Contact tracing is a fundamental part of controlling the covid-19 pandemic. The vaccination rollout continues at pace, but basic public health strategies will be needed for months, possibly years. Finding the contacts of patients who are in hospital presents a particular challenge to NHS test and trace. Hospital cases of covid-19 have been at a peak over the past few weeks, so untraced contacts of in-patients is a problem which needs addressing. Many in-patients are too ill to answer phone calls or engage with remote contact tracers. Clinically diagnosed cases of covid-19 without laboratory confirmation are not approached by NHS test and trace.
Ward based conversations with an initial group of 37 in-patients suggested a substantial unmet need for both forward and backward contact tracing. Full provision is made in many Trusts for contact tracing of hospital-acquired infections. However, the only contact tracing provision for community acquired cases admitted with covid-19 is through NHS test and trace.
We conducted a pilot study to assess the effectiveness of NHS test and trace for in-patients and undertook contact tracing for a sample of 20 ward-based patients. We involved medical students, community volunteers, hospital staff, the local Health Protection Team and local authority staff. Medical student volunteers were recruited as part of their placement with a local community contact tracing group.  Training included use of patient databases, ward-based personal protective equipment, and procedures as well as contact tracing. Twenty in-patients were identified on infectious diseases wards of a major teaching hospital. Entry criteria were either clinical or laboratory confirmed covid-19. Clinically diagnosed patients met criteria which included epidemiological, radiological and/or haematological changes consistent with covid-19. Face to face interviews were conducted on the ward. Data on close contacts were sent to the local Health Protection Team for referral to NHS test and trace.
Of the 20 cases enrolled, 11 (55%) patients had been contacted by NHS Test and Trace of whom only seven (35%) engaged and gave details of their close contacts. 13 (65%) of the 20 patients studied were not engaged by NHS Test and Trace. Seven (35%) of the patients failed to engage with NHS Test and Trace because of illness, language, hearing, or cognitive barriers. four (20%) were not engaged because their clear clinical diagnosis had no laboratory confirmation.
All clinically diagnosed cases should be formally notified to the local “Proper Officer” in Public Health England and also to NHS Test and Trace and their contacts traced.  The failure to isolate these patients’ contacts will contribute to further outbreaks. Forty nine close contacts were identified of whom 33 were from the 13 (65%) cases that NHS Test and Trace had failed to engage. This means that 67% of all the contacts identified had not been traced by NHS Test and Trace. If this is the pattern nationally, it suggests a major omission. Volunteers collected contact details for a majority of close contacts.
“Backward” contact tracing information on possible sources of infection was given by 11 (55%) of whom 8 (40%) gave detailed tracking information including individuals, named schools, or health and care settings.  Two had been in a care home or hospital outbreak (detected in the community after discharge). These data can be crucial for outbreak control.
Medical student volunteers spent an average of one hour to use PPE, interview patients and collect data on a hospital database. With limited time, they did not undertake contact tracing themselves. Data on contacts were transferred to the City Council Contact Tracing service who had difficulties using information because they can only add data if patients already have an NHS Test and Trace computerised record. This means that in-patients and their contacts that have been missed by NHSTT cannot be registered by NHS Test and Trace. Local authority staff have repeatedly complained about this feature of CTAS, the national IT system. Financial and other support, normally available, cannot be offered to these index cases or their contacts.
UK mortality from covid-19 is high and contact tracing will be needed for months.  NHS Test and Trace misses the untested, false negatives as well as those cases it fails to reach.  Many of these missed cases are readily identifiable in hospital but hard for NHS Test and Trace to reach. Problems associated with contact tracing of in-patients have received little national publicity, despite rapidly rising inpatient numbers.  This has widespread implications for outbreak control and should be addressed urgently.
This small, zero budget study demonstrates the need for hospital based in-patient contact tracing, that a volunteer-based service is practical and effective, that a sustainable, salaried service should be considered and demonstrates an urgent need for increased flexibility and connectivity between hospital trusts, local Health Protection Teams and NHS Test and Trace. It is unacceptable that, at this late stage of the epidemic, two thirds of the contacts of in-patients with covid-19 appear not to have systematic advice and support.
Rachel Foster, Consultant in Infectious Diseases Sheffield Teaching Hospital Foundation Trust.
Andzelika Duda, 3rd year medical student University of Sheffield.
Abigail Reynolds, 3rd year medical student, University of Sheffield.
Ian Carey, RGN Sheffield, working as a volunteer.
Jack Czauderna, retired GP Sheffield.
Bing Jones, retired Associate Specialist in Haematology Sheffield.
Alex Westran, Operational Manager for Sheffield City Council Track and Trace.
Competing interests: none declared.
1. Sheffield Community Contact Tracers available at: http://www.communitycontacttracers.com/
2. Department of Health and Social Care (5th March 2020). Coronavirus (COVID-19) listed as a notifiable disease. [online] Available from: https://www.gov.uk/government/news/coronavirus-covid-19-listed-as-a-notifiable-disease
3. Endo A et al. (October 2020) Implication of backward contact tracing in the presence of overdispersed transmission in COVID-19 outbreaks. Wellcome Open Res 2020, 5:239 (https://doi.org/10.12688/wellcomeopenres.16344.1)
4. European Centre for Disease Prevention and Control, COVID-19 situation update worldwide, as of week 50 2020. Available from: https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-cases [accessed 21st December 2020]
5. Jones, B; Czauderna, J; Regrave, P. (2020) We must stop being polite about Test and Trace—there comes a point where it becomes culpable. London: BMJ Publishing Group Limited 2020. https://blogs.bmj.com/bmj/2020/11/10/we-must-stop-being-polite-about-test-and-trace-there-comes-a-point-where-it-becomes-culpable/
6. NHS (2020). COVID-19 Hospital Activity. [online] Available at: https://www.england.nhs.uk/statistics/statistical-work-areas/covid-19-hospital-activity/ [accessed 23/12/2020]