Comment on “Comparative Effectiveness of Two Methods for Assigning Care Coordinators to Person With Dementia”

We read with great interest the article by Kern et al., which evaluated two strategies for assigning care coordinators to Person with Dementia within a Medicare accountable care organization [1]. We commend the authors for conducting a pragmatic randomized trial embedded in real-world clinical practice and for innovatively using fragmented ambulatory care as an inclusion criterion—a setting often underrepresented in dementia care coordination studies. The authors' use of the reversed Bice–Boxerman index (≥ 0.86) to identify individuals with highly fragmented care is methodologically sound, yet it may inadvertently exclude those with significant coordination challenges but lower recorded fragmentation, such as patients with advanced dementia receiving care concentrated within one health system [2]. Combining claims-based fragmentation indices with caregiver-reported burden or functional dependence measures could improve sensitivity and better capture those at risk for coordination failure [3]. Second, the relatively low acceptance rate (38%) of proactive care management highlights engagement barriers. As the authors note, urgency and perceived need were stronger after hospitalization. Future interventions could enhance uptake by integrating longitudinal, trust-based contacts—potentially through primary care–linked teams or community health workers—and by addressing caregiver stress and digital literacy [4]. Structured, culturally attuned communication may further enhance the perceived relevance of care coordination services. Third, although the primary composite of emergency department (ED) visit or hospitalization showed no between-group difference, this may partly reflect the limited intervention reach and the use of distal outcomes. Incorporating intermediate metrics—such as timely follow-up, medication reconciliation, or caregiver reported experience—might detect earlier benefits [5]. Adaptive or stepped-wedge trial designs could also improve statistical power while preserving pragmatic generalizability [6]. Finally, this study has broader implications for policy. Given the limited supply of care coordinators, aligning resource allocation with predictive analytics, caregiver-reported needs, and value-based payment incentives could improve efficiency [7]. Moreover, supporting family caregivers—often the “hidden workforce” performing informal coordination—should become an explicit quality metric in dementia-related initiatives [8]. In conclusion, Kern et al. make an important contribution by demonstrating the feasibility of embedding a pragmatic randomized trial within a large to address care fragmentation in Person with Dementia. Their work underscores that the challenge lies not only in identifying who needs coordination but also in ensuring that the support offered is both acceptable and actionable to patients and families. Jianguo Wang: writing – original draft. Hua Yan: investigation, formal analysis. Xiaoting Ye: data curation, validation. Qiang Guo: supervision, writing – review and editing. Qiang Guo contributed to revising the manuscript. The authors have nothing to report. The authors have nothing to report. This study had no sponsor. The authors declare no conflicts of interest. This publication is linked to a related reply by Kern and Bell. To view this article, visit https://doi.org/10.1111/jgs.70351.