Predictors of p factor scores in children with chronic physical illness.

Background: The p factor represents the overall liability for the development of mental illness within individuals and we have previously validated a bi-factor model of the p factor in children with chronic physical illness.

Objective: In this next phase, we modelled predictors of the p factor in this sample of children.

Methods: Data come from the ongoing Multimorbidity in Children and Youth Across the Life-course study. Data from 263 children with a chronic physical illness aged 2-16 years and their parents were collected over 24 months. The parent-reported Emotional Behavioural Scales was used to develop a bi-factor model of the p factor. Subsequently, p factor scores were extracted from the model and standardized (Mean = 100, SD = 15). Analysis of variance compared p factor scores across different physical illnesses. Multiple regression was used to identify multilevel baseline predictors of p scores.

Results: There was no significant difference in p scores across categories of physical illness (F = 0.44, p = 0.849). Factors predictive of elevated p scores were older child age (B = 0.44), higher level of disability (B = 1.03), elevated parent psychopathology (B = 0.22) and stress (B = 0.21), and living in communities with older age and lower labor force participation (B = 1.66) and higher concentrations of racialized/newcomer populations (B = 2.05). Lower p scores were associated with being female (B = -3.85) and having immigrant parents (B = -5.43).

Conclusion: Factors predicting psychopathology, measured using p scores, in children with physical illness are multilevel. Fixed characteristics can inform targeted screening efforts, whereas modifiable characteristics are opportunities for upstream intervention in the context of family-centered integrated physical-mental health services.