Objectives/purpose
The 70-item Truth-Telling Questionnaire (TTQ-70) is widely used to investigate cancer patients’ preferences regarding truth-telling. However, its length and high internal consistency suggest a need for item reduction to reduce patient response burden and enhance healthcare providers’ efficiency in understanding patient preferences. This study aimed to develop and validate two short versions of the TTQ-70 based on classical test theory (CTT) and item response theory (IRT).
Sample and setting
This secondary analysis used data from a cross-institutional truth-telling project. Cancer patients were recruited via convenience sampling. The data were generated as derivation and validation cohorts.
Procedures
The derivation cohorts were used to develop two short-form versions of the TTQ using CTT and IRT, respectively, ensuring representative item selection. The validation cohort was employed to evaluate and compare the psychometric properties of the short forms. The TTQ-70 was measured across four subscales—methods of disclosure, emotional support, additional information, and setting. Item-level tests, a graded response model, and confirmatory factor analysis were used for data analysis.
Results
The CTT-based and IRT-based short forms retained 23 and 43 items, respectively. Both versions demonstrated acceptable reliability (CTT-based: Cronbach’s alpha = 0.89; IRT-based: Marginal reliability = 0.92). Regarding construct validity, the IRT-based short form exhibited better model fit (standardized root mean squared residual: CTT-based = 0.0866; IRT-based = 0.0580). The item distribution across subscales in the IRT-based version effectively preserved the conceptual structure of the original TTQ, demonstrating better construct validity.
Conclusion and Clinical Implications
The 43-item IRT-based short-form TTQ is recommended as the preferred instrument for assessing cancer patients’ truth-telling preferences. It reduces response burden while maintaining robust validity and reliability, offering a practical tool to support healthcare providers in delivering sensitive information in alignment with patient needs and preferences.