The esteemed journal Quality of Life Research proudly featured a seminal paper titled “Canadian-French, German and UK versions of the Child Health Questionnaire: methodology and preliminary item scaling results” in 1998, authored by J. M. Landgraf, E. Maunsell, K. Nixon Speechley, M. Bullinger, S. Campbell, L. Abetz, and J. E. Ware. This publication provides an in-depth look at the rigorous development and evaluation of cross-culturally adapted versions of the Child Health Questionnaire – Parent Form 50 (CHQ-PF50), a critical tool for measuring the health status of children and adolescents worldwide.
The Child Health Questionnaire (CHQ) and its Purpose
Initiated in 1990 as part of the Child Health Assessment Project, the CHQ was developed to advance methods for measuring the physical and psychosocial health status and well-being of children and adolescents. The CHQ-PF50, a 50-item, parent-completed short form, was empirically derived from a longer 98-item version to meet the industry’s demand for more practical length instruments. This instrument is designed to measure two fundamental components of child and adolescent health: physical and psychosocial functioning and well-being. It captures 14 distinct concepts, with ten focused on the child’s health-related quality of life (e.g., physical functioning, mental health, self-esteem) and four on the emotional impact of the child’s health on the parent (e.g., parental time impact, family activities). The scales are scored using Likert’s method, with raw scores converted to a 0-100 scale, where higher scores consistently indicate better functioning and well-being.
Rigorous Cross-Cultural Translation and Evaluation
A cornerstone of this study was the stringent methodology used for translation and adaptation, adhering to emerging international guidelines in health outcomes research. The authors employed a meticulous protocol involving both forward and backward iterations of the questionnaire for Canadian-French, German, and UK versions. A key benefit of this early translation work, before widespread dissemination, was the identification and adaptation of problematic items, ambiguous phrases, and conceptual inequivalences, leading to a more culture-free instrument. Emphasis was placed on conceptual equivalence, ensuring that the connotative meaning of items was captured while accounting for cultural nuances.
Country-specific efforts included:
- Germany: A single forward and backward translation was conducted for a project assessing the impact of airport noise.
- United Kingdom: An Anglicization process involved reviewing each item for comparability with colloquial expressions. A notable modification was changing “sick” to “ill” in a general health item to accurately convey the intent of a severe illness.
- Canada: The Canadian-French translation involved multiple professional and lay translators, with the lay translator introduced earlier in the process to ensure the use of common language. Extensive harmonization meetings and pre-testing with French and English speakers were conducted. The American-English CHQ was deemed appropriate for English-speaking Canadians.
Psychometric Validation and Key Findings
The psychometric evaluation employed multitrait analysis to assess convergent and discriminant validity and Cronbach’s α coefficient to estimate internal consistency reliability. The results were highly encouraging:
- 79% of all item–scale correlations achieved acceptable internal consistency (0.40 or higher).
- Tests of item convergent and discriminant validity were successful at least 87% of the time across all scales and countries.
- Reliability coefficients (Cronbach’s α) ranged from 0.43 to 0.97, with a median of 0.80, demonstrating strong internal consistency. Most scales met or exceeded the recommended 0.70 threshold for group comparisons.
- Negligible floor effects were observed across all samples, suggesting the instrument can capture even low levels of health problems.
- As expected, noteworthy ceiling effects were observed for physical scales (mean 73%), indicating high levels of optimal physical functioning in the study populations. Conversely, psychosocial scales showed fewer ceiling effects (3–17%).
A particular highlight was the encouraging results obtained for the Canadian-French developmental version, despite its relatively small sample size (n=16). Despite the challenge of a higher standard error due to the small sample, the item correlations still largely met the success criteria. The structural integrity and psychometric properties of the American-English CHQ-PF50 and its respective translations were strongly supported by these pilot studies.
Conclusion and Future Impact
This groundbreaking work underscores the value of using stringent protocols to develop conceptually robust translations for health assessment instruments. The findings confirm that the German, UK, and Canadian-French developmental versions of the CHQ-PF50, along with the American-English version for English-speaking Canadians, are robust enough to be recommended for subsequent investigations. These future studies will facilitate further psychometric evaluations, collection of normative data, and development of comprehensive scoring and interpretation guidelines. By adopting rigorous methods for translation and evaluation, this research significantly facilitates efforts to better understand the health and well-being of children and adolescents and their families globally.
Reference for the Article:
Landgraf, J. M., Maunsell, E., Nixon Speechley, K., Bullinger, M., Campbell, S., Abetz, L., & Ware, J. E. (1998). Canadian-French, German and UK versions of the Child Health Questionnaire: methodology and preliminary item scaling results. Quality of Life Research, 7, 433–445.