A previous study in Pakistan assessed the effectiveness of delivering responsive stimulation and enhanced nutrition interventions to young children. Responsive stimulation significantly improved children's cognitive, language, and motor development at 2 years of age. Both interventions significantly improved parenting skills, with responsive stimulation showing larger effects. In this follow-up study, we investigated whether interventions had benefits on children's healthy development and care at 4 years of age. We implemented a follow-up study of the initial, community-based cluster-randomised effectiveness trial, which was conducted through the Lady Health Worker programme in Sindh, Pakistan. We re-enrolled 1302 mother-child dyads (87% of the 1489 dyads in the original enrolment) for assessment when the child was 4 years of age. The children were originally randomised in the following groups: nutrition education and multiple micronutrient powders (enhanced nutrition; n=311), responsive stimulation (n=345), combined responsive stimulation and enhanced nutrition (n=315), and routine health and nutrition services (control; n=331). The data collection team were masked to the allocated intervention. The original enrolment period included children born in the study area between April 1, 2009, and March 31, 2010, if they were up to 2·5 months old without signs of severe impairments. The primary endpoints for children were development and growth at 4 years of age. Interventions were given in monthly group sessions and in home visits. The primary endpoint for mothers was wellbeing and caregiving knowledge, practices, and skills when the child was 4 years of age. Analysis was by intention to treat. The original trial is registered with ClinicalTrials.gov, number NCT00715936. 1302 mother-child dyads were re-enrolled between Jan 1, 2013, and March 31, 2013, all of whom were followed up at 4 years of age. Children who received responsive stimulation (with or without enhanced nutrition) had significantly higher cognition, language, and motor skills at 4 years of age than children who did not receive responsive stimulation. For children who received responsive stimulation plus enhanced nutrition, effect sizes (Cohen's d) were 0·1 for IQ (mean difference from control 1·2, 95% CI -0·3 to 2·7), 0·3 for executive functioning (0·18, -0·07 to 0·29), 0·5 for pre-academic skills (7·53, 5·14 to 9·92) and 0·2 for pro-social behaviours (0·08, 0·03 to 0·13). For children who received responsive stimulation alone, effect sizes were 0·1 for IQ (mean difference with controls 1·7, -0·3 to 3·7), 0·3 for executive functioning (0·17, 0·07 to 0·27), 0·2 for pre-academic skills (3·86, 1·41 to 6·31), and 0·2 for pro-social behaviours (0·07, 0·02 to 0·12). Enhanced nutrition improved child motor development, with effect size of 0·2 for responsive stimulation plus enhanced nutrition (0·56, -0·03 to 1·15), and for enhanced nutrition alone (0·82, 0·18 to 1·46). Mothers who received responsive stimulation (with or without enhanced nutrition) had significantly better responsive caregiving behaviours at 4 years of child age than those who did not receive intervention. Effect size was 0·3 for responsive stimulation plus enhanced nutrition (1·95, 0·75 to 3·15) and 0·2 for responsive stimulation (2·01, 0·74 to 3·28). The caregiving environment had a medium effect size of 0·3 for all interventions (responsive stimulation plus enhanced nutrition 2·99, 1·50 to 4·48; responsive stimulation alone 2·82, 1·21 to 4·43; enhanced nutrition 3·52, 1·70 to 5·34). Responsive stimulation delivered in a community health service can improve child development and care, 2 years after the end of intervention. Future analyses of these data are needed to identify which children and families benefit more or less over time. Grand Challenges Canada.