Woodlands Health Campus

Frailty is highly prevalent in hospitalized older adults and predicts adverse health outcomes but remains under-recognized. Manual screening tools previously employed were challenging in high-volume settings and reliant on individuals' knowledge on frailty. With the development of electronic health records (EHRs), there is a potential to automate screening for frailty in hospitalized older adults. We introduce a quality improvement initiative that utilizes an EHR-automated Best Practice Advisory (BPA) alert to identify inpatients who may benefit from geriatric intervention and encourage timely Geriatric referral to a Mobile Frailty Intervention Team (MFIT). MFIT was piloted at Woodlands Hospital, an integrated acute and community hospital in Singapore. BPA was automatically triggered to encourage referral to MFIT if any of the following criteria were met a) Clinical Frailty Scale (CFS) ≥ 7 b) CFS 5-6 with presence of either of cognitive impairment / high falls risk / high readmission risk c) Age ≥60 with presence of delirium, regardless of CFS status. The MFIT team conducted Comprehensive Geriatric Assessments (CGA) as part of routine review with the diagnosed geriatric syndromes and discharge dispositions recorded. On MFIT review, 81.3 % (N = 248) of patients referred had a geriatric syndrome and 68.5 % (N = 209) had multiple syndromes. MFIT further identified syndromes which may be neglected during acute admission such as underlying dementia (87 %, N = 83), osteoporosis (13.4 %, N = 41) and urinary incontinence (9.8 %, N = 30). MFIT also provided discharge recommendations which were adhered to in 79.2 % (N = 232) of patients. 32.8 % (N = 100) were given specialized outpatient clinic follow-up with Geriatric medicine, suggesting the potential to divert patients from away from primary care services and unplanned readmissions. Further studies are needed to investigate whether this transition effectively optimizes resource allocation. A CFS-based BPA alert may be feasible in providing an automated and scalable method to identify hospitalized older adults with frailty that would benefit from timely geriatric intervention.

Orthopaedic surgeons are the custodians of the musculoskeletal system in Singapore, with >350 orthopaedic surgeons looking after the whole continuum of musculoskeletal disease in a population of 6 million. Orthopaedic research in Singapore currently has 4 focus areas: tissue engineering, biomechanics, clinical registries and cohorts, and population health and health services research. We have identified 4 key enablers of the continued development of orthopaedic research: talent development and academic clinical programs; shared data infrastructure, national cohorts, and artificial intelligence; innovation; and interdisciplinary, industry, and international collaboration. This is an exciting time for orthopaedic research in Singapore, where we find ourselves at the cusp of a new wave of talent, ideas, and resources. We stand ready and excited to partner with the world to advance musculoskeletal care globally.

Summary: While armed assailant attacks in a healthcare setting are rare, they pose a potential risk to healthcare staff, patients, visitors, and the infrastructure. They can disrupt health care services and deprive people of life-saving treatment. Many hospitals have well-developed disaster plans to respond to a mass casualty incident (MCI) occurring outside the hospital. Hospital staff receive emergency preparedness training to respond to MCI as part of their job scope, either through online modules or participation in disaster drills. A lack of an armed assailant incident response plan can significantly reduce the hospital’s ability to appropriately respond to such an incident. The random, short-lasting but violent nature of an armed assailant incident makes healthcare response extremely challenging. Along with the response to the incident, the other challenges also include the management of the assailant as well as the safety and management of the patients, public, and healthcare staff during the incident. The primary objectives of the response plan should be: • threat assessment and pre-incident planning • initiating immediate response during an incident to minimize the assailant’s access to potential victims • damage and post-incident recovery and support plans During the attack, based on the best available evidence, adopting the model of “Run, Hide, Fight” or “Secure, Preserve, Tell” are recommended. Care provision to the injured during this event should be remembered by the acronym THREAT. • T- Threat Suppression • H- Hemorrhage Control • RE - Rapid Extrication to safety • A - Assessment by medical providers • T- Transport to definitive care Once the attack has ended, there must be an assessment of the department/hospital’s ability to continue routine operations. Various stakeholders of the health care facility should work together to develop an armed assailant incident response plan that best suits their needs and capacity. This will be of immense benefit for keeping healthcare facilities safe.

