Skip to main content
. Author manuscript; available in PMC: 2017 Jun 5.
Published in final edited form as: Clin Surg. 2016 Dec 7;1:1236.

Table 2.

Logistic regression predicting odds of visual impairment using imputed datasetsa.

Covariate Value OR (95% CI) p-value
Gender Female 0.62 (0.33, 1.17) 0.10
Male REF
Homeless Homeless 1.71 (0.20, 14.40) 0.58
Not Homeless REF
Marital Status Never Married 1.41 (0.53, 3.76) 0.44
Previously Married 1.37 (0.60, 3.08) 0.41
Currently Married REF
Ethnicity Not White 0.68 (0.33, 1.37) 0.23
White REF
First Language Not English 1.38 (0.49, 3.93) 0.50
English REF
Citizenship Status Permanent Resident 0.61 (0.15, 2.57) 0.46
Visitor 0.94 (0.23, 3.78) 0.92
US Citizen REF
Employed Not Employed 3.05 (1.19, 7.87) 0.01
Employed REF
Insurance Not Insured 1.40 (0.61, 3.23) 0.39
Insured After Private Plans Effective REF
Number of Dependents 0.95 (0.74, 1.22) 0.67
Age, Five-Year Increments 1.03 (0.88, 1.20) 0.69
Income, $100 Increments 1.01 (0.98, 1.04) 0.50

REF: Reference category; OR: Odds Ratio; CI: Confidence Interval; p-values< 0.05 considered statistically significant.

a

Odds ratios, confidence intervals, and p-values calculated by combining point estimates and variances from analyses on each imputed dataset using the formulae given in Rubin (1987b) and Li, Raghunathan, and Rubin (1991).