Retina Society 2026: Inside the Mary Tyler Moore diabetic retinal disease PRO measure

Visual acuity (VA) has long served as the primary measure of vision loss in diabetic retinal disease (DRD), but it captures only part of what patients experience. K. Thiran Jayasundera, MD, MS, FACS, Natalie Fosse Endowed Chair in Vision Science Research and Chair of the Department of Ophthalmology & Vision Science at the UC Davis School of Medicine, discusses the development of the Mary Tyler Moore Diabetic Retinal Disease-Specific Patient-Reported Outcome (PRO) Measure. The measure—built from patient interviews and designed to capture the functional and psychological effects of DRD that existing instruments miss—was presented by Jayasundera at the 59th Annual Scientific Meeting of The Retina Society, September 23-26, 2026, in Los Angeles, California.

In this Q&A conversation with Modern Retina, Jayasundera shares the story behind the measure’s development, from the qualitative work that shaped its domains to what comes next for validation and adoption.

Note: Transcript edited lightly for clarity and length.

MR: What drove the need for a diabetic retinal disease-specific PRO measure, and what were existing generic or vision-related PRO tools missing for this patient population?

K. Thiran Jayasundera, MD, MS, FACS: Although reduced VA is the most common recognized manifestation in patients with diabetic macular edema (DME), vitreous hemorrhage, and tractional retinal detachments, patients with DRD frequently experience limitations of functional vision arising from visual pathways governing contrast sensitivity, peripheral vision, mesopic and scotopic function, light and dark adaptation, and photosensitivity which go unrecognized by clinicians. Existing PRO instruments used in DRD include the Retinopathy-Dependent Quality of Life (RetDQoL), RetCAT (Diabetic Retinopathy Computerized Adaptive Testing), and NEI VFQ-25 (National Eye Institute Visual Function Questionnaire). However, each of these PROs have distinct limitations. Most focus on general quality of life or visual function but tend to provide limited information about specific visual pathways being affected, psychological challenges experienced by our patients, or the burden of treatment.

MR: Please walk us through how the Mary Tyler Moore measure was developed. Was this built from patient interviews and focus groups, and what domains or themes emerged as most important to patients living with diabetic retinal disease?

Jayasundera: The following was the step-by-step process of developing the MTM DRD Questionnaire:

Four expert panels comprising 15 attendees including general ophthalmologists, retina specialists, low-vision rehabilitation specialists, ophthalmic technicians, psychiatrists, endocrinologists, patient advocates, and patients with diabetes discussed themes pertaining to DRD. Meetings were held on Zoom and were guided by a trained study team member, utilizing discussion topics selected through literature review and expert opinion. The meetings were recorded, transcribed, and analyzed using qualitative analysis software (Atlas.ti, version 9.5.4) to extract common themes and occurrences. This analysis combined with further literature review was used to create an in-depth interview (IDI) guide, which contained open-ended questions designed to probe patients to further discuss aspects of their experience living with DRD. Discussion guide topics included functional vision effects of DRD, psychological impacts of DRD and diabetes, impact of vision on daily activities, and the burden of both diabetes and vision treatments. The creation of the discussion guide was an iterative process, allowing for the addition of further questions throughout the interview process if new themes emerged.

One hundred fourteen total participants were interviewed in 2 phases: in-depth interviews (n = 91) and cognitive interviews (n = 23). A subset of in-depth interview participants (n = 15) also participated in patient focus groups.

• In-Depth Patient Interviews
  Patients were interviewed using the open-ended IDI discussion guide. Interviews were conducted by a trained study team member and took place over the phone or through video Zoom. IDIs were recorded and transcribed for subsequent qualitative analysis.
• Patient Focus Groups
  Four patient focus groups were conducted over Zoom and recorded, with each consisting of 3-4 participants who had previously completed an IDI. 15 total participants completed focus group interviews. Focus groups were moderated by a trained study team member and utilized the same discussion guide as IDIs. Groups were intended to be open-ended, focusing on the topics most relevant to participants with the intent of ensuring content saturation of previous IDIs. In addition, the group setting was desired to facilitate further discussion and discover topics that potentially went unrealized in an individual setting.
• Qualitative Data Analysis
  Transcripts were uploaded into Atlas.ti qualitative analysis software and separated into quotations according to the topics discussed. Quotations were further coded and analyzed using grounded theory of selective coding. A code dictionary was created to organize codes and quotations. Study team members consistently met to review the code dictionary, ensuring accuracy and that all codes were unique. Individual interview transcripts were reviewed by five independent coders. IDIs continued until content saturation of coded interviews, which was determined by the emergence of no new codes over several consecutive interviews.

