COVID-19 diagnoses in South East London peaked before the UK suggesting early measures reduced transmission.

Dear Editor,

The COVID-19 pandemic has followed differing trajectories both in amplitude and timing between nations and even within countries). In the UK, London was affected more severely and earlier than some other parts of the country. Measures undertaken by individual governments to control the pandemic require analysis in relation to their timing in order to assess their impact on the spread and mortality of the disease. For example, the UK government introduced its most stringent measures to control transmission of COVID-19 on March 23, 2020. However, prior to this date several important steps had been advocated and followed including avoidance of travel to work, voluntary self-isolation of symptomatic individuals and a reduction in public gathering. Analysis of COVID-19 data from London may elucidate the extent to which the early measures were effective in slowing the local pandemic.

King's College Hospital (KCH) in South-East London has treated amongst the highest number of COVID-19 cases to date in the UK (N = 2339 on April 17, 2020). Since February 2020, data relating to COVID-19 diagnoses (positive RT-PCR) and hospitalisations to KCH were gathered prospectively. We analysed the frequency distribution of daily new COVID-19 diagnoses and hospitalisations at KCH from March 18 to April 17, compared these to UK national data and explored the relationship between the stringency of preventive measures and local COVID-19 epidemiology. We used non-linear regression to curve-fit the data and found that the daily incidence data conformed to a Gaussian (normal) distribution (Shapiro-Wilk test of normality). One-way ANOVA with Tukey's test for multiple comparisons was used to compare the peak incidence dates to the UK national case-load trajectory. Data are presented as mean and 95% confidence intervals.

New COVID-19 diagnoses, new COVID-19 hospitalisations (both at KCH) and new cases (UK national data) are shown in Fig. 1 (a)–(c) respectively. At KCH, the peak in COVID-19 diagnoses and admissions occurred 12.4 (11.4–13.3) days 12.8 (11.9–13.6) days respectively after March 18. The peaks in both new diagnoses (p<0.0001) and admissions (p<0.0001) occurred 12–13 days in advance of the calculated national peak (25.2 [22.9–29.8] days after March 18). There was no difference in the peak incidence dates between COVID-19 diagnoses and admissions, which is to be expected as at our institution and within the UK as a whole, testing for COVID-19 during this time had been performed almost exclusively on patients requiring hospital admission.

Fig. 1

Daily number of COVID-19-related new diagnoses (a) and new admissions to King's College Hospital (b), and new diagnoses across the UK (c). Curves show the best-fit and 95% confidence band.

Our data illustrate that new diagnoses and hospitalisations within South-East London peaked almost two weeks in advance of the rest of the UK. Following transmission, symptoms of COVID-19 may develop after an incubation period of 4–5 days. , The median time from onset of symptoms to hospitalisation is 11 (IQR 8–14) days. Therefore, it can be estimated that the time interval from transmission to hospitalisation for the majority of patients is 12–19 days. Our data indicate that the peak in new admissions occurred only 8 days after the formal ‘lockdown’ within the UK, and that the deceleration and subsequent decline in new admissions to KCH occurred before the effect of the formal ‘lockdown’ could have had its full impact. Moreover, our data suggest that the early measures to reduce transmission had a major beneficial impact on the COVID-19 epidemic in South London. These data emphasise the importance of assessing the dynamics of the UK COVID-19 pandemic at a local level to understand the impact of measures taken to control transmission.

Declaration of Competing Interest

None of the authors have a conflict of interest.