Dynamic linkage of COVID-19 test results between Public Health England's Second Generation Surveillance System and UK Biobank

Abstract

UK Biobank (UKB) is an international health resource enabling research into the genetic and lifestyle determinants of common diseases of middle and older age. It comprises 500 000 participants. Public Health England's Second Generation Surveillance System is a centralized microbiology database covering English clinical diagnostics laboratories that provides national surveillance of legally notifiable infections, bacterial isolations and antimicrobial resistance. We previously developed secure, pseudonymized, individual-level linkage of these systems. In this study, we implemented rapid dynamic linkage, which allows us to provide a regular feed of new COVID-19 (SARS-CoV-2) test results to UKB to facilitate rapid and urgent research into the epidemiological and human genetic risk factors for severe infection in the cohort. Here, we have characterized the first 1352 cases of COVID-19 in UKB participants, of whom 895 met our working definition of severe COVID-19 as inpatients hospitalized on or after 16 March 2020. We found that the incidence of severe COVID-19 among UKB cases was 27.4 % lower than the general population in England, although this difference varied significantly by age and sex. The total number of UKB cases could be estimated as 0.6 % of the publicly announced number of cases in England. We considered how increasing case numbers will affect the power of genome-wide association studies. This new dynamic linkage system has further potential to facilitate the investigation of other infections and the prospective collection of microbiological cultures to create a microbiological biobank (bugbank) for studying the interaction of environment, human and microbial genetics on infection in the UKB cohort.

Keywords: COVID-19; Public Health England; SARS-CoV-2; UK Biobank; bugbank; database linkage.

Fig. 1.

Information flow for identifying infection events among UKB participants in SGSS and issuing laboratory alerts to retrieve those micro-organisms. (1) A laboratory results file is received by SGSS from an NHS or PHE laboratory. (2) Hourly, an agent checks SGSS for new UKB infection events, and adds any to a separate database in PHE (Bugbank). The agent copies specimen and AMR susceptibility records. (3) Periodically, an extract of the data is transferred securely to UKB for incorporation into their system. (4) Daily, the agent sends an email alert to each active NHS or PHE laboratory. The email contains minimal information necessary for the laboratory to retrieve micro-organisms from UKB participant infections. (5) A secure SharePoint site provides a front end to view each laboratory’s specimens in PHE’s Bugbank database and records whether each micro-organism has been recovered, is missing or the record veracity has been questioned by the laboratory.

Fig. 2.

Example email alert to retrieve micro-organisms cultured from UKB participants’ infections. The alert is sent automatically from PHE Colindale to the NHS or PHE laboratory. It contains minimal information necessary to retrieve the micro-organisms.

Fig. 3.

Summary of UKB infection events at the John Radcliffe Hospital, Oxford, during September 2019. The summary indicates the volume of events by date (left), the time of day (middle) and the time to record receipt in SGSS (right).

Fig. 4.

Demographic features of individuals with COVID-19 among English UKB participants (top) and all individuals in the PHE SGSS (bottom). Only hospital inpatients are shown, since these cases can be inferred as severe COVID-19, because only severe cases were admitted to hospital from 16th March 2020 onwards. COVID-19 is determined by positive PCR for SARS-CoV-19. Panels show the total number of new cases per day (left), cumulative number of cases (middle), and the age and sex distribution of cases (right). The dark grey shaded region (left and middle panels) highlights the reporting lag period for some cases, assessed as around 5 days.

Fig. 5.

Incidence of SARS-CoV-2 positive individuals in England, 16th March – 19th May 2020. Per capita incidence is show for all English inpatients (a). Relative incidence (per capita incidence in each age group divided by mean per capita incidence across age groups) is compared between non-UKB inpatients, UKB inpatients and UKB non-inpatients for females (b) and males (c). Vertical black lines indicate 95 % CI values calculated assuming a Poisson distribution for the underlying counts. Incidence was calculated using the known age and sex distribution for England as a whole, and English UKB participants.

Fig. 6.

Impact of alternative definitions of severe COVID-19 on estimated effects of age and sex on incidence. PCR-positive hospitalized inpatients are considered to represent severe cases. Two methods of identifying inpatients are compared, the proposed working definition (green) and a more specific but less sensitive method (grey). For each definition, the effects of age and sex on the odds of severe COVID-19 were estimated using fisher.test in R and compared.

Fig. 7.

Power calculations for GWASs. The smallest odds ratio (case/control status versus risk/protective allele) detectable with 80 % power is shown as the number of cases increases from 200 to 10 000, assuming a genome-wide significance threshold of 5×10⁻⁸.