th

e re

com

men

ded

prev

entiv

e se

rvic

es

rece

ived

(7.3

% v

s 2.

7%; P

=0.0

33),

num

ber o

f offi

ce v

isits

mad

e (1

.74

vs 0

.71;

P=0

.014

), an

d nu

mbe

r of

pres

crip

tions

writ

ten

(1.7

0 vs

0.5

2;

P= 0

.001

) and

fille

d (2

.38

vs 0

.88;

P<

0.00

1).

ER v

isits

wer

e sa

me

for b

oth

grou

ps.

Serv

ice

Util

izat

ion

Com

pare

d w

ith E

nglis

h sp

eaki

ng p

atie

nts,

pat

ient

s w

ho u

sed

the

inte

rpre

ter

serv

ices

mad

e si

gnifi

cant

ly

mor

e off

ice

visi

ts.

Prev

entio

n (S

ervi

ces,

Pr

escr

iptio

ns a

nd P

rimar

y C

are)

Com

pare

d w

ith E

nglis

h sp

eaki

ng p

atie

nts,

pa

tient

s w

ho u

sed

the

inte

rpre

ter s

ervi

ces

rece

ived

sig

nific

antly

m

ore

reco

mm

ende

d pr

even

tive

serv

ices

and

ha

d m

ore

pres

crip

tions

w

ritte

n an

d fil

led.

Betw

een

year

1 an

d ye

ar

2 of

the

stud

y, p

reve

ntiv

e se

rvic

es, p

rimar

y ca

re a

nd

tota

l cos

ts in

crea

sed

for

both

gro

ups.

ER c

osts

incr

ease

d in

th

e co

mpa

rison

gro

up

and

decr

ease

d fo

r the

in

terp

reta

tion

grou

p.

Mod

erat

e

12Im

pact

of

Inte

rpre

tatio

n M

etho

d on

Clin

ic

Visi

t Len

gth

Faga

n, M

.J.,

Dia

z, J.

A.,

Rein

ert,

S.E.

, Sc

ianm

anna

, C

.N.,

& F

agan

, D.

M. (

2003

).

613

visi

ts –

72%

(n=4

41)

of th

e vi

sits

requ

ired

no

inte

rpre

ters

and

28%

(n

=172

) use

d so

me

form

of

inte

rpre

ter.

15%

of a

ll vi

sits

use

d a

patie

nt s

uppl

ied

inte

rpre

ter

(n=9

0); 8

% o

f all

visi

ts u

sed

a ho

spita

l int

erpr

eter

(n=5

1),

and

5% o

f all

visi

ts u

sed

a te

leph

one

inte

rpre

ter

(n=3

1).

Sett

ing:

Rho

de Is

land

, USA

Stud

y D

esig

n: C

ross

se

ctio

nal

Com

parin

g Tr

aine

d In

terp

rete

rs

vs. U

ntra

ined

Inte

rpre

ters

vs.

No

Inte

rpre

ters

Com

pare

d to

pat

ient

s no

t usi

ng

any

inte

rpre

ters

, pat

ient

s us

ing

a te

leph

one

inte

rpre

ter h

ad

sign

ifica

ntly

long

er m

ean

prov

ider

tim

es (3

6.3

min

utes

vs.

28.

0 m

in) a

nd

clin

ic ti

mes

(99.

9 m

inut

es v

s. 8

2.4

min

utes

)

Patie

nts

usin

g a

patie

nt s

uppl

ied

inte

rpre

ter h

ad s

igni

fican

tly lo

nger

m

ean

prov

ider

tim

es (3

4.4

min

vs.

28

.0) a

nd m

ean

clin

ic ti

mes

(92.

8 m

in v

s. 8

2.4

min

, P =

0.02

7) w

hen

com

pare

d to

pat

ient

s no

t req

uirin

g an

inte

rpre

ter.

In c

ontr

ast,

patie

nts

usin

g a

hosp

ital

inte

rpre

ter d

id n

ot h

ave

sign

ifica

ntly

di

ffere

nt m

ean

prov

ider

tim

es (2

6.9

min

vs.

28.

0 m

in, P

=.51

) or m

ean

clin

ic ti

mes

(91.0

min

vs

82.4

min

, P=

0.16

) tha

n pa

tient

s no

t req

uirin

g an

in

terp

rete

r.

Leng

th o

f App

oint

men

t

In th

e m

odel

, use

of t

elep

hone

in

terp

rete

r was

ass

ocia

ted

with

long

er m

ean

prov

ider

tim

e (8

.3 m

in, 9

5% C

I, 3.

94

to 12

.7.a

s w

as u

se o

f a p

atie

nt

supp

lied

inte

rpre

ter (

4.58

m

in, 9

5% C

I, 1.8

4 to

7.3

3). I

n co

ntra

st, h

ospi

tal i

nter

pret

er

use

was

not

ass

ocia

ted

with

si

gnifi

cant

ly lo

nger

mea

n pr

ovid

er ti

me.

Whe

n co

mpa

red

to p

atie

nts

not r

equi

ring

inte

rpre

ters

, pat

ient

s us

ing

som

e fo

rm o

f int

erpr

eter

had

lo

nger

mea

n pr

ovid

er ti

mes

(3

2.4

min

utes

vs.

28

min

utes

) an

d cl

inic

tim

es (9

3.6

min

vs.

82

.4 m

in).

Mod

erat

e


13Er

rors

of M

edic

al

Inte

rpre

tatio

n an

d Th

eir P

oten

tial

Clin

ical

Con

sequ

ence

s:

A C

ompa

rison

of

Pro

fess

iona

l Ve

rsus

Ad

Hoc

Vers

us N

o In

terp

rete

rs

Flor

es, G

., A

breu

, M.,

Baro

ne, C

.P.,

Bach

ur, R

., &

Li

n, H

. (20

12).

57 e

ncou

nter

s in

clud

ed

with

20

with

pro

fess

iona

l in

terp

rete

rs, 2

7 w

ith a

d ho

c in

terp

rete

rs, a

nd 10

with

no

inte

rpre

ters

.

Inte

rpre

ters

wer

e pr

esen

t du

ring

thes

e en

coun

ters

in

clud

ed p

rofe

ssio

nal

inte

rpre

ters

, 20

(35%

); ad

ho

c in

terp

rete

rs, 2

7 (4

7%);

and

no in

terp

rete

r, 10

(18%

).

Sett

ing:

Bos

ton,

MA

, USA

Stud

y D

esig

n: C

ross

Se

ctio

nal

Com

parin

g Tr

aine

d In

terp

rete

rs v

s.

Unt

rain

ed v

s. N

o In

terp

rete

rs

The

prop

ortio

n of

err

ors

was

si

gnifi

cant

ly lo

wer

for p

rofe

ssio

nal

(12%

) vs.

ad

hoc

(22%

) vs.

no

inte

rpre

ters

(20%

).

Com

pare

d w

ith p

rofe

ssio

nal

inte

rpre

ters

, ad

hoc

inte

rpre

ters

and

ha

ving

no

inte

rpre

ter r

esul

ted

in

sign

ifica

ntly

hig

her p

ropo

rtio

ns o

f om

issi

ons

and

fals

e flu

ency

err

ors,

w

hile

pro

fess

iona

l int

erpr

eter

s ha

d hi

gher

err

ors

in th

e le

ss fr

eque

nt

erro

r cat

egor

ies.

Med

ical

Err

ors

The

mos

t com

mon

err

or w

as

omis

sion

(47%

of a

ll er

rors

), fo

llow

ed b

y fa

lse

fluen

cy (2

6%),

addi

tion

(10%

), ed

itoria

lizin

g (9

%) a

nd s

ubst

itutio

n (9

%).

Am

ong

prof

essi

onal

in

terp

rete

rs, p

revi

ous

hour

s of

inte

rpre

ter t

rain

ing,

but

no

t yea

rs o

f exp

erie

nce

wer

e si

gnifi

cant

ly a

ssoc

iate

d w

ith

erro

r num

bers

, typ

es a

nd

pote

ntia

l con

sequ

ence

s.

Thos

e w

ith g

reat

er th

an o

r eq

ual t

o 10

0 ho

urs

of tr

aini

ng

com

mitt

ed s

igni

fican

tly

low

er p

ropo

rtio

ns o

f err

ors

of

pote

ntia

l con

sequ

ence

ove

rall

(2%

vs.

12%

) in

ever

y er

ror

cate

gory

Mod

erat

e

14Ex

plor

ing

the

Impa

ct

of L

angu

age

Serv

ices

on

Util

izat

ion

and

Clin

ical

O

utco

mes

for

Dia

betic

s

Hac

ker,

K.,

Cho

i, Y.

S.,

Treb

ino,

L.,

Hic

ks, L

.R.,

Frie

dman

, E.,

Blan

chfie

ld, B

., &

Gaz

elle

, G.S

. (2

012)

.

