SummaryCambridge Health Alliance enhanced its round-the-clock interpreter services program, which provides face-to-face interpreting in 60 languages through onsite staff, per diem staff, and agency interpreters and remote telephonic and video interpreting through a call center as well as external telephonic interpreting vendors, with access to 140 languages. Working as part of the Robert Wood Johnson Foundation's Speaking Together collaborative, the alliance tracks key indicators that gauge the quality of its interpreter services on an ongoing basis. Based on analyses of these data, along with the desire to harmonize its operations with the Institute of Medicine's Six Aims for Improvement, the alliance made a number of enhancements to the program, including increasing the capacity of its telephone-based services, improving staff efficiency through use of a database and electronic dispatching system, and instituting linguistically appropriate depression screening for limited English proficiency patients. The program expansion has resulted in an increase in the percentage of time that interpreters spend interpreting and led to minimal wait times and few abandoned calls for patients seeking interpreters, as well as high rates of depression screening for limited English proficiency patients at East Cambridge Health Center. The program enhancements may also have the potential to reduce readmissions, although further study is needed.Suggestive: The evidence consists of post-implementation data and/or anecdotal reports on the percent of staff time spent interpreting, patient wait times, abandoned calls, and depression screening rates, as well as a small pilot comparison study of readmission rates among Cambridge Health Alliance patients.
Developing OrganizationsCambridge Health Alliance
Date First Implemented2006
Vulnerable Populations > Non-english speaking/limited english proficiency
Problem AddressedAccording to the 2007 American Community Survey, more than 50 million people speak a language other than English in their homes.1 Communication barriers due to limited English proficiency can negatively affect access to and quality of care, and these barriers are common in the absence of trained medical interpreters.
- Communication barriers affect access to care: Individuals with limited English proficiency are less likely to have a regular source of primary care2 and to receive preventive services than are individuals whose first language is English.3 In a national survey of insured nonelderly adults, Spanish-speaking Hispanic patients were less likely to visit physicians and mental health professionals and to receive influenza vaccinations and mammograms than were non-Hispanic White patients.4
- Communication barriers affect quality of care: Communication barriers impede the understanding and compliance of treatment plans and therapies among patients with limited English proficiency.5 Studies show that patients who need but do not have access to an interpreter are less likely to understand instructions for taking medications, receive information on medication side effects, and be satisfied with their care.6,7
Description of the Innovative ActivityCambridge Health Alliance enhanced its round-the-clock interpreter services program, which provides face-to-face or remote interpreting in 140 languages. Based on analyses of key data gathered as a part of the alliance's participation in the Robert Wood Johnson Foundation's Speaking Together collaborative, the alliance made a number of enhancements to the program, including increasing the capacity of telephone-based services, improving staff efficiency through use of a database, and instituting linguistically appropriate depression screening for all patients. As a followup in fiscal year 2009, the alliance developed new measures to integrate interpreter services with efficiency, timeliness, patient-centeredness, and equity of care provided to limited English proficient patients. Key elements of this expansion are described below:
- Background on Cambridge Health Alliance interpreter program: At all Cambridge Health Alliance sites, providers who need assistance with limited English proficiency patients can dial internal extension 3333 at any time of the day to specify the patient's preferred language and whether face-to-face or remote interpreting is required. The program offers interpreter services in 140 different languages, with Brazilian Portuguese, Spanish, and Haitian Creole being the most commonly requested. The interpreter services office determines whether onsite interpreters can cover the need or outside interpreters or remote interpreters are needed. A team of trained dispatchers manages the roster of all interpreters and informs the appropriate interpreters where they are needed by means of a custom electronic dispatching system. Reports from outpatient scheduling modules inform the interpreter office of scheduled upcoming needs.
