Table 4. Multivariable logistic regression model predicting development of proliferative diabetic retinopathy or related complications among adult patients with diabetic retinopathy.

Variable	Adjusted Odds Ratio (95% Confidence Interval)	p-value
Number of eye doctor visits in the past 12 months	1.32 (1.16–1.49)	<0.001
Diagnosed with kidney disorder due to diabetes mellitus (those with diagnosis vs. those without diagnosis of kidney disorder due to diabetes)	1.94 (1.24–3.01)	0.003
Delayed care due to differences in race, religion, or language between patient and provider (“yes” responses vs. “no” responses)	1.20 (1.02–1.41)	0.03
Delayed care due to having to provide care for another adult (“yes” responses vs. “no” responses)	3.14 (1.01–9.50)	0.04
Skipped medication in order to save money (“yes” responses vs. “no” responses)	1.69 (0.97–2.89)	0.06
Delayed care due to inadequate healthcare coverage (“yes” responses vs. “no” responses)	1.35 (0.97–1.95)	0.09
Age (in years)	0.99 (0.97–1.00)	0.09