Increase in hepatitis cases of unknown aetiology in children: a rapid OpenSAFELY analysis to support the national incident response¶
In April 2022, the World Health Organisation was notified of 10 cases of severe acute hepatitis of unknown aetiology in children aged under 10 in Scotland. Following further investigations, as of 8 April, 74 cases were identified across the United Kingdom since January 2022, 49 of which were in England (1). Possible causes included COVID-19 and adenovirus; it was thought that changes in patterns of circulating virus due to COVID-19 pandemic regulations could have also contributed to this outbreak (1). In response, the UK Health Security Agency (UKHSA) initiated a national incident and alerted the public to be on the lookout for signs and symptoms in children (2).
OpenSAFELY is a secure health analytics platform that allows near real-time analysis of pseudonymised primary care patient records in England. We have previously shown that the OpenSAFELY platform can be used for rapid audit and feedback for national health incidents (3). To support the incident team's investigation, we were asked to use OpenSAFELY to provide contextual information on children presenting in general practice for symptoms related to hepatitis. This included any changes in the recording of gastrointestinal (GI) illness or liver functions tests that may be ordered in the diagnosis of hepatitis.
With the approval of NHS England, we therefore conducted a rapid analysis to describe the recording of hepatitis-related activity in primary care in children for consideration by the incident team formed at UKHSA.
We used the OpenSAFELY framework to conduct a retrospective cohort study of hepatitis, gastrointestinal illness and liver function tests across the full pseudonymised patient primary care records held by the EHR provider TPP, covering 40% of general practices in England.
Between 1st April 2017 and March 31st 2022, at the beginning of each month, we identified individuals aged <=30 years registered at a GP practice. In these patients, we then identified anyone with a SNOMED CT code recorded indicating hepatitis, GI illness or any of the following liver function tests: alanine aminotransferase (ALT); aspartate aminotransferase (AST); bilirubin. To allow rapid analysis, codelists for identifying hepatitis and GI illness were pragmatically developed using the SNOMED CT hierarchy; hepatitis was defined as any child code of âInflammatory disease of liverâ (SNOMED CT: 128241005) and GI illness defined as any child code of âDisorder of the digestive systemâ (SNOMED CT: 53619000). All of the codelists used in this analysis are shown in Table 1 and are openly available on opencodelists.org. To investigate any relationship with COVID-19 infection, any patients with a positive COVID-19 test in the 3 months prior were identified.
Table 1. Codelists used in this study
|Alanine aminotransferase (ALT)||Link|
|Aspartate aminotransferase (AST)||Link|
For the above indications, we describe the absolute count and rate per 1000 patients by age group. For AST and ALT tests, we identified the number of tests out of range (>500 IU/L), as set out in the WHO working case definition (4). For bilirubin tests, the number of tests out of range were calculated using the upper reference range attached to the lab result. We also investigate the number of tests out of range by COVID-19 status, as well as the mean test value by age group, to potentially indicate any shift in test result values.
The count of recording of GI illness codes between April 2017 and March 2022 is shown in Figure 1 (top). There was a large decrease following the onset of the COVID-19 pandemic across all age groups which gradually returned to pre-pandemic levels by March 2022. The monthly count of recording of hepatitis diagnostic codes between April 2017 and March 2022 is shown in Figure 1 (bottom). There was a small decrease in those aged 21-30 years following the onset of the COVID-19 pandemic in April 2020 but there was otherwise no marked change. The monthly number of diagnostic codes recorded in patients aged 6-10 years, 3 months-5 years and 0-3 months were all <=7, and are consequently redacted from Figure 1. There was no obvious change in the pattern of recorded hepatitis or GI illness since January 2022.
Figure 1. The monthly count of recorded codes relating to (top) GI illness and (bottom) hepatitis diagnosis between April 2017 and March 2022 by age group. Note: age bands with monthly counts <=7 have been redacted. All counts are rounded to the nearest 5.
The rate of recording of liver function tests between April 2017 and March 2022 is shown in Figure 2. As with hepatitis and GI illness diagnoses, a sharp decrease in coding rate was observed in April 2020 for each of the liver function tests observed. Coding of ALT and bilirubin tests returned to pre-pandemic levels by November 2021. The rate of AST tests remain below pre-pandemic levels in those aged 21-30 and 11-20, but have recovered in other age groups. There was a small increase in the rate of ALT and bilirubin tests across all age groups between January 2022 and March 2022.
Figure 2. The monthly rate per 1000 patients of recorded codes relating to (top) ALT tests, (middle) AST tests and (bottom) bilirubin tests between April 2017 and March 2022 by age group. Note: age bands with monthly counts <=7 have been redacted. All counts are rounded to the nearest 5 before rates are calculated.
