Early detection of patients at risk for MET (Medical Emergency Team) activations is crucial for timely intervention. Threshold-based reference model used in general wards has low sensitivity in detecting clinical deterioration. Its fixed thresholds make early trend prediction difficult, whereas a predicted probability approach provides a more dynamic assessment, where a higher probability indicates a higher risk of deterioration. We describe a machine learning model that enables the identification of patients at risk of clinical deterioration or death 22 h prior to the event, which is 10 h earlier than the current reference (baseline). A retrospective single-center Asian cohort study was conducted from 2019 to 2020 of inpatients in Singapore. A total of 2755 patients with 6496 cases/admissions were included. An XGBoost (XGB) model was trained and tested to compare the performance of the machine learning model to the reference model using sensitivity, specificity, and area under the curve for receiver operating characteristic (AUC-ROC). The XGB model with all vital signs shows highest ROC-AUC (91.7%) and good balance between sensitivity (88.7%) and specificity (88%). The reference model shows highest specificity (98.5%), but very low sensitivity (74.2%) and moderate ROC-AUC (86.4%). The key advantages of machine learning models are they can predict clinical deterioration with higher accuracy (more than 12%) and 10 h earlier than the reference model. In addition, machine learning models require less vital sign inputs for prediction where the reference model requires all vital sign signals. This study also finds that respiration rate and oxygen saturation are primary features in early prediction of clinical deterioration. This aligns with clinical knowledge that abnormalities in these vital signs often serve as early indicators of critical conditions such as respiratory failure or sepsis.

Objective:To evaluate the characteristics, antecedents, and outcomes of in-hospital cardiac arrests (IHCAs) in a Singaporean tertiary hospital. Design:We conducted a retrospective review of electronic medical records of all IHCAs that occurred in the general ward from January to December 2022 at Changi General Hospital. Based on pre-arrest characteristics, IHCAs were categorised as "potential suboptimal end-of-life planning (SELP)", "potentially predictable", or "sudden and unexpected". Main outcome measures:Primary outcomes were IHCA incidence and survival to hospital discharge. Secondary outcomes included the proportion of potential SELP cases and the frequency of pre-arrest extent of care (EOC) documentation. Results:There were 50 IHCAs among 52 617 admissions (0.95/1000 admissions). The median age was 72.5 (interquartile range [IQR], 63-81) years, and 82% had a Clinical Frailty Scale score of ≥ 4. Thirty (60%) arrests fulfilled criteria for potential SELP, while EOC documentation was present in 30% of patients. Five patients (10%) survived to hospital discharge. Factors associated with survival included shorter cardiopulmonary resuscitation duration (p = 0.004), witnessed arrests (p = 0.019), and sustained return of spontaneous circulation (ROSC) (p = 0.005). Conclusions:Our data corroborate that witnessed arrests, shorter downtime, and sustained ROSC are associated with better outcomes. However, the high prevalence of potential SELP and low rates of documented EOC reveal critical gaps in end-of-life planning within a frail inpatient population. Earlier, routine EOC discussions may help align care with patient goals, optimising resource utilisation and reducing potentially avoidable IHCAs.

Rationale: The optimal duration of systemic corticosteroid therapy for hospitalized asthma exacerbations remains unclear. Our objective was to assess whether the blood eosinophil count can be used to safely shorten the duration of systemic corticosteroid therapy in hospitalized asthma exacerbations. Methods: In an open-label, randomized trial at two centers, adults hospitalized for asthma exacerbation were assigned, in a 1:1 ratio, to usual care or blood eosinophil-directed care. Before receiving any systemic corticosteroids, all participants had their blood eosinophil count measured. The usual care group received 5 days of prednisolone, while the eosinophil-directed group received prednisolone based on eosinophil count: a 3-day course if the count was <300 cells/µL, or a 5-day course if ≥300 cells/µL. The primary outcome was non-inferiority of eosinophil-directed care compared to usual care in treatment failure rates, defined as the need to extend the duration of systemic corticosteroid therapy, mechanical ventilation, or death within the index admission. A priori, the study was designed with 80% power to ensure that a one-sided 97.5% confidence interval would exclude a treatment failure rate difference favoring the usual care group by more than 20%. Results: In total, 110 patients were randomized, with 55 assigned to the eosinophil-directed group and 55 to usual care. 60% of exacerbations were eosinophilic, and 40% non-eosinophilic. The mean cumulative dose of systemic steroid prescribed was significantly lower in non-eosinophilic patients compared to eosinophilic patients for the eosinophil-directed group (136 vs. 214 mg prednisolone-equivalent, p=0.0004), but not for the usual care group (186 vs. 211 mg prednisolone-equivalent, p=0.18). Treatment failure occurred in 4/55 or 7.3% of the usual care group, and in 6/55 or 10.9% of the eosinophil-directed group. The primary noninferiority outcome was met, with a 3.6% difference in treatment failure rates between the eosinophil-directed and usual care groups (95% confidence interval: -8.9% to 16.2%). No deaths occurred in this study. Comparing eosinophil-directed care to usual care, there were no significant differences in length of hospital stay, or in changes in Asthma Control Questionnaire-5 from baseline at 7, 14, 30, and 90 days. Additionally, there was no significant difference between the groups in the rate of additional systemic corticosteroid prescriptions within 14 days. The 30- and 90-day cumulative incidences of pneumonia, sepsis, venous thromboembolism, fractures, and gastrointestinal bleeding were also similar between groups. Conclusion: Compared to usual care, eosinophil-directed care is non-inferior for guiding the duration of systemic corticosteroid therapy in hospitalized asthma exacerbations.