1425

LEP

pat

ient

s in

C

ambr

idge

Hea

lth A

llian

ce

Dia

bete

s Re

gist

ry.

Ther

e ar

e 7

sam

ple

grou

ps w

hich

are

bas

ed

on th

e am

ount

and

the

com

bina

tion

of la

ngua

ge

serv

ices

(lan

guag

e co

ncor

dant

pro

vide

rs,

form

al in

terp

reta

tion,

no

inte

rpre

ters

)

Sett

ing:

Cam

brid

ge, M

A,

USA

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ohor

t D

esig

n.

Freq

uenc

y of

Inte

rpre

tatio

n

In u

nadj

uste

d an

alys

es, p

atie

nts

who

rece

ived

a m

ixtu

re o

f lan

guag

e se

rvic

es (P

atie

nts

with

1–49

% o

f th

eir v

isits

with

form

al in

terp

retin

g se

rvic

es a

nd 1–

99%

of t

heir

visi

ts

with

lang

uage

-con

cord

ant p

rovi

ders

an

d Pa

tient

s w

ith 5

0–10

0% o

f the

ir vi

sits

with

form

al in

terp

retin

g se

rvic

es a

nd 1–

99%

of t

heir

visi

ts w

ith

lang

uage

- con

cord

ant p

rovi

ders

) w

ere

mor

e lik

ely

to h

ave

expe

rienc

ed

a ho

spita

lizat

ion

or E

D v

isit

rela

ted

to

diab

etes

(ove

r 19.

5%/2

3 an

d 19

.2%

/10

resp

ectiv

ely)

com

pare

d to

oth

er

grou

ps in

the

outc

ome

perio

d.

Prev

entio

n

Patie

nts

who

rece

ived

10

0% o

f the

ir pr

imar

y ca

re v

isits

with

lang

uage

co

ncor

dant

pro

vide

rs

wer

e le

ast l

ikel

y to

ha

ve d

iabe

tes

rela

ted

emer

genc

y de

part

men

t vi

sits

com

pare

d to

oth

er

grou

ps in

the

follo

win

g 6

mon

ths.

Patie

nts

who

rece

ived

10

0% o

f the

ir vi

sits

with

la

ngua

ge-c

onco

rdan

t pr

ovid

ers

wer

e le

ss li

kely

to

hav

e an

ED

vis

it re

late

d to

dia

bete

s or

poo

rly

cont

rolle

d di

abet

es

com

pare

d to

pat

ient

s re

ceiv

ing

no la

ngua

ge

serv

ices

.

Mod

erat

e


15Eff

ect o

f La

ngua

ge B

arrie

rs

on F

ollo

w-u

p A

ppoi

ntm

ents

A

fter

an

Emer

genc

y D

epar

tmen

t Vis

it

Sarv

er, J

. &

Bake

r D. W

. (2

000)

.

Thre

e gr

oups

. Gro

up 1

- Lan

guag

e co

ncor

dant

- n

ativ

e En

glis

h an

d Sp

anis

h sp

eake

rs w

ho

com

mun

icat

ed w

ithou

t the

ai

d of

an

inte

rpre

ter.

Gro

up

2 –

inte

rpre

ter w

as u

sed

– na

tive

Span

ish-

spea

kers

w

ho c

omm

unic

ated

with

th

eir p

rovi

der i

n Sp

anis

h.

Gro

up 3

– in

terp

rete

r ne

eded

but

not

use

d –

nativ

e Sp

anis

h-sp

eake

rs

who

sai

d an

inte

rpre

ter

was

not

use

d bu

t tho

ught

an

inte

rpre

ter s

houl

d ha

ve

been

use

d.

Sett

ing:

Tor

ranc

e, C

A, U

SA

Stud

y D

esig

n: C

ohor

t Stu

dy

Com

parin

g Tr

aine

d In

terp

rete

rs v

s.

Unt

rain

ed In

terp

rete

rs

The

prop

ortio

n of

pat

ient

s w

ho

rece

ived

a fo

llow

up

appo

intm

ent

was

83%

for t

hose

with

out

lang

uage

bar

riers

, 75%

for t

hose

w

ho c

omm

unic

ated

thro

ugh

an

inte

rpre

ter,

and

76%

for t

hose

w

ho s

aid

they

sho

uld

have

an

inte

rpre

ter b

ut d

id n

ot h

ave

one

(P

= 0.

05).

App

oint

men

t com

plia

nce

rate

s w

ere

sim

ilar f

or p

atie

nts

who

com

mun

icat

ed th

roug

h an

in

terp

rete

r, th

ose

who

sai

d an

in

terp

rete

r sho

uld

have

bee

n us

ed

but w

as n

ot, a

nd th

ose

with

out

lang

uage

bar

riers

(P =

0.7

8).

Serv

ice

Util

izat

ion

The

prop

ortio

n of

pat

ient

s w

ho re

ceiv

ed a

follo

w u

p ap

poin

tmen

t was

83%

for

thos

e w

ithou

t lan

guag

e ba

rrie

rs, 7

5% fo

r tho

se w

ho

com

mun

icat

ed th

roug

h an

inte

rpre

ter,

and

76%

for

thos

e w

ho s

aid

they

sho

uld

have

an

inte

rpre

ter b

ut d

id

not h

ave

one

(P =

0.0

5).

App

oint

men

t com

plia

nce

rate

s w

ere

sim

ilar f

or p

atie

nts

who

co

mm

unic

ated

thro

ugh

an

inte

rpre

ter,

thos

e w

ho s

aid

an

inte

rpre

ter s

houl

d ha

ve b

een

used

but

was

not

, and

thos

e w

ithou

t lan

guag

e ba

rrie

rs (P

=

0.78

).

Mod

erat

e

16D

oes

a Vi

deo-

Inte

rpre

ting

Net

wor

k

Impr

ove

Del

iver

y of

Car

e in

the

Emer

genc

y D

epar

tmen

t?

Jaco

bs, E

. A.,

Fu J

r, P.

C.,

&

Rath

ouz,

P. J

. (2

012)

.

LEP

Span

ish

spea

king

vs

non-

LEP.

ED o

f 2 n

etw

ork

hosp

itals

in

Cal

iforn

ia (H

ospi

tal A

, in

a ru

ral a

rea,

and

Hos

pita

l B,

in a

n ur

ban

area

in)

Sett

ing:

Cal

iforn

ia, U

SA

Stud

y D

esig

n: P

re-p

ost,

Obs

erva

tiona

l stu

dy

Leng

th o

f Sta

y

Mea

n ho

spita

l tim

e in

the

ED

for b

oth

lang

uage

gro

ups

at

both

hos

pita

ls w

ent d

own

in

the

post

-vid

eo-in

terp

retin

g ne

twor

k tim

e pe

riod

com

pare

d w

ith th

e pr

e-vi

deo-

inte

rpre

ting

netw

ork

time

perio

d, b

y 16

m

inut

es fo

r Eng

lish

spea

kers

an

d 31

min

utes

for S

pani

sh

spea

kers

at H

ospi

tal A

, and

by

34 a

nd 8

7 m

inut

es fo

r the

se

two

grou

ps, r

espe

ctiv

ely,

at

Hos

pita

l B. [

Not

sta

tistic

ally

si

gnifi

cant

]

Adm

issi

on R

ates

The

perc

enta

ge o

f ED

pat

ient

s ad

mitt

ed to

the

hosp

ital

wen

t dow

n in

the

post

-vid

eo-

netw

ork

time

perio

d fo

r bot

h la

ngua

ge g

roup

s at

Hos

pita

l A

and

was

ess

entia

lly u

ncha

nged

fo

r bot

h la

ngua

ge g

roup

s at

Hos

pita

l B. T

here

was

no

sign

ifica

nt d

iffer

ence

in th

e ch

ange

in a

dmis

sion

rate

s fo

r En

glis

h sp

eake

rs c

ompa

red

with

Spa

nish

spe

aker

s at

eith

er

stud

y ho

spita

l.

Mod

erat

e


17Po

stop

erat

ive

Pain

Man

agem

ent

in C

hild

ren,

Pa

rent

al E

nglis

h Pr

ofic

ienc

y,

and

Acce

ss to

In

terp

reta

tion

Jim

enez

, N.,

Jack

son,

D. L

., Zh

ou, C

., Ay

ala,

N

. C.,

& E

bel,

B.

E. (2

014)

.

