. 2021 Nov 16;4(11):e2134753. doi: 10.1001/jamanetworkopen.2021.34753

Table 2. Associations of Lower RNFL Thickness (per SD) With Incidence of Clinically Relevant Depressive Symptoms and Depressive Symptoms Over Time.

Regression coefficient^a	Crude	Model 1^b	Model 2^c
Incidence of clinically relevant depressive symptoms, HR (95% CI)	1.08 (0.99-1.19)	1.11 (1.01-1.22)^d	1.11 (1.01-1.23)^d
Depressive symptoms over time, RR (95% CI)	1.02 (1.00-1.05)	1.04 (1.01-1.06)^d	1.04 (1.01-1.06)^d

Abbreviations: HR, hazard ratio; RNFL, retinal nerve fiber layer thickness; RR, rate ratio.

^{^a}

Regression coefficients represent the HR for the incidence of clinically relevant depressive symptoms or RR for depressive symptoms over time per SD lower RNFL thickness, where a higher HR or RR indicates higher incidence of clinically relevant depressive symptoms or more depressive symptoms over time. One SD of RNFL thickness corresponds with 10.9 µm in the incidence of clinically relevant depressive symptoms study population and 11.0 µm in the depressive symptoms over time study population.

^{^b}

Model 1 includes age, sex, glucose metabolism status, and educational status.

^{^c}

Model 2 includes model 1 plus waist circumference, total cholesterol to high-density lipoprotein cholesterol ratio, use of lipid-modifying medication, office systolic blood pressure, use of antihypertensive medication, smoking, alcohol consumption, and partner status.

^{^d}

P < .05.