Purpose:The objective of this study was to compare the effectiveness and safety of short versus standard riboflavin induction times in cross-linking (CXL) for keratoconus.Methods:A retrospective comparative study was conducted with data from the Save Sight Keratoconus Registry. Inclusion criteria were epithelium-off technique, standard UVA CXL protocol (3 mW/cm2 for 30 minutes), riboflavin induction for 15 minutes (short) or 30 minutes (standard), and 1 year of follow-up data after CXL. Outcome measures included changes in best-corrected visual acuity (BCVA), keratometry in the steepest meridian (K2), maximum keratometry (Kmax), thinnest pachymetry (TCT), and adverse events. Analysis was conducted using mixed-effects regression models adjusted for age, sex, visual acuity, keratometry, pachymetry, practice, and eye laterality.Results:Two hundred eighty eyes (237 patients; mean, 27.3 ± 10.5 years old; 30% female) were included. The riboflavin induction time was short in 102 eyes (82 patients) and standard in 178 eyes (155 patients). The baseline characteristics (sex, mean age, BCVA, keratometry, and pachymetry [TCT]) were similar between the groups. At the 1-year follow-up visit, no statistically significant differences were observed in flattening in K2 and improvement in BCVA. Greater Kmax flattening [−1.5 diopters (D) vs. −0.5D, P = 0.031] and a greater proportion of >2% increase in TCT (23.5 vs. 11.3, P = 0.034) and haze (29 vs. 15, P = 0.005) were observed with short riboflavin induction.Conclusions:Short and standard riboflavin induction times achieved similar degrees of flattening in K2 and improvement in vision. Greater improvements in Kmax and TCT were seen with short riboflavin times; however, this group had higher rates of haze.