Chi

ldre

n w

ith L

EP p

aren

ts

(n=2

37);

Chi

ldre

n w

ith

EP p

aren

ts (n

=237

). O

bjec

tive

was

to e

xam

ine

the

asso

ciat

ion

betw

een

pare

ntal

lang

uage

pr

ofic

ienc

y, in

terp

rete

d ca

re, a

nd p

osts

urgi

cal

pedi

atric

pai

n m

anag

emen

t

Sett

ing:

USA

Stud

y D

esig

n:

Retr

ospe

ctiv

e M

atch

ed

Coh

ort

Freq

uenc

y of

Inte

rpre

tatio

n

With

in th

e LE

P gr

oup,

chi

ldre

n w

ith ≥

2 in

terp

reta

tions

per

day

had

lo

wer

pai

n sc

ores

aft

er m

edic

atio

n ad

min

istr

atio

n (P

< .0

5) a

nd w

ere

mor

e lik

ely

to re

ceiv

e op

ioid

s at

pai

n le

vels

sim

ilar t

o th

ose

of E

P fa

mili

es.

Chi

ldre

n of

fam

ilies

who

rece

ived

le

ss th

an 2

inte

rpre

ted

visi

ts p

er d

ay

had

high

er m

ean

post

ana

lges

ic p

ain

scor

es (P

- 0.

04) r

elat

ive

to c

hild

ren

with

mor

e fr

eque

nt in

terp

reta

tion.

Mod

erat

e

18C

onve

nien

t Ac

cess

to

Prof

essi

onal

In

terp

rete

rs in

the

Hos

pita

l D

ecre

ases

Re

adm

issi

on

Rate

s an

d Es

timat

ed

Hos

pita

l Ex

pend

iture

s fo

r Pat

ient

s w

ith

Lim

ited

Engl

ish

Prof

icie

ncy

Karli

ner,

L. S

., Pe

rez-

Stab

le, E

. J.

, & G

rego

rich,

S.

E. (

2017

).

Patie

nts

aged

50+

yea

rs --

- re

ceiv

ing

the

inte

rven

tion:

th

e du

al-h

ands

et

inte

rpre

ter t

elep

hone

at

ever

y be

dsid

e (b

edsi

de

inte

rpre

ter i

nter

vent

ion)

.

Sett

ing:

USA

Stud

y D

esig

n: P

re-P

ost

Read

mis

sion

Rat

es

Obs

erve

d 30

-day

read

mis

sion

de

crea

sed

durin

g th

e in

terv

entio

n pe

riod

and

incr

ease

d ag

ain

post

in

terv

entio

n. T

he in

terv

entio

n w

as p

rogr

amm

ed to

al

low

24-

hour

acc

ess

to a

pr

ofes

sion

al (t

rain

ed a

nd

test

ed) m

edic

al in

terp

rete

r fo

r mor

e th

an 10

0 la

ngua

ges.

Th

e eff

ect o

f the

inte

rven

tion

on re

adm

issi

on ra

tes

was

si

gnifi

cant

ly m

odifi

ed b

y pa

tient

lang

uage

gro

up; t

hat

is, t

he e

ffect

of t

he s

tudy

pe

riods

on

read

mis

sion

rate

s si

gnifi

cant

ly d

iffer

ed a

cros

s th

e 2

lang

uage

gro

ups

(P =

0.

040

for t

est o

f int

erac

tion)

. Th

e od

ds o

f rea

dmis

sion

for

the

LEP

com

pare

d w

ith E

P gr

oup

was

low

er d

urin

g th

e in

terv

entio

n pe

riod;

whi

le it

w

as ro

ughl

y eq

uiva

lent

dur

ing

both

the

pre-

inte

rven

tion

and

post

inte

rven

tion

perio

ds

(Tab

le 4

).

Mod

erat

e

19U

se o

f In

terp

rete

rs b

y Ph

ysic

ians

for

Hos

pita

lized

Lim

ited

Engl

ish

Prof

icie

nt P

atie

nts

and

Its Im

pact

on P

atie

nt

Out

com

es

Lope

z, L

., Ro

drig

uez,

F.

, Hue

rta,

D.,

Souk

up, J

., &

H

icks

, L. (

2015

).

Gen

eral

LEP

vs.

Non

-LEP

. Fo

ur g

roup

s:

1): i

nter

pret

er u

sed

by n

on-

phys

icia

n (i.

e., n

urse

); 2)

inte

rpre

ter u

sed

by a

non

-H

ospi

talis

t phy

sici

an;

3) in

terp

rete

r use

d by

H

ospi

talis

t; an

d 4)

no

inte

rpre

ter u

sed

durin

g ho

spita

lizat

ion.

Sett

ing:

USA

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ohor

t A

naly

sis

Leng

th o

f Sta

y

Com

pare

d to

Eng

lish

spea

kers

, LEP

pat

ient

s w

ith n

o in

terp

rete

rs h

ad

sign

ifica

ntly

sho

rter

leng

th o

f st

ay. L

EP p

atie

nts

who

had

an

inte

rpre

ter a

nd a

phy

sici

an

pres

ent h

ad th

e lo

nges

t len

gth

of s

tay.

Mod

erat

e


20U

se o

f In

terp

rete

rs b

y Ph

ysic

ians

for

Hos

pita

lized

Lim

ited

Engl

ish

Prof

icie

nt P

atie

nts

and

Its Im

pact

on P

atie

nt

Out

com

es

Ham

pers

, L.C

., &

McN

ulty

, J.E

. (2

002)

.

LEP

child

ren.

LEP

with

pr

ofes

sion

al in

terp

rete

rs v

s.

LEP

with

bili

ngua

l pro

vide

r

Exam

ined

vis

its o

f chi

ldre

n (a

ges

2mon

ths

to 10

ye

ars)

with

a p

rese

ntin

g te

mpe

ratu

re o

f 38.

5C o

r hi

gher

or a

com

plai

nt o

f vo

miti

ng o

r dia

rrhe

a. In

17

0 ca

ses,

the

trea

ting

phys

icia

n w

as b

iling

ual.

In 2

39, a

pro

fess

iona

l in

terp

rete

r was

use

d.

In th

e re

mai

ning

141,

a pr

ofes

sion

al m

edic

al

inte

rpre

ter w

as u

nava

ilabl

e

Sett

ing:

Chi

cago

, IL,

USA

Stud

y D

esig

n: P

rosp

ectiv

e C

ohor

t

Com

parin

g Tr

aine

d In

terp

rete

rs v

s.

Unt

rain

ed In

terp

rete

rs

Dec

isio

n m

akin

g w

as m

ost c

autio

us

whe

n no

n-En

glis

h-sp

eaki

ng c

ases

w

ere

trea

ted

in th

e ab

senc

e of

bi

lingu

al p

hysi

cian

s or

pro

fess

iona

l in

terp

rete

rs.

Leng

th o

f Sta

y

Cas

es w

ith a

n in

terp

rete

r had

lo

nger

leng

th o

f sta

y vi

sits

(+16

m

inut

es; 9

5% C

I, 6.

2-26

min

s).

The

barr

ier c

ohor

t (no

n-En

glis

h) w

ithou

t a p

rofe

ssio

nal

inte

rpre

ter s

how

ed n

o di

ffere

nce

in le

ngth

of v

isit.

Whe

n a

barr

ier w

as p

rese

nt,

and

a pr

ofes

sion

al in

terp

rete

r w

as u

nava

ilabl

e, p

hysi

cian

s pe

rfor

med

mor

e ex

tens

ive

eval

uatio

ns (m

ore

freq

uent

an

d m

ore

expe

nsiv

e di

agno

stic

te

stin

g) a

nd tr

eate

d ch

ildre

n m

ore

cons

erva

tivel

y (m

ore

intr

aven

ous

hydr

atio

n an

d m

ore

freq

uent

hos

pita

l ad

mis

sion

s.

Serv

ice

Util

izat

ion

Com

pare

d w

ith th

e En

glis

h-sp

eaki

ng c

ohor

t, no

n-En

glis

h sp

eaki

ng c

ases

with

bili

ngua

l ph

ysic

ians

had

sim

ilar r

ates

of

reso

urce

util

izat

ion.

Cas

es

with

an

inte

rpre

ter w

ere

leas

t lik

ely

to b

e te

sted

(odd

s ra

tio

[OR]

, mor

e lik

ely

to a

dmitt

ed

(OR,

1.7;

95%

CI,

1.1-2

.8),

and

no m

ore

likel

y to

rece

ive

intr

aven

ous

fluid

s. T

he b

arrie

r co

hort

(non

-Eng

lish

with

out a

pr

ofes

sion

al in

terp

rete

r) h

ad a

hi

gher

inci

denc

e (O

R, 1.

5; 9

5%

CI,

1.04-

2.2)

for t

estin

g an

d w

as

mos

t lik

ely

to b

e ad

mitt

ed (O

R,

2.6;

95%

CI,

1.4-4

.5) a

nd to

re

ceiv

e in

trav

enou

s hy

drat

ion

(OR,

2.3

; 95%

CI,

1.2-4

.3).