- Ongoing tracking of services to facilitate quality improvement: As part of Robert Wood Johnson Foundation's Speaking Together program, Cambridge Health Alliance tracked key measures of the quality of interpreter services, focusing on how interpreters communicate with patients and how hospital staff can better structure and manage language services programs to ensure efficient, equitable, patient-centered and timely communications with limited English proficient patients. As of 2009, Cambridge Health Alliance revised and expanded these measures based on the Institute of Medicine's Six Aims for Improvement. Current measures being tracked include the following:
- Percent of work time spent interpreting for full-time and part-time interpreters
- Percent of all interpreted encounters with a delay over 20 minutes, measured from the time the interpreter is contacted with the request to the time the interpreter is ready and available to interpret
- Percent of all calls abandoned for a measured time period
- Percent of interpreter requests completed
- Mean score on a custom survey question asking emergency room patients to rate interpreters from 1 to 5
- Percent of emergency room patients answering 5 on 1 to 5 scale (indicating "very good") for a custom survey question about interpreter services
- Percent of documents translated with use of a TRADOS software program
- Average number of days from the assignment of a document translation to the return of the finished translation to the customer
- Enhancements to interpreter services: In response to participating in the Speaking Together collaborative and tracking quality measures, the alliance has implemented three major enhancements to its interpreter program:
- Improved remote interpreting: Information provided in February 2013 indicates that interpreting at the alliance is primarily accessed via internal phone extension 3333. The caller selects face-to-face or telephonic interpreting. Telephonic interpreting is available for Portuguese, Spanish, Haitian Creole, Hindi, and Bengali via internal phone queues. If a different language is needed, the caller has the option to select a number that links them directly to an outside telephonic interpreting vendor. In 2012, telephonic intepreting surpassed face-to-face interpreting as the primary interpreting modality across the alliance. Historically, interpreting via the internal phone queues was provided by interpreters at various alliance sites who logged onto a virtual automatic call distribution system to handle requests for remote services when they were not busy providing face-to-face interpretation. Because this system sometimes resulted in delays, the Alliance implemented two major enhancements to the system.
- Small, dedicated call center for telephonic interpreting: In addition to having interpreters log into the virtual system, the alliance instituted a call center in 2008 where dedicated Portuguese and Spanish interpreters provided telephonic interpreter services Monday through Friday between 8 a.m. and 5 p.m. These interpreters no longer had to split their time between providing face-to-face and remote services. Eventually, a Haitian Creole interpreter was added to the call center. Information provided in February 2013 indicates that in June 2012, the alliance added video interpreting capability to its call center and rolled out videophone units to its three Emergency Rooms, Surgical Day Care, and a number of inpatient and ambulatory sites. Callers can request video interpreter services by pressing the desired language on a video touchscreen. The average connect time for video interpreter services is less than 40 seconds. The popularity of video interpreting with emergency room providers and patients led the alliance to extend its call center hours to 11 p.m. and offer remote interpreting on weekends.
- Automatic call rollover system: Information provided in February 2013 indicates that the alliance instituted a call rollover system that automatically transfers unanswered telephonic or video calls to external vendors who provide telephonic or video interpreting. This change made the system more user-friendly for providers, who no longer had to deal with unanswered calls or long waits when alliance interpreters were busy.
- Database to monitor, improve staffing efficiency: To improve interpreter productivity, the alliance linked the interpreter database to payroll information on staffing hours. The database contains information on each interpreter encounter (gathered by interpreters on encounter tracking forms and entered through an easy-to-use, Web-based tracking system), including request time, arrival time, start time, end time, type of service provided, and an explanation if services could not be provided. Interpreters enter data after each encounter via the tracking system, which can be accessed from any phone on any of the alliance's three main campuses. The database allows managers to analyze the percent of each interpreter's time spent providing interpreter services. In addition to identifying the level of activity of each individual interpreter, this tool allows the alliance to examine volumes and service types by hour (e.g., the number of Spanish encounters from 11 a.m. to 12 p.m.) and site and thus reconfigure staffing based on demand patterns.
- Improved written translation service: In 2008, the alliance began to use TRADOS software to support timely turnaround of requests for written translation. TRADOS allows translators to leverage past translations to improve efficiency, achieve greater customer satisfaction through accuracy and consistency, share glossaries and specialized terminology, and reduce translation time. By the end of fiscal year 2009, 60 percent of all translated documents were completed through use of the TRADOS software. By the end of fiscal year 2012, the percent of documents translated with TRADOS had risen to 90 percent (updated February 2013).
