OBJECTIVES: To assess the minimal clinically important difference (MCID) of the Ocular Surface Disease Index® (OSDI®) using multiple computation methods. METHODS: The OSDI is a 12-item patient-reported outcomes (PRO) questionnaire specifically designed to quantify ocular disability due to dry eye disease. The Overall OSDI score ranges from 0 to 100; with scores categorized by ocular disease severity, i.e. normal (0-12), mild (13-22), moderate (23-32) and severe (33-100). To estimate the MCID for the OSDI, we used data from the RESTORE Registry study, which collects clinical, efficacy, PRO, including the OSDI, and safety data from patients with dry eye disease. We applied three anchor-based approaches to a clinician's global impression rating (CGI) and a subject global assessment (SGA). In each analysis, we computed one-way ANOVA with different subsets of patients responding with the smallest clinical change (minimal improvement/worsening for the CGI and improved/worse for the SGA). Patients were included in our analysis if they completed the OSDI at baseline and at a subsequent follow-up visit and completed the SGA or their clinician completed the CGI. RESULTS: Data from 160 patients were available. 84% of patients were Caucasian, 85% female, and mean age was 59.7 years. Both the CGI and the SGA were significantly correlated with the OSDI change score (r=-0.350, p<0.0001 and r=-0.443, p<0.0001). Most patients reported an improvement in OSDI scores. The mean change in OSDI score was similar for both anchors: -10.6 (N=154, SD=17.8) and -10.6 (N=144, SD=18.1), for the CGI and SGA, respectively. The MCID ranged from -4.9 to -8.7 for the CGI anchor and from -7.7 to -9.0 for the SGA anchor. CONCLUSION: Although preliminary, these results will assist clinicians and researchers when interpreting change in OSDI scores. The ongoing data collection from RESTORE will allow future analyses of larger sample sizes to finalize the MCID estimates.