Predictors of narrow angle detection rate—a longitudinal study of Massachusetts residents over 1.7 million person years

Abstract

Abstract Background/Objectives To determine the predictors of narrow angle detection in a United States population-based cohort. Materials and methods This was a retrospective cohort study using the Massachusetts All-Payer Claims Database. Demographic information of all patients and eye care provider information during the years 2011–2015 were extracted from Massachusetts All Payers Claims Data. All payers who received eye care during 1/1/2012–12/31/2015 without any previous eye visit during 2011 were included in the analyses. Laser peripheral iridotomy was identified by Current Procedural Terminology code 66761. Narrow angle detection was defined as the diagnosis of narrow angles by diagnosis code followed by a laser peripheral iridotomy procedure. Different predictors of narrow angle detection were evaluated using Kaplan–Meier curves with the log rank and Cox regression modeling. Results A total of 1,082,144 patients were included. The hazard ratio of narrow angle detection increased with age compared to the reference group of 0–10 years: 21–30 years of age (hazard ratio = 4.5), 31–40 (10.5), 41–50 (27.9), 51–60 (46.1), 61–70 (68.4), 71–80 (56.8) (all p < 0.0002), was 1.47 times higher in women and 1.85 times higher if evaluated by ophthalmologists compared to optometrists, after controlling for provider × time interaction. Conclusion Older age and female sex are associated with narrow angles. The rate of narrow angle detection was significantly higher if patients are seen by ophthalmologists compared to optometrists only. Evaluation with an ophthalmologist may be important for patients at high risks for developing primary angle closure glaucoma.