Readability of Patient Discharge Instructions

INTRODUCTION

Safe transitions from hospital to home require multicomponent strategies including patient engagement and education to mitigate preventable adverse events¹. Standard hospital discharge practice includes providing tailored discharge instructions to patients. The American Medical Association, Department of Health and Human Services, and National Institutes of Health recommend patient facing instructions be written at a sixth-grade level to ensure adequate comprehension by the average adult. Ensuring discharge instructions are written at the recommended level allows patients to receive information at their health literacy level and improve overall health status².

Prior literature demonstrates discharge instructions are written above the recommended level for pediatric, emergency department, and post-surgical care patients³. The readability of instructions for patients discharged from general internal medicine services at a public safety-net hospital has not been previously reported. We analyzed the reading level of discharge instructions provided to hospitalized adults discharged from a public safety-net hospital to understand the current state and guide opportunities to improve transitions of care.

METHODS

The study was conducted at a 284-bed safety-net hospital and level one trauma center. We retrospectively analyzed electronic medical record After Visit Summary instructions. We included patients 18 years or older discharged from the general internal medicine floor service between November 2019 and May 2020. We excluded instructions in languages other than English, those not written by a physician, and those of patients transferred to another medical service or skilled nursing facility. Patient discharge instructions were analyzed using the Flesch-Kincaid Grade Level (FKGL) and Reading Score (FKRS) via Microsoft Word (Seattle, WA). We obtained patient age, gender, primary language, length of stay, and 30-day readmission status. Statistical significance was defined as a P value of ≤.05. Analysis was performed using JMP version 9.0 (SAS Institute., NC).

RESULTS

Of 388 discharge instructions analyzed, 69.6% of patients identified as male; 26.9% as White, 25.3% as Hispanic, 19.1% as Black or African American, 13.4% as Asian, and 6.7% as mixed race. The average age of patients was 55.9 ± 17.1 years and average length of hospitalization 5.2 ± 7.3 days.

Mean FKGL and FKRS were 9.1 ± 2.0 and 57.1 ± 11.6, respectively. An FKRS score of 50–60 indicates 10–12^th grade readability. Only 11.3% (n= 44) of discharge instructions were written below the seventh-grade level (within recommended guidelines).

Despite instructions written by the provider in English, 12.4% of patients required a Spanish interpreter service, 2.7% a Cantonese interpreter, 1.3% a Tagalog interpreter, 1.1% a Mandarin interpreter, and 2.0% another language interpreter.

Seventy (18%) of the examined patients experienced readmission within 30 days. There was no correlation between FKGL and FKRS and 30-day readmission (FKRS, P=.8075: FKGL, P=.6960).

DISCUSSION

Health literacy is a predictor of health status. Patients with low health literacy are more likely to be hospitalized, use emergency services, have poor medication adherence, and experience higher mortality rates². Providing appropriate patient facing materials can attenuate consequences of poor health literacy⁴. The majority of patients hospitalized at our safety-net institution are underinsured or uninsured, portending lower health literacy⁵.

Despite recommended readability levels for patient-facing materials, our results show the vast majority (88.7%) of discharge instructions from our internal medicine services may be inaccessible to the patients we serve. This could be related to provider lack of awareness of recommended guidelines, challenges translating medical jargon into patient accessible language, and implicit assumptions around patient literacy.

As more of the electronic health record is patient-facing under the 21st Century Cures Act, additional research is needed on how patients understand medical documents written at variable reading levels, including those explicitly written for patient instruction and those written for clinician-to-clinician communication. Discharge documents composed at an appropriate readability level may help frame other accessible information for patients, such as notes and test results, leading to improved patient-clinician communication, self-management, and trust⁶.

Our study was conducted at a single institution, readability analyses may not be nationally representative. We did not examine clinician characteristics as a factor in readability levels. We did not examine morbidity or mortality, besides hospital readmission.

Future initiatives may include the dissemination of readability guidelines to clinicians, including examples of sixth-grade reading levels, advisory panels to review patient-facing documents, and integration of readability software into electronic medical records to provide real-time feedback on patient-facing materials.

Declarations

Conflict of Interest

Authors have no conflict of interests to report.