Validity, Reliability, and Responsiveness of the Thai Palliative Care Outcome Scale Staff and Patient Versions Among Cancer Patients

Abstract

© 2018 American Academy of Hospice and Palliative Medicine Background: Palliative care is now part of Universal Health Coverage goals. Measurement of person-centered outcomes is central to determining quality and effectiveness. Guidance in psychometrics requires tools applied in new settings to have their properties tested. Objectives: To translate staff and patient versions of the Palliative care Outcome Scale (POS, version 2) into Thai, and to determine its psychometric properties among cancer patients in a Thai public hospital. Design: The Thai POS was subjected to cross-cultural translation: forward translation, backward translation, review by experts, and content validity index measurement. The patient-rated and staff-rated versions were completed by 379 nurses. We tested internal consistency, known-group comparison, responsiveness, and agreement. Setting/Participants: About 379 Thai cancer patients were admitted to Maharaj Nakorn Chiang Mai Hospital. Results: We found good internal consistency (Cronbach's alpha 0.9), good discrimination between known groups (significant difference in scores between high and low performance status groups; Z ranged from −9.95 to −7.80; P < 0.001), good responsiveness (improvements in at Time 2; Z ranged from −14.01 to −6.31; P < 0.001), and acceptable to good patient-staff agreement on ratings (weighted kappa range 0.31–0.73). Conclusion: The Thai POS is valid and reliable. These findings enable researchers and clinicians to apply the POS in primary research and routine clinical practice to both determine the effectiveness of interventions and improve care. This is the first validation in the region of a multidimensional person-centered outcome measure designed specifically for patients and families with advanced disease.