Mod

erat

e

21Pa

tient

-Rep

orte

d Q

ualit

y of

Pai

n Tr

eatm

ent a

nd

Use

of In

terp

rete

rs

in S

pani

sh-

Spea

king

Pat

ient

s H

ospi

taliz

ed

for O

bste

tric

and

G

ynec

olog

ical

C

are

Jim

enez

, N.,

Mor

eno,

G.,

Leng

, M.,

Buch

wal

d, D

., &

Mor

ales

L. S

. (2

012)

.

LEP

wom

en w

ith a

n in

terp

rete

r vs.

LEP

with

out

an in

terp

rete

r

Sett

ing:

USA

Stud

y D

esig

n: C

ross

-Se

ctio

nal

Freq

uenc

y of

Inte

rpre

tatio

n

Patie

nts

who

did

not

alw

ays

rece

ive

inte

rpre

tatio

n re

port

ed

sign

ifica

ntly

low

er s

core

s in

all

3 su

rvey

item

s: le

ss li

kely

to h

ave

thei

r pa

in c

ontr

olle

d (O

R=0.

4, p

<0.0

5),

have

a ti

mel

y re

spon

se to

thei

r pa

in (O

R=0.

4, p

<0.0

5), a

nd le

ss

help

fuln

ess

from

sta

ff (O

R=0.

03,

p<0.

05).

Dia

gnos

tic

Patie

nts

who

did

no

t alw

ays

rece

ive

inte

rpre

tatio

n re

port

ed

sign

ifica

ntly

low

er s

core

s in

all

3 su

rvey

item

s:

less

like

ly to

hav

e th

eir

pain

con

trol

led

(OR=

0.4,

p<

0.05

), ha

ve a

tim

ely

resp

onse

to th

eir p

ain

(OR=

0.4,

p<0

.05)

, and

less

he

lpfu

lnes

s fr

om s

taff

(OR=

0.03

, p<0

.05)

.

Mod

erat

e


22In

terp

reta

tion

in

Con

sulta

tions

w

ith Im

mig

rant

Pa

tient

s

With

Can

cer:

How

Ac

cura

te Is

It?

Buto

w, P

. N. e

t al

. (20

11).

To c

ompa

re a

ccur

acy

of in

terp

reta

tion

usin

g pr

ofes

sion

al v

s. a

d-ho

c in

terp

rete

rs. L

EP p

atie

nts

with

new

can

cer d

iagn

oses

w

ere

reco

rded

dur

ing

thei

r firs

t tw

o on

colo

gy

cons

ulta

tions

aft

er

diag

nosi

s, w

ith a

d-ho

c/fa

mily

inte

rpre

ters

or

prof

essi

onal

inte

rpre

ters

. G

ener

al L

EP (n

=49

enco

unte

rs; s

ome

of w

hich

ha

ve m

ultip

le in

terp

rete

rs

pres

ent)

VS.

LEP

with

an

othe

r int

erpr

eter

mod

e

Sett

ing:

Aus

tral

ia

Stud

y D

esig

n: C

ross

-se

ctio

nal

Com

parin

g Tr

aine

d In

terp

rete

rs v

s.

Unt

rain

ed In

terp

rete

rs

Prof

essi

onal

inte

rpre

ters

(pho

ne

or fa

ce-t

o-fa

ce) w

ere

less

like

ly to

pr

esen

t non

-equ

ival

ent i

nter

pret

atio

n th

an fa

mily

mem

bers

(50%

vs.

65%

, p=

0.05

). H

owev

er, a

cros

s bo

th

grou

ps th

ere

was

no

sign

ifica

nt

diffe

renc

e in

the

rate

of n

egat

ive

non-

equi

vale

nt in

terp

reta

tion.

No

diffe

renc

es b

etw

een

tele

phon

e an

d fa

ce-t

o-fa

ce in

terp

reta

tion.

Mod

erat

e

23C

ompa

ring

In-P

erso

n, V

ideo

, an

d Te

leph

onic

M

edic

al

Inte

rpre

tatio

n

Loca

tis, C

., et

al

. (20

10).

Gen

eral

LEP

(n =

241

; fac

e-to

-fac

e=80

; pho

ne=8

0;

vide

o =8

1) v

s. L

EP w

ith

anot

her i

nter

pret

er m

ode

Sett

ing:

Sou

th C

arol

ina,

U

SA

Stud

y D

esig

n: R

ando

miz

ed

Con

trol

Tria

l

Com

parin

g In

terp

reta

tion

Mod

es

Wai

t tim

es a

re s

igni

fican

tly lo

nger

(2

.5 m

in, p

=0.0

1) fo

r vid

eo v

s.

in-p

erso

n in

terp

reta

tion.

Inte

rvie

w

times

are

sig

nific

antly

long

er (7

.4

min

, p=0

.01)

for i

n-pe

rson

com

pare

d to

pho

ne, n

ot s

igni

fican

tly lo

nger

(5.3

m

in, p

=0.1)

for i

n-pe

rson

com

pare

d to

vid

eo.

Wea

k

24C

ompa

ring

the

Use

of P

hysi

cian

Ti

me

and

Hea

lth

Car

e Re

sour

ces

Am

ong

Patie

nts

Spea

king

Eng

lish,

Sp

anis

h, a

nd

Russ

ian

Krav

itz, R

L.,

Hel

ms,

J.L.

, A

zari,

R.,

Ant

oniu

s, m

D.,

& M

elni

kow

, J.

(200

0).

Dis

tingu

ish

betw

een

patie

nts

usin

g “h

ealth

sy

stem

inte

rpre

ters

” (p

aid

inte

rpre

ters

or b

iling

ual

phys

icia

ns) a

nd th

ose

usin

g “p

erso

nal i

nter

pret

ers”

(f

riend

s or

fam

ily m

embe

rs)

Gen

eral

LEP

(18

+ ye

ars)

w

ho s

peak

Spa

nish

and

Ru

ssia

n (n

=62

Span

ish)

(n

=111

Rus

sia)

com

pare

d to

N

on -

LEP

(18+

yea

rs) (

n=11

2 En

glis

h)

Sett

ing:

Cal

iforn

ia, U

SA

Stud

y D

esig

n: P

rosp

ectiv

e O

bser

vatio

nal S

tudy

Leng

th o

f App

oint

men

t

Com

pare

d w

ith E

nglis

h-sp

eaki

ng

patie

nts,

Spa

nish

and

Rus

sian

sp

eake

rs w

ho u

sed

heal

th

syst

em in

terp

rete

rs

aver

aged

12.2

and

7.1

addi

tiona

l m

inut

es o

f phy

sici

an ti

me,

re

spec

tivel

y (P

=

0.28

for S

pani

sh s

peak

ers

and

P =

0.82

for R

ussi

an

spea

kers

). Fo

r Spa

nish

and

Ru

ssia

n sp

eake

rs u

sing

pe

rson

al in

terp

rete

rs, t

here

w

as n

o st

atis

tical

ly s

igni

fican

t di

ffere

nce

betw

een

them

and

En

glis

h-sp

eaki

ng p

atie

nts.

Wea

k


25N

o D

iffer

ence

in

Em

erge

ncy

Dep

artm

ent

leng

th o

f sta

y fo

r Pat

ient

s w

ith

Lim

ited

Prof

icie

ncy

in

Engl

ish

Wal

lbre

cht,

J.,

Hod

es-V

illam

ar,

L., W

eiss

, S. J

., &

Ern

st, A

. A.

(201

4).

Gen

eral

LEP

(124

LEP

pa

tient

s); N

on -

LEP

(ES)

: Fo

r eac

h ES

pat

ient

, one

LE

P pa

tient

with

the

sam

e ac

uity

and

sim

ilar a

ge w

as

sele

cted

. (12

1 ES

patie

nts)

Sett

ing:

New

Mex

ico,

USA

Stud

y D

esig

n: P

rosp

ectiv

e C

onve

nien

ce S

ampl

ing

Coh

ort S

tudy

Leng

th o

f Sta

y

The

use

of in

terp

rete

rs w

as

asso

ciat

ed w

ith a

sig

nific

ant

incr

ease

in le

ngth

of s

tay

(incr

ease

d le

ngth

of s

tay

of

237

min

utes

from

the

time

of

arriv

al to

tim

e of

dis

char

ge

or a

dmis

sion

requ

est)

. No

diffe

renc

e in

leng

th o

f sta

y be

twee

n ES

pat

ient

s an

d pa

tient

s w

ith L

EP. O

nly

53%

of

pat

ient

s w

ith L

EP w

ere

repo

rted

to h

ave

used

an

inte

rpre

ter d

urin

g th

eir E

D

visi

t.

Wea

k

26Th

e Im

pact

of

Med

ical

In

terp

reta

tion

Met

hod

on T

ime

and

Erro

rs

Gan

y, F

., Ka

pelu

szni

k,

L., P

raka

sh,

K., G

onza

lez,

J.

