METHODS: Longitudinal data from two different sources were analysed: a pre-post study of an online parenting programme conducted across six countries - the ePLH Evaluation Study, and a prospective cohort study in the United States - LONGSCAN. Cross-lagged panel model on parenting stress and child maltreatment was used in each dataset.

RESULTS: Based on repeatedly measured data of 484 caregivers in the ePLH study across five time points (every two weeks), we found that parenting stress at an earlier time point predicted later child maltreatment (IRR = 1.14, 95 % CI: 1.10,1.18). In addition, the occurrence of child maltreatment was associated with higher subsequent short-term parenting stress (IRR = 1.04, 95 % CI: 1.01,1.08) and thus could form a vicious circle. In the LONGSCAN analysis with 772 caregivers who were followed up from child age of 6 to child age of 16, we also found parenting stress at an earlier time point predicted later child maltreatment (β = 0.11, 95 % CI: 0.01,0.20), but did not observe an association between child maltreatment and subsequent long-term parenting stress.

LIMITATIONS: Potential information bias on the measurements.

DESIGN: An exploratory qualitative study with 55 parent-participants of Good Enough Parenting was conducted.

METHODS: One-to-one interviews were conducted with participants, using critical incident technique and guided by semi-structured interview schedule, to explore their experiences with the program. Transcripts were then analyzed using thematic analysis.

RESULTS: Coding showed a high degree of inter-rater reliability (kappa value of 0.78). The themes that emerged were Cultivating Awareness of Parents' Own Schemas, Cultivating Intentionality, Working through Developmental Issues, Responses to Challenges at Home, Performing Multiple Roles, and the Learning Process. Participants overwhelmingly reported satisfaction within these key themes.