Face scales for the assessment of pediatric pain: A critical review LEORA KUTTNER TONY LEPAGE B.C. Children's Hospital and AND B.C. Children's Hospital and Columbia University of British Columbia Centre for Integrated Health Services ABSTRACT Self-report instruments in current pediatric pain research were reviewed noting their limitations and advantages. Early instruments were examined and face scales in current use discussed. Specifically, the work of Beyer (1984), Kuttner and LePage (1983), and McGrath et al. (1985) was highlighted as indicative of the trend towards more reliable and valid instruments that also have psychometric sophistication. The trend towards face scales with more complex stimuli and possessing greater empathic value was noted. Suggestions were offered towards resolving research problems across developmental ages by the creation of different instruments for different ages. Areas for future research were dis- cussed and the role of face scales in the further development of pain research in children was underlined. Recent efforts to investigate the nature of pediatric pain have made it clear that the assessment of children's pain remains a difficult, indirect, and challenging task. Since no single or group of physiological indices provides an objective measure of a child's pain, observational, behavioural, and self-report measures have been developed to provide multi-dimensional assessment of pediatric pain. Of these, self-report measures, used singly or in conjunction with the observa- tional measures, uniquely record the important subjective and personal experience of the child in pain. Increasingly, self-report instruments are being valued for research in general (Wolff, 1979) and in pediatric research (Johnson & Melamed, 1979) in particular. For a comprehensive discussion of the assessment of chil- dren's pain, the McGrath (1987) review on behavioural, physiological, and direct scaling techniques is highly recommended. The purpose of this article is to review critically a subgroup of the self-report instruments for the assessment of pain in children, the face scales, and to suggest possible new research directions. Children's Self-Report Children, particularly those of pre-school age, have often been regarded as unreliable or incapable of providing an accurate measure of their inner experience. Johnson and Melamed (1979) reported that self-report measures for children under seven years of age were unreliable. Katz (1979) found an insignificant correla- tion between the seven-face scale (a bi-dimensional scale with happy and sad faces) and observer measures for children seven years and younger, which contrasted with the significant correlation found for the older groups. He also noted that 85% of the younger age group sample reported extreme distress on this self-report measure, which resulted in a restrictive response range. Jay, Ozolins, Elliott, and CANAD. J. BEHAV. SCI./REV. CANAD. SCI. COMP. 21(2), 1989