Seeing the Light in Self-Reported Glare

We conducted a secondary analysis to evaluate two glare items versus a composite score from the two items in a sample of 544 patients (38% women; Mdn = 29 years old; 13% high school education or less) before and after eye surgery. The first question was from a National Eye Institute (NEI) survey and the second question included a definition and picture of glare. At baseline, 28% of participants reported glare on the NEI item versus 39% on the question with the definition and picture. There was 76% agreement between the two questions (K = .46). Three months after baseline, there was no change in the percentage of participants who reported glare on the NEI question (27%), but a significant decrease in participants reporting glare on the question with the definition and picture (38% at baseline to 25% 3 months later). A 2-item glare scale was more reliable and highly correlated with multi-item measures of halos and starbursts than the individual items. Change in the scale was more strongly associated than the items with the change in satisfaction with vision and with satisfaction with surgery. The scale may provide a better indicator of glare than either item alone.