We read with great interest the article by Lee and Ryu (2025), ‘Factors Associated With the Survival of Older Patients With Pneumonia in the Emergency Department: A Retrospective Observational Study’. In this single-centre cohort of 327 patients aged 65 years or older, the authors used routinely collected triage and clinical variables to identify factors associated with in-hospital survival, including sex, Korean Triage and Acuity Scale (KTAS) level, oxygen supplementation, level of consciousness, body temperature, and the presence of malignancy. By focusing on information that emergency nurses already obtain at first contact, rather than on complex pneumonia-specific scores, the study offers a pragmatic picture of how risk is currently recognised at the front door and highlights concrete levers for refining geriatric pneumonia triage in everyday practice. Interestingly, the paper does more than enumerate prognostic factors; it gently challenges several long-held assumptions. It is striking that a body temperature above 37.5°C was associated with better survival, whereas normothermia or hypothermia predicted worse outcomes, echoing evidence that blunted febrile responses in older adults can signal immune frailty rather than mild disease. Equally thought-provoking is the strong association between oxygen supplementation in the emergency department and mortality, a pattern that mirrors findings from other emergency cohorts in which more invasive initial oxygen support tracked with in-hospital death across age groups (Chongthanadon et al. 2023). The prominent role of consciousness level—especially the ‘painful response’ category—as an independent predictor suggests that even subtle changes in mental status, often dismissed as ‘baseline confusion’, may in fact point to critically reduced physiological reserve. Finally, the marked impact of underlying malignancy on survival dovetails with studies showing that cancer history is a central determinant of pneumonia outcomes in very old patients, not just a background comorbidity (Baek et al. 2020). Together, these observations invite us to see geriatric pneumonia less as a uniform entity and more as a syndrome in which host vulnerability, rather than infection alone, drives prognosis. At the same time, we believe the authors' important work opens several avenues for refinement that could help future research move from association toward actionable tools for triage and care planning. Our intention is not to criticise the study retrospectively but to suggest directions that build on its strengths. First, from a methodological point of view, the use of in-hospital survival as a binary outcome is a logical starting point for a retrospective analysis, yet it under-uses the rich temporal information embedded in the data. Time-to-event approaches could capture not only who dies but also when death occurs, distinguishing very early deterioration from later in-hospital mortality and allowing exploration of how emergency department length of stay and timing of ward transfer interact with triage variables (Baek et al. 2020). Such models would complement the current logistic regression and align with recent geriatric pneumonia work showing that traditional scores lose discrimination in very elderly patients, suggesting that age-tailored dynamic risk models are needed (Baek et al. 2020). Second, although the study shows that initial KTAS level correlates with survival, KTAS is not formally evaluated as a predictive tool in its own right. Future work might report discrimination and calibration for KTAS-based risk stratification and compare this performance with widely used pneumonia indices, while also testing whether simple ‘add-on’ variables—such as malignancy status, body temperature pattern and consciousness level—meaningfully improve risk classification in older adults. This would speak directly to the clinically important question of how much value nurse-led triage categories add beyond pneumonia-specific tools for an ageing emergency population. The need to re-examine KTAS performance in older adults is underscored by recent data showing that its association with severity and outcomes is weaker in patients over 65 years than in younger adults, with a higher rate of up-triage in the older group (Chung et al. 2023). Third, the authors appropriately acknowledge that limiting the cohort to admitted patients and excluding those who died in the emergency department may bias estimates toward survivors of the most critical phase. A natural extension would be a parallel analysis including emergency department deaths, comparing factors associated with very early mortality and later in-hospital death. Such an approach would clarify whether oxygen requirement and depressed consciousness are primarily markers of imminent collapse or indicators of a broader, prolonged vulnerability. At the same time, incorporating measures of baseline function and frailty into triage models may further sharpen prognostic value. Recent work on a Clinical Frailty Scale–adjusted KTAS has shown that combining frailty assessment with standard triage levels improves prediction of hospital outcomes in older patients, suggesting a promising direction for frailty-informed pneumonia triage (Chung et al. 2024). Finally, the single-centre nature of the present study is an understandable starting point but also a reminder that the most powerful implications of this work are system-level. As many health systems move rapidly into a super-aged era, there is a strong case for multicentre validation across hospitals with different resources and case-mix profiles, using harmonised definitions and shared analytic code. Such studies could test geriatric-specific KTAS cut-offs, evaluate frailty-adjusted or comorbidity-adjusted triage schemes, and examine how proposed refinements perform in urban and rural emergency departments, high- and low-volume settings, and institutions with varying nurse-to-patient ratios (Chung et al. 2023). If coupled with prospective data on early interventions—such as timing of antibiotics, escalation to higher levels of care and goals-of-care discussions—this research agenda would move from describing risk to triggering timely, evidence-based responses for the growing population of older adults presenting with pneumonia. In summary, Lee and Ryu provide a timely and clinically grounded analysis of survival predictors among older pneumonia patients using variables that are immediately accessible to emergency nurses and physicians. Their work highlights that consciousness level, body temperature, oxygen requirement and oncologic history are not mere background details but key signals in geriatric pneumonia care. By layering time-to-event methods, formal performance metrics, inclusion of emergency department deaths, frailty-adjusted triage and multicentre validation onto this solid foundation, future studies can translate these insights into refined triage tools and protocols that better serve older adults with acute respiratory infection. Wenqing He: conceptualization, writing – original draft and editing. Jinpeng Weng: conceptualization, writing – original draft and editing. Zilin Zhao: conceptualization, writing – review and editing. Thank all the authors for their contributions to this article. The authors have nothing to report. Generative AI Statement: The author(s) declare that no Generative AI was used in the creation of this manuscript. Ethical approval was not required for this work as it did not involve human or animal subjects. No data collection from human participants or animals occurred. All authors listed above have reviewed the final version of this manuscript and consent to its submission for publication. We affirm that this work is original, has not been published previously, and is not under consideration elsewhere. If accepted, all authors approve its publication in its present form and will grant the Journal an exclusive licence to publish the article. The authors declare no conflicts of interest. The authors have nothing to report.
He et al. (Fri,) studied this question.