Background: The quality of caregiving in mothers with substance abuse problems appears to be compromised. However, divergent findings, methodological variability, and sample characteristics point to the need for research synthesis. Methods: A comprehensive systematic search was undertaken. Studies were eligible if they (1) compared substance-misusing mothers with non–substance-misusing mothers, (2) involved children from birth to 3 years, and (3) maternal sensitivity and child responsiveness were measured using observational methodology. Results: A global meta-analysis for maternal sensitivity (n = 24 studies) and child responsiveness (n = 16 studies) on 3433 mother-infant dyads yielded significant population effect sizes and significant heterogeneity. Subgroup analyses found reduced heterogeneity when the meta-analysis was conducted on studies where groups were matched on key demographic characteristics; although the effect size was small, it was still significant for maternal sensitivity but not child responsiveness. Conclusions: Compromised quality of caregiving is found in high-risk, substance-misusing mothers, emphasising the importance of early intervention that draws from attachment-based interventions.