A cohort study of the incidence and risk factors for negative behavior changes in children after general anesthesia

BACKGROUND: Hospitalization and anesthesia can have a substantial psychological impact on children, which may be manifested by negative behavioral change. The primary aims of this study were to estimate the incidence of negative behavior change postanesthesia in a large cohort of children, and to identify the possible risk factors. METHODS: One thousand two hundred and fifty children aged from 3 to 12 years scheduled for anesthesia for a variety of procedures were enrolled. The absolute version of the Vernon Post Hospitalization Behavior Questionnaire (PHBQ) was used to assess behavior change at 3 and 30 days postanesthesia. Deterioration in seven or more items of behavior was defined to be a significant negative behavior change. Demographic data, anesthesia details, type and extent of preparation, details of procedure and length of hospitalization were recorded. Child temperament, child anxiety and parental anxiety were also measured. RESULTS: Twenty-four percent of children had significant negative behavior change at day 3 and 16% at day 30. After logistic regression, factors associated with significant negative behavior change at 3 days were increased parental state anxiety, younger age, overnight admission, lower birth order and preparation via having a discussion with the anesthetist. At day 30, longer hospitalization, younger age, reading the anesthesia preparation book and a previous difficult anesthesia experience were associated with significant negative behavior change. Also at day 30, reading the anesthesia preparation book was strongly associated with negative behavior change in children having short procedures, but not longer procedures. However, at both 3 and 30 days, the amount of variability explained by factors included in the models was low. CONCLUSIONS: Significant negative behavior change can occur in children after anesthesia. It is difficult to precisely predict in which children this will occur, however, some individual, family and procedural variables are associated with significant negative behavior change. If used, preparation should be considered according to level of surgical complexity.