Table 4.

Results of the Age-Adjusted Proportional Hazards Regression of the Effect of Statin Use on AMD Progression Based on AMD Subtype: Hazard Ratios and 95% Confidence Limits for Matched Statin Use Propensity Scores.

	All participants N = 2512*		Bilateral Large Drusen N = 1546		Unilateral Late AMD N = 944
	Hazard Ratio	95% CI	Hazard Ratio	95% CI	Hazard Ratio	95% CI
Any late age-related macular degeneration
Without accounting for the competing risk of death	0.92	0.62–1.36	0.63	0.37–1.05	1.22	0.67–2.20
Accounting for the competing risk of death	0.81	0.55–1.20	0.53	0.31–0.89	1.12	0.62–2.01
Any geographic atrophy
Without accounting for the competing risk of death	1.07	0.63–1.82	0.74	0.38–1.43	1.86	0.76–4.56
Accounting for the competing risk of death	0.94	0.55–1.62	0.64	0.32–1.26	1.75	0.72–4.24
Central geographic atrophy
Without accounting for the competing risk of death	0.88	0.43–1.82	0.90	0.38–2.15	1.18	0.32–4.41
Accounting for the competing risk of death	0.76	0.37–1.56	0.76	0.32–1.83	1.14	0.32–4.09
Neovascular age-related macular degeneration
Without accounting for the competing risk of death	1.07	0.65–1.76	0.86	0.44–1.69	1.36	0.66–2.82
Accounting for the competing risk of death	0.92	0.57–1.52	0.73	0.38–1.42	1.25	0.06–2.59

CI = confidence interval.

^*includes 22 participants who were matched based on propensity score with one of the pair from the bilateral large drusen group and the other from the unilateral AMD group.