The Development of Medication Adherence Scale for Persons with Coronary Artery Disease (MAS-CAD): A Nursing Perspective

Abstract

The purpose of this study was to develop a Medication Adherence Scale for Persons with Coronary Artery Disease (MAS-CAD) based on a nursing perspective. This study was guided by scale development procedures, which consisted of 7 steps: 1) clarifying and determining the concept, 2) generating an item pool, 3) determining the format for measurement, 4) reviewing the initial item pool by 7 professional experts, 5) conducting preliminary item pilot study with 30 Thais with CAD, 6) conducting field-testing for psychometric property testing among 457 Thais with CAD who attended heart clinics of 7 tertiary hospitals in Thailand, and 7) developing scoring and interpretation of the scale scores. Data were analyzed by using descriptive statistics, content validity index, Cronbach’s alpha coefficient, confirmatory factor analysis by LISREL, and Pearson product moment correlation. The results showed that the MAS-CAD is a self-report instrument, comprised of 5 rating Likert-scale categorical format, composed of 25 items covering 4 constructs; proper knowledge of prescribed medication (7 items), storing medication appropriately (3 items), self-regulating medication administration correctly and continuously (11 items), and participating in the medication treatment plan (4 items). Psychometric property testing demonstrated satisfactory internal consistency with Cronbach’s alpha coefficient of 0.81, and test-retest reliability was acceptable (r = .62, p < .01). Construct validity was supported by confirmatory factor analysis. In conclusion, the MAS-CAD is a reliable and valid instrument. Nurses can use this scale as a tool for assessing medication adherence for persons with CAD.

Measurement Properties of Patient-Reported Outcome Measures for Medication Adherence in Cardiovascular Disease: A COSMIN Systematic Review

“…Two asked about intentional reasons only [ 74 , 120 ] and one [ 73 ] asked about unintentional reasons only. Group-3: items asks about both the extent of, and reasons for, non-adherence [ 45 , 50 , 54 – 56 , 65 , 67 , 77 , 78 , 81 , 85 , 89 , 91 – 93 , 95 , 98 , 103 , 110 , 113 , 114 , 117 , 126 ]. Six MA-PROMs [ 50 , 54 , 81 , 92 , 93 , 98 ] in Group-3 were comprehensive enough to capture information on the extent of both medication-taking and prescription-filling and at least two additional domains from both intentional and unintentional reasons for non-adherence from the MAM framework [ 40 ] (Table 2 ).…”

“…Patient-reported outcome measures development was rated for the 24 original MA-PROMs [ 45 , 60 , 77 , 78 , 81 , 86 , 89 , 96 , 97 , 102 , 103 , 105 , 113 , 114 , 116 , 119 – 122 , 127 – 130 ] and 13 MA-PROMs [ 46 , 50 , 51 , 53 – 56 , 65 , 67 , 70 , 80 , 85 , 92 , 95 , 110 , 117 ] were modified versions of the original MA-PROMs. Patient-reported outcome measures development was not rated for 3 MA-PROMs [ 66 , 71 , 91 , 125 ], the development studies for two were published in a language other than English [ 71 , 91 ], and a development study could not be found for one [ 66 ].…”

“…The quality PROM development was obtained from PROM design (concept elicitation) and cognitive interview studies (Table 3 ). The concept elicitation for 21 MA-PROMs did not involve patients in their development and was therefore deemed inadequate [ 45 , 50 , 56 , 60 , 77 , 78 , 81 , 97 , 102 , 103 , 105 , 113 , 114 , 119 , 120 , 122 , 126 ]. The concept elicitation for one study [ 98 ] was rated very good as patients representative of the target population were involved in a qualitative study [ 130 ] of very good quality.…”

“…Cognitive interviews with patients were reported for the development of ten original MA-PROMs [ 45 , 60 , 77 , 81 , 102 , 113 , 116 , 119 , 121 , 122 ]. Cognitive interviews were rated doubtful for all of these MA-PROMs because at least one of the standards for cognitive interviews was rated doubtful.…”

“…Of the 35 studies that evaluated the content validity of an existing MA-PROM, 21 involved only patients [ 42 , 44 , 47 , 52 , 55 , 59 , 61 , 62 , 68 , 70 , 72 , 75 , 76 , 79 , 84 , 88 , 90 , 99 , 109 , 115 , 131 ], 4 involved only experts [ 67 , 89 , 103 , 124 ], and 10 involved both patients and experts [ 46 , 54 , 63 , 64 , 73 , 74 , 86 , 108 , 117 , 123 ] (Table 4 ). No studies that were not development studies were found on content validity for 21 of the identified MA-PROMs [ 45 , 50 , 51 , 56 , 60 , 66 , 71 , 78 , 80 , 85 , 91 – 93 , 96 , 97 , 105 , 110 , 113 , 119 – 121 , 125 ]. All 31 content validity studies involving patients were of doubtful quality for relevance and comprehensibility, and none evaluated the comprehensiveness aspect of the content validity.…”