- Linguistically appropriate depression screening: East Cambridge Health Center, which is part of Cambridge Health Alliance, conducts linguistically appropriate depression screenings of patients using translated versions of the Patient Health Questionnaire-9 form. The health center screens all new patients, patients coming in for annual physicals, diabetic patients, expecting and new mothers, and any other patient believed to be at risk for depression. Patients typically receive the screening form at the front desk and fill it out while waiting to see their provider. The health center uses a special touch-screen computer with audio to administer the questionnaire to patients with low-literacy skills in Brazilian Portuguese, European Portuguese, Spanish, and Haitian Creole, the primary languages served. Patients' scores are entered into the electronic medical record. Patients who screen positively for depression are, for the most part, managed within primary care, with primary care providers supported by psychiatric consultation liaison. In some cases, patients are referred to a Cambridge Health Alliance mental health clinic, where, as in primary care, language assistance is provided via interpreters and bilingual, bicultural providers.
Context of the InnovationCambridge Health Alliance is an academic public health care system that operates three acute care hospitals, more than 20 primary care sites, and the public health department for the city of Cambridge, MA. Approximately 30 to 35 percent of alliance patients speak a language other than English, with Portuguese, Spanish, and Haitian Creole being the commonly preferred languages. Recognizing that proper communication with patients is crucial for patient safety and good quality health care, the alliance first implemented medical interpreter services at Cambridge Hospital in the 1970s and eventually established a system-wide Interpreter Services Department. In an effort to improve its interpreter services by focusing on quality measurement and learning from others, Alliance leaders applied for and received a grant in 2006 to participate in Speaking Together: National Language Services Network, a national collaborative program funded by the Robert Wood Johnson Foundation aimed at improving the quality and availability of health care language services for patients with limited English proficiency. In fiscal year 2009, the alliance followed up on its work with Speaking Together by implementing a number of quality measures designed to align its operations with the Institute of Medicine's Six Aims for Improvement, with the goal of creating a health care experience that is safe, effective, efficient, timely, patient centered, and equitable.
ResultsPost-implementation and anecdotal data suggest that the program expansion has resulted in an increase in the percentage of time that interpreters spend interpreting and led to minimal wait times and few abandoned calls for patients seeking interpreters. It has also led to high rates of depression screening for limited English proficiency patients at East Cambridge Health Center. A small pilot study of congestive heart failure patients suggests that the program enhancements may have the potential to reduce readmissions, although further study is needed.
Suggestive: The evidence consists of post-implementation data and/or anecdotal reports on the percent of staff time spent interpreting, patient wait times, abandoned calls, and depression screening rates, as well as a small pilot comparison study of readmission rates among Cambridge Health Alliance patients.
- Many patients served: In fiscal year 2009, Cambridge Health Alliance supported more than 171,000 interpreter encounters; 80 percent were face to face. Information provided in February 2013 indicates that in 2012, telephonic interpreting surpassed face to face interpreting as the primary requested interpreting modality, with video interpreting also growing in popularity. The introduction of video interpreting in Emergency Rooms in June 2012 has made it possible to bring the wait time for limited English proficient patients in line with that of English speakers; on average less than four minutes between check-in and triage.
- More time spent interpreting: From 2007 to early 2008, alliance interpreters spent an average of 33 to 45 percent of their time providing medical interpretation in clinical encounters; by June of 2009, this figure increased to 61 percent as a result of new productivity goals. Information provided in February 2013 indicates that, with the introduction of video interpreting to the call center and the requirement that campus-based interpreters log on to provide remote interpreting services when not assigned to a face to face request, department productivity continued to rise and was at 70 percent in January 2013. Although pre-implementation comparison data are unavailable (the alliance started collecting data on this measure as part of the Speaking Together initiative), anecdotal reports from interpreters suggest that the staffing efficiency improvements, such as the use of the Web-based tracking system, have also allowed them to spend more time interpreting.
- Short wait times and few abandoned calls: According to data gathered from January to March 2009, only 2.5 percent of the 42,294 patients requesting interpreter services encountered a wait time of longer than 20 minutes, and only 4 percent of all phone calls seeking interpreting were abandoned (i.e., dropped before being answered by a dispatcher or interpreter). From July 2009 through June 2011, the percentage of wait times longer than 20 minutes has consistently been less than 2.32 percent of total requests and call abandonment has ranged between 3.5-3.9 percent (updated February 2013).
