The study is designed to answer two research questions: What is the generalizability of pain assessment procedures used clinically with hospitalized children after surgery; and what cues do observers use when making judgments about the level of children's pain? The specific aims related to the fist question are: 1) to estimate of generalizability of pain measurement across three observers (parents, children, and nurses) and across time for each of two methods: Hester's Poker Chip Tool and the Pain Ladder, and 2) to compare the generalizability coefficients derived.
The specific aims pertinent to the second question are: 1) to develop and descibe categories of cues used by parents and nurses and create classification systems of parents and nurses cues, 2) to compare the two classification systems, and 3) to correlate categories of pain with the level of pain assessed. The first research question will be addressed through a pair of generalizability studies and the second through a descriptive correlational study. The sample for the study is 200 children stratified on age and sex who are randomly assigned to two groups: the Poker Chip Group and the Pain Ladder Group. Pain ratings will be obtained from the child, parent, and nurse on four different occasions on the first post-operative day. Parents and nurses will identify the cues they used in making the pain judgments. Analysis of variance of the pain ratings analyses will give estimated variance components for the main effects (subjects, observers, and time) and their interactions. Comparison of these components will permit explanation related to the variables. The analysis will also give a generalizability coefficient for each method across observers and time analagous to classical reliability. Content analysis will be used for the cue data. From this two year study, implications, for pain assessment procedures will be drawn. The cues used by parents and nurses in the assessment of children's pain will be examined for use in nursing.