Chronic obstructive pulmonary disease (COPD) has a high burden in Asia. These patients are also susceptible to various cardiovascular diseases (CVD). A panel of expert Asian pulmonologists explored the published literature to understand the impact of COPD and CVD on each other and to identify the cardiopulmonary risk factors in the region. The experts concluded that an elevated risk of all-cause mortality and acute cardiovascular events persists for up to 2 years following moderate and severe COPD exacerbations, with the risk of death being highest in the first 30 days after the exacerbation. High smoking rate (especially in males), high indoor and outdoor air pollution in Asia, relatively low vaccination rate in Asia (especially in low- and middle-income countries), and relatively low rate of utilisation of inhaler medications impact the cardiopulmonary risk in Asia.

Background A set of interventions in a hand hygiene change package was developed in a pilot ward by the end of 2017. In 2018, Changi General Hospital embarked on scaling up the change package to other wards with the intention to eventually spread the hand hygiene change package hospital-wide.Methods Changi General Hospital conducted a quality improvement project on hand hygiene with the intention to effect organisation-wide improvement in hand hygiene. Spread methodologies such as the Institute for Healthcare Improvement’s framework for Spread and various complementary spread concepts such as having an organisational strategy, which plans for spread as early as possible, and addressing social aspects of change were applied in order to scale up and spread a change package.Setting A general tertiary care hospital in Singapore.Results Overall hospital-wide hand hygiene compliance improved from a median of 66% during the pilot phase to 73% in the scale-up phase (p<0.05) to 82% during the spread phase (p<0.05).Conclusions A systematic approach to hand hygiene improvement based on spread literature successfully improved and sustained hospital-wide hand hygiene compliance. Success factors included the development of a change package that had clear guiding principles, with the intent to create proactive learning cycles within units which could be adapted to work in various contexts.

Introduction: The 30-day readmission rate for chronic obstructive pulmonary disease (COPD) is a common performance metric but may be confounded by factors unrelated to quality of care. Our aim was to assess how sociodemographic factors, multimorbidity and frailty impact 30-day readmission risk after COPD hospitalisation, and whether risk adjustment alters interpretation of temporal trends. Method: This is a retrospective analysis of administrative data from October 2017 to June 2023 from Changi General Hospital, Singapore. Multivariable mixed-effects logistic regression models were used to estimate unadjusted and risk-adjusted 30-day readmission odds. Covariates included age, sex, race, Charlson Comorbidity Index (CCI), Hospital Frailty Risk Score (HFRS) and year. Temporal trends in readmission risk were compared across unadjusted and adjusted models. Results: Of the 2774 admissions, 749 (27%) resulted in 30-day readmissions. Higher CCI (CCI≥4 versus [vs] CCI=1: adjusted odds ratio [aOR] 2.00, 95% confidence interval [CI] 1.33–2.99, P=0.003; CCI 2–3 vs CCI=1: aOR 1.50, 95% CI 1.15–1.96, P=0.001) and higher HFRS (≥5 vs <5: aOR 1.29, 95% CI 1.01–1.65, P=0.04) were independently associated with increased readmission risk. While unadjusted analyses showed no significant temporal trends, the risk-adjusted model revealed a 32–35% reduction in readmission odds in 2021–2023 compared to baseline. Conclusion: Multimorbidity and frailty significantly impact COPD readmissions. Risk adjustment revealed improvements in readmission risk not evident in unadjusted analyses, emphasising the importance of applying risk adjustments to ensure valid performance metrics.

