Home care patients with language barriers at higher risk for rehospitalization

Speaking a language other than English raises the chance of hospital readmission among U.S. patients

Patients with language barriers who are receiving home health care are at an increased risk for hospital readmission, according to a new study published in the International Journal of Nursing Studies.

“Language preference as a social determinant of health is not a new factor in health care delivery,” said Allison Squires, PhD, RN, FAAN, associate professor at NYU Rory Meyers College of Nursing and the study’s lead author. “Our research suggests that patients with language barriers in home health care may be particularly vulnerable to poor outcomes.”  

Language barriers between patients and providers can make effective communication challenging and have been shown to have a negative impact on the quality of care and patient safety. Research shows that patients who are most comfortable speaking a language other than the dominant local language are at an increased risk for longer hospital stays, falls, and infections, among other negative outcomes.  

Language gaps may be particularly detrimental in home health care—in which health providers, primarily nurses, visit patients’ homes to deliver care—given that interpreter services can be more difficult to implement and access in the home setting. For example, many home care patients, who are often older adults over the age of 65, do not have high-speed internet access, or sometimes even a phone, which makes accessing phone or video interpreter services difficult.

A prior study by Squires found that home health care services are not meeting the demand for language services among diverse populations; that study of one large urban agency found only 20 percent of patients speaking a language other than English were visited by a provider who spoke their language. For smaller agencies, the numbers are likely far worse.

The goal of home health care services is to maintain a person’s ability to care for themselves in their home and avoid costly emergency or hospital services. As a result, hospital readmission from home health care is a key quality indicator for U.S. home health agencies.  

To determine whether a patient’s language preference influences hospital readmission risk from home health care, the researchers examined electronic health records for more than 87,000 patients living in New York City who were receiving home health care after being discharged from a hospital. The records captured what language patients preferred to communicate in; the researchers focused the study on the five most common languages spoken among the patients studied: English, Spanish, Russian, Chinese, and Korean. They then looked at who was readmitted to a hospital within 30 days.  

Home health patients who preferred a language other than English had a higher hospital readmission rate (20.4%) than English-speaking patients (18.5%). Risk of readmission also varied by language, with higher risk among Spanish and Russian speakers and lower risk among Chinese and Korean speakers.

The researchers recommend using a variety of approaches to address this disparity, including team-based care transition programs from hospital to home health care that account for patient language preferences, improving translation capabilities, and employing health care teams that speak the same languages as their patients.  

“Ensuring that patients have clinical interactions in their preferred language across the care continuum is important, as it can increase access to care at the earliest stages, reduce the risk for readmission, and improve care transitions throughout the health care system,” said Squires.

In addition to Squires, study authors include Chenjuan Ma of NYU Meyers, Sarah Miner of St. John Fisher College, Penny Feldman of Visiting Nurse Service of New York, Elizabeth Jacobs of MaineHealth, and Simon Jones of NYU Grossman School of Medicine. The research was funded by the Agency for Healthcare Research and Quality (R01-HS023593).