Aims People experiencing homelessness (PEH) have been identified as being increasingly susceptible to Coronavirus disease (COVID-19), with policies enacted to test, isolate, increase hygiene practices and prioritise vaccines among this population. Here, we conduct a scoping review of the current evidence-base pertaining to the prevalence and presentation of COVID-19 in PEH, COVID-vaccine hesitancy rates and government interventions enacted within the first year of the pandemic for PEH. Materials and methods A systematic search was conducted on Pubmed, Cochrane, Embase and MedRxiv databases for studies reporting primary data on COVID-19 prevalence and clinical characteristics in PEH, vaccine uptake for PEH and policies enacted targeting PEH. Study qualities were assessed with The National Heart, Lung and Blood Institute’s set of Study Quality. Results Eighty-three studies were included in our final analysis. The overall prevalence of symptomatic COVID-19 infection in PEH is estimated at 35%. The most common symptoms found were cough and shortness of breath, followed by fever. Concerns regarding vaccine hesitancy amongst PEH related to thoroughness of COVID-19 vaccine clinical trials, side effects and mistrust of the government. The main strategies implemented by governments were mass testing, adaption of healthcare service provision, provision of alternative housing, encouraging personal hygiene (hand sanitation and mask wearing), and inter-organisational communication. Discussion In our meta-analysis, 35% of PEH with a COVID-19 infection presented symptomatically; the low prevalence of symptomatic COVID-19 infection suggests widespread testing following outbreaks would be beneficial for this group of individuals. Temporary recuperation units and measures for housing stability in the pandemic, namely provision of alternative housing and stopping evictions, were found to be highly effective. High rates of vaccine hesitancy means that education and encouragement towards vaccination would be beneficial for this vulnerable population, where comorbidities are common. Finally increased focus in research should be placed on the mental health burden of COVID-19 and the pandemic on PEH moving forwards.

Background Patients with anorexia nervosa (AN) are at higher risk of sudden cardiac death. Although the underlying aetiology for this association remains unclear. It may be related to prolongation of the QT interval, which can degenerate into fatal ventricular arrhythmias. However, the presence of prolonged heart rate-corrected QT interval (QTc) in AN remains controversial, and two previous meta-analyses on AN and QTc showed contradictory findings [1,2]. Purpose In this systematic review and meta-analysis, we aimed to evaluate if AN was associated with changes in the QTc interval and dispersion. Methods MEDLINE, EMBASE and COCHRANE databases were systematically searched from inception to January 2021. Random-effects meta-analysis and meta-regression were used. The inclusion criteria were (i) confirmed diagnosis of AN, (ii) measurement of QTc on electrocardiogram and (iii) peer-reviewed articles. The primary endpoint of the study was the duration of the QTc interval calculated using the Bazett (QTcB), Hodges (QTcH), Fridericia (QTcF) and Framingham (QTcFr) formulae. The secondary endpoints were QT dispersion (QTd) and QTc dispersion (QTcd). Results The 25 eligible studies included 5687 patients (1862 AN, 3825 control) (Figure 1: PRISMA diagram). The majority of patients were female (96.3%) with a mean age between 14.3 to 31.0 years and mean duration of disease ranging from 9.1 to 129.6 months. The mean BMI ranged from 13.7 to 18.5 kg/m2. Pooled analysis did not show significant prolongation between AN versus control in QTcB (mean difference (MD) MD 4.9ms, 95% CI −3.2, 13.1ms, p=0.23; I2=95%; n=24/25 studies; Figure 2A), QTcH (MD 1.3ms, 95% CI −8.5, 11.2ms, p=0.79; I2=71%; n=3/25 studies), and QTcF (MD 3.1ms, 95% CI −21.6, 27.7ms, p=0.81; I2=97%; n=3/25 studies). Only two studies reporting QTcFr showed a significant prolongation between AN and control (MD 15.9ms, 95% CI 0.0, 31.8ms, p=0.05, I2=65%; n=2/25 studies; Figure 2B). However, QTd and QTcd were significantly greater in AN than control (MD 21.3ms, 95% CI 10.4, 32.3ms, p=0.0001, I2=94%; Figure 2C and MD 16.9ms, 95% CI 4.5, 29.3ms, p=0.007 I2=93%; Figure 2D, respectively). Conclusion To the best of our knowledge, this is the largest meta-analysis of QTc in AN and the first meta-analysis exploring the significance of QTd and QTcd in AN. AN was not found to be associated with prolongation of QTc calculated using the Bazett, Fridericia and Hodges formulae. However, an association of AN with prolonged QTc was observed in the studies using the Framingham formula. More pronounced dispersion (QTd and QTcd) was also observed in patients with AN. Funding Acknowledgement Type of funding sources: None.