Background:In 2019 and 2023, the Global Initiative for Chronic Obstructive Lung Disease (GOLD) provided updated strategies for modifying the therapy of patients with chronic obstructive pulmonary disease (COPD) and high exacerbation risk. A key update since the 2019 guidelines recommends considering blood eosinophil count to guide decisions on inhaled corticosteroid (ICS) treatment. To evaluate the potential impact of these updated recommendations, this study aimed to assess how extensively future practice would diverge from contemporaneous prescribing practices at a single center in Singapore, assuming adherence to the 2019 and 2023 GOLD guidelines. Methods:Retrospective cohort analysis of the Changi General Hospital COPD data warehouse involving patients aged ≥40 years hospitalized for a COPD exacerbation (October 2018-April 2020) receiving long-acting muscarinic antagonist (LAMA), LAMA plus a long-acting beta2-agonist (LABA), or an ICS plus LABA at admission. The proportion of patients eligible for treatment escalations per GOLD 2019 and 2023 recommendations was calculated. Results:In total, 268 patients were included (mean age 73 years; 91% male). At admission, 19%, 59%, and 22% of patients were receiving LAMA, LAMA + LABA, and ICS + LABA, respectively. Overall, 226 patients would have been eligible for treatment escalation per GOLD 2019 or 2023 recommendations; 31 (13.7%) had treatment escalations consistent with GOLD 2019 guidelines and 34 (15%) received treatment escalations consistent with GOLD 2023 guidelines. A total of 205 patients (76.5%) remained on the same treatment regimen at hospital discharge as they were receiving at admission. Lower measured post-bronchodilator forced expiratory volume in 1 second was associated with treatment escalations that would have been GOLD-concordant (P=0.028), as was increased number of emergency department/hospital visits in the last year (P=0.048). Conclusions:Compared with real-world clinical practice, a significantly higher proportion of patients may be eligible for treatment escalation under the GOLD 2019 and 2023 eosinophil-directed algorithms.

BACKGROUND:Sensitisation to Aspergillus fumigatus is linked to worse outcomes in patients with COPD; however, its prevalence and clinical implications in domestic (residential) settings remains unknown. METHODS:Individuals with COPD (n=43) recruited in Singapore had their residences prospectively sampled and assessed by shotgun metagenomic sequencing including indoor air, outdoor air and touch surfaces (a total of 126 specimens). The abundance of environmental A. fumigatus and the occurrence of A. fumigatus (Asp f) allergens in the environment were determined and immunological responses to A. fumigatus allergens determined in association with clinical outcomes including exacerbation frequency. Findings were validated in 12 individuals (31 specimens) with COPD in Vancouver, Canada, a climatically different region. RESULTS:157 metagenomes from 43 homes were assessed. 11 and nine separate Aspergillus spp. were identified in Singapore and Vancouver, respectively. Despite climatic, temperature and humidity variation, A. fumigatus was detectable in the environment from both locations. The relative abundance of environmental A. fumigatus was significantly associated with exacerbation frequency in both Singapore (r=0.27, p=0.003) and Vancouver (r=0.49, p=0.01) and individuals with higher Asp f 3 sensitisation responses lived in homes with a greater abundance of environmental Asp f 3 allergens (p=0.037). Patients exposed and sensitised to Asp f 3 allergens demonstrated a higher rate of COPD exacerbations at 1-year follow-up (p=0.021). CONCLUSION:Environmental A. fumigatus exposure in the home environment including air and surfaces with resulting sensitisation carries pathogenic potential in individuals with COPD. Targeting domestic A. fumigatus abundance may reduce COPD exacerbations.

Background: Airway mycobiomes in COPD demonstrating increased Aspergillus, Curvularia and Penicillium associated with poorer clinical outcomes, however, their relationship to environmental exposure in the home remains unexplored. Methods: Patients with stable COPD (n=43) and healthy participants (n=28) were prospectively recruited and underwent clinical data collation and home studies including deep shotgun metagenomic assessment of host (sputum), indoor air (bedroom), outdoor air (balcony) and surface dust (fan or air-conditioner filter) (4 metagenomes per home; Total 284 metagenomes). Results: Shotgun metagenomics reveals increased Aspergillus fumigatus, Pseudomonas alcaligenes and Brevibacterium luteolum in indoor air (bedroom) and Malassezia restrica and Aspergillus spp. in surface dust and outdoor air (balcony) in the homes of individuals with COPD. Relative abundance of indoor air Aspergillus fumigatus positively correlates with exacerbation frequency in COPD (r=0.4, p=0.009) however no differences were observed in symptom burden or lung function. Conclusion: Aspergillus spp., present at increased levels in the home environment is associated to COPD exacerbations. Environmental exposures in the home represent a source of Aspergillus-related outcomes in COPD. Funding: Singapore Ministry of Health's NMRC under its Clinician Scientist Award (CSA) (MOH-000710) (S.H.C) & Singapore Ministry of Education under its AcRF Tier 1 Grant (RT1/22) (S.H.C).

