The puppet interview to measure illness perceptions in paediatric oncology: development and psychometric properties in acute treatment and follow-up care.

Background: Illness perceptions comprise assumptions about symptoms, timeline, consequences, controllability, and emotional responses to an illness. Recent evidence shows that illness perceptions are associated with coping and well-being. So far, assessment in paediatric care was based on parental report only, since no instrument for the direct assessment of young children was available. We aim to describe the development (incl. indication and contraindication) of an innovative puppet interview to assess illness perceptions in children with cancer from the age of four years. Moreover, we investigate longitudinal trajectories and examine psychometric properties.

Methods: The puppet interview was developed based on the Illness-Perception-Questionnaire-Revised and the Berkeley-Puppet-Interview. Longitudinal trajectories and psychometric properties were examined in a sample of patient-parent dyads (N = 75) in a prospective longitudinal study in acute treatment with a 1-year follow-up (study 1: n = 41, n = 27) and in a cross-sectional study in follow-up care (study 2, n = 34).

Results: The puppet interview is comprehensible and well-received by children in acute treatment and follow-up care. There were significant differences in perceptions of a chronic timeline (U = 301.00, p = .008), consequences (U = 251.00, p = .008), and emotional representations (U = 244.50, p = .020) between children in acute treatment and children in follow-up care. Over the course of one year, children in acute treatment perceived more symptoms as part of their illness (M = 3.6, SD = 2.9, M = 4.5, SD = 3.1, n = 27, Z = -2.603, p = .009). We found expected intercorrelations between illness perception dimensions, e.g. between perception of consequences and emotional representations (r = .27, p = .033), and between perception of a chronic timeline and consequences (r = .38, p = .001). Moreover, we found confirming results regarding construct validity, as child's perceptions of symptoms correlated with their self-rated HRQoL (r = -.32, p = .014). Also parent-rated subscale on illness-specific aspects of child's HRQoL correlated with child's perception of symptoms (r = -.26, p = .016), cyclicity (r = -.28, p = .016), and consequences (r = -.34, p = .014). Acceptable internal consistency was shown for the dimensions timeline-acute/chronic and personal control.

Conclusions: Parental report can now be complemented by a self-report of illness perceptions in children aged four years and older. This will allow for the further adaptation of medical and psychosocial care during and after acute cancer treatment.

Trial Registration: The study has been pre-registered at the German Clinical Trials Register (registered 30/06/2020; DRK-S00022034) and at the Open Science Foundation ( https://osf.io/7xr6z ).