- High rates of screening limited English proficiency patients for depression: From 2007 to early 2008, East Cambridge Health Center achieved monthly depression screening rates for limited English proficiency patients that range from 67 to 87 percent, comparable to the rates achieved for English-speaking patients.
- Possible reduction in readmission rates: A small pilot study of patients with congestive heart failure found that those who did not receive discharge instructions from an interpreter were more likely to be rehospitalized; in fact, none of the patients who received interpreter services at discharge required readmission. However, because the data were collected from a small sample of patients, further studies are needed to evaluate the impact of the program on readmission rates.
Planning and Development ProcessKey steps in the planning and development process are outlined below:
- Identifying an executive and physician champion: To join Speaking Together, the alliance was required to identify an executive and physician champion. The chief executive officer participated in the interviewing process for the program. The senior vice president for quality and performance improvement was the executive champion.
- Choosing areas of focus and collecting baseline data: Before joining Speaking Together, the alliance had already created a data entry system to track the number of interpreter encounters per day. On joining the collaborative, the alliance engaged in discussions with physicians and nurses that led to the decision to focus on two specific performance measures that were required by the Speaking Together Collaborative: the percentage of limited English proficiency patients receiving initial assessment and discharge instructions from assessed and trained interpreters or from bilingual providers assessed for language proficiency, and the percent of time interpreters spend providing medical interpretation in clinical encounters with patients. The alliance then modified its electronic medical record to include reportable fields for how patients' language needs are met.
- Collaborating to screen for depression: After Speaking Together chose East Cambridge Health Center as a site for an outpatient project, the alliance decided to focus on mental health because of its strong psychiatry program and the health center's previous efforts to improve screening for depression among diabetic patients. The psychiatric consultation liaison team developed a system to treat patients who screen positively for depression and, with the Community Affairs Department, depression screening software for patients with low-literacy skills. The team also educated providers on screening, treatment, and evaluation processes and held regular meetings to get their feedback on the program. Screening began with one provider; once the "kinks" were worked out of the system, the alliance rolled the program out across East Cambridge Health Center.
- Training: Providers and administrative staff at the Health Center received training on how to conduct linguistically appropriate depression screening. The practice manager in charge of the front desk led the sessions for administrative employees, whereas the clinic administrator led the provider training sessions. Providers, care coordinators, medical assistants, administrative employees, a nutritionist, and a social worker attend ongoing monthly meetings to discuss how the program is working.
- Choosing further quality measures to track: In fiscal year 2009, the quality and performance improvement team worked closely with the interpreter services department on the definition and development of quality measures aligned with efficient, timely, patient-centered, and equitable care to all patients.
Resources Used and Skills Needed
- Staffing: Information provided in February 2013 indicates that Cambridge Health Alliance's interpreter program has a combined staff of approximately 50 full-time and part-time interpreters, and 70 per diem interpreters. A senior director supervises three managers. The managers are primarily responsible for overseeing dispatchers and interpreters at the alliance's three hospitals, the call center, and at a number of alliance health centers. In 2010 and 2011, respectively, the department assumed oversight of Patient Transport and Patient Information Desks.
- Costs: The initial enhancements to the program cost roughly $60,000, primarily for the project coordinator's compensation and data analysis. No new staff were hired as a part of the initial enhancements. Costs associated with subsequent initiatives to align the department's operations with the Institute of Medicine's Six Aims and the development of additional quality measures were included in the department's general operating budget. Additional capital funding was secured for the 2012 introduction of video interpreting to the call center and across campuses; these funds were applied to infrastructure in the installation of a secure wireless network and to videophone equipment for interpreters and end-users (updated February 2013).
Funding SourcesRobert Wood Johnson Foundation
The Robert Wood Johnson Foundation provided a $60,000 grant to Cambridge Health Alliance as part of the Speaking Together program.
Getting Started with This Innovation
- Assess capacity for documenting language needs of your patients: It is essential to assess how patients' language needs are being identified in the hospital system and to identify ways that information might be streamlined with other patient data. For example, if patients' preferred languages of care are collected as part of the hospital's registration process, it might be possible to link that information electronically with scheduled visits.
- Begin by expanding on existing programs: Because change often meets with resistance, it may make sense to implement the program in a stepwise approach, beginning with modest expansions to existing services. This approach can help to secure the buy-in necessary for long-term success.
