The American Journal of Managed Care
July 2021
Volume 27
Issue 7

The Role of Race, Ethnicity, and Language in Care Transitions

Diverse patients experience disparities in care transitions. A survey of 224 patients showed differences by race, ethnicity, and language in technology access and in patient worries post discharge.


Objectives: To identify areas where transition from hospital to community could be improved, with a special focus on racial, ethnic, and language differences.

Study Design: A cross-sectional survey administered by postal mail and bilingual telephone interviewers.

Methods: Patients were randomly selected within strata by race, ethnicity, and language proficiency. A total of 224 patients (response rate: 63.5%) who had recently experienced a hospital stay completed the survey.

Results: Overall, 1 in 4 patients were alone at discharge. More than half of patients with limited English proficiency reported lack of access to medical interpreters and translated materials. We noted significant differences by race, ethnicity, and language in technology access and in patient-reported worries in the posthospital period. Hispanic or Latino patients and patients with limited English proficiency were less likely to report access to a computer and less likely to access the Patient Gateway portal. Black or African American patients were also less likely to use the Patient Gateway portal. Asian patients were more likely to be worried about getting home health services.

Conclusions: Our findings highlight the enhanced difficulties that diverse patients may experience when transitioning from hospital to community-based settings. When considering how to best address the complex needs of diverse populations, interventions must be sensitive to the presence or absence of others, potential digital divides, and medical interpretation.

Am J Manag Care. 2021;27(7):e221-e225.


Takeaway Points

System challenges in providing care to diverse patients at time of discharge include:

  • Lack of interpreters and appropriately translated materials;
  • Lack of patient access to technology for needed posthospital telehealth visits; and
  • Lack of access to patient portals, which may hinder health care engagement.


Racial, ethnic, and linguistic minority patients experience disparities in hospital care and risk for emergency department readmissions.1,2 Providing consistent and adequate support at care transitions, including certified medical interpreter and certified translation services when indicated, may be a challenge for health systems during the hospital course, at discharge, and in the posthospital period. Although not recommended, family members and caregivers may provide enhancements to communication; certified medical interpreters are essential to patients with limited English proficiency (LEP). Both in-person and remote interpreters have been proven effective in improving quality for patients with LEP.3

In our hospital, patient-reported experience data are regularly assessed for disparities. In 2019, we noted racial and ethnic differences in our Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) surveys, including several measures related to care transitions in the postacute period (patient/family preferences taken into account at discharge, patient had good understanding of responsibilities for managing health at discharge, and patient understood purpose of taking each medication at discharge).4 We conducted surveys of recently discharged patients to explore factors prior to, during, and after discharge that had been shown in prior studies to inform variations in diverse patients’ experience of care, care transitions, and readmission risk.



Massachusetts General Hospital (MGH) is a Harvard University–affiliated teaching hospital in Boston with more than 900 beds. MGH serves a population that is ethnically diverse, although (like Massachusetts) not racially diverse (4% Black, 9% Hispanic or Latino, 84% White, 3% other). Annually, interpreters provide more than 100,000 encounters in more than 70 languages. Data on patient race, ethnicity, and language are gathered at the time of patient registration into our system and are routinely updated at the point of care, along with insurance and personal contact information.

Sample and Data Collection

We conducted the MGH Care Transitions Survey from July to December 2019 using self-administered mail and telephone-assisted interviews with a sample of 400 patients. Patients were randomly selected within strata by race, ethnicity, and language using the MGH Research Patient Data Registry,5 which is a continuously updated hospital registry of clinical and billing data used to identify patients discharged in the 3 months prior to the start of the field period. Surveys were conducted in English, Spanish, Portuguese, Chinese, Khmer, Haitian Creole, and Vietnamese. For patients with a documented preferred language other than English, we mailed a bilingual letter and translated questionnaire. We followed up the mail survey with calls from bilingual telephone interviewers. Additional patient characteristics were obtained through electronic health records. The response rate was 63.5%.


The brief questionnaire included 12 items (taking less than 10 minutes to complete by either paper or telephone administration) and was written by the research team, drawing from prior studies, HCAHPS items, and committee needs. Measures included information on whether patients were accompanied or alone, access to technology, and worries experienced post hospital that might inform readmission risk. Patients with LEP were asked 3 questions about language needs. Content was reviewed by the Care Transitions Task Force assembled to improve experiences transitioning from hospital to community. The questionnaire was approved by the Mass General Brigham Institutional Review Board.


Here, we compare key measures by patient sample group (ie, cross-section, Black or African American, Asian, Hispanic or Latino) and language proficiency. The cross-section sample consisted of 91% White non-Hispanic patients.


A total of 224 patients completed the survey (paper: n = 106; telephone: n = 118) between July and December 2019. Among the 400 contacted patients, 29 could not be interviewed due to serious illness or death; 18 could not be located. Seventy-one patients were interviewed in a language other than English. As shown in Table 1, 52.7% of patients were female and 47.3% were male; 19.6% were aged 18 to 39 years, 39.3% were aged 40 to 64 years, and 41.1% were 65 years or older. Of all respondents interviewed, 46% had private insurance, 33% had Medicare coverage, and 18% had MassHealth (Medicaid). Among patients with LEP, 38% had MassHealth compared with 9% of non-LEP patients.

