Abstract
Purpose :
To compare the 12-month outcomes of individual practitioners’ patients with neovascular age-related macular degeneration and investigate the difference in treatment regimens of outliers and non-outliers.
Methods :
A retrospective multi-centre cohort study of data from the Fight Retinal Blindness! (FRB!) registry. Treatment-naïve eyes with nAMD that started VEGF inhibitors between January 2017–November 2022 in Australia were eligible. A funnel plot on the mean change in visual acuity [VA] (adjusted by age, sex, baseline VA and initial drug) 12 months after the first treatment classified the practitioners into better, normal and worse performing groups. Treatment regimens were compared between the three groups using linear models.
Results :
A total of 1748 eyes (male, 36%; mean age, 81 years old) from 46 practitioners were included. Funnel plot analysis identified 2, 43 and 1 practitioners[MG1] in the better, normal and worse performance groups (change in VA, 7.2, 5.0 and 3.6 letters, respectively). The 2 [MG2] better performing administered significantly more injections over the 12 months than the normal group (8.8 vs. 8.3; P = 0.004) and the worse practioner[MG3] administered significantly fewer injections than the normal group (7.7 vs. 8.3; P = 0.004). The last interval of the injections was similar between the better and normal performance groups (9.7 vs. 10.1 weeks; P = 0.27), whereas the worse practitioner had a significantly longer interval than the normal group (11.4 vs. 10.1 weeks; P =0.015). The proportion of visits when the CNV was active was similar between the better and normal groups (48% vs. 49%; P= 0.56), while the worse practitioner’s patients had a significantly higher proportion of CNV-active visits than the normal group (70% vs. 49%; P < 0.001).
Conclusions :
We found evidence that underperforming practitioners gave fewer injections for nAMD. Anonymously informing underperforming practitioners of their outcomes would help to improve the quality of care as a whole.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.