OBJECTIVE: A practice change intervention demonstrated improvements in the provision of antenatal care addressing alcohol consumption. The aim of this study was to explore whether the effectiveness of the intervention differed between subgroups of pregnant women and types and location of maternity services. DESIGN AND SETTING: Post-hoc exploratory subgroup analyses of the outcomes from a randomised stepped-wedge controlled trial conducted with all public maternity services within three sectors of a local health district in Australia. MEASUREMENTS: Two outcomes (receipt of alcohol assessment and complete care) measured at two visit types (initial and subsequent) were included in analyses. Logistic regression models explored interactions between pre-post differences and subgroups of women (age, Aboriginal origin, education level, disadvantage, gravidity and alcohol consumption in pregnancy) and services (geographic remoteness, service and provider type/s) that have been reported to be associated with variation in guideline implementation. FINDINGS: Surveys from 5694 women were included in the analyses. For the initial visit, no significant differential intervention effects between subgroups of women or type/location of services were found for either outcome. For subsequent visits, the intervention effect differed significantly only between Aboriginal origin subgroups (Aboriginal OR: 1.95; 95% CI: 0.99-3.85; non-Aboriginal OR: 5.34; 95% CI: 4.17-6.83; p<0.01) and women's alcohol consumption in pregnancy subgroups (consumed alcohol OR: 1.28; 95% CI: 0.59-2.78; not consumed alcohol OR: 5.22; 95% CI: 4.11-6.65; p<0.001) for assessment of alcohol consumption. KEY CONCLUSIONS: These exploratory results suggest that the intervention may have had similar effects between different subgroups of women and types and location of services, with the exception of women who were non-Aboriginal and women who had not consumed alcohol, for whom the intervention was potentially more effective. IMPLICATIONS FOR PRACTICE: The practice change intervention could be implemented with different maternity service and provider types to effectively support improvements in antenatal care addressing alcohol consumption. These exploratory results provide further data for hypothesis generation regarding targeted areas for the testing of additional strategies that enable Aboriginal women to benefit equally from the intervention, and to ensure those women most in need of care, those consuming alcohol during pregnancy, have their care needs met.