Social determinants of recovery from ongoing symptoms following COVID-19 in two UK longitudinal studies: a prospective cohort study

Background Social gradients in COVID-19 exposure, illness severity, and mortality have been observed in multiple international contexts. Whether pre-existing social factors affect recovery from ongoing symptoms following COVID-19 and long COVID is less well understood. Methods We analysed data on self-perceived recovery following self-reported COVID-19 illness in two United Kingdom community-based cohorts, COVID Symptom Study Biobank (CSSB) (N = 2548) and TwinsUK (N = 1334). Composite variables quantifying socio-demographic advantage and disadvantage prior to the COVID-19 pandemic were generated from sex, ethnic group, education, local area deprivation and employment status. Associations between self-perceived recovery and composite variables were tested with multivariable logistic regression models weighted for inverse probability of study participation, adjusting for potential confounding by age, region and pre- pandemic health factors, and potential mediation by COVID-19 illness characteristics and adverse experiences during the pandemic. Further analyses tested associations between recovery and individual socio-demographic variables reflecting status prior to and during the COVID-19 pandemic. Findings Socio-demographic gradients in recovery were observed, with unadjusted recovery rate varying between 50% and 80% in CSSB and 70% and 90% in TwinsUK based on composite socio-demographic variables. Likelihood of recovery was lower for individuals with more indicators of pre-pandemic social disadvantage in both cohorts (CSSB: odds ratio, OR = 0.74, 95% confidence interval, CI: 0.62-0.88, TwinsUK: OR = 0.79, 95% CI: 0.64-0.98 per disadvantage) and higher with more social advantages (CSSB: OR = 1.26, 95% CI: 1.08-1.47, TwinsUK: OR = 1.36, 95% CI: 1.09-1.70 per advantage). Associations were neither explained by differences in COVID-19 illness severity or timing, nor adverse social experiences during the pandemic, which were themselves inversely associated with recovery. Interpretation Strong social inequalities in the likelihood of recovery from COVID-19 were observed, with ongoing symptoms several months after coronavirus infection more likely for individuals with multiple indicators of social disadvantage. Work is needed to identify modifiable biopsychosocial factors to enable interventions that address inequalities. Funding Chronic Disease Research Foundation, National Institute for Health and Care Research, Medical Research Council, Wellcome LEAP, Wellcome Trust, Engineering & Physical Sciences Research Council, Biotechnology and Biological Sciences Research Council, Versus Arthritis, European Commission, Zoe Ltd. Plain language summary Across the world acute COVID-19 illness has affected the most disadvantaged in society the most. However, we have not looked in detail whether people’s social circumstances affect their recovery from COVID-19. In our study, we asked people from two UK-based health studies if they still had symptoms after having COVID-19. We looked at how advantaged or disadvantaged they were at the start of the pandemic, based on information about their sex, ethnic group, education level, local area, and employment. In both studies, people who were more disadvantaged were more likely to still have symptoms long after having COVID-19. In contrast, more advantaged people were more likely to have fully recovered. We also saw that people who had negative experiences during the pandemic such as losing their job, being unable to afford their bills or not being able to access health & social care services were less likely to recover. More work is needed to understand how and why recovery was so different for people with different circumstances. Evidence before this study To search for previous reports on associations between recovery from COVID-19 and socio-demographic factors, we screened abstracts identified from the PubMed search query on December 21, 2023: “((COVID-19) AND ((recovery) OR (convalescence) OR (“ ongoing symptoms”)) AND ((socioeconomic) OR (sociodemographic) OR (social) OR (gradient))) AND LitCLONGCOVID[filter]”, where LitCLONGCOVID is a filter for articles relating to long COVID (), which returned 210 results published between July, 2020 and December, 2023. A small number (N = 11) of studies contained direct measures of recovery from COVID-19 in terms of presence/absence of ongoing symptoms relating to COVID-19 illness, either as perceived by the individual or inferred from current symptom reports. Of these, most focused on associations with COVID-19 illness factors such as severity and symptomatology, and prior health indicators. Socio-demographics were mostly used for sample description and adjustments in models rather than as exposures of interest. Of the few studies (N = 8) that tested associations with socio-demographic