RESUMEN
BACKGROUND: Persons with profound intellectual and multiple disabilities (PIMD) are vulnerable when it comes to experiencing pain. Reliable assessment of pain-related behaviour in these persons is difficult. Aim To determine how pain items can be reliably scored in adults with PIMD. METHODS: We developed an instruction protocol for the assessment of pain-related behaviour in four phases. We used videos of 57 adults with PIMD during potentially painful situations. The items were assessed for inter-rater reliability (Cohen's kappa or percentage of agreement). RESULTS: The developed instruction protocol appeared to be adequate. Twelve items had satisfactory inter-rater reliability (n = 9: .30-1.00; n = 3: 85%-100%). DISCUSSION: Calibrating and adjustments to the instructions and item set appeared to be crucial to reliably score 12 items in adults with PIMD. Further research should focus on creating an assessment instrument based on these reliably scored items.