Abstract
Purpose :
Individuals are often lost to follow-up after initiation of treatment with anti-vascular endothelial growth factor (VEGF) injections for Diabetic Retinopathy (DR). The purpose of this study is to identify whether demographic and social factors within the electronic medical record (EMR) or common indices of social determinants of health are predictive of being lost to follow-up (LTFU).
Methods :
This study includes individuals with diabetic retinopathy who received anti-VEGF injections from an urban, academic ophthalmology center that were seen between May 2013 and March 2023. We abstracted demographic characteristics from the EMR at Yale Eye Center (YEC) and Dana Clinic through the Joint Data Analytics Team (JDAT). We evaluated age, sex, race, ethnicity, smoking status, alcohol use, preferred language, insurance status, and distance from YEC and Dana clinic. We then utilized individual addresses to obtain Area Deprivation (ADI) and Distressed Communities (DCI) Indices via online tools for each individual.
We compared LTFU DR individuals who did not return within 6 months after anti-VEGF treatment to returners and performed univariate followed by multivariate logistic regression using R. We then compared mean ADI national and state percentiles and DCI values between the two groups with Welch’s t-test.
Results :
A total of 335 individuals were included in this study. Sixty-six of those individuals were LTFU while 269 returned to clinic. Both groups were similar in age, racial breakdown, and insurance coverage. Univariate logistic regression of demographic factors and socioeconomic indices revealed no significant predictors of being LTFU. After multivariate logistic regression, it was found that English being the preferred language of a individual increased the odds of being LTFU (OR= 2.60,p = 0.033). Other factors including race, ethnicity, ADI, and DCI, did not significantly increase the odds of being lost to follow-up in our individuals.
Conclusions :
Our results show that individuals who prefer English have higher odds of being lost to follow-up. This is likely just one of many contributing factors, and highlights the need to evaluate more demographic and social factors when considering ophthalmic care. We are now utilizing qualitative interviews of some of these individuals lost to follow-up to better learn from their experiences.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.