, Ort

a, L

.Y.,

Tsen

g, C

.H.,

&

Cha

ngra

ni, J

. (2

007)

.

4 sc

ripte

d cl

inic

al

enco

unte

rs

Mod

es o

f int

erpr

etat

ion:

Re

mot

e si

mul

tane

ous,

Re

mot

e co

nsec

utiv

e,

Prox

imat

e co

nsec

utiv

e,

Prox

imat

e ad

hoc

Sett

ing:

New

Yor

k, N

Y, U

SA

Stud

y D

esig

n: C

ross

Se

ctio

nal

Com

parin

g M

odes

of I

nter

pret

atio

n an

d Tr

aine

d vs

. Unt

rain

ed

Inte

rpre

ters

Rem

ote

sim

ulta

neou

s m

edic

al

inte

rpre

ting

(RSM

I) en

coun

ters

av

erag

ed 12

.72

vs. 1

8.24

min

utes

for

the

next

fast

est m

ode

(pro

xim

ate

ad

hoc)

(p=0

.002

). Th

ere

wer

e 12

tim

es

mor

e m

edic

al e

rror

s of

mod

erat

e or

gr

eate

r clin

ical

sig

nific

ance

am

ong

utte

ranc

e in

non

-RSM

I enc

ount

ers

com

pare

d to

RSM

I enc

ount

ers

(p=0

.000

2).

Med

ical

Err

ors

RSM

I pro

duce

d fe

wer

err

ors

than

the

othe

r mod

es, w

hich

ha

d er

ror r

ates

that

wer

e cl

uste

red

at th

e a

sign

ifica

ntly

hi

gher

rate

. RSM

I had

a m

ean

of 1.

139

lingu

istic

err

ors

per

utte

ranc

e an

d 0.

019

med

ical

er

rors

of m

oder

ate

or g

reat

er

clin

ical

sig

nific

ance

per

ut

tera

nce.

Non

–RS

MI i

nter

pret

ing

mod

aliti

es w

ere

asso

ciat

ed

with

12-fo

ld g

reat

er ra

te o

f po

tent

ial m

edic

al e

rror

s (o

f mod

erat

e or

gre

ater

si

gnifi

canc

e (p

er u

tter

ance

co

mpa

red

to R

SMI (

P=0.

0002

).

Wea

k


27Th

e Ro

le

of M

edic

al

Inte

rpre

tatio

n on

Bre

ast a

nd

Cer

vica

l Can

cer

Scre

enin

g A

mon

g A

sian

Am

eric

an

and

Paci

fic

Isla

nder

Wom

en

Dan

g, J.

, Lee

, J.

, Tra

n, J.

H.,

Kaga

wa-

Sing

er,

M.,

Foo,

M.A

., &

Ngu

yen,

T.N

. (2

010)

.

1,708

Asi

an A

mer

ican

and

Pa

cific

Isla

nder

wom

en in

N

orth

ern

and

Sout

hern

C

alifo

rnia

. Cam

bodi

ans

n =

344,

Lao

tians

n =

353

, To

ngan

s n

= 28

6.

Sett

ing:

Cal

iforn

ia, U

SA

Stud

y D

esig

n: C

ross

-se

ctio

nal

Com

parin

g Tr

aine

d vs

. Unt

rain

ed v

s.

No

Inte

rpre

ters

Peop

le p

refe

rred

to s

peak

to th

eir

doct

or o

r med

ical

pro

vide

r in

a no

n-En

glis

h la

ngua

ge –

bili

ngua

l sta

ff (9

4.3%

), m

edic

al in

terp

rete

r (93

.6%

) an

d fa

mily

/frie

nds

(88.

9%).

Prev

entio

n

Wom

en w

ho ty

pica

lly

used

a m

edic

al in

terp

rete

r ha

d a

grea

ter o

dd

of h

avin

g re

ceiv

ed a

m

amm

ogra

m (O

R =

1.85;

95

% c

onfid

ence

inte

rval

=

1.21,

2.83

), a

clin

ical

br

east

exa

m (O

R 3.

03,

95%

CI =

1.82,

5.0

3) a

nd

a Pa

p Sm

ear (

OR

=2.3

4;

95%

CI =

1.38

to 3

.97)

than

th

ose

who

did

not

use

an

inte

rpre

ter.

Patie

nts

who

had

bili

ngua

l st

aff w

ere

twic

e as

like

ly

to h

ave

a cl

inic

al b

reas

t ex

am (2

.23

OR)

and

a

pap

smea

r (2.

82 O

R)

com

pare

d to

pat

ient

s w

ho d

id n

ot u

se a

ny

type

of i

nter

pret

atio

n.

Thos

e w

ho u

sed

bilin

gual

st

aff m

embe

rs h

ad th

e hi

ghes

t rat

e of

Pap

Sm

ears

(85.

8%) c

ompa

red

with

the

othe

r gro

ups

(non

e, 6

5.1%

; fam

ily o

r fr

iend

s, 7

7.1%

; med

ical

in

terp

rete

rs, 8

1.8%

).

Did

not

find

gre

ater

odd

s of

rece

ivin

g br

east

or

cerv

ical

can

cer s

cree

ning

af

ter a

ccou

ntin

g fo

r oth

er

varia

bles

in th

e m

odel

.

Wea

k

28Q

ualit

y of

Re

prod

uctiv

e H

ealth

Ser

vice

s to

Lim

ited

Engl

ish

Prof

icie

nt (L

EP)

Patie

nts

De

Boca

negr

a,

H.T.

D.,

Rost

ovts

eva,

D.

, Cet

inka

ya,

M.,

Rund

el, C

. &

Lew

is, C

. (20

11).

Mod

es o

f Int

erpr

etat

ion

– st

aff in

terp

rete

rs, l

angu

age

disc

onco

rdan

t (LD

I);

bilin

gual

clin

icia

n, la

ngua

ge

conc

orda

nt (L

C)

831 c

lient

s fo

r 158

9 re

prod

uctiv

e he

alth

vis

its

of fe

mal

e an

d m

ale

LEP

clie

nts

whe

re th

eir p

rimar

y la

ngua

ge w

as S

pani

sh a

nd

in w

hich

the

serv

ices

wer

e pr

ovid

ed b

y a

clin

icia

n w

ho

spok

e th

eir l

angu

age

or

thro

ugh

a st

aff in

terp

rete

r.

Sett

ing:

Cal

iforn

ia, U

SA

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ohor

t (Pr

e-Po

st)

Com

parin

g D

iffer

ent M

odes

of

Inte

rpre

tatio

n LD

I vis

its w

ere

sign

ifica

ntly

less

like

ly th

an L

C

visi

ts to

con

tain

doc

umen

tatio

n of

the

prov

isio

n of

edu

catio

n an

d co

unse

lling

ser

vice

s an

d le

ss li

kely

to

hav

e do

cum

enta

tion

of s

exua

lly

tran

smitt

ed in

fect

ion

(STI

) ris

k as

sess

men

t am

ong

new

fem

ale

clie

nts.

Fem

ale

clie

nts

in L

DI a

nd L

C

visi

ts w

ere

equa

lly li

kely

to b

e te

sted

fo

r Chl

amyd

ia.

Am

ong

wom

en, L

C v

isits

wer

e si

gnifi

cant

ly m

ore

likel

y th

an L

DI

visi

ts to

hav

e do

cum

enta

tion

of S

TI

risk

asse

ssm

ent.

How

ever

, fem

ale

clie

nts

in L

DI a

nd L

C v

isits

wer

e ab

out e

qual

ly li

kely

to b

e te

sted

for

Chl

amyd

ia, b

oth

for w

omen

und

er

the

age

of 2

6 an

d fo

r wom

en a

ge 2

6 an

d ol

der.

Dia

gnos

tic

Less

freq

uent

do

cum

enta

tion

of

educ

atio

n an

d co

unse

lling

on

any

topi

c w

hen

usin

g an

inte

rpre

ter c

ompa

red

to th

at fo

r clie

nts

seen

by

a la

ngua

ge c

onco

rdan

t pr

ovid

er.

Prov

ider

s w

ere

sign

ifica

ntly

less

like

ly

to d

ocum

ent a

n ST

I as

sess

men

t win

an

LDI

visi

t tha

n in

an

LC v

isits

–

new

clie

nts

seen

in

LDI w

ere

sign

ifica

ntly

le

ss li

kely

to h

ave

docu

men

tatio

n of

a S

TI

asse

ssm

ent w

hile

ther

e w

as n

o di

ffere

nce

in ri

sk

asse

ssm

ent b

etw

een

LDI a

nd L

C v

isits

am

ong

esta

blis

hed

clie

nts.

