The purpose of this logistic regression study was to identify predictors of teacher-reported sustained use of data-based instruction (DBI) during the COVID-19 pandemic and assess the extent to which the identified predictors explained teachers' sustained use after completing programmatic support for intensive early writing instruction. We surveyed 58 teachers who participated in a professional development efficacy trial regarding their sustained use of DBI in writing. The model indicated a higher predicted probability of sustaining DBI for teachers who received the full treatment (tools, learning modules, and coaching for 20 weeks of intervention during their year of participation in the efficacy trial) compared to teachers in the control group who only received tools and learning modules at the end of their participation year. In addition, teachers who taught in-person were more likely to sustain compared to those who taught in remote or hybrid models, controlling for other variables. Furthermore, as the number of facilitators that teachers reported increased, teachers were more likely to sustain their use of DBI components, controlling for other variables. Further research could shed light on the relative impact of different types of facilitators and different levels of support.