Background: Diabetic foot ulcers (DFUs) and lower-extremity amputations (LEAs) complicating diabetes mellitus are significant contributors to morbidity, mortality, and disease burden. There is insufficient evidence, however, linking podiatric care to mortality and healthcare resource use. There is, concurrently, inadequate access to podiatric care, particularly in Asia. This study evaluated the clinical and healthcare use outcomes of patients with DFUs who received podiatric care. Methods: A longitudinal study involving patients in Diabetic Foot in Primary and Tertiary (DEFINITE) Care followed over a period of 2 years reviewed the effectiveness of podiatric care with regard to 1-year outcomes using multiple logistic and zero-inflated negative binomial regressions, adjusting for covariates. Clinical outcomes were minor and major LEA, mortality, and LEA-free survival rates; healthcare use outcomes were the number of admissions, number of visits to clinics and emergency departments, and length of stay. Results: Eligible patients (n = 2798 [65.5%]) completed at least 12 months of follow-up. Comparisons were made with patients without access to podiatric care. The overall mean ± SD patient age was 65.7 ± 12.7 years. Most patients receiving podiatric follow-up were of Malay or Indian ethnicities and had poor diabetic control and chronic kidney disease. There were more admissions (p < 0.01) and visits to the emergency department (p < 0.01) and hospital outpatient clinics (p < 0.01), but a shorter length of stay (incidence rate ratio, 0.833; p < 0.01), without any reported differences regarding visits to public primary care clinics (p = 0.68). There were more minor LEAs (p < 0.01) but fewer deaths (p < 0.01) and greater LEA-free survival (odds ratio, 1.26; p = 0.02). Conclusions: Podiatric care improved clinical outcomes for patients with DFU, particularly in terms of LEA-free survival, with a positive impact on healthcare utilization. This is a crucial contribution to the limited evidence on podiatric care in diabetic limb salvage in Asian populations.

Background/Aims Fear of childbirth, also known as tokophobia, significantly impacts a woman's mental health during pregnancy and birth, but how this fear changes or manifests through pregnancy and childbirth is under-researched. This study aimed to determine factors associated with fear of childbirth among pregnant women in Malaysia. Methods A cross-sectional study was conducted with a convenience sample of 243 pregnant women at Hospital Universiti Sains Malaysia. Data were collected using the Childbirth Fear Questionnaire. Pearson's Chi-squared test and Fisher's exact test were used to examine the association between fear and participants' demographic and obstetric characteristics. Results More than one-third (38.3%) of participants had moderate fear of childbirth, while 11.1% had severe fear. Fear of childbirth was associated with age (P=0.044). Almost all participants reported that their fear had a low impact on their life (93.4%). This impact was associated with age (P=0.008) and parity (P=0.019). Conclusions Almost half of the participants experienced moderate or severe fear of childbirth, which was linked with their age. However, most did not find that it seriously impacted their daily life. Implications for practice Healthcare professionals should incorporate routine screening for fear of childbirth, particularly among younger and first-time mothers, and integrate tailored antenatal education with brief psychological interventions to reduce fear and promote positive birth experiences.

<h3></h3> The Medical Oncology department in TTSH is one of the largest Oncology departments in Singapore. ACP improves quality of life and is especially relevant for patients with cancer. ACP rates have however been on the downward trend, with only one ACP completed among inpatients between 2022–2023. We aimed to improve ACP referral rate among Oncology inpatients with advanced cancer. We focused on patients admitted to the Oncology inpatient service who had metastatic cancer or on palliative cancer treatment. Patients in collusion, lacking mental capacity or critically ill and those with existing ACP were excluded. At baseline in June 2024, no patient was referred for ACP despite 71 inpatients being suitable. 56 causes for low ACP uptake were identified. Following voting and plotting of Pareto chart, the top causes by 80–20 rule included (a) No screening criteria, (b) Focus on acute medical issues and (c) Heavy inpatient and clinic workload. We addressed these causes sequentially in Plan-Do-Study-Act cycles. In July 2024, we defined screening criteria and added a section on ACP to daily rounding notes. A mean referral rate of 24.4% was achieved. Following feedback from junior doctors, our intervention in August 2024 involved weekly reminders to senior doctors. The mean referral rate however dropped to 12.9%. In September 2024, we switched focus to addressing causes (b) and (c). We adopted a multidisciplinary approach with a senior doctor screening cases weekly and an Advanced Practice Nurse (APN) assisting with ACP discussion. There was provision of dedicated time for the APN and majority of inpatients were referred for completion of ACP in our outpatient ACP clinic. Mean referral rate improved to 28.7%. Over 4 months, referral rate for ACP successfully increased from near 0% to 23% and ACP completion improved significantly. With this submission, we share our experience and challenges and highlight interventions that were successful in improving referral rates in our Oncology unit in an attempt to increase knowledge and guide strategies to improve ACP uptake among Oncology patients worldwide. We propose that interventions helpful for improving ACP rates include clear definition of referral criteria, multidisciplinary involvement and provision of dedicated time, having majority of ACPs completed as outpatient and having ACP champions to sustain efforts. Reminders to the healthcare team were insufficient to improve ACP rates. Oncology APNs have now been integrated into our inpatient teams and this represents a significant step in our department’s efforts to improve ACP rates.