Wea

k


29Sc

reen

ing

of

men

tal d

isor

ders

in as

ylum

-see

kers

fr

om K

osov

o

Eyta

n, A

., Bi

scho

ff, A

., Rr

uste

mi,

I.,

Dur

ieux

, S.,

Lout

an, L

., G

ilber

t, M

., &

Boi

ver,

P.

(200

2).

319

stru

ctur

ed in

terv

iew

s co

nduc

ted

by n

urse

s w

ith a

sylu

m s

eeke

rs in

Ko

sovo

dur

ing

a sy

stem

atic

m

edic

al s

cree

ning

at t

ime

of e

ntry

, con

sist

ing

of

ques

tions

abo

ut h

ealth

co

nditi

ons,

pas

t exp

osur

e to

trau

mat

ic e

vent

s, a

nd

post

-tra

umat

ic s

ympt

oms.

Sett

ing:

Gen

eva,

Sw

itzer

land

Stud

y D

esig

n: C

ross

-se

ctio

nal

Com

parin

g Tr

aine

d In

terp

rete

rs

vs. U

ntra

ined

Inte

rpre

ters

vs.

No

Inte

rpre

ters

Subj

ectiv

e ra

ting

of c

omm

unic

atio

n w

as p

oore

st w

hen

ther

e w

as n

o in

terp

rete

r pre

sent

, bet

ter w

hen

rela

tives

wer

e us

ed a

nd b

est w

hen

trai

ned

inte

rpre

ters

wer

e us

ed (t

rend

te

st, p

<0.

0001

).

Serv

ice

Util

izat

ion

The

use

of b

oth

type

of

inte

rpre

ters

(tra

ined

and

re

lativ

e) le

d to

incr

ease

d re

ferr

al to

med

ical

car

e (n

on-

inte

rpre

ter 3

1%, r

elat

ives

as

inte

rpre

ter 4

8%, t

rain

ed

inte

rpre

ter 4

2%, p

=0.0

3.

Incr

ease

d re

ferr

al to

ps

ycho

logi

cal c

are

with

the

use

of a

trai

ned

inte

rpre

ter

(no

inte

rpre

ter 4

%, r

elat

ives

as

inte

rpre

ter 3

% a

nd tr

aine

d in

terp

rete

r 15%

, p =

0.0

7)

Dia

gnos

tic –

Rep

ortin

g of

Sy

mpt

oms

The

use

of re

lativ

es a

nd

trai

ned

inte

rpre

ters

si

gnifi

cant

ly in

fluen

ced

the

prop

ortio

n of

per

sons

re

port

ing

trau

mat

ic

even

ts a

nd p

sych

olog

ical

sy

mpt

oms.

Wea

k

30H

ow a

cces

sibl

e ar

e in

terp

rete

r se

rvic

es to

di

alys

is p

atie

nts

of N

on-E

nglis

h Sp

eaki

ng?

Back

grou

nd?

Zim

budz

i, E.

, Th

omps

on,

S., &

Ter

rill B

. (2

010)

.

To d

eter

min

e th

e le

vel

of in

terp

rete

r util

izat

ion

by d

ialy

sis

patie

nts;

to

asse

ss h

ow th

e fr

eque

ncy

of in

terp

rete

r util

izat

ion

affec

ts p

atie

nt o

utco

mes

. G

ener

al L

EP b

roke

n do

wn

into

num

ber o

f tim

es a

n in

terp

rete

r has

bee

n us

ed

(0-5

+)

Sett

ing:

Vic

toria

, Aus

tral

ia

Stud

y D

esig

n:

Retr

ospe

ctiv

e C

ohor

t

Freq

uenc

y of

Inte

rpre

tatio

n

Saw

redu

ctio

ns in

hos

pita

l ad

mis

sion

s du

e to

hyp

erka

lem

ia,

hype

rten

sion

& fl

uid

over

load

co

rrel

ated

with

freq

uenc

y of

in

terp

rete

r use

. Tho

se w

ith 0

in

terp

rete

rs s

aw 10

adm

issi

ons

with

flu

id o

verlo

ad a

nd h

yper

kalie

mia

, do

wn

to 5

and

2 (r

espe

ctiv

ely)

whe

n pa

tient

had

5 in

terp

rete

d en

coun

ters

. H

yper

tens

ion

mov

ed fr

om 6

(no

inte

rpre

ter)

to 2

adm

issi

ons

(5

inte

rpre

ter e

ncou

nter

s). N

o st

atis

tical

si

gnifi

canc

e or

con

foun

ders

de

scrib

ed.

Wea

k


Endnotes 1 Toronto Public Health. (2011). The Global City: Newcomer Health in Toronto. Retrieved from: http://www.toronto.ca/

legdocs/mmis/2011/hl/bgrd/backgroundfile-42361.pdf

2 Bowen, S. (2015). The Impact of Language Barriers on Patient Safety and Quality of Care. Société Santé en Francais. Retrieved from: https://santefrancais.ca/wp-content/uploads/SSF-Bowen-S.-Language-Barriers-Study.pdf

3 Flores, G. (2006). Language barriers to health care in the United States. New England Journal of Medicine, 355: 229-231. Retrieved from: http://www.nejm.org/doi/full/10.1056/NEJMp058316

4 Sears, J., Khan, K., Ardern, C.I., & Tamim, H. (2013). Potential for patient-physician language discordance in Ontario. BMC Health Services Research, 13:535. Retrieved from: https://bmchealthservres.biomedcentral.com/articles/10.1186/1472-6963-13-535

5 Toronto Public Health. (2011). The Global City: Newcomer Health in Toronto. Retrieved from: http://www.toronto.ca/legdocs/mmis/2011/hl/bgrd/backgroundfile-42361.pdff

6 Bowen, S. (2001). Language Barriers in Access to Health Care. Health Canada. Retrieved from: https://www.canada.ca/content/dam/hc-sc/migration/hc-sc/hcs-sss/alt_formats/hpb-dgps/pdf/pubs/2001-lang-acces/2001-lang-acces-eng.pdf

7 Jacobs, E., Chen, A.H.M., Karliner, L.s., Agger-Gupta, N., & Mutha, S. (2006). The Need for More Research on Language Barriers in Health Care: A Proposed Research Agenda. The Milbank Quarterly, 84(1), 111-133. Retrieved from: http://onlinelibrary.wiley.com/doi/10.1111/j.1468-0009.2006.00440.x/full

8 Naish, J., Brown, J., & Denton, B. (1994). Intercultural consultations: investigation of factors that deter non-English speaking women from attending their general practitioners for cervical screening. British Medical Journal, 309, 1126-1128.

9 Fortier, J.P., Strobel, C., Aguilera, E. (1998). Language barriers to health care: federal and state initiatives, 1990-1995. Journal of Health Care for the Poor and Underserved, 9(5), S81-S100.

10 Wilson, E., Chen, A.H., Grumbach, K., Wang, F., & Fernandez, A. (2005). Effects of limited English proficiency and physician language on health care comprehension. Journal of General Internal Medicine, 20(9), 800-806.

11 Lee, E.D., Rosenberg, C.R., Sixsmith, D.M., Pang, D., & Abularrage, J. (1998). Does a Physician-Patient Language Difference Increase the Probability of Hospital Admission? Academic Emergency Medicien, 5(1), 86-89.

12 Brown, A.F., Perez-Stable, E.J., Whitaker, E.E., Alexander, M., Gathe, J., & Washington, A.E. (1999). Ethnic differences in hormone replacement prescribing patterns. Journal of General Internal Medicine, 14, 663-669.

13 Gill, P., Scrivener, G., Lloyd, d., & Dowell, T. (1995). The effect of patient ethnicity on prescribing rates. Health Trends, 27, 111-114.

14 Chan, A., & Woodruff, R.K. (1999). Comparison of palliative care needs of English- and non-English-speaking patients. Journal of Palliative Care, 15, 26-30.

15 Hampers, L.C., Cha, S., Gutglass, D.J., Binns, H.J., & Krug, S.E. (1999). Language barriers and resource utilization in a pediatric emergency department. Pediatrics, 103, 1253-1256.

16 Morales, L. S., Cunningham, W. E., Brown, J. A., Liu, H., & Hays, R. D. (1999). Are Latinos less satisfied with communication by health care providers? Journal of general internal medicine, 14(7), 409-417.

17 Carrasquillo, O., Orav, E. J., Brennan, T. A., & Burstin, H. R. (1999). Impact of language barriers on patient satisfaction in an emergency department. Journal of General Internal Medicine, 14(2), 82-87.

18 Hornberger, J., Itakura, H., & Wilson, S. R. (1997). Bridging language and cultural barriers between physicians and patients. Public Health Reports, 112(5), 410.

