Repository logoRepository logo

การพัฒนาแบบวัดความร่วมมือในการใช้ยาสำหรับคนไทย

Date
Publication

Journal Title

Journal ISSN

Volume Title

Publisher

มหาวิทยาลัยสงขลานครินทร์
Abstract

Abstract

The objective of this research was to develop the Medication Adherence Scale in Thais (MAST) and to find the cut off score to determine the level of medication adherence (MA). The measure consisted of 8 questions on a 6 points scale. The measure was tested for content validity by the review of 3 experts. The measure was pilot-tested in 30 hypertensive patients. It was further tested in 250 patients with hypertension at Pattani Hospital selected by convenient sampling. The study administered the MAST, the SF-36, and the Marlowe-Crowne Social Desirability Scale (MCSDs) to the subjects at month 0 and 6 at the end of the research. The study also collected the data on blood pressure and counted the remaining pills in the month 0, 2, 4 and 6. Cut-off score was established using analysis of receiver operator characteristics curve (ROC curve) Exploratory factor analysis revealed the measure consisted one dimension. The MAST showed criterion-related validity as evidenced by its correlation coefficients with MA assessed by pill counts (r = 0.559 ถึง 0.645; P <0.001). Cronbach's alpha coefficients of the MAST in month 0 and month 6 were 0.871 and 0.868, respectively. The test-retest reliability assessed from correlation coefficient of the MAST scores in the first and sixth months was 0.896. The MAST scores were related to SF-36 in the direction as predicted by the theory (r = 0.442-0.436, P < 0.001) and showed no relationship with the MCSDs (r=-0.086, P=0.175). Those with higher MAST scores at month 0 had a better blood pressure control in the next 2, 4 and 6 months. These results of the study indicated validity of the measure. Appropriate cut-off score of the 8 items MAST was 34. The area under the ROC of MAST scores and the blood pressure (measured at month 0) was 0.848, with sensitivity of 88.00%, specificity of 80.00%, positive predictive value (PPV) of 74.58% and negative predictive value (NPV) of 90.91%. Positive likelihood ratio (LR+) was 4.40, while negative likelihood ratio (LR-) was 0.15. In conclusion, the MAST as a measure of MA shows good reliability and validity. Furthermore, it exhibited high sensitivity, specificity and predictive value, suitable for assessing MA in hypertensive patients. Future studies should test this measure and find its cut-off score in other chronic diseases.
Details

Description

วิทยานิพนธ์ (ภ.ม. (เภสัชศาสตร์สังคมและการบริหาร))--มหาวิทยาลัยสงขลานครินทร์, 2562

Citation

Collections

Endorsement

Review

Supplemented By

Referenced By