Informatics Grand Rounds with Dr. Cindy Cai

Why It Matters

Integrating informatics with social‑determinant insights enables earlier detection and targeted outreach, directly reducing preventable blindness and advancing health‑equity in diabetic eye care.

Key Takeaways

• Social determinants significantly reduce eye‑care utilization among diabetics
• EHR lacks alerts for patients at risk of care lapses
• Multi‑dimensional interventions needed for different social risk groups
• Predictive models using EHR data can identify high‑risk patients
• Standardizing ophthalmology data to OMOP improves research equity

Summary

Johns Hopkins’ Grand Rounds featured Dr. Cindy Cai, an ophthalmologist‑researcher who uses biomedical informatics to tackle diabetic retinopathy, a leading cause of vision loss in working‑age adults. She outlined how gaps in routine eye‑care—often driven by social determinants of health (SDOH)—lead to preventable disease progression, and described her dual role bridging clinical practice with data‑science initiatives. Cai presented multiple analyses showing that adverse SDOH—such as financial insecurity, lack of transportation, and limited health‑insurance coverage—correlate with lower odds of annual eye‑exam attendance. National Health Interview Survey data revealed that 43% of adults with diabetes reported no eye‑care visit in the prior year, and Medicaid‑based studies identified three distinct risk clusters ranging from financially secure patients to those facing multiple socioeconomic barriers. She highlighted concrete examples: a latent‑class analysis that grouped patients into three social‑risk profiles, and the use of the Area Deprivation Index to predict higher rates of proliferative diabetic retinopathy in deprived neighborhoods. Cai also explained how the lack of EHR‑driven alerts hampers early intervention, prompting her team to develop predictive models that flag patients likely to lapse in care. The talk underscored the need for multi‑dimensional, equity‑focused interventions and for standardizing ophthalmology data to the OMOP common data model. By sharing predictive tools across institutions, clinicians can proactively reach high‑risk patients, reduce vision‑loss disparities, and accelerate large‑scale observational research.

Original Description

This session with Dr. Cindy Cai will discuss the role of social determinants of health and big data in dealing with vision loss from diabetic retinopathy. #diabeticretinopathy #johnshopkins #bioinformatics #healthinformatics https://bids.jhmi.edu/about/events-also-past-grand-rounds/

This session covers:

-Identifying the critical role that social determinants of health play in vision outcomes

-Identifying how common data models are critical for big data research

-Examining how SDoH and big data research can intersect to improve vision loss from Diabetic Retinopathy