Background This study evaluated the psychometric properties of the Non-communicating Children's Pain Checklist-Postoperative Version (NCCPC-PV) when used with children with severe intellectual disabilities. Methods The caregivers of 24 children with severe intellectual disabilities (aged 3-19 yr) took part. Each child was observed by one of their caregivers and one of the researchers for 10 min before and after surgery. They independently completed the NCCPC-PV and made a visual analog scale rating of the child's pain intensity for those times. A nurse also completed a visual analog scale for the same observations. Results The NCCPC-PV was internally reliable (Cronbach alpha = 0.91) and showed good interrater reliability. A repeated-measures analysis of variance indicated NCCPC-PV total and subscale scores were significantly higher after surgery and did not differ by observer. Postoperative NCCPC-PV scores correlated with visual analog scale ratings provided by caregivers and researchers, but not with those of nurses. A score of 11 on the NCCPC-PV, by caregivers, provided 0.88 sensitivity and 0.81 specificity for classifying children with moderate to severe pain. Conclusions The NCCPC-PV displayed good psychometric properties when used for the postoperative pain of children with severe intellectual disabilities and has the potential to be useful in a clinical setting. The results suggest familiarity with an individual child with intellectual disabilities is not necessary for pain assessment.