Previously, in 30 Bangladeshi villages, 2 groups of children with iron-deficiency anemia (IDA) and nonanemic (NA) iron sufficiency aged 6 to 24 months participated in 2 parallel cluster randomized controlled trials of the effect of psychosocial stimulation on neurodevelopment. The intervention was composed of weekly play sessions at home for 9 months. All children with anemia received iron treatment of 6 months. The intervention improved the mental development of NA but not IDA groups. Six years after end line when the children were aged 8 to 9 years, we aimed to determine if benefits were sustained in the NA group or late-onset benefits emerged in the IDA group. We relocated 372 (90%) of the initial 412 children from all the clusters (villages), and assessed their IQ with the Wechsler Abbreviated Scale of Intelligence-II, motor development, and school achievement including math, spelling, and reading. Analyses were by intention-to-treat, adjusting for clustering. There was a significant interaction between anemia groups (IDA/NA) and intervention on IQ. The intervention benefitted the NA group's Full-Scale IQ (effect size, 0.43 [95% confidence interval, 0.08-0.79]) and Perceptual Reasoning Index (effect size, 0.48 [95% confidence interval, 0.08-0.89]) but did not affect the IDA group's outcomes. No other outcomes were significant. The benefits from early childhood psychosocial stimulation on the NA group's IQ, 6 years after intervention ended, adds to the limited evidence on the sustainability of benefits in low- and middle-income countries. Reasons for lack of effect in children with anemia are unknown.