Phantipa Sakthong1, Chanadda Chinthammit2, Pattarin Sukarnjanaset3, Nontapat Sonsa-Ardjit4, Wipaporn Munpan5. 1. Faculty of Pharmaceutical Sciences, Department of Pharmacy Practice, Chulalongkorn University, Bangkok, Thailand. Electronic address: phantipa.s@pharm.chula.ac.th. 2. Department of Pharmaceutical Sciences, College of Pharmacy, University of Arizona, Tucson, AZ, USA. 3. Faculty of Pharmaceutical Sciences, Department of Pharmaceutical Care, Rangsit University, Pathumthani, Thailand. 4. Pharmacy Unit, King Chulalongkorn Memorial Hospital, Bangkok, Thailand. 5. Pharmacy Unit, Phramongkutklao Hospital, Bangkok, Thailand.
Abstract
OBJECTIVES: To assess the psychometric properties of a novel instrument for medication management, the Patient-Reported Outcomes Measure of Pharmaceutical Therapy for Quality of Life (PROMPT-QOL), using both conventional psychometrics and Rasch analysis in a large sample. METHODS: This study was conducted with 1156 adult outpatients continuously taking any medicines at least 3 months from three university hospitals in Bangkok, Thailand, between July 2014 and March 2015. The psychometric properties were assessed in five steps: 1) assessment of dimensional structure, 2) item selection, 3) assessment of practicality, 4) assessment of reliability, and 5) assessment of criterion and known-groups validity. RESULTS: The PROMPT-QOL contained 43 items including nine domains, and their five-point Likert scale functioning worked well. Most items fulfilled the item selection criteria. The PROMPT-QOL took an average administration time of 13.4 ± 5.8 minutes. Only two items had missing data of 0.1% to 0.2%. All domains provided good to excellent test-retest reliability with intraclass correlation coefficients between 0.67 and 0.83. All domains of the PROMPT-QOL yielded high Cronbach's α values between 0.77 and 0.89, greater than an acceptable level of 0.70, except for the Availability and Accessibility domain (0.58). A multiple regression showed that the Medication Effectiveness domain was the strongest predictor of the overall QOL of the PROMPT-QOL, followed by the Therapeutic Relationships, Psychological Impacts, Convenience, and Availability and Accessibility domains (adjusted R2 ~ 52%). As expected, patients with higher PROMPT-QOL domain scores were associated with being younger, more educated, having a lower number of medicines, patients' perceptions of better disease control, having no adverse drug reactions, and medicine preference. CONCLUSIONS: The PROMPT-QOL was practical, reliable, and valid for Thai patients.
OBJECTIVES: To assess the psychometric properties of a novel instrument for medication management, the Patient-Reported Outcomes Measure of Pharmaceutical Therapy for Quality of Life (PROMPT-QOL), using both conventional psychometrics and Rasch analysis in a large sample. METHODS: This study was conducted with 1156 adult outpatients continuously taking any medicines at least 3 months from three university hospitals in Bangkok, Thailand, between July 2014 and March 2015. The psychometric properties were assessed in five steps: 1) assessment of dimensional structure, 2) item selection, 3) assessment of practicality, 4) assessment of reliability, and 5) assessment of criterion and known-groups validity. RESULTS: The PROMPT-QOL contained 43 items including nine domains, and their five-point Likert scale functioning worked well. Most items fulfilled the item selection criteria. The PROMPT-QOL took an average administration time of 13.4 ± 5.8 minutes. Only two items had missing data of 0.1% to 0.2%. All domains provided good to excellent test-retest reliability with intraclass correlation coefficients between 0.67 and 0.83. All domains of the PROMPT-QOL yielded high Cronbach's α values between 0.77 and 0.89, greater than an acceptable level of 0.70, except for the Availability and Accessibility domain (0.58). A multiple regression showed that the Medication Effectiveness domain was the strongest predictor of the overall QOL of the PROMPT-QOL, followed by the Therapeutic Relationships, Psychological Impacts, Convenience, and Availability and Accessibility domains (adjusted R2 ~ 52%). As expected, patients with higher PROMPT-QOL domain scores were associated with being younger, more educated, having a lower number of medicines, patients' perceptions of better disease control, having no adverse drug reactions, and medicine preference. CONCLUSIONS: The PROMPT-QOL was practical, reliable, and valid for Thai patients.