Psychometric Properties of the Altarum Consumer Engagement (ACE) Measure of Activation in Patients with Prediabetes

Abstract

BACKGROUND: Patient activation is associated with better outcomes in chronic conditions. OBJECTIVE: We evaluated the psychometric properties of the 12-item Altarum Consumer Engagement™ Measure (ACE-12) in patients with prediabetes. PARTICIPANTS: ACE-12 was administered to patients in the Prediabetes Informed Decisions and Education Study. MAIN MEASURES: We conducted an exploratory factor analysis followed by confirmatory factor analytic models. We evaluated item response categories using item characteristic curves. Construct validity was assessed by examining correlations of the ACE-12 scales with education, depressive symptoms, self-rated health, hemoglobin A1c, body mass index, and weight loss. KEY RESULTS: Participants (n = 515) had a median age of 58; 56% were female; 17% Hispanic; 54% were non-White. The scree plot and Tucker and Lewis reliability coefficient (0.95) suggested three factors similar to the original scales. One item loaded on the navigation rather than the informed choice scale. Ordinal alpha coefficients for the original scales were commitment (0.75); informed choice (0.71); and navigation (0.54). ICCs indicated that one or more of the response categories for 5 of the 12 items were never most likely to be selected. Patients with lower education were less activated on the commitment (r = − 0.124, p = 0.004), choice (r = − 0.085, p = 0.009), and overall score (r = − 0.042, p = 0.011). Patients with depressive symptoms had lower commitment (r = − 0.313, p ≤ 0.001) and overall scores (r = − 0.172, p = 0.012). Patients with poorer health scored lower on the Commitment (r = − 0.308, p ≤ 0.001), Navigation (r = − 0.137, p ≤ 0.001), and overall score (r = − 0.279, p ≤ 0.001). CONCLUSION: The analyses provide some support for the psychometric properties of the ACE-12 in prediabetic patients. Future research evaluating this tool among patients with other chronic conditions are needed to determine whether Q1 (I spend a lot of time learning about health) should remain in the informed choice or be included in the navigation scale. Additional items may be needed to yield acceptable reliability for the navigation scale.

Psychometric evaluation of the 8-item Altarum Consumer Engagement (ACE) Measure™ in community-dwelling adults in Singapore

“…It was subsequently shortened to 12 items consisting of three subscales: Commitment (4 items), Informed Choice (4 items), and Navigation (4 items). The Ownership subscale was not included due to its similarity in outcome prediction with that of the Commitment subscale [ 24 ].…”

“…The 12-item ACE Measure has been validated among United States prediabetes patients [ 24 ] and has been used in many ways including population benchmarking, generating patient’s personalized engagement reports, triaging resources, determining course of treatment and health coaching since its development [ 25 ]. Each subscale of the 12-item ACE Measure can be used independently.…”

Capsule Commentary on Castellon-Lopez et al.: “Psychometric Properties of the Altarum Consumer Engagement (ACE) Measure of Activation in Patients with Prediabetes”

“…An important limitation is the lack of longitudinal measures of A1C. 1 Patient activation interventions have previously shown a modest improvement in A1C. 5 The lack of tools to measure and monitor activation has limited our understanding of what "activates" individuals 4 and hindered the development of evidence-based interventions.…”