We will use data from general practices linked to hospital data to investigate how infection with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), which causes COVID-19, has impacted the health of children, in both the short and long-term in the UK. We will analyse primary (GP consultations or drug prescriptions) and secondary healthcare use (hospital admissions, accident & emergency visits, or outpatient bookings) in children up to 3 months prior to and after a SARS-CoV-2 infection. We will also explore if these patterns of healthcare use are different to use before and after other respiratory infections, like influenza. Focusing on persistent, post-infection symptoms (like post-viral fatigue), we will describe how rates of these post-infection symptoms have changed over time since 2006, and if rates of persistent post-infection symptoms reported in GP data in children have increased significantly after the COVID-19 pandemic. National guidelines recommend that children over 5 years old have receive are vaccinated against COVID vaccines-19, particularly to reduce their risk of being admitted to hospital. In the last part of the study, we will look at whether available COVID-19 vaccines can prevent symptoms of SARS-CoV-2 infection that require children to use either primary or secondary care services.
SARS-CoV-2 infections are generally milder in children than adults. This may explain why research has focused comparatively less on children, and, until much later in the pandemic, in the development and evaluation of vaccines.
The overall aim is to examine healthcare use before and after SARS-CoV-2 infection in children, and to what extent COVID-19 vaccines prevent primary or secondary healthcare use following infection. A particular focus will be on persistent post-infection symptoms requiring healthcare contact following infection. We plan to analyse data from the Clinical Practice Research Datalink (CPRD) linked to Hospital Episode Statistics (HES), COVID-19 test results from the NHS Digital Second-Generation Surveillance system, and deprivation indicators for this study.
We will develop a cohort of children within CPRD and use this to examine rates of primary care and hospital admission rates up to 3 months following SARS-CoV-2 infection compared to 3 months prior. Using difference-in-difference analyses we will compare patterns of healthcare use 3 months before and after a SARS-CoV-2 positive test with patterns of use before and after influenza and RSV-related consultations.
By carrying out a time trends analysis using Poisson or negative binomial regression models, we will examine any changes during and after the H1N1 influenza and COVID-19 pandemics in the rate of persistent post-infection symptoms, defined as recorded symptoms relating to post-infectious fatigue or other post-infectious sequalae in primary care.
We will examine the effectiveness of COVID-19 vaccines in preventing healthcare use following SARS-CoV-2 infection, using Cox proportional hazards regression models in a matched cohort design, and self-controlled case series analyses.
This study will provide evidence to inform healthcare resource allocation for under-18s in the event of a new or existing pandemic-prone pathogen or emerging SARS-CoV-2 variant. Understanding the effectiveness of COVID-19 vaccines in preventing healthcare use could change or validate current COVID-19 vaccine recommendations.
For objectives 1, and 2, our outcomes will be (before/after infection) primary care consultation rates (overall; and for conditions grouped into the following categories: digestive; circulatory; respiratory; endocrine, metabolic, or nutritional; genitourinary; eye or ear; musculoskeletal; mental; skin; blood; and general or unspecified conditions). These will be derived from SNOMED/Read codes; prescription rates in primary care (for any prescription, antibiotics, asthma inhalers); these will be derived from therapy codes linked to BNF chapters; hospital admission rates (overall, total planned admissions, total unplanned admissions, and for specific diagnoses (grouped into chapters of the International Classification of Diseases version 10; (ICD-10); rates of accident and emergency attendances and outpatient department bookings (overall;- due to inconsistent diagnostic coding in these datasets we will not break down analyses by type of condition);
For objective 3, our outcome will be rates of consultations for persistent, post infection symptoms, defined using Read or SNOMED codes.
For objective 4, our outcomes will be primary care consultation rates related to SARS-CoV-2 infection; prescribing rates for antibiotics and asthma inhalers; hospital admission rates related to SARS-CoV-2 infections; hospital admission rates related to any respiratory condition.
Pia Hardelid - Chief Investigator - University College London ( UCL )
Vishnuga Raveendran - Corresponding Applicant - University College London ( UCL )
Claire Thorne - Collaborator - University College London ( UCL )
Deenan Pillay - Collaborator - University College London ( UCL )
Linda Wijlaars - Collaborator - University College London ( UCL )
Vishnuga Raveendran - Collaborator - University College London ( UCL )
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;Patient Level Index of Multiple Deprivation;SGSS (Second Generation Surveillance System);COVID-19 Linkages