Diabetes patients with low literacy more likely to have poorly controlled disease

Diabetes patients with low literacy are nearly twice as likely as patients with higher literacy to have poorly-controlled blood sugar and serious long-term diabetes complications, according to UCSF researchers.

This study, which appears in the July 24 issue of the Journal of the American Medical Association (JAMA), is the first to demonstrate that lower literacy is associated with worse health outcomes for patients with a chronic medical condition—even after accounting for patients’ educational attainment, age, race, insurance, and diabetes drug regimen.

“Having diabetes and difficulty reading creates a double bind. Diabetes patients rely on a number of tools to manage their disease and prevent serious health problems. For patients with low literacy, it’s as though they have received the toolkit, but not the operating instructions,” said Dean Schillinger, MD, UCSF assistant professor of medicine at San Francisco General Hospital Medical Center (SFGHMC) and lead author of the study.

“Because our health system expects patients to be able to read at a very high level, we may be leaving a lot of patients in the dark.”

The diabetes care regimen is one of the most challenging of any for chronic illness, he explained. Patients often must monitor their own blood sugar, manage multiple medications, visit many providers, maintain foot hygiene, adhere to diet and meal plans, and engage in exercise. Patients also have to be able to identify when they are having problems with these functions, problem-solve to prevent crises, and advocate for themselves in what is often an overwhelming health care system. For this reason, diabetes outcomes may be especially sensitive to problems in communication, empowerment, and self-management, Schillinger said.

“When a patient doesn’t speak English, the communication problems are obvious. But even if a patient speaks the same language as the health care provider, literacy is often a hidden barrier. We’re just opening our eyes to this and discovering that we need to partner with patients to figure out how to communicate more effectively to lower systemic barriers.”

The researchers assessed patients’ “functional health literacy,” a measure of a patient’s ability to perform basic health-related reading and numerical tasks. Those with poor health literacy had difficulties that ranged from reading labels on pill bottles and interpreting blood sugar values or dosing schedules, to comprehending appointment slips, educational brochures and websites, or informed-consent documents. Prior research has demonstrated that poor health literacy is common in health care settings, affecting one in three Medicare patients and one in two patients at public hospitals like SFGHMC.

The study enrolled 408 patients with adult-onset diabetes, one of the most common diseases in the United States, affecting more than 16 million people and 18% of all people 65 years of age and older.

Patients with dementia or poor vision were excluded. Patients with inadequate health literacy were more likely than those with adequate literacy to be older, female, non-white, having received only high school education or less and having Medicare coverage.

Thirty-six percent of patients with inadequate health literacy, as opposed to 19% of patients with adequate health literacy, had diabetic retinopathy, an eye condition that can be a precursor to blindness. While prior studies have shown that optimal blood sugar control can prevent retinopathy, only 20% of patients with inadequate health literacy had optimal blood sugar control, as opposed to 33% of those with better health literacy. The extent of the associations between health literacy and other diabetes complications - such as kidney disease, amputation, and heart disease - was similar to that of eye disease. This study did not allow researchers to ascertain whether inadequate health literacy was causally associated with poor diabetes outcomes.

Co-authors on this study include: Kevin Grumbach, MD, vice chair and professor, UCSF department of family and community medicine; John Piette PhD, associate professor of internal and general medicine, University of Michigan School of Medicine; Dennis Osmond, PhD, UCSF associate professor of epidemiology; Francis Wang, MS, senior biostatistician, UCSF Primary Care Research Center; and Andrew Bindman, MD, UCSF professor of medicine and epidemiology and chief, division of general internal medicine at SFGHMC.  Carolyn Daher, MPH, formerly affiliated with the UCSF Primary Care Research Center; Gabriela Diaz Sullivan, MD, UCSF resident; and Jorge Palacios, MA, research associate in the UCSF Primary Care Research Center,  also contributed to the study.

The UCSF Primary Care Research Center seeks to improve access and quality of care for vulnerable populations. The study was funded by the UCSF Hellman Family Early Career Research Award; the Pfizer Health Literacy Research Award; the National Center for Research Resources (NCRR), and The Agency for Healthcare Research and Quality (AHRQ).

\###