Key Takeaways

• An analysis of more than 4.6 million ophthalmic clinical letters found that 95.3% were less readable than recommended for patient-facing information
• Readability worsened between 2013 and 2025, with Flesch Reading Ease scores declining by an average of 0.64 points annually
• Only 2.25% of letters were classified as being addressed directly to patients, highlighting an opportunity to make ophthalmic correspondence more accessible and patient-centered

Clinical letters sent following ophthalmology appointments may be considerably more difficult for patients to understand than recommended, according to a large-scale analysis of more than 4.6 million letters.¹

Investigators found that 95.3% of ophthalmic clinical letters were less readable than recommended for patient-facing materials. Moreover, readability declined over the 12-year study period, suggesting that clinical correspondence has become increasingly complex despite efforts to promote patient-centered communication.

The findings were published in BMJ Health & Care Informatics by Ariel Yuhan Ong and colleagues from University College London Institute of Ophthalmology, Moorfields Eye Hospital NHS Foundation Trust, Oxford Eye Hospital, and collaborating institutions.

Effective communication between clinicians and patients can affect engagement with care, treatment adherence, and clinical outcomes. However, the investigators noted that the extent to which routine clinical correspondence meets recommended readability standards has not been well quantified. To examine the issue at scale, researchers analyzed more than 4.6 million outpatient clinic letters written for 804,986 patients at Moorfields Eye Hospital NHS Foundation Trust in the United Kingdom between 2013 and 2025. The letters represented 17 ophthalmic subspecialty services.

The researchers assessed the correspondence using established measures of readability, including Flesch Reading Ease (FRE), the SMOG Index, and Automated Readability Index. They also evaluated more complex linguistic characteristics, including lexical density and type-token ratio. FRE served as the study's primary outcome. The investigators additionally examined differences among ophthalmic subspecialties and changes in readability over time. Rule-based measures were used to determine whether letters were addressed directly to patients.

Across the study population, the median FRE score was 51.1, with an interquartile range of 41.9 to 58.5. The researchers characterized this as approximately a college-level reading difficulty and substantially more complex than recommended for patient-facing health information.

Overall, 95.3% of letters fell outside recommended readability levels. Similar patterns emerged with the other readability measures included in the analysis. Readability also differed significantly among ophthalmic subspecialties (P<.001), whereas patient sociodemographic characteristics such as age and ethnicity showed limited associations with readability.

The findings indicate that communication complexity was not confined to an individual ophthalmic service or particular patient population but represented a broader pattern across clinical correspondence.

Rather than improving during the 12-year period, the readability of ophthalmic letters declined. FRE scores decreased by an average of 0.64 points per year (P<.001), indicating progressively more difficult text. The rate of change varied among subspecialties. Neuro-ophthalmology showed one of the largest declines, at 1.24 FRE points per year (P<.001), whereas some services, including optometry, demonstrated no statistically significant change.

The researchers also examined the extent to which correspondence appeared to speak directly to patients. Use of second-person terminology was uncommon, with a median of 0 second-person terms per 100 words and an interquartile range of 0 to 1.1. Only 2.25% of letters overall were classified as being addressed directly to patients. There was, however, evidence of gradual change. The likelihood that a letter was addressed directly to a patient increased annually, although the overall proportion remained low.

The authors suggested that their findings could provide an empirical foundation for local quality-improvement efforts aimed at making correspondence easier for patients to understand. The analytical framework could also be applied beyond ophthalmology to evaluate clinical communication in other specialties.

Reference

1. Ong AY, Kiire CA, Wagner SK, Nguyen Q, Thomas PBM, Keane PA. How well do we write for patients? Longitudinal analysis of the readability and complexity of 4.6 million ophthalmic clinical letters. BMJ Health Care Inform. 2026;33(1):e102086. doi:10.1136/bmjhci-2026-102086