Impact of social, economic, and demographic factors in disparity in clinical coding of long COVID in national electronic health records

Long COVID is a spectrum of new or persistent symptoms that can occur following infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) also known as COVID-191. In the UK context, this includes ongoing symptomatic COVID-19 and PostCOVID-19 syndrome2. Ongoing symptomatic COVID-19 is defined as experiencing signs and symptoms of COVID-19 between 4 weeks and 12 weeks following infection. PostCOVID-19 syndrome is defined as experiencing signs and symptoms of COVID-19 that develop during or after an infection consistent with COVID-19, continue for more than 12 weeks and are not explained by an alternative diagnosis. Both of these long COVID conditions can negatively impact an individual's quality of life3. Accurately recording long COVID clinical codes has become a priority to better support those living with the condition4. The use of long COVID codes in Electronic Health Records (EHRs) has improved over time5. Despite this, long COVID appears to remain under recorded in EHRs6. Disparities have been reported in recording of long COVID codes. For example, the prevalence of coded long COVID was found to be different in Primary Care Practices using EMIS and TPP (SystmOne) EHR software7. Patient demographic factors also appear to be associated with recording of long COVID codes6-8. Previously difference in the prevalence of long COVID code recording had been reported in different geographical areas across England7. In this case the disparity is likely in part due to the differences in EHR Software being used. However, to ensure the use of these codes improves over time, more research is needed to highlight any inequalities that may exist in the use of these codes. This project aims to determine the impact of social, economic, and demographic factors on long COVID coding in EHRs. We aim to investigate potential disparity in coding of long COVID conditions across urban and rural locations as well as coastal and non-coastal locations across England. Furthermore, we aim to assess potential disparity in coding of long COVID conditions between areas with varying levels of deprivation. We also aim to investigate how patient characteristics impact coding of long COVID. Objectives: The specific objectives of this research are as follows. 1. To estimate the incidence of clinically coded long COVID in the population 2. To determine how incidence of clinically coded long COVID may vary by social, demographic and economic factors. 3. To determine factors associated with potentially missed coding of long COVID. This is a retrospective cohort study using electronic health records. The study period is from the 29th of January 2020 to the 31st of March 2022. We will use the OpenSAFELY platform, which provides electronic health records data with full population coverage. Within OpenSAFELY, electronic health records data have been linked to national SARS-COVID-2 testing records (secondary General Surveillance System), vaccination data (National Immunisation Management Service), Index of Multiple Deprivation (IMD), and the ONS death registry. Admitted Patient Care Spells (APCS) is part of Hospital Episode Statistics (HES) and is provided to OpenSAFELY within NHS Digital's Secondary Use Service (SUS). The study population will be living adults aged 18 to 105 years old and registered with an English general practice which uses the TPP system on the study start date. Additionally, individuals must have 1-year prior follow-up with their registered general practice to be included in this study. This is to ensure baseline characteristics are accurately captured. Further inclusion criteria will be: Known age, Known sex, Known region, No record of long COVID prior to study start date. The primary outcome will be clinically coded long COVID in electronic health records. The secondary outcomes will be long lasting symptoms following a COVID-19 diagnosis.