"This book provides a comprehensive look at many different aspects of response in Singapore to the pandemic in the crucial first several months, including clinical, laboratory, epidemiology, research, community engagement and the unprecedented challenge of outbreak involving migrant workers in dormitory settings. On a personal note, it has first-hand accounts of staff at the NCID who were at the forefront of battling COVID-19 in Singapore. It also gives a global perspective of the pandemic, together with insights into the unique Singapore experience of managing the pandemic. The Singapore response to the pandemic has been something which the global community has been very interested in and this book is the first to comprehensively describe that response from a number of different angles which will be useful to scientists, clinicians, public health professionals and policy makers"--
IntroductionRecent clinical trials have provided clear evidence to support early anti‐retroviral therapy (ART) in patients with HIV/TB co‐infection and low CD4 counts. We investigated how this has changed treatment and outcomes in Singapore.Materials and MethodsA retrospective review was performed with inpatient and outpatient records for all subjects diagnosed with HIV/TB co‐infection from 2006 to 2011 attending the Tuberculosis Control Unit, Tan Tock Seng Hospital, Singapore. Data for subjects with a presenting CD4<200 cell/mm3 were extracted and split into two groups, "Delayed": ART more than 8/52 after starting TB treatment, and "Early": ART within 8/52 of starting TB treatment.ResultsOne hundred thirty‐four out of 180 subjects in the database met the inclusion criteria for this study, 89 in the delayed group and 45 in the early. No statistically significant differences in baseline demographics between the two groups were identified. Both groups presented with markedly low CD4 counts, with overall 60% <50cells/mm3. Median CD4 counts were lower in the delayed ART group (37 vs 50, p=0.015). Prevalence of other opportunistic infections at TB diagnosis was not significantly different (20%), but TB in the early ART group was more likely to include extra‐pulmonary disease (46% vs 57%, p=0.038). Four cases were culture negative, 2 multi‐drug resistant and 10 (7.8%) were isoniazid mono‐resistant. There was a significant trend to earlier ART with more recent TB diagnosis (p<0.001). In the first 365 days after TB diagnosis, 11 deaths occurred in the delayed ART group, and 0 in the early (p=0.033). A Kaplan‐Meier time‐to‐event analysis demonstrated a clear separation in the frequency of death or opportunistic infections at eight weeks (Figure 1, p<0.001). Immune reconstitution disease was significantly more likely in the early ART group, but did not result in death (9% vs 38%, p<0.001). Treatment interruptions due to adverse drug events (ADE) developed in a median of 25 days (IQR 15–43). Interestingly, early ART was associated with a significantly lower number of treatment interruptions attributed to ADEs, with a higher proportion of patients completing two months of pyrazinamide induction (66% vs 85%, p=0.054) and rifampicin consolidation (79% vs 95%, p=0.03) – after excluding resistance or death. A trend to longer duration TB treatment was observed with delayed ART.ConclusionsSignificant improvements in HIV/TB infection outcomes correlate with the switch to earlier ART.
BACKGROUND: COVID-19 is a novel pandemic affecting almost all countries leading to lockdowns worldwide. In Singapore, locally-acquired cases emerged after the first wave of imported cases, and these two groups of cases may have different health-seeking behavior affecting disease transmission. We investigated differences in health-seeking behavior between locally-acquired cases and imported cases, and within the locally-acquired cases, those who saw single versus multiple healthcare providers. METHODS: We conducted a retrospective study of 258 patients who were diagnosed with COVID-19 from 23 January to 17 March 2020. Variables related to health-seeking behavior included number of visits prior to hospitalization, timing of the first visit, duration from symptom onset to admission, and places where the cases had at least one visit. RESULTS: Locally-acquired cases had longer duration from onset of symptoms to hospital admission (median 6 days, interquartile range [IQR] 4–9) than imported cases (median 4 days, IQR 2–7) (p < 0.001). Singapore residents were more likely to have at least one visit to private clinics and/or government-subsidized public clinics than non-residents (84.0% vs. 58.7%, p < 0.001). Among locally-acquired cases, those who sought care from a single healthcare provider had fewer visits before their hospital admissions compared with those who went to multiple providers (median 2 vs. 3, p = 0.001). CONCLUSION: Our study indicates the need to encourage individuals to seek medical attention early on in their patient journey, particularly from the same healthcare provider. This in turn, would facilitate early detection and isolation, hence limiting local transmission and enabling better control of the COVID-19 outbreak.
In: Bulletin of the World Health Organization: the international journal of public health = Bulletin de l'Organisation Mondiale de la Santé, Band 99, Heft 2, S. 92-101
Background. Few studies have validated the effectiveness of public health interventions in reducing influenza spread in real-life settings. We aim to validate these measures used during the 2009 pandemic. Methods. From 22 June to 9 October 2009, we performed a prospective observational cohort study using paired serum samples and symptom review among 3 groups of Singapore military personnel. "Normal" units were subjected to prevailing pandemic response policies. "Essential" units and health care workers had additional public health interventions (eg, enhanced surveillance with isolation, segregation, personal protective equipment). Samples were tested by hemagglutination inhibition; the principal outcome was seroconversion to 2009 influenza A(H1N1). Results. In total, 1015 individuals in 14 units completed the study, with 29% overall seroconversion. Seroconversion among essential units (17%) and health care workers (11%) was significantly lower than that in normal units (44%) (P < .001). Symptomatic illness attributable to influenza was also lower in essential units (5%) and health care workers (2%) than in normal units (12%) (P = .06). Adjusted for confounders, unit type was the only significant variable influencing overall seroconversion (P < .05). From multivariate analysis within each unit, age (P < .001) and baseline antibody titer (P = .012) were inversely related to seroconversion risk. Conclusions. Public health measures are effective in limiting influenza transmission in closed environments.
Background. Few studies have validated the effectiveness of public health interventions in reducing influenza spread in real-life settings. We aim to validate these measures used during the 2009 pandemic. Methods. From 22 June to 9 October 2009, we performed a prospective observational cohort study using paired serum samples and symptom review among 3 groups of Singapore military personnel. "Normal" units were subjected to prevailing pandemic response policies. "Essential" units and health care workers had additional public health interventions (eg, enhanced surveillance with isolation, segregation, personal protective equipment). Samples were tested by hemagglutination inhibition; the principal outcome was seroconversion to 2009 influenza A(H1N1). Results. In total, 1015 individuals in 14 units completed the study, with 29% overall seroconversion. Seroconversion among essential units (17%) and health care workers (11%) was significantly lower than that in normal units (44%) (P < .001). Symptomatic illness attributable to influenza was also lower in essential units (5%) and health care workers (2%) than in normal units (12%) (P = .06). Adjusted for confounders, unit type was the only significant variable influencing overall seroconversion (P < .05). From multivariate analysis within each unit, age (P < .001) and baseline antibody titer (P = .012) were inversely related to seroconversion risk. Conclusions. Public health measures are effective in limiting influenza transmission in closed environments.