Psychometric Properties of MMAS-8 Question 8

Academic journals indicate that question 8 of the MMAS-8, with its five-point Likert response scale, contributes positively to the scale’s psychometric properties, particularly in terms of construct validity and sensitivity to adherence nuances. It aligns with the scale’s unidimensional structure, supports adequate reliability (though sometimes below optimal thresholds), and enhances convergent validity through correlations with related psychological constructs.


Key Findings from Academic Journals
1. Construct Validity and Factor Structure:
• Studies consistently report that the MMAS-8 question 8 demonstrates adequate construct validity with the Likert-scored question 8, contribute to a single adherence index. For instance, a 2014 study in a psychiatric outpatient setting in Spain (n=967) found that question 8 aligned with the one-factor structure, supporting its role in measuring medication adherence. The item’s Likert format did not disrupt the unidimensionality, as all items loaded onto a single factor explaining a significant portion of variance (e.g., 50.7% in some analyses).
• In a study validating the Spanish version of the MMAS-8 for type 2 diabetes patients, question 8 was part of a principal component (PC 1) alongside items 1 and 2, suggesting it measures a specific aspect of adherence related to memory or cognitive barriers. However, the study noted that the three-dimensional structure (PC 1: items 1, 2, 8; PC 2: items 3, 5; PC 3: items 4, 6, 7) explained 50.7% of the variance, indicating question 8’s contribution to a distinct facet of adherence behavior.
2. Reliability:
• A study in sub-Saharan Africa with type 2 diabetes patients reported a Cronbach’s alpha of 0.83 for the MMAS-8, suggesting that question 8’s inclusion in the scale supports adequate reliability in certain populations, though specific item-level reliability for question 8 was not isolated.
3. Convergent Validity:
• Question 8’s Likert response format enhances the scale’s sensitivity to detect variations in adherence behavior. Studies show significant correlations between MMAS-8 total scores (including question 8) and related constructs, such as the 10-item Drug Attitude Inventory (DAI-10), the Multidimensional Health Locus of Control scale (MHLC-C), and the Hong Psychological Reactance Scale (HPRS). These correlations indicate that question 8 contributes to the scale’s ability to capture psychological and behavioral factors influencing adherence.
• For example, in the psychiatric outpatient study, the MMAS-8, including question 8, differentiated adherence levels across mental disorder diagnosis groups, suggesting that the Likert item helps capture nuanced differences in adherence behavior.
4. Sensitivity and Specificity:
• The MMAS-8, with question 8’s Likert scale, has shown acceptable sensitivity and specificity in identifying low, medium, and high adherence (scores <6, 6–7, and 8, respectively). Question 8’s graded responses allow for finer distinctions in adherence challenges, particularly related to memory, which is critical in populations with cognitive or psychiatric impairments.
• However, a study in sub-Saharan Africa noted that criterion validity was not strong when using a cutoff score of 6, suggesting that question 8’s contribution to overall scale validity may depend on the population and context.
5. Challenges with Likert-Type Responses:
• The use of a Likert scale for question 8 introduces potential issues inherent to Likert-type items, such as response biases (e.g., acquiescence or social desirability). A 2014 study cited Hartley (2014) to acknowledge that Likert scales can confound results due to these biases, but the MMAS-8’s overall psychometric properties remained robust despite this limitation.
• Research on Likert scales in general suggests that the five-point format used in question 8 provides adequate sensitivity without overwhelming respondents, though some studies argue that increasing response options (e.g., to 7 or 11 points) could enhance normality and reduce skewness. In the context of MMAS-8, the five-point scale for question 8 is considered practical and effective for clinical settings.
Specific Considerations for Question 8
• Scoring and Standardization: Question 8’s Likert responses (0–4) are standardized by dividing by 4 to align with the 0–1 scoring of dichotomous items, ensuring a total MMAS-8 score range of 0–8. This standardization maintains the item’s contribution to the overall adherence index but may reduce its unique variability in some analyses.
• Population-Specific Performance: In psychiatric populations, question 8’s focus on memory difficulties is particularly relevant, as cognitive impairments may exacerbate adherence challenges. The item’s Likert format allows for capturing varying degrees of this difficulty, enhancing the scale’s utility in such settings.
• Cultural Adaptations: Studies validating translated versions (e.g., Spanish, Malaysian, Korean) note that question 8’s Likert scale performs consistently across cultures, though internal consistency may vary slightly due to cultural differences in response styles or interpretation of the item.