BackgroundVariable clinical outcomes are reported with fungal sensitisation in chronic obstructive pulmonary disease (COPD), and it remains unclear which fungi and what allergens associate with the poorest outcomes. The use of recombinant as opposed to crude allergens for such assessment is unknown.MethodsA prospective multicentre assessment of stable COPD (n=614) was undertaken in five hospitals across three countries: Singapore, Malaysia and Hong Kong. Clinical and serological assessment was performed against a panel of 35 fungal allergens including crude and recombinantAspergillusand non-Aspergillusallergens. Unsupervised clustering and topological data analysis (TDA) approaches were employed using the measured sensitisation responses to elucidate if sensitisation subgroups exist and their related clinical outcomes.ResultsAspergillus fumigatussensitisation was associated with increased exacerbations in COPD. Unsupervised cluster analyses revealed two “fungal sensitisation” groups. The first was characterised byAspergillussensitisation and increased exacerbations, poorer lung function and worse prognosis. Polysensitisation in this group conferred even poorer outcome. The second group, characterised byCladosporiumsensitisation, was more symptomatic. Significant numbers of individuals demonstrated sensitisation responses to only recombinant (as opposed to crude)A. fumigatusallergens f 1, 3, 5 and 6, and exhibited increased exacerbations, poorer lung function and an overall worse prognosis. TDA validated these findings and additionally identified a subgroup withinAspergillus-sensitised COPD of patients with frequent exacerbations.ConclusionAspergillussensitisation is a treatable trait in COPD. Measuring sensitisation responses to recombinantAspergillusallergens identifies an important patient subgroup with poor COPD outcomes that remains overlooked by assessment of only crudeAspergillusallergens.

Aims This study sought to elucidate the occupational health risk perception and psychological impact during the early phase of the COVID-19 pandemic on healthcare workers in a general hospital in Singapore, and factors that influenced risk perception and psychological impact. Methods Healthcare workers from a general hospital in Singapore were invited to participate in an online survey in June 2020. It posed questions on demographic and occupational information (age, gender, nationality, marital status, profession, working area, length of working experience in healthcare), 20 items on occupational health risk perception and psychological impact of COVID-19, and the Depression Anxiety and Stress Scale-21 (DASS-21). The 20 items were adapted from a previous study during the 2003 Severe Acute Respiratory Syndrome (SARS) outbreak and designed to assess participants’ perceived exposure risk, risk acceptance, families’ perception, stigmatisation, feelings of appreciation, workload, and perceived effectiveness of workplace protective measures. Participants’ responses were obtained on a 6-point Likert scale (strongly agree, agree, somewhat agree, somewhat disagree, disagree, strongly disagree). For data analysis, responses on occupational risk perception were regrouped into three levels. Depression, anxiety, and stress scores were categorised into quartiles. Ordinal logistics regression was used to compare the association of occupational risk perception with DASS-21 scores, and demographic factors with occupational risk perception. Variables that showed statistical significance (set at P <0.05) in univariate analysis were included in the multivariate ordinal logistics regression model to identify independent predictors. Results There were 1252 respondents (92 doctors, 661 nurses, 318 allied health professionals, 181 administrative and support personnel). 85% felt an increased risk of exposure to COVID-19 while 90% accepted the risk as part of their jobs. Stigmatisation against healthcare workers was present, with 45% reported they were shunned and 21% reported their families were avoided. 78% experienced increased workload. Fortunately, most (94%) found workplace protective measures adequate, and felt appreciated by their employer (87%) and society (81%). Increased perception of occupational health risk was significantly associated with nursing profession, workers in patient-facing areas, and staff with shortest working experience in healthcare. The mean DASS-21 scores were 9.2 (borderline normal) for Depression, 8.5 (borderline mild) for Anxiety, and 10.9 (normal) for Stress. Increased DASS-21 scores were significantly associated with greater occupational risk perception, younger age, and less years of working experience. Conclusion Occupational risk perception amid the early COVID-19 pandemic is associated with adverse mental health among healthcare workers. Nurses, younger staff, and staff with least working experience are more vulnerable.

