Aim: To conduct a psychometric analysis of the Polish version of the CHQ-PF28 questionnaire for measuring the health-related quality of life (HRQL) of children and adolescents, on the basis of two population studies carried out in 2003 and 2008.

Material and methods: The CHQ-PF28 questionnaire was completed by parents of 1093 adolescents surveyed in 2003 and by parents of 605 adolescents surveyed in 2008. The average age of children was similar (13.4 yrs). An exploratory and confirmatory factor analysis was carried out on a combined sample from those studies. The HRQL profile of adolescents displaying and not displaying health problems, was compared. Mean standardized quality of life scores were calculated for all 13 partial dimensions and a two summary dimensions of physical and psychosocial health where high CHQ score always means better quality of life.

Results: In the 2003 study, the mean scores of partial scales ranged between 59.78 (SD=20.39) for the scale of family coherence and 97.07 (SD=11.88) for the scale of emotional limitations in playing social roles. In the 2008 study, mean scores for the same subscales were as follows: 59.17 (SD=19.50) and 96.17 (SD=14,45). It was concluded that parents assessed children’s psychosocial health as worse than their children’s physical health. In girls, higher mean scores were noted in the scales of self-esteem and behavioural disorders; in boys, higher mean scores were recorded in the scale of pain sensation. The reliability of partial scales was measured using α-Cronbach, which only for four scales was above 0,7. Six items were identified, which severely impair psychometric characteristics of the Polish version of the CHQ-PF28. A clear link was found between mean CHQ-PF28 indeces and the results of screening tests which identify physical and mental health abnormalities, such as the SDQ (Strengths and Difficulties Questionnaire), as well as the CSHCN (Children with Special Health Care Needs) test.

Conclusions: The Polish version of the CHQ-PF28 questionnaire has satisfactory psychometric characteristics. A wider use of the CHQ-PF28 in clinical research and population studies in Poland is fully justified. However, we would suggest implementing its abbreviated version, excluding items, which may affect the reliability and validity of the subscales and of the overall research tool.