There was no evidence of any changes in the number of test values out of range for any of the liver function tests since January 2022. Due to the small counts of test values out of range a breakdown by COVID-19 status is not presented, but no clear difference was observed.
We rapidly provided contextual information on children presenting in general practice for symptoms related to hepatitis. We presented the findings to the incident response team at UKHSA for manual review to inform their response.
Findings in context¶
The changes observed in primary care activity following the onset of the COVID-19 pandemic are reflective of wider changes in primary care activity which we have observed previously (5). Despite the increase in identified cases of hepatitis of unknown aetiology across the UK, we did not observe any clear indication of changes in primary care activity in TPP practices since January 2022. This could be explained by insufficient representation of cases in TPP practices for a signal to be observed. It may also be an indication of the severity of the case definition; it is likely that any patients matching the specified case definition presented directly in secondary care, which could have resulted in reduced or delayed recording of cases within primary care. Indeed, UKHSA reported a small increase in the number of children aged 1 to 4 years admitted to hospital with a diagnosis code that may represent non-A-E hepatitis in February and March 20222 (6). Similarly, there is some indication of increased emergency care department attendance with clinical codes related to liver disease in children aged 1 to 4 and to a lesser extent in children aged 5 to 14 years (6).
Strengths and weaknesses¶
The main strength of this study is the speed and scale of its delivery. This study was rapidly implemented across the full EHR coded data covering 40% of the English population. Such a timely audit would have been challenging prior to the establishment of the OpenSAFELY platform. A limitation of this study is the use of non-specific codelists. Use of the SNOMED CT hierarchy was a practical decision that allowed prompt development of codelists for this analysis in the context of a rapid incident response during a global pandemic, but it likely captured non-specific codes such as hepatitis of known cause, so the results presented should be interpreted accordingly. Improved identification of hepatitis indications can be developed.
This study demonstrates that OpenSAFELY can be used for near real time surveillance. The approach taken here can be further developed to allow future responses.
We are very grateful for all the support received from the TPP Technical Operations team throughout this work, and for generous assistance from the information governance and database teams at NHS England and the NHS England Transformation Directorate.
Information Governance and Ethics¶
All data were linked, stored and analysed securely using the OpenSAFELY platform, https://www.opensafely.org/, as part of the NHS England OpenSAFELY COVID-19 service. Data include pseudonymised data such as coded diagnoses, medications and physiological parameters. No free text data are included. All code is shared openly for review and re-use under MIT open license https://github.com/opensafely/hepatitis_in_children. Detailed pseudonymised patient data is potentially re-identifiable and therefore not shared.
NHS England is the data controller of the NHS England OpenSAFELY COVID-19 Service; TPP is the data processor; all study authors using OpenSAFELY have the approval of NHS England (7). This implementation of OpenSAFELY is hosted within the TPP environment which is accredited to the ISO 27001 information security standard and is NHS IG Toolkit compliant (8).
Patient data has been pseudonymised for analysis and linkage using industry standard cryptographic hashing techniques; all pseudonymised datasets transmitted for linkage onto OpenSAFELY are encrypted; access to the NHS England OpenSAFELY COVID-19 service is via a virtual private network (VPN) connection; the researchers hold contracts with NHS England and only access the platform to initiate database queries and statistical models; all database activity is logged; only aggregate statistical outputs leave the platform environment following best practice for anonymisation of results such as statistical disclosure control for low cell counts (9).
The service adheres to the obligations of the UK General Data Protection Regulation (UK GDPR) and the Data Protection Act 2018. The service previously operated under notices initially issued in February 2020 by the the Secretary of State under Regulation 3(4) of the Health Service (Control of Patient Information) Regulations 2002 (COPI Regulations), which required organisations to process confidential patient information for COVID-19 purposes; this set aside the requirement for patient consent (10). As of 1 July 2023, the Secretary of State has requested that NHS England continue to operate the Service under the COVID-19 Directions 2020 (11). In some cases of data sharing, the common law duty of confidence is met using, for example, patient consent or support from the Health Research Authority Confidentiality Advisory Group (12).
Taken together, these provide the legal bases to link patient datasets using the service. GP practices, which provide access to the primary care data, are required to share relevant health information to support the public health response to the pandemic, and have been informed of how the service operates.
3. Fisher L, Speed V, Curtis HJ, Rentsch CT, Wong AY, et al. Potentially inappropriate prescribing of DOACs to people with mechanical heart valves: a federated analysis of 57.9 million patientsâ primary care records in situ using OpenSAFELY. 2021 Jul [cited 2021 Nov 10] p. 2021.07.27.21261136.