Abstract Background This study reports the outcomes of Communities for Healthy Living (CHL), a cluster randomized obesity prevention trial implemented in partnership with Head Start, a federally-funded preschool program for low-income families. Methods Using a stepped wedge design, Head Start programs (n = 16; Boston, MA, USA) were randomly assigned to one of three intervention start times. CHL involved a media campaign and enhanced nutrition support. Parents were invited to join Parents Connect for Healthy Living (PConnect), a 10-week wellness program. At the beginning and end of each school year (2017-2019), data were collected on the primary outcome of child Body Mass Index z-score (BMIz) and modified BMIz, and secondary outcomes of child weight-related behaviors (diet, physical activity, sleep, media use) and parents’ weight-related parenting practices and empowerment. Data from 2 years, rather than three, were utilized to evaluate CHL due to the COVID-19 pandemic. We used mixed effects linear regression to compare relative differences during intervention vs. control periods (n = 1274 vs. 2476 children) in (1) mean change in child BMIz and modified BMIz, (2) the odds of meeting child health behavior recommendations, (3) mean change in parenting practices, and (4) mean change in parent empowerment. We also compared outcomes among parents who chose post-randomization to participate in PConnect vs. not (n = 55 vs. 443). Results During intervention periods (vs. control), children experienced greater increases in BMIz and modified BMIz (b = 0.06, 95% CI = 0.02,0.10; b = 0.07, 95% CI = 0.03, 0.12), yet were more likely to meet recommendations related to three of eight measured behaviors: sugar-sweetened beverage consumption (i.e., rarely consume; Odds Ratio (OR) = 1.5, 95% CI = 1.2,2.3), water consumption (i.e., multiple times per day; OR = 1.6, 95% CI = 1.2,2.3), and screen time (i.e., ≤1 hour/day; OR = 1.4, 95% CI = 1.0,1.8). No statistically significant differences for intervention (vs. control) periods were observed in parent empowerment or parenting practices. However, parents who enrolled in PConnect (vs. not) demonstrated greater increases in empowerment (b = 0.17, 95% CI = 0.04,0.31). Conclusions Interventions that emphasize parent engagement may increase parental empowerment. Intervention exposure was associated with statistically, but not clinically, significant increases in BMIz and increased odds of meeting recommendations for three child behaviors; premature trial suspension may explain mixed results. Trial registration ClinicalTrials.gov, NCT03334669, Registered October 2017.