Comment on “Development of a Novel Frailty Trigger for Use at Triage in the Emergency Department”

We read with great interest Moloney et al.'s work developing a two-question Frailty Trigger administered at triage and compared with established screens and a Comprehensive Geriatric Assessment (CGA) [1]. The direct triage administration, the CGA reference, and the observed associations with length of stay and one-year survival speak to practical relevance for front-door emergency care. To support translation, three considerations may help focus subsequent studies and implementation. First, to preserve independence between index test and reference, it may be helpful to minimize construct overlap. Because the Trigger queries activities of daily living (ADL) dependence and cognition, a CGA that includes multi-domain self-report on similar constructs could introduce criterion contamination and inflate apparent area under the receiver operating characteristic curve (AUC) [2]. A prespecified CGA adjudication that prioritizes elements independent of ADL self-report, with blinded multidisciplinary consensus, and sensitivity analyses excluding overlapping domains would clarify incremental diagnostic value at triage. Clinically, this helps ED (Emergency Department) teams understand whether Trigger-positive status identifies patients beyond those already evident on functional history. Second, reporting measures that inform bedside decision-making would aid adoption. Discrimination (AUC) is useful, yet calibration, decision-curve analysis (net benefit across threshold probabilities), and likelihood ratios at pragmatic cut-points are needed to estimate post-test probabilities at typical ED frailty prevalences [3]. Presenting calibration plots and simple nomograms or tables that map pre-test probability to post-test probability would let ED leaders weigh misclassification costs (e.g., consult demand, observation bed use) against potential gains in safer disposition. Third, reproducibility and a capacity-aware pathway will determine real-world performance. Triage environments are noisy, with proxy responses and evolving training [4]. A multicentre reproducibility study that reports inter-rater agreement (e.g., Cohen's kappa for the dichotomous Trigger items and intraclass correlation coefficient [ICC] where appropriate), with drift checks at three to 6 months and stratification by nurse experience, would separate signal from measurement noise. For implementation, a tiered pathway may be helpful: Trigger as step one, followed by a continuous measure such as the Clinical Frailty Scale (CFS) to guide standardized actions (early therapy assessment, medication review, delirium-prevention bundle). Embedding the Trigger in the Electronic Health Record (EHR) with automated audit and equity monitoring across deprivation strata would enable feedback loops without alert fatigue [5]. In summary, this study advances frailty identification at the first clinical contact. Ensuring independence from overlapping constructs, adding calibration and net-benefit reporting to quantify clinical utility, and confirming inter-rater reliability within a capacity-aware pathway may support safe, scalable adoption and—crucially—translate to improvements in patient outcomes and ED flow. Shyam Sundar Sah: conceptualization, methodology, writing – original draft, writing – review and editing. Abhishek Kumbhalwar: validation, supervision, project administration, writing – original draft, writing – review and editing. Generative AI tools, including Paperpal and ChatGPT 5, were utilized solely for language, grammar, and stylistic refinement. These tools had no role in the conceptualization, data analysis, interpretation of results, or substantive content development of this manuscript. All intellectual contributions, data analysis, and scientific interpretations remain the sole work of the authors. The final content was critically reviewed and edited to ensure accuracy and originality. The authors take full responsibility for the accuracy, originality, and integrity of the work presented. The authors have nothing to report. The authors declare no conflicts of interest. This article is linked to https://doi.org/10.1111/acem.70165. The authors have nothing to report.