19 Statistics Canada. (2017). An Increasingly Diverse Linguistics Landscape: Highlights from the 2016 Census. Retrieved from http://www.statcan.gc.ca/daily-quotidien/170802/dq170802b-eng.htm?HPA=1

http://www.toronto.ca/legdocs/mmis/2011/hl/bgrd/backgroundfile-42361.pdf


https://santefrancais.ca/wp-content/uploads/SSF-Bowen-S.-Language-Barriers-Study.pdf

http://www.nejm.org/doi/full/10.1056/NEJMp058316

https://bmchealthservres.biomedcentral.com/articles/10.1186/1472-6963-13-535

https://bmchealthservres.biomedcentral.com/articles/10.1186/1472-6963-13-535




https://www.canada.ca/content/dam/hc-sc/migration/hc-sc/hcs-sss/alt_formats/hpb-dgps/pdf/pubs/2001-lang-acces/2001-lang-acces-eng.pdf



http://onlinelibrary.wiley.com/doi/10.1111/j.1468-0009.2006.00440.x/full

http://onlinelibrary.wiley.com/doi/10.1111/j.1468-0009.2006.00440.x/full

http://www.statcan.gc.ca/daily-quotidien/170802/dq170802b-eng.htm?HPA=1


20 Statistics Canada. (2017). Data Tables, 2016 Census: Mother Tongue (8), Knowledge of Official Languages (5), Language Spoken Most Often at Home (8), Other Language(s) Spoken Regularly at Home (9), Age (7) and Sex (3) for the Population Excluding Institutional Residents of Canada, Provinces and Territories, Census Divisions and Census Subdivisions, 2016 Census - 100% Data. Retrieved from http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/dt-td/Rp-

21 Statistics Canada. (2017). Data tables, 2016 Census: Mother Tongue (8), Knowledge of Official Languages (5), Language Spoken Most Often at Home (8), Other Language(s) Spoken Regularly at Home (9), Age (7) and Sex (3) for the Population Excluding Institutional Residents of Canada, Provinces and Territories, Census Divisions and Census Subdivisions, 2016 Census – 100% Data. Retrieved from: http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/dt-td/Rp-eng.cfm?LANG=E&APATH=3&DETAIL=0&DIM=0&FL=A&FREE=0&GC=0&GID=0&GK=0&GRP=1&PID=109668&PRID=10&PTYPE=109445&S=0&SHOWALL=0&SUB=0&Temporal=2016&THEME=118&V-ID=0&VNAMEE=&VNAMEF=

22 Statistics Canada. (2016). Mother Tongue (8), Knowledge of Official Languages (5), Language Spoken Most Often at Home (8), Other Language(s) Spoken Regularly at Home (9), Age (7) and Sex (3) for the Population Excluding Institutional Residents of Canada, Provinces and Territories, Census Divisions and Census Subdivisions, 2016 Census - 100% Data. Retrieved from http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/dt-td/Rp

23 Bowen, S. (2001). Language Barriers in Access to Health Care. Health Canada. Retrieved from: https://www.canada.ca/content/dam/hc-sc/migration/hc-sc/hcs-sss/alt_formats/hpb-dgps/pdf/pubs/2001-lang-acces/2001-lang-acces-eng.pdf

24 McMurtray, A., Saito, E., & Nakamoto, B. (2009). Language preference and development of dementia among bilingual individuals. Hawaii Medical Journal, 68(9), 223.

25 Ministry of Health and Long-Term Care. (2017). Patients First: Action Plan for Health Care. Retrieved from: http://www.health.gov.on.ca/en/ms/ecfa/healthy_change/

26 Health Quality Ontario. (2017). What is Health Quality – Health Equity and Quality. Retrieved from: Language barriers pose significant impediments to accessing health care services.

27 Statistics Canada. (2015). Linguistic Characteristics of Canadians. Retrieved from: http://www12.statcan.gc.ca/census-recensement/2011/as-sa/98-314-x/98-314-x2011001-eng.cfm#a2

28 City of Toronto. (2012). 2011 Census – Language – Backgrounder. Retrieved from: https://www1.toronto.ca/city_of_toronto/social_development_finance__administration/files/pdf/language_2011_backgrounder.pdf

29 Riddick, S. (1998). Improving Access to Limited English-Speaking Consumers: A Review of Strategies in Health Care Settings. Journal of Health Care for the Poor and Underserved, 9(1998), S40-S61.

30 Aery, A., Kumar, N., Laher, N., & Sultana, A. (2017). Interpreting Consent – A Rights Based Approach to Language Accessibility in Ontario’s health care system. Wellesley Institute. Retrieved from: http://www.wellesleyinstitute.com/publications/interpreting-consent-a-rights-based-approach-to-language-accessibility-in-ontarios-health-care-system/

31 O’Campo, P., Devotta, K., Dowbor, T., & Pedersen, C. (2014). Reducing the language accessibility gap – Language Services Toronto Program Evaluation Report. St. Michael’s Hospital. Retrieved from: http://stmichaelshospitalresearch.ca/wp-content/uploads/2016/12/LST_Program_Evaluation_Report_July31_one-up.pdf

32 Bauer, A.M., & Alegria, M. (2010). The Impact of Patient Language Proficiency and Interpreter Service Use on the Quality of Psychiatric Care: A Systematic Review. Psychiatric Serv, 61(8), 765-773.

33 Flores, G. (2005). The Impact of Medical Interpreter Services on the Quality of Health Care: A Systematic Review. Medical Care Research and Review, 62(3), 255-299.

34 Arksey, H., & O’Malley, L. (2007). Scoping studies: towards a methodological framework. International Journal of Social Research Methodology, 8(1), 19-32. Retrieved from: http://www.tandfonline.com/doi/abs/10.1080/1364557032000119616

35 Effective Public Health Practice Project. (n.d.). Quality Assessment Tool for Quantitative Studies. Retrieved from:

http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/dt-td/Rp-

http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/dt-td/Rp-

Retrieved from: http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/dt-td/Rp-eng.cfm?LANG=E&APATH=3&DETAIL=0&DIM=0&FL=A&FREE=0&GC=0&GID=0&GK=0&GRP=1&PID=109668&PRID=10&PTYPE=109445&S=0&SHOWALL=0&SUB=0&Temporal=2016&THEME=118&VID=0&VNAMEE=&VNAMEF=




http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/dt-td/Rp




http://www.health.gov.on.ca/en/ms/ecfa/healthy_change/

http://www.health.gov.on.ca/en/ms/ecfa/healthy_change/

http://www12.statcan.gc.ca/census-recensement/2011/as-sa/98-314-x/98-314-x2011001-eng.cfm#a2

http://www12.statcan.gc.ca/census-recensement/2011/as-sa/98-314-x/98-314-x2011001-eng.cfm#a2

https://www1.toronto.ca/city_of_toronto/social_development_finance__administration/files/pdf/language_2011_backgrounder.pdf

https://www1.toronto.ca/city_of_toronto/social_development_finance__administration/files/pdf/language_2011_backgrounder.pdf

http://www.wellesleyinstitute.com/publications/interpreting-consent-a-rights-based-approach-to-language-accessibility-in-ontarios-health-care-system/

http://www.wellesleyinstitute.com/publications/interpreting-consent-a-rights-based-approach-to-language-accessibility-in-ontarios-health-care-system/

http://stmichaelshospitalresearch.ca/wp-content/uploads/2016/12/LST_Program_Evaluation_Report_July31_one-up.pdf

http://stmichaelshospitalresearch.ca/wp-content/uploads/2016/12/LST_Program_Evaluation_Report_July31_one-up.pdf

http://www.tandfonline.com/doi/abs/10.1080/1364557032000119616

http://www.tandfonline.com/doi/abs/10.1080/1364557032000119616


http://www.ephpp.ca/tools.html

36 Bernstein, J., Bernstein, E., Dave, A., Hardt, E., James, T., Linden, J., ... & Safi, C. (2002). Trained medical interpreters in the emergency department: effects on services, subsequent charges, and follow-up. Journal of Immigrant Health, 4(4), 171-176.

37 Jacobs, E. A., Lauderdale, D. S., Meltzer, D., Shorey, J. M., Levinson, W., & Thisted, R. A. (2001). Impact of Interpreter Services on Delivery of Health Care to Limited–English proficient Patients. Journal of General Internal Medicine, 16(7), 468-474.

38 Jacobs, E. A., Shepard, D. S., Suaya, J. A., & Stone, E. L. (2004). Overcoming language barriers in health care: costs and benefits of interpreter services. American Journal of Public Health, 94(5), 866-869.

39 Fagan, M. J., Diaz, J. A., Reinert, S. E., Sciamanna, C. N., & Fagan, D. M. (2003). Impact of interpretation method on clinic visit length. Journal of General Internal Medicine, 18(8), 634-638.