Aim: Rapid response systems (RRS) are present in many acute hospitals in western nations but are not widely adopted in Asia. The influence of healthcare culture and the effect of implementing an RRS over time are infrequently reported. We describe the introduction a RRS into a Singa-porean hospital and the barriers encountered. The efferent limb activation rates, cardiac arrest rates and unplanned intensive care unit (ICU) admis-sions are trended over eleven years.Methods: We conducted a retrospective observational study using prospectively collected data derived from administrative and Medical Emergency Team (MET) databases.Results: The RRS used a MET with a single parameter track and trigger and physician led efferent limb. Barriers encountered included clinical leadership buy-in, assembling and equipping the efferent team, maintaining a non-punitive mindset, improving accessibility to MET and communi-cating the impact of the MET. Over an 11-year period with 488,252 hospital admissions, MET activation rates increased from 1.6/1000 admissions (2009) to 14.1/1000 admissions (2019). Code blue activations and unplanned ICU admission rates decreased from 2.9 to 1.7 and from 8.8 to 2.0/1000 admissions, respectively over the 11 years. There were associations between increasing MET activation rate and reduction in code blue activations (p = 0.013) and unplanned medical ICU admission rates (p = 0.001).Conclusion: Implementing, sustaining and continued improvement of an RRS in Singapore is possible despite challenges encountered. With increasing activation rates over a decade, there were reductions in cardiac arrest rates and unplanned medical ICU admissions.

BACKGROUND:The coronavirus disease 2019 (COVID-19) affects almost all countries in the world and it impacts every aspect of people's life-physically, mentally, and socio-economically. There are several research studies examining the impact of this pandemic on health, however, very few studies examining the impact of this pandemic on quality of life. This study aimed to investigate the association between proximity to the COVID-19 and quality of life of healthcare workers and identify factors influencing quality of life.METHODS:A cross-sectional study was conducted among hospital staff in a tertiary hospital in Singapore. Data on demographic, medical history, lifestyle factors, psychosocial factors, and quality of life were collected using online self-administered questionnaire. Quality of life (QoL) was measured by the WHOQOL-BREF questionnaire. Robust linear regression was used to determine factors associated with quality of life.RESULTS:A total of 1911 participants were included in the analysis. The average age of participants was 38.25 (SD = 11.28) years old. 26.90% of participants had been quarantined, hospitalised, being suspected or diagnosed of having COVID-19 infection and they were found to have the lowest levels of QoL across all four domains (physical, psychological, social, and environmental domains). Participants who were singles or nurses, worked in shifts or worked longer hours, had chronic diseases were likely to have lower QoL scores compared to participants in other categories. Healthy lifestyle, social connectivity, resilience, social and workplace support were associated with higher QoL scores.CONCLUSIONS:In planning of measures which aim to improve QoL of healthcare workers, priority should be given to individuals who have been quarantined, hospitalised, being suspected, or diagnosed of having COVID-19 infection. In addition to the proximity of the COVID, lifestyle and psychosocial factors contribute to QoL of healthcare workers. Hence, multifaceted interventions are needed to improve QoL of healthcare workers.

ObjectivesThe long-term clinical trajectory of chronic obstructive pulmonary disease (COPD) in terms of year-to-year hospital utilisation rates can be highly variable and is not well studied. We investigated year-to-year trends of hospitalisation or emergency department (ED) visits among patients with COPD over 3 years, identified distinct trajectories and examined associated predictive factors.DesignA retrospective cohort study.SettingData were extracted from the Changi General Hospital, Singapore COPD data warehouse.ParticipantsPatients with COPD aged ≥40 years with 3 years of follow-up data.Primary and secondary outcome measuresThe yearly rates of hospitalisations or ED visits, stratified by COPD-related or all-cause, were described. Group-based trajectory modelling was used to identify clinically distinct trajectories year-by-year. Baseline predictive factors associated with different trajectories were examined.ResultsIn total, 396 patients were analysed (median age 70 years; 87% male). Four trajectories were generated for year-to-year trends in COPD-related hospitalisations/ED visits (C1–C4: consistently frequent, consistently infrequent, improving and worsening); post-bronchodilator forced expiratory volume in 1 second (FEV1) was a significant predictor of trajectory, with worse lung function being the main factor associated with less favourable trajectories. For all-cause hospitalisations/ED visits, four trajectories were identified (A1–A4: infrequent and stable, frequent and stable, frequent and decreasing, frequent and increasing); significant differences in age (p=0.041), sex (p=0.016) and ethnicity (p=0.005) were found between trajectories. Higher overall comorbidity burden was a key determinant in less favourable trajectories of all-cause hospitalisations/ED visits.ConclusionsDistinct trajectories were demonstrated for hospitalisations/ED visits related to COPD or all causes, with predictive associations between FEV1and COPD trajectory and between comorbidities and all-cause trajectory. Trajectories carry nuanced prognostic information and may be useful for clinical risk stratification to identify high-risk individuals for preventative treatments.