Is Low Health Literacy a Significant Risk Factor for Hospital Readmission

Background:

Even though some readmissions are unavoidable, there is an implicit concern that they represent a failure in the health care system or inappropriate wishful thinking. Readmissions are not only expensive but suggest we have placed the patient at risk. In numerous studies, a variety of factors such as noncompliance, adverse drug reaction, and lack of access to care have been identified as major contributors to readmissions. We serve a predominantly Hispanic population that is underinsured and has less access to care. We speculated that low health literacy might contribute to these factors and was related to our readmission rate.

Methods:

This was a prospective cohort study. We assessed all of the following about patients discharged over a 1‐month period: demographic data, insurance data, prescription plan, and discharge planning, including follow‐up appointments and posthospitalization services. Patients' health literacy was evaluated by a modified 3‐question S‐TOFHLA tool. For those patients who were readmitted, we surveyed both the admitting physician and the patient about whether the readmission was related to deterioration of the previous conditions, a new medical condition, patient noncompliance, drug interaction or side effects, inadequate follow‐up appointment, or service.

Results:

Of the 220 patients discharged, 23 (10.5%) were readmitted in the following month. Demographically, Hispanics (91% vs. 64%) and women (57% vs. 44%) were more likely to be readmitted, but age was not a factor. Patients with Medicaid or Medicare were more likely to be readmitted, but there were no differences among those who were uninsured. Although some degree of impairment of health literacy based on responses to the 3 questions was found in 63%‐75% of our patients, severely impaired health literacy was found significantly more frequently in those patients who were readmitted (52%‐57% vs. 31%‐35%). The most common reason for readmission cited by both physicians and patients was deterioration of the previous condition (57%); noncompliance was cited by 18% and lack of access by 13%.

Conclusions:

In this small study we tried to determine modifiable elements associated with readmission. Our principle finding was that most patients who were readmitted had low health literacy. We suspect that low health literacy is associated with the perceived noncompliance of a patient and is likely related to deterioration of a patient's medical condition. It is unclear if these findings apply to other populations, but in other studies basic or below basic health literacy has been found in 36% of adults. Although studies to confirm and extend these findings are necessary, one approach to reducing readmission rates is to focus on assessing health literacy and improving communication with those with low health literacy.

Author Disclosure:

J. F. Hanley, None; C. Wade, None.