Background: Interventions facilitated by caregivers have gained popularity among those caring for young children with autism. Instructing caregivers on specific techniques to foster social communication skills in their at-risk or diagnosed autistic children has the potential to alleviate concerns about their children's development. Moreover, it can offer a more intensive early intervention compared to what community providers alone can deliver. This study seeks to explore the correlation between caregiver strategies employed prior to participating in a caregiver-mediated intervention and the caregiver's fidelity to the intervention, as well as its sustainability during the follow-up period and child outcomes. This study constitutes a secondary analysis of a randomized controlled trial that compared the joint attention, symbolic play, engagement, and regulation (JASPER) and Psychoeducational Education Intervention (PEI), revealing significant advancements in children's social communication skills with the JASPER intervention. Methods: Eighty-six children (average age 31.5 months) with ASD and their primary caregivers enrolled in the two armed randomized trial evaluating the effect of JASPER versus PEI. Generalized linear mixed models were used to model the longitudinal trajectories of the outcomes. Results: Results indicated that caregivers in the JASPER intervention made more gains in overall JASPER strategies and individual domain strategies (environment, prompt, communication, mirrored pacing) compared to the caregivers in PEI (p's < 0.01) from baseline to exit. While both groups regressed some in overall and subdomain strategies at follow-up, caregivers in the JASPER intervention maintained more overall, and specifically in communication, and mirrored pacing strategies compared to PEI group (p's < 0.05). Further, baseline caregiver strategies moderated the treatment effect of child's joint attention skills from exit to follow-up (p = 0.002), where JASPER dyads with high caregiver strategy use at baseline continued to improve in JA skills post exit, whereas all other children did not. Conclusion: In summary, understanding caregiver style of interaction before intervention on caregiver fidelity and maintenance from exit to follow up and child progress is important to improving intervention uptake and child outcomes.