论文部分内容阅读
Aim:To study the validity and liability of a Swedish translation of the Post Hospitalization Behavior Questionnaire(PHBQ)-in children in Sweden.Methods:The PHBQ was translated using a back-translation method.The subjects were 340 children,ages 2-13 y,admitted for elective surgery or diagnostic procedure with anaesthesia.The results were analysed using exploratory factor analysis with principal component analysis with Oblimin rotation.The fit to data was examined using confirmative factor analysis with a good measure of fit for the model(P > 0.09 for all factors).Results:Five factors emerged as being most consistent:general anxiety-withdrawal,eating disturbances,separation anxiety,regression-aggression and sleep anxiety.A panel of child psychologists confirmed the face validity of factors.Internal consistency(Chronbach’s alpha)was adequate(0.75-0.87)for subscales and excellent for total score(0.93).Children less than 5 y old had higher scores than older children(mean 0.046±0.018 vs-0.0089±0.014,P < 0.001).There were no gender differences.Conclusion:The results support a conclusion that a five-factor model better fits data from Swedish children than the original six-factor model.
Aim: To study the validity and liability of a Swedish translation of the Post Hospitalization Behavior Questionnaire (PHBQ) -in children in Sweden. Methods: The PHBQ was translated using a back-translation method. The subjects were 340 children, ages 2-13 y, admitted for elective surgery or diagnostic procedure with anaesthesia. the results were analysed using exploratory factor analysis with principal component analysis with Oblimin rotation. fit to data was examined using confirmative factor analysis with a good measure of fit for the model (P> 0.09 for all factors). Results: Five factors emerged as being most consistent: general anxiety-withdrawal, eating disturbances, separation anxiety, regression-aggression and sleep anxiety. A panel of child psychologists confirmed the face validity of factors. Internal consistency (Chronbach’s alpha) was adequate (0.75-0.87) for subscales and excellent for total score (0.93) .Children less than 5 y old had higher scores than older children (mean 0.046 ± 0.018 vs-0.0 089 ± 0.014, P <0.001). Where were no gender differences. Conflusion: The results support a conclusion that a five-factor model better fits data from Swedish children than the original six-factor model.