40 Hacker, K., Choi, Y. S., Trebino, L., Hicks, L., Friedman, E., Blanchfield, B., & Gazelle, G. S. (2012). Exploring the impact of language services on utilization and clinical outcomes for diabetics. PloS One, 7(6), e38507.

41 Sarver, J., & Baker, D. W. (2000). Effect of language barriers on follow‐up appointments after an emergency department visit. Journal of General Internal Medicine, 15(4), 256-264.

42 d’Ardenne, P., Ruaro, L., Cestari, L., Fakhoury, W., & Priebe, S. (2007). Does interpreter-mediated CBT with traumatized refugee people work? A comparison of patient outcomes in East London. Behavioural and Cognitive Psychotherapy, 35(3), 293-301.

43 Nápoles, A. M., Santoyo-Olsson, J., Karliner, L. S., Gregorich, S. E., & Pérez-Stable, E. J. (2015). Inaccurate Language Interpretation and Its Clinical Significance in the Medical Encounters of Spanish-speaking Latinos. Medical Care, 53(11), 940-947.

44 Wagner, J., Abbott, G., & Lacey, K. (2005). Knowledge of heart disease risk among Spanish speakers with diabetes: the role of interpreters in the medical encounter. Ethnicity & Disease, 15(4), 679-684.

45 Flores, G., Abreu, M., Barone, C. P., Bachur, R., & Lin, H. (2012). Errors of medical interpretation and their potential clinical consequences: a comparison of professional versus ad hoc versus no interpreters. Annals of Emergency Medicine, 60(5), 545-553.

46 Butow, P.N., Goldstein, D., Bell, M.L., Sze, M., Aldridge, L.J., Abdo, S., Tanious, M., Dong, S., Iedema, R., Vardy, J., Ashgari, R., Hui, R., & Eisenbruch, M. (2011). Interpretation in Consultations with Immigrant Patients with Cancer: How Accurate Is It? Journal of Clinical Oncology, 29(20), 2801-2807.

47 Gany, F., Kapelusznik, L., Prakash, K., Gonzalez, J., Orta, L. Y., Tseng, C. H., & Changrani, J. (2007). The impact of medical interpretation method on time and errors. Journal of General Internal Medicine, 22(2), 319-323.

48 Leng, J. C., Changrani, J., Tseng, C. H., & Gany, F. (2010). Detection of depression with different interpreting methods among Chinese and Latino primary care patients: a randomized controlled trial. Journal of Immigrant and Minority health, 12(2), 234-241.

49 Lion, K. C., Brown, J. C., Ebel, B. E., Klein, E. J., Strelitz, B., Gutman, C. K., ... & Mangione-Smith, R. (2015). Effect of telephone vs video interpretation on parent comprehension, communication, and utilization in the pediatric emergency department: a randomized clinical trial. JAMA Pediatrics, 169(12), 1117-1125.

50 Grover, A., Deakyne, S., Bajaj, L., & Roosevelt, G. E. (2012). Comparison of throughput times for limited English proficiency patient visits in the emergency department between different interpreter modalities. Journal of Immigrant and Minority Health, 14(4), 602-607.

51 Locatis, C., Williamson, D., Gould-Kabler, C., Zane-Smith, L., Detzier, I., Roberson, J., Maisiak, R., & Ackerman, M. (2010). Comparing In-Person, Video, and Telephonic Medical Interpretation. J Gen Intern Med, 25(4), 345-50.

52 de Bocanegra, H. T., Rostovtseva, D., Cetinkaya, M., Rundel, C., & Lewis, C. (2011). Quality of reproductive health services to limited English proficient (LEP) patients. Journal of Health Care for the Poor and Underserved, 22(4), 1167-1178.

http://www.ephpp.ca/tools.html


53 Zimbudzi, E., Thompson, S., & Terrill, B. (2010). How accessible are interpreter services to dialysis patients of Non-English-Speaking Background? Australasian Medical Journal, 1(3), 205-212.

54 Wallbrecht, J., Hodes-Villamar, L., Weiss, S. J., & Ernst, A. A. (2014). No difference in emergency department length of stay for patients with limited proficiency in English. Southern Medical Journal, 107(1), 1-5.

55 López, L., Rodriguez, F., Huerta, D., Soukup, J., & Hicks, L. (2015). Use of interpreters by physicians for hospitalized limited English proficient patients and its impact on patient outcomes. Journal of General Internal Medicine, 30(6), 783-789.

56 Hampers, L. C., & McNulty, J. E. (2002). Professional interpreters and bilingual physicians in a pediatric emergency department: effect on resource utilization. Archives of Pediatrics & Adolescent Medicine, 156(11), 1108-1113.

57 Jacobs, E. A., Fu, P. C., & Rathouz, P. J. (2012). Does a Video Interpreting Network Improve Delivery of Care in the Emergency Department? Health Services Research, 47(1pt2), 509-522.

58 Davies, S. E., Dodd, K. J., Tu, A., Zucchi, E., Zen, S., & Hill, K. D. (2016). Does English proficiency impact on health outcomes for inpatients undergoing stroke rehabilitation? Disability and Rehabilitation, 38(14), 1350-1358.

59 Lindholm, M., Hargraves, J. L., Ferguson, W. J., & Reed, G. (2012). Professional language interpretation and inpatient length of stay and readmission rates. Journal of General Internal Medicine, 27(10), 1294-1299.

60 Eytan, A., Bischoff, A., Rrustemi, I., Durieux, S., Loutan, L., Gilbert, M., & Bovier, P. A. (2002). Screening of mental disorders in asylum-seekers from Kosovo. Australian & New Zealand Journal of Psychiatry, 36(4), 499-503.

61 Karliner, L. S., Pérez-Stable, E. J., & Gregorich, S. E. (2017). Convenient access to professional interpreters in the hospital decreases readmission rates and estimated hospital expenditures for patients with limited English proficiency. Medical Care, 55(3), 199-206.

62 Jimenez, N., Jackson, D. L., Zhou, C., Ayala, N. C., & Ebel, B. E. (2014). Postoperative pain management in children, parental English proficiency, and access to interpretation. Hospital Pediatrics, 4(1), 23.

63 Dang, J., Lee, J., Tran, J. H., Kagawa-Singer, M., Foo, M. A., Nguyen, T. U. N., ... & Tanjasiri, S. P. (2010). The role of medical interpretation on breast and cervical cancer screening among Asian American and Pacific Islander women. Journal of Cancer Education, 25(2), 253-262

64 Price, E. L., Pérez-Stable, E. J., Nickleach, D., López, M., & Karliner, L. S. (2012). Interpreter perspectives of in-person, telephonic, and videoconferencing medical interpretation in clinical encounters. Patient education and counseling, 87(2), 226-232.

65 Chen, L.M., Farwell, W.R. & Jha, A.K. (2009). Primary care visit duration and quality: does good care take longer? Arch Intern Med, 169(20):1866–72. doi:10.1001/archinternmed.2009.341.

66 Abbo, E.D., Zhang, Q., Zelder, M., & Huang, E.S. (2008). The increasing number of clinical items addressed during the time of adult primary care visits. J Gen Intern Med, 23(12):2058–65. doi:10.1007/s11606-008-0805-8.

67 Lopez, L., Rodriquez, F., Huerta, D., Soukup, J. &Hicks, L. (2015). Use of Interpreters by Physicians for Hospitalized Limited English Proficient Patients and Its Impact on Patient Outcomes. Journal of General Internal Medicine, 30(6), 783-789.

68 Karliner, L.S., Kim, S.E., Meltzer, D.O., & Auerback, A.D. (2010). Influence of language barriers on outcomes of hospital care for general medicine inpatients. Journal of Hospital Medicine, 5(5), 276-282.


70 Bauer, A.M., & Alegria, M. (2010). The Impact of Patient Language Proficiency and Intrepreter Service Use on the Quality of Psychiatric Care: A Systematic Review. Psychiatric Serv, 61(8), 765-773.



72 Bauer, A.M., & Alegria, M. (2010). The Impact of Patient Language Proficiency and Intrepreter Service Use on the Quality of Psychiatric Care: A Systematic Review. Psychiatric Serv, 61(8), 765-773.

73 Grant, M. J., & Booth, A. (2009). A typology of reviews: an analysis of 14 review types and associated methodologies. Health Information & Libraries Journal, 26(2), 91-108.

74 Madore, O. (1992). The Canadian and American Health Care Systems. Economics Division. Retrieved from: http://publications.gc.ca/Collection-R/LoPBdP/BP/bp300-e.htm

http://publications.gc.ca/Collection-R/LoPBdP/BP/bp300-e.htm

http://publications.gc.ca/Collection-R/LoPBdP/BP/bp300-e.htm

Documents

Access to Language Interpretation Services and its Impact ......the following: if language barriers are addressed, if the intervention involved a trained interpreter, and if clinical