Table 2 shows a comparison by patient sample group. We noted racial and ethnic differences in technology access in the past year. Hispanic or Latino patients were less likely to report access to a laptop, desktop, or tablet computer (25.5%). Despite availability of our electronic health record patient portal in Spanish, only 7.8% reported access compared with patients in the cross-section sample (72.1% reporting access to a laptop, desktop, or tablet computer and 63.2% reporting access to the patient portal) (P < .001). Black or African American patients (33.3%) were also less likely to use the patient portal compared with patients in the cross-section sample (63.2%) (P < .001).

Table 3 shows total responses and a comparison by language proficiency. Overall, approximately 1 in 4 patients was alone at discharge. More than half of patients with LEP reported lack of access to medical interpreters and translated materials at discharge. We noted significant differences by language proficiency in technology access in the past year. Patients with LEP were less likely to report access to a smartphone (56.3%), to a computer (25.4%), and to our patient portal (9.9%) compared with non-LEP patients (85.0%, 72.5%, and 58.2%, respectively) (P < .001).

Figure 1 shows patient-reported worries in the posthospital period by language proficiency. Patients with LEP were more likely to be worried about getting home health care services (23.9%), food (19.7%), and transportation to appointments (28.2%) than non-LEP patients (9.8%, 9.2%, and 11.1%, respectively) (P < .05).

Figure 2 shows patient-reported worries in the posthospital period by sample group. The most frequently cited worry by patients in the cross-section sample was understanding medications (16.2%), and the least frequently cited worry was getting home health care services (2.9%). The most frequently cited worry by patients in the Asian sample was getting medical supplies or equipment (29.4%), and the least frequently cited worry was getting enough food (11.8%). In the Black or African American patient sample, the most frequently cited worries were getting medical supplies or equipment, getting enough food, and getting transportation to appointments (16.7%); the least frequently cited worries were understanding medications and getting home health care services (13%). The most frequently cited worry by patients in the Hispanic or Latino sample was getting transportation to appointments (21.6%), and the least frequently cited worry was understanding medications (9.8%). Statistical testing by sample group revealed that Asian patients were more likely to be worried about getting home health care services (27.5%) than patients in the cross-section sample (2.9%) (P < .05). There were no statistical differences by sample group in the other patient-reported worries.


This survey, conducted before the COVID-19 pandemic, revealed system challenges and failures in providing care to diverse patients at time of discharge and beyond. The data gathered here provide a more rich and nuanced view of patient perspectives than some standard quality improvement surveys (eg, the HCAHPS). However, since we reported these data internally, and particularly since the COVID-19 pandemic, our system has begun to implement new workflows and assets to address lack of interpreters, failure to provide appropriate translated materials, lack of translated materials in the patient portal, and barriers to technology and internet access. The scientific literature makes clear that these barriers exist, and must be addressed, in many health systems.6,7 Our significant findings emphasize the importance of examining nuanced patient concerns beyond traditional quality improvement measures. Doing so can enhance and expand system knowledge related to the experience of diverse patient populations. These findings invite an implementation science approach that can better address unmet patient needs and improve the patient experience. Future research should continue to assess whether lack of interpreters at discharge instructions, lack of translated materials, and barriers to technology are associated with patients’ lack of understanding of the purpose of taking medications at discharge, care plans, and important follow-up care arrangements.8

These data underscore the challenges faced by diverse patients, specifically those with LEP and especially when they are unaccompanied by family members.9 Further, these data reveal the particular difficulties faced by patients who are experiencing a digital divide. These include a lack of access to technology for patients and families for needed posthospital telehealth visits and the challenges of real-time communication. Since collecting these data in fall 2019, we have witnessed the impact of barriers to access, particularly when family members are in isolation from one another due to the COVID-19 pandemic.

Interventions to better support this population could include a more widely available bilingual clinical and community health workforce who can provide clinical, cultural, and linguistic competence10,11; improved technology to distribute translated materials through the patient portal or on paper at point of care; and use of virtual video technology to improve rates of family involvement and interpretation at discharge. These same staffing tools could be useful in addressing common needs including food, medication access, home health care services, and transportation directly in communities where they arise. We considered, if interpreters could not be at discharge, whether scheduling a postdischarge call from the hospitals could serve to answer patient questions. Technology access may be a barrier to provision of interpreter services and access to the patient portal outside of the hospital. Lack of technology and language proficiency may hinder systemwide efforts to provide accessible ways for patients to engage in their own care. Health systems aiming to encourage adoption of patient portals may reduce communication disparities by making the portals available in different languages and by providing patient education addressing technological and health literacy.


Care transitions may be especially complex for diverse patients, including those with LEP or other communication barriers.Health care systems should continually follow up with patients to ascertain and meet needs and limit preventable complications and readmissions. These communications must be sensitive to the presence or absence of family or community support, technology access, and availability of medical interpreters so that major barriers to postacute care may be overcome.


