Readability Formulas and User Perceptions of Electronic Health Records Difficulty: A Corpus Study

Area of Science:

• Health Informatics
• Patient Engagement
• Medical Communication

Background:

• Electronic health records (EHRs) offer potential for patient engagement and activation.
• Patient access to EHR notes can improve understanding of conditions and treatments.
• Technical language in EHRs often hinders patient comprehension.

Purpose of the Study:

• To examine the correlation between standard readability formulas and laypeople's perceived difficulty of general health information and EHR notes.
• To validate the generalizability of readability formulas on complex technical documents.

Main Methods:

• Collected 140 Wikipedia articles and 242 EHR notes on diabetes.
• Recruited 15 Amazon Mechanical Turk users to rate document difficulty.
• Measured correlations between perceived difficulty and readability formula scores.

Main Results:

• Readability formula scores and laypeople's perceptions differed significantly between text genres.
• Correlations between readability predictions and perceived difficulty were weak.
• Documents with similar readability scores were perceived as having different difficulty levels.

Conclusions:

• Readability formulas' predictions did not align with perceived difficulty for EHR notes or general health information.
• Widely used readability formulas are not appropriate for assessing the readability of EHR notes.
• Relying on current readability formulas may not effectively improve patient understanding of their health information.