The Effectiveness Satisfaction Questionnaire for COPD (ESQ-COPD) previously developed by us based on classical test theory has proven to have some limitations. The combination use of classical test theory and item response theory may achieve complementary advantages, which may be a theoretical basis and methodological support for the development or revision of instruments.

To develop a modified ESQ-COPD (mESQ-COPD) based on the combination of disease and syndrome, and to evaluate its psychometric properties.

By predefining basic characteristics of the instrument, normalizing the essential concepts and terminologies, revising the conceptual framework, generating an item pool, and reviewing the items, the draft of the mESQ-COPD was developed by our revision group. Then the draft was improved to be a mESQ-COPD by screening items based on the results of a pretest conducted during October 2017 to February 2018 with a convenience sample of 265 stable COPD patients from the respiratory clinic of 10 grade A tertiary hospitals in China, in combination with item screening using the classical test theory and item response theory. After that, the psychometric properties of the mESQ-COPD were tested using an onsite investigation carried out in a convenience sample of 365 stable COPD patients from the respiratory clinic of six grade A tertiary hospitals in China from November 2018 to February 2019.

The final mESQ-COPD includes four domains (clinical symptoms, capabilities for work and life, ability of environmental adaptation, and therapeutic effect) and 19 items. The Cronbach's α of the questionnaire was 0.949, and that of the four domains was 0.943, 0.869, 0.829, and 0.767, respectively. The correlation coefficient between the domain of clinical symptoms, capabilities for work and life, ability of environmental adaptation, or therapeutic effect and its component items ranged from 0.667 to 0.798, 0.855 to 0.882, 0.795 to 0.907, or 0.857 to 0.934, respectively. Confirmatory factor analysis showed that the comparative fit index, incremental fit index, non-normed fit index, standardized root mean square residual, and root mean square error of approximation measuring the construct validity of the mESQ-COPD were 0.94, 0.94, 0.93, 0.07, and 0.16, respectively. The correlation coefficients of the mESQ-COPD and its domains with COPD assessment test ranged from 0.371 to 0.538, while those of the mESQ-COPD and its domains with the modified Medical Research Council scale ranged from 0.329 to 0.564. The differences in the total score and domain scores of the mESQ-COPD between mild/moderate and severe/extremely severe COPD patients were both statistically significant (P<0.01). The acceptance rate of the mESQ-COPD was 99.5%, and the completion rate was 99.2%, with a mean completion time of (6.13±4.59) min.

The 19-item mESQ-COPD (consists of four domains: clinical symptoms, ability of work and life, capability for environmental adaptation, and therapeutic effect) has good reliability, validity, and applicability, which could be used to support the efficacy evaluation in COPD.