variables, the range of socio-demographics tested was limited and/or follow-up time typically restricted to 6-12 months since symptom onset. In these studies, associations with recovery were reported for age (N = 4), sex (N = 7), race/ethnicity (N = 2), local area deprivation (N = 1), and education level (N = 1). Associations between long-term symptoms and education or income have been reported in single separate studies. Monthly bulletins up to March 2023 from the UK Coronavirus Infection Survey highlighted prevalence of individuals reporting current effects on daily activities due to long COVID was associated with age, sex, race/ethnicity, local area deprivation and economic activity. No studies were identified that tested for associations of multiple socio-demographics in combination with the likelihood of recovery following COVID-19. Added value of this study This is the first study to testing the effects of multiple socio-demographics on self-perceived recovery in combination. Measures that attempt to quantify social advantage and disadvantage were generated from multiple known social determinants of health. We tested a wider range of socio-demographic factors than previous studies, including UK geographic region, educational qualification level, employment status and income. Our study has a longer follow-up time than previous comparable reports, with most participants assessed more than one year after infection onset. Detailed data on health before the coronavirus pandemic and COVID-19 illness allowed models to be adjusted extensively and mediation effects to be tested. Implications of all the available evidence The likelihood of full recovery following COVID-19 appears to follow a social gradient, higher for individuals with multiple indicators of social advantages and fewer disadvantages, and lower for those with multiple social disadvantages and fewer advantages prior to the coronavirus pandemic. This reflects and reaffirms the established cycle of social inequalities in health, between individuals’ status within social hierarchies and ill-health. More work is needed to understand the pathways through which this inequality operates so that interventions can be made. ### Competing Interest Statement NJC is supported by NIHR via their institution. MPG is supported by UKRI and NIHR via their institution, is chair of the TwinsUK Volunteer Advisory Board, and declares accommodation and registration fees paid for by conference organisers for International Society for Twin Studies (ISTS); Twins Congress 2023. JDC is supported by NIHR via their institution. AG is supported by a UKRI Future Leaders Fellowship. EJT is supported by NIHR via their institution, and project grants from NIHR and EU Hospital Association. CHS is supported by Alzheimer's Society via their institution, is Scientific Advisor to and stock holder in BrainKey. EM is supported by NIHR and MRC via their institution. MA is supported by NIHR via their institution. RSP was supported by an NIHR Academic Clinical Fellowship and is currently supported by a Wellcome Trust Personal PhD fellowship grant. NRH is supported by NIHR via their institution. LSC is supported by Wellcome Trust. BM is supported by NIHR via their institution. EK is supported by Wellcome EPSRC Centre for Medical Engineering via their institution. ELD was supported by Chronic Disease Research Foundation via their institution. CJS is supported by UKRI and NIHR via their institution, and previously consulted for ZOE Ltd. All other authors have nothing to declare. ### Funding Statement The CSS Biobank is supported by the Chronic Disease Research Foundation. TwinsUK is funded by the Medical Research Council (MRC), Wellcome LEAP, Wellcome Trust, Engineering & Physical Sciences Research Council, Biotechnology and Biological Sciences Research Council, Versus Arthritis, European Commission, Chronic Disease Research Foundation (CDRF), Zoe Ltd, the National Institute for Health and Care Research (NIHR) Clinical Research Network (CRN) and Biomedical Research Centre based at Guy's and St Thomas' NHS Foundation Trust in partnership with King's College London. NJC, EJT, CJS and JDC were supported by the NIHR CONVALESCENCE grant [COV-LT-0009]. RSP is a fellow on the Multimorbidity Doctoral Training Programme for Health Professionals, which is supported by the Wellcome Trust [223499/Z/21/Z]. EM was supported by MRC [MR/R016372/1] and NIHR [NIHR134293]. LSC is supported by Wellcome Trust grant [215010/Z/18/Z]. Authors affiliated with King's College London are also supported by the Wellcome Trust / Engineering and Physical Sciences Research Council Centre for Medical Engineering at King's College London (KCL, [203148/Z/16/Z]) and the UK Department of Health via the NIHR comprehensive Biomedical Research Centre award to Guy's & St Thomas' NHS Foundation Trust (GSTT) in partnership with KCL and King's College Hospital NHS Foundation Trust. ZOE Ltd provided in-