Long COVID in non-hospitalised individuals: symptoms, risk factors and syndromes

Date of ISAC Approval: 
2021-04-27 00:00:00
Lay Summary: 
Some people who have survived COVID-19 develop longer-lasting symptoms, known as Long COVID. There is a lack of understanding of the cause and consequences of these symptoms. An understanding of the prevalence and causes of these symptoms may help to develop interventions and design health services tailored to address specific symptoms. We will look at anonymised electronic health records, held within a secure national database, to find non-hospitalised patients with COVID-19 and describe their demographic and clinical features. We will find out what symptoms they have and how prevalent the symptoms are in the immediate and longer-term after their infection with SARS CoV-2 (the virus causing COVID-19). We will examine if factors such as age, ethnic group, social background, weight, and co-existing health conditions influence whether a patient with COVID-19 develops Long COVID, i.e., longer-term symptoms from the infection. In patients with Long COVID, we will also examine which symptoms group together (cluster) and which clusters are most common. We will compare health service use and the number of deaths between patients from different symptom clusters. This work will inform a follow-up study that will invite patients to report their Long COVID symptoms, quality of life and work capability with the aim of developing supportive interventions and treatments. This study is part of the NIHR/UKRI funded Therapies for Long COVID in non-hospitalised individuals (TLC) Study.
Technical Summary: 
We will undertake a retrospective cohort study of individuals with COVID-19 who have had a positive reverse transcriptase-polymerase chain reaction (RT-PCR) test result for SARS-CoV-2, without a record of hospitalisation in the 28-day period following COVID-19 diagnosis, from 30th January 2020 to the most recent available date. We will describe the demographic and clinical characteristics of the cohort. We will estimate the prevalence of symptoms during periods of “acute COVID-19” (within four weeks of diagnosis), “ongoing symptomatic COVID-19” (four to twelve weeks from diagnosis), and “post-COVID-19 syndrome” or “Long COVID” (after twelve weeks of diagnosis) and compare the symptom prevalence to propensity score-matched patients without a diagnosis of COVID-19 within similar time intervals. A list of >60 symptoms has been developed through a rapid review of 24 primary research studies. These relate to multiple organ systems including cardiopulmonary, naso-oropharyngeal, gastroenterological, musculoskeletal and neuro-psychological. Those presenting with at least one symptom at 12 weeks or longer after COVID-19 diagnosis will be classified as having Long COVID for preliminary analyses. We will investigate risk factors for developing Long COVID among the cohort of non-hospitalised individuals with COVID-19 using logistic regression. We will then define unique symptom clusters using unsupervised clustering methods. We will compare clinical characteristics, hospital admissions after 28 days, intensive care unit (ICU) admissions and mortality rates between patients from different symptom clusters. This work will enable us to define a cohort of patients with confirmed non-hospitalised Long COVID-19 and matched controls who will be subsequently invited to a patient-consented prospective cohort study via their GPs with support from CPRD’s Interventional Research services. The follow-up study will be covered by a separate research ethics approval.
Health Outcomes to be Measured: 
(1) Symptoms in patients with COVID-19 (coded in health records) (2) Long COVID symptom clusters (3) A&E attendances (4) Hospital admissions (5) ICU admissions (6) All-cause and cause-specific mortality
Collaborators: 

Shamil Haroon - Chief Investigator - University of Birmingham

Shamil Haroon - Corresponding Applicant - University of Birmingham

Anuradhaa Subramanian - Collaborator - University of Birmingham

Georgios Gkoutos - Collaborator - University of Birmingham

Grace Turner- Collaborator - University of Birmingham

Joht Singh Chandan - Collaborator - University of Birmingham

Krishna Gokhale - Collaborator - University of Birmingham

Krishnarajah Nirantharakumar - Collaborator - University of Birmingham

Melanie Calvert - Collaborator - University of Birmingham

Nicola Adderley - Collaborator - University of Birmingham

Nikita Simms-Williams - Collaborator - University of Birmingham

Olalekan Lee Aiyegbusi - Collaborator - University of Birmingham

Puja Myles - Collaborator - CPRD

Tom Taverner - Collaborator - University of Birmingham

Linkages: 
CHESS, HES A&E, HES APC, ONS, Patient IMD, SGSS