Items to be utilized in a draft PRO were generated by reviewing common themes discovered from qualitative coding analysis. This process was supplemented by reviewing the previously created database of previously validated PROs, adapting relevant items. PRO draft items were grouped into template domains to help organize content and ensure all relevant information from coding was extracted. The language of items was designed to reflect an 8^th grade literacy level, and the exact wording utilized by participants in interviews.

Cognitive interviews were conducted to assess participant understanding of the draft PRO items. Participants were asked to provide open-ended feedback on the wording, clarity, quality, order, response burden, and overall changes they would suggest to the PRO. Interviews were recorded and potential changes were documented. The draft PRO was refined based on interviewee suggestions and finalized for subsequent mass administration and validation.

A parent PRO with 424 unique items was created based on recurrent themes from qualitative analysis.

The parent PRO was administered to 264 participants with 31 participants completing retest administrations. Participants represented the full disease spectrum of DRD, including diabetes without retinopathy, mild, moderate, and severe non-proliferative diabetic retinopathy (DR), proliferative DR, and DME.

Factor analysis and Item Response Theory (utilizing a Graded Response Model) was used to create the final short form PRO.

The functional vision domains created are Central Vision Photopic, Central Vision Mesopic, Central Vision Scotopic, Contrast Sensitivity, Peripheral Vision Photopic, Peripheral Vision Mesopic, Peripheral Vision Scotopic, Light Adaptation, Dark Adaptation, Scotopic Vision, Photosensitivity, Depth Perception, Diurnal Vision Variability, Driving Central Vision Mesopic, and Driving Peripheral Vision Scotopic. These 15 domains collectively encompass the 83 questionnaire items and assess a broad range of visual functions across different lighting conditions, visual fields, adaptation states, driving-related tasks, and daily vision experiences.

Nine domains comprising 51 items addressing psychological or psychosocial aspects of DRD were; Intravitreal Injection Anxiety; Intravitreal Injection Tolerability; Panretinal Photocoagulation Laser Anxiety; Panretinal Photocoagulation Laser Tolerability; Vision-Related Negative Thinking and Rumination; Vision-Related Loneliness; Vision-Related Social Anxiety; Diabetes Burnout; and Diet Attitudes and Emotions.

The marginal reliability for each finalized domain was ≥0.80.

MR: How does this PRO measure account for the full spectrum of diabetic retinal disease, from early non-proliferative changes to more advanced vision-threatening complications?

Jayasundera: Every phase of PRO construction and validation included patients with diabetes without retinopathy, mild, moderate, and severe non-proliferative DR, proliferative DR, and DME.

MR: What role do you see this measure playing going forward? Is the goal to use it in clinical trials as a companion to anatomic and visual acuity endpoints, or also to bring it into routine clinical care?

Jayasundera: This PRO is intended to serve as a standardized outcome measure for future DRD clinical trials as well as patient-centered care. With a holistic and integrated approach, this PRO aims to recognize the interconnectedness of visual, physical, and psychological health, and inform the development of more effective treatments and better health outcomes.

MR: Now that the measure has been developed, what is next in terms of validation? How far is this from being ready for broader adoption?

Jayasundera: This PRO has utilized the framework recommended by the FDA for PRO development. It is available for incorporation in both a clinic setting as well as clinical trials.

The questionnaire is copyrighted material of the University of California Davis and University of Michigan. All rights reserved.

For licensing information contact the University of California Davis Technology Transfer Office at [email protected]

K. Thiran Jayasundera, MD, MS, FACS

E: [email protected]

Jayasundera is the Natalie Fosse Endowed Chair in Vision Science Research and Chair of the Department of Ophthalmology & Vision Science at the UC Davis School of Medicine in Sacramento, California.