Duke-NUS Critical Revisits Study: Rare ED Risks in Singapore's Elderly

Overview of Duke-NUS's Groundbreaking Study on ED Revisits

Singapore's healthcare system faces mounting pressure from rising emergency department (ED) visits, particularly among its aging population. A recent study from Duke-NUS Medical School's Prehospital and Emergency Research Centre (PERC) sheds new light on a critical yet rare phenomenon: critical revisits after ED discharge. Published in the International Journal of Emergency Medicine, the research analyzed over one million patient discharges from Singapore General Hospital (SGH) between 2008 and 2020, revealing that while 72-hour ED revisits occur in 4.2% of cases, only 0.12% escalate to critical outcomes like intensive care unit (ICU) admission, high-dependency unit (HDU) admission, or death.

This finding underscores the generally high quality of ED decision-making in Singapore but highlights vulnerabilities, especially for elderly patients with multiple comorbidities. Charlson Comorbidity Index (CCI) scores—a validated measure assessing the burden of chronic conditions like diabetes, heart disease, and cancer—emerged as a key predictor, with patients having CCI ≥2 facing significantly higher risks.

Background: ED Revisits in Singapore's Evolving Healthcare Landscape

Emergency departments in Singapore handle a staggering volume of cases annually, with SGH alone processing around 350 daily attendances, 70% resulting in discharge. Amid an aging population—projected to see one in four Singaporeans aged 65+ by 2030—EDs grapple with complex presentations from patients with multimorbidity. Traditional quality metrics focus on 72-hour revisit rates, but most are non-critical and discharged again. Duke-NUS researchers shifted focus to 'critical revisits,' defined as those leading to death (68 cases), ICU (214), or HDU (1,039) admissions, comprising just 3% of all revisits.

This aligns with broader Singapore trends: all-cause 30-day readmission rates hover at 11.6% overall and 19% for elderly patients, straining resources in a system emphasizing efficient discharges. The Duke-NUS study provides granular insights, urging targeted interventions beyond blanket metrics.

Study Design and Methodology: Rigorous Data-Driven Analysis

Led by researchers from SGH's Department of Emergency Medicine and Duke-NUS's PERC and Centre for Quantitative Medicine, the retrospective cohort study mined electronic health records (EHR) from 1,057,533 ED discharges. Exclusions ensured focus on true discharges: no ward admissions, observation stays, or undocumented left-against-medical-advice cases without diagnosis. De-identified data captured demographics, vital signs (e.g., heart rate, mean arterial pressure), labs (bicarbonate, chloride, platelets, troponin, urea, white cell count), CCI, triage class, and diagnoses.

Multivariable Firth's logistic regression—suited for rare events—identified independent risk factors, with backward selection (p≥0.2 removal). This approach minimized bias in a skewed dataset where critical events were scarce.

Key Findings: Rarity and Demographics of Critical Revisits

Of 44,506 72-hour revisits (4.2%), 1,321 (0.12%) were critical. Critical patients averaged 56 years old (vs. 48 overall), with CCI 1.59 (vs. 0.396). Males predominated (aOR 1.30), as did those over 65 (aOR 1.68). Triage P2 cases (semi-urgent) showed elevated risk (aOR 1.35), while lower acuity reduced it.

• Deaths: 68 (0.006% of discharges)
• ICU admissions: 214 (0.02%)
• HDU admissions: 1,039 (0.10%)

These stats affirm robust ED processes but pinpoint at-risk groups for proactive care.

Risk Factors: Vital Signs, Labs, and Comorbidities Unpacked

The study pinpointed actionable predictors. Abnormal vitals like elevated heart rate (aOR 1.52) and mean arterial pressure (aOR 1.31) signaled instability. Labs flagged metabolic derangements: low bicarbonate (aOR 1.57), low chloride (aOR 2.11), low platelets (aOR 1.49), abnormal troponin (aOR 1.36), deviant urea, and high white cells (aOR 2.25).

Comorbidities like peripheral vascular disease and stroke amplified odds, while dementia and pulmonary issues oddly lowered them—possibly due to cautious discharge protocols.Explore research roles at Duke-NUS

Photo by Brett Jordan on Unsplash

High-Risk Diagnoses: Cerebrovascular and Abdominal Red Flags

Index visit diagnoses strongly predicted critical revisits. Acute cerebrovascular disease topped (OR 38.00), followed by gastrointestinal disorders (OR 3.10), unclassified codes (OR 2.69), genitourinary issues (OR 2.34), and abdominal pain (OR 1.41). Critical revisit diagnoses mirrored: cerebrovascular (15.4%), septicemia (5.8%), biliary disease (4.0%). Abdominal categories accounted for disproportionate burden.

• Step-by-step process: Clinicians triage, diagnose, discharge; revisit signals potential oversight or rapid deterioration.
• Cultural context: Singapore's fast-paced urban life delays symptom recognition in busy elderly.

Unclassified codes highlight diagnostic challenges, urging better documentation.

Duke-NUS PERC: Pioneering Prehospital and Emergency Integration

Launched in 2020 as Singapore's first such center, PERC at Duke-NUS bridges prehospital, hospital, and community care. Collaborating with SingHealth, it advances pathways via data-driven research like this study. Achievements include stroke care models and ED prediction tools, impacting national guidelines. PERC's quantitative expertise, via tools like Firth's regression, exemplifies Duke-NUS's role in translational med research.Duke-NUS faculty positions Visit PERC website

Implications for Singapore's Healthcare Policy and Practice

In a system with rising ED demands, this Duke-NUS research advocates risk-stratified follow-up. High-risk elderly could benefit from telehealth, hospital-at-home, or phone checks—proven to cut readmissions elsewhere. Aligns with Healthier SG initiatives emphasizing preventive care. Policymakers might integrate CCI into ED protocols, enhancing discharge safety without overburdening resources.

Stakeholder Perspectives: Clinicians, Patients, and Policymakers

Clinicians hail the study's nuance, moving beyond crude revisit rates. Patient advocates stress elderly support amid Singapore's filial piety culture. MOH data shows ED admissions rising with age, comorbidities fueling cycles. Duke-NUS PERC's work positions Singapore as a leader in evidence-based emergency care.

Future Outlook: Innovations and Actionable Insights from Duke-NUS

Ongoing PERC projects explore AI for readmission prediction, prehospital interventions. Actionable steps: Screen high-CCI elderly for follow-up; educate on abdominal symptom vigilance. Duke-NUS's ecosystem fosters such innovations, training clinician-scientists via its MD-PhD programs.Academic CV tips for med researchers

Photo by Brett Jordan on Unsplash

• Benefits of tele-follow-up: 20-30% readmission drop.
• Risks of inaction: Escalating costs, mortality.

Conclusion: Enhancing Patient Safety Through Research Excellence

Duke-NUS's critical revisits study illuminates safe ED discharges while flagging elderly multimorbidity risks. By leveraging PERC's expertise, Singapore can refine pathways, ensuring vulnerable patients thrive post-discharge. Explore opportunities at Rate My Professor, Higher Ed Jobs, University Jobs, or Career Advice to join this vital field.