Individuals with geographic atrophy (GA) are at greater risk for fall and fracture events compared to individuals without GA: a US claims data analysis

Abstract

Purpose : There is limited evidence on the risk of falls/fractures in GA patients. We assessed the risk of falls/fractures in individuals with GA compared to a matched cohort of individuals without GA.

Methods : This retrospective study analyzed 3 US claims datasets (Optum Clinformatics, MarketScan Commercial and Medicare, and IQVIA US Pharmetrics). Patients with GA were defined as having ≥1 GA ICD-10 code (H35.31x3, H35.31x4) since Oct 2015. Patients with GA or wet age-related macular degeneration codes before Oct 2015 were excluded. GA patients were matched 1:2 exactly to patients without GA on age, sex, and year of diagnosis; then, patients were propensity score (PS) matched on disease history, including conditions in the Charlson Comorbidity index, other eye conditions, presence of osteopenia and osteoporosis, and certain mental conditions. Patient follow up was capped at 4 years to minimize survival bias. Relative risk (RR) and 95% CIs were calculated for all GA patients then stratified by eye and subfoveal involvement. Estimates were calculated separately for each dataset, then pooled.

Results : A total of 81,489 GA patients were evaluated after PS matching across all databases. Across datasets, the mean (SD) age of GA patients ranged from 75.9 (8.3) to 80.5 (7.2) years, and the proportion of females ranged from 61.1% to 64.0%. Most (35.6%) had bilateral GA without subfoveal involvement; 23% bilateral, subfoveal GA; 22.7% unilateral GA without subfoveal involvement; 16% unilateral, subfoveal GA; and 2.6% bilateral GA with and without subfoveal involvement. Among all GA patients, falls (overall RR: 1.32 [95%CI: 1.28-1.35]) and fractures (overall RR: 1.27 [1.24-1.29]) were higher in the 4 years following the index date, compared to matched controls; this was consistent among the GA subgroups. GA patients with bilateral and subfoveal involvement had the most pronounced difference in falls (overall RR: 1.47 [1.40-1.54]) and fractures (overall RR: 1.42 [1.37-1.48]) compared to controls.

Conclusions : GA patients were at higher risk of falls/fractures than individuals without GA, indicating an additional burden of illness. This burden was most pronounced for the GA cohort with bilateral GA and subfoveal involvement. Given the large number of GA patients and high costs to manage falls/fractures, our analysis indicates large impact of GA on public health.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.