The authors gratefully acknowledge the support of the Edward P. Lawrence Center for Quality and Safety at Massachusetts General Hospital (MGH) as well as the members of the Equity in Care Transitions Committee. They also would like to acknowledge Christopher Kirwan, Anna Pandolfo, and Melissa Calverley from MGH Interpreter and Translation Services, whose support helped make this multilingual effort possible. The complete phone interviewer team included Jonah Estabrook (English), Surie Johnson (English), Oluwagbotemi Olafunmiloye (English), Shafi Rubbani (English), Frederick Chin (English, Chinese), Chantha Long (Khmer), Trang Nguyen (Vietnamese), Paulo Chaves (Portuguese), Flavia Smith (Portuguese), Edaniel Beauplant (Haitian Creole), and Beatriz Oropeza Urquhart (Spanish). The authors received additional staff support from Kim Phan.

Author Affiliations: MGH Equity and Community Health (EAB, JRB), Division of General Internal Medicine (JC), and Edward P. Lawrence Center for Quality and Safety (AT), Massachusetts General Hospital, Boston, MA; Harvard Medical School (EAB, JC, JRB, KD), Boston, MA; Health Policy Research Center (SG, CM, KD) and Disparities Solutions Center (AT-M), Mongan Institute, Massachusetts General Hospital, Boston, MA.

Source of Funding: This project was funded by the Edward P. Lawrence Center for Quality and Safety at Massachusetts General Hospital.

Author Disclosures: Dr Donelan has received payment for completion of this study from the institutional Center for Quality and Safety. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (KD); acquisition of data (EAB, SG, KD); analysis and interpretation of data (EAB, CM, KD); drafting of the manuscript (EAB, SG, JRB, AT-M, KD); critical revision of the manuscript for important intellectual content (EAB, SG, CM, JC, JRB, AT, AT-M, KD); statistical analysis (EAB, CM); obtaining funding (AT); administrative, technical, or logistic support (AT, AT-M, KD); developing key survey measures (JC); providing expertise on approaches to eliminating racial and ethnic disparities in health care (JRB, AT-M); providing expertise on improving the coordination and continuity of care transitions (JC, AT, KD); and supervision (EAB).

Address Correspondence to: Esteban A. Barreto, PhD, MGH Equity and Community Health, Massachusetts General Hospital, 55 Fruit St, BUL 127, Boston, MA 02114. Email:


1. Ngai KM, Grudzen CR, Lee R, Tong VY, Richardson LD, Fernandez A. The association between limited English proficiency and unplanned emergency department revisit within 72 hours. Ann Emerg Med. 2016;68(2):213-221. doi:10.1016/j.annemergmed.2016.02.042

2. Boyle J, Speroff T, Worley K, et al. Low health literacy is associated with increased transitional care needs in hospitalized patients. J Hosp Med. 2017;12(11):918-924. doi:10.12788/jhm.2841

3. Semere W, Nápoles AM, Gregorich S, Livaudais-Toman J, Karliner L. Caregiving for older adults with limited English proficiency: transitioning from hospital to home. J Gen Intern Med. 2019;34(9):1744-1750. doi:10.1007/s11606-019-05119-y

4. Carter J, Ward C, Thorndike A, Donelan K, Wexler D. Social factors and patient perceptions associated with preventable hospital readmissions. J Patient Exp. 2020;7(1):19-26. doi:10.1177/2374373518825143

5. Nalichowski R, Keogh D, Chueh HC, Murphy SN. Calculating the benefits of a Research Patient Data Repository. AMIA Annu Symp Proc. 2006;2006:1044.

6. López L, Rodriguez F, Huerta D, Soukup J, Hicks L. Use of interpreters by physicians for hospitalized limited English proficient patients and its impact on patient outcomes. J Gen Intern Med. 2015;30(6):783-789. doi:10.1007/s11606-015-3213-x

7. Tieu L, Schillinger D, Sarkar U, et al. Online patient websites for electronic health record access among vulnerable populations: portals to nowhere? J Am Med Inform Assoc. 2017;24(e1):e47-e54. doi:10.1093/jamia/ocw098

8. Velasquez D, Uppal N, Perez N. Equitable access to health information for non-English speakers amidst the novel coronavirus pandemic. Health Affairs. April 2, 2020. Accessed August 12, 2020.

9. Garg S, Kim L, Whitaker M, et al. Hospitalization rates and characteristics of patients hospitalized with laboratory-confirmed coronavirus disease 2019 – COVID-NET, 14 states, March 1-30, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(15):458-464. doi:10.15585/mmwr.mm6915e3

10. Carter J, Walton A, Donelan K, Thorndike A. Implementing community health worker–patient pairings at the time of hospital discharge: a randomized control trial. Contemp Clin Trials. 2018;74:32-37. doi:10.1016/j.cct.2018.09.013

11. ¿Habla Español? Mass General creates Spanish-language care group to assist COVID care teams. News release. Massachusetts General Hospital; May 15, 2020. Accessed August 12, 2020.

Related Videos
Related Content
CH LogoCenter for Biosimilars Logo