Predictors of decisional capacity in AD biomarker test candidates

Abstract

AbstractBackgroundDeciding to learn one’s Alzheimer’s disease (AD) biomarker status requires comprehending and deliberating about complex and nuanced information. Assessing capacity to consent to AD biomarker testing and disclosure is further complicated when candidates have cognitive impairment and present with family care partners. The objective of this analysis was to identify predictors of decisional capacity for an amyloid PET disclosure study among persons with Mild Cognitive Impairment (PwMCI) and their care partners (CPs).MethodThis investigation was a secondary analysis, using baseline data of a randomized controlled trial of amyloid PET results disclosure. Multiple linear regression modeling was used to examine predictors of decisional capacity, measured by the University of California, San Diego Brief Assessment of Capacity to Consent (UBACC). Focusing on PwMCI (n = 82), the predictor variables of interest were sociodemographic characteristics, and level of knowledge on MCI/AD and normal aging, global cognition measured by Mini Mental Status Exam (MMSE), and neuropsychological tests of language and executive functioning. Dyadic analyses were conducted with CP data (n = 82), using the Actor Partner Interdependence Model (APIM) to assess for presence of partner effects among key variables.ResultParticipants and CPs were mostly White (91.5%, 90.2%), highly educated (16.5, 15.5 years on average), with mean age 72.7 and 66.8 years, respectively. CPs tended to be female, and spouses of the participants (70.7%). MMSE scores (b = 0.215, SE = 0.082, p = 0.011) and MCI/AD knowledge scores (b = 0.241, SE = 0.101, p = 0.020) positively predicted decisional capacity for PwMCI. Dyadic analysis using APIM showed that level of MCI/AD knowledge (p = 0.019) and level of education (p<0.001) were highly correlated within the dyad. Furthermore, analyses revealed significant positive correlations for MCI/AD knowledge for actor relations for decisional capacity for both the participant (a11 = 0.328, p = 0.005) and the CP (a22 = 0.262, p = 0.025). CP’s level of MCI/AD knowledge had a negative association for partner effects for decisional capacity (p21 = ‐0.207, p = 0.043) of the PwMCI.ConclusionAssessment of decisional capacity is imperative when considering participation in AD research, including biomarker studies involving results disclosure. Our findings highlight that not only do individual characteristics put PwMCI at risk for lower decisional capacity, but dyadic effects from their CPs may also be present.