- Start small and then roll out: Starting small provides the opportunity to identify and address workflow issues before they affect a large population. Achieving success on a small scale may also increase confidence in a new process before it is rolled out across the organization.
- Get buy-in from the executive level: Senior executives need to recognize language barriers as a patient safety issue and should commit to meeting patients' language needs. At Cambridge Health Alliance, getting buy-in from the chief executive officer was crucial to the organization's decision to join Speaking Together.
- Collaborate with other departments and use existing resources: Enlisting other departments for help can raise awareness about existing resources. For example, Cambridge Health Alliance psychiatrists collaborated with staff of the Community Affairs Department; this collaboration allowed them to learn about existing information technology resources, which were then were used to create the touchscreen version of the Patient Health Questionnaire-9 form. Similarly, collaboration between the alliance's Multilingual Interpreter Services and information technology departments laid the groundwork for a lasting partnership in the customization of software and identification of reportable fields for key quality measures.
Sustaining This Innovation
- Make it easy for providers to access interpreters: Facilitate the use of medical interpreters by making it easy for providers to use them. For example, the alliance developed the call rollover system as a way to reduce delays experienced by providers trying to access telephonic interpreter services.
- Track performance measures related to language needs: Measuring performance on key indicators is critical to identifying language gaps. For each encounter, examine whether an interpreter was present, the provider spoke directly with the patient, the patient refused an interpreter, or the provider asked a bilingual employee for help. In this process, it is important to evaluate the data available in the hospital's current system in comparison to additional data that may need to be collected.
- Screen and train interpreters: The alliance was able to screen and hire well-trained interpreters due to the presence of excellent training programs in the Boston area.
- Expect some hesitancy related to depression: There was initially some nervousness on the part of providers in dealing with patients who test positively for depression. This discomfort led some providers to resist screening patients for the disease. Over time, however, providers became more comfortable administering the translated Patient Health Questionnaire-9 form and using interpreters to assist in treating patients who test positively.
Contact the InnovatorMursal Khaliif
Senior Director, Multilingual Patient Services
Cambridge Health Alliance
1493 Cambridge St
Cambridge, MA 02139
Phone: (617) 591-6792
Fax: (617) 591-6990
Innovator DisclosuresMr. Khaliif has not indicated whether he has financial interests or business/professional affiliations relevant to the work described in this profile.
References/Related ArticlesSpeaking Together - National Language Services Network. Available at: http://www.speakingtogether.org
More information about the Robert Wood Johnson Foundation's promising practices is available at http://pweb1.rwjf.org/qualityequality/pp.jsp.
Weinick RM, Krauss NA. Racial/ethnic differences in access to care. Am J Public Health. 2000;90(11):1771-4. [PubMed]
Woloshin S, Schwarts LM, Katz SJ, et al. Is language a barrier to the use of preventive services? J Gen Intern Med. 1997;12:472-7. [PubMed]
Fiscella K, Franks P, Doescher MP, et al. Disparities in health care by race, ethnicity, and language among the insured. Med Care. 2002;40:52-9. [PubMed]
Ku L. How race/ethnicity, immigration status and language affect health insurance coverage, access to care and quality of care among the low-income population. Washington, DC: Kaiser Family Foundation; August 2003. Available at: http://www.kff.org/uninsured/kcmu4132report.cfm
6 Andrulis D, Goodman N, Pryor N. What a difference an interpreter can make: health care experiences of uninsured with limited English proficiency. Boston, MA: The Access Project; April 2003.
David RA, Rhee B. The impact of language as a barrier to effective health care in an underserved urban Hispanic community. Mt Sinai J Med. 1998;65(5,6):393-7. [PubMed]
|Disclaimer: The inclusion of an innovation in the Innovations Exchange does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or Westat of the innovation or of the submitter or developer of the innovation. Read more.|
Original publication: October 10, 2008.
Original publication indicates the date the profile was first posted to the Innovations Exchange.
Last updated: March 26, 2014.
Last updated indicates the date the most recent changes to the profile were posted to the Innovations Exchange.
Date verified by innovator: February 14, 2013.
Date verified by innovator indicates the most recent date the innovator provided feedback during the annual review process. The innovator is invited to review, update, and verify the profile annually.