<h3>Aim or Purpose of the Presentation</h3> To share a ward-based quality improvement initiative aimed at increasing Advance Care Planning (ACP) completion in a newly established tertiary hospital, using the ACP-F framework to guide patient identification and promote integration of ACP into routine clinical care. <h3>Background</h3> Woodlands Health, Singapore’s newest tertiary hospital, began operations in May 2024. A baseline review showed that only 1% of inpatients had a completed Preferred Plan of Care (PPC), despite 30–50% of patients being appropriate for ACP. In Singapore, ACP includes General ACP (GACP) for future preferences and PPC for current care decisions. The ACP-F framework—Advanced Organ Failure, Advanced Cancer, Progressive Neurodegenerative Diseases, and Frailty—was used to screen patients suitable for PPC. This project was motivated by the need to embed ACP early into a new system’s clinical workflows and culture. <h3>Method</h3> A six-month quality improvement project began in September 2024 in a 40-bedded general medical ward. Interventions included the development of screening workflows using the ACP-F framework, mentoring junior doctors, engaging ward-based palliative resource nurses, and reinforcing ACP in daily multidisciplinary rounds. Staff received education and bedside coaching to enhance skills and confidence. By March 2025, the ACP/PPC completion rate in the ward rose from 1% to a peak of 4.4%. Although the original 10% target was not met, a culture shift was evident, with more consistent patient screening and initiation of discussions. Temporary dips occurred during festive periods and bed crunch due to system-level pressures. Institution-wide, 279 ACP discussions (226 PPCs, 53 GACPs) were completed. Nurses led 48% of all ACPs, with APNs driving PPC. General ACP was primarily facilitated by nurses (64%), followed by allied health (32%) and doctors (4%). <h3>Unique Contribution</h3> This submission provides insight into the early integration of ACP in a new tertiary hospital using a structured, condition-based screening approach (ACP-F). It demonstrates the impact of engaging nurses and embedding ACP into daily workflows. It also highlights real-world barriers such as staffing pressure and ward transitions, often overlooked in early implementation phases. <h3>Implications</h3> This experience suggests that ACP integration is feasible from the outset in new healthcare systems. Embedding ACP into routine care, supported by a multidisciplinary approach and nurse empowerment, enhances sustainability. Policymakers and clinical leaders should consider structured screening tools and nurse-led facilitation models to scale ACP uptake and normalize conversations across care settings.

Misdiagnosis causes significant harm to healthcare systems worldwide, leading to increased costs and patient risks. MedRAG is a smart multimodal healthcare copilot equipped with powerful large language model (LLM) reasoning, designed to enhance medical decision-making. It supports multiple input modalities, including non-intrusive voice monitoring, general medical queries, and electronic health records. MedRAG provides recommendations on diagnosis, treatment, medication, and follow-up questioning. Leveraging retrieval-augmented generation enhanced by knowledge graph-elicited reasoning, MedRAG retrieves and integrates critical diagnostic insights, reducing the risk of misdiagnosis. It has been evaluated on both public and private datasets, outperforming existing models and offering more specific and accurate healthcare assistance. A demonstration video of MedRAG is available at: https://www.youtube.com/watch?v=PNIBDMYRfDM. The source code is available at: https://github.com/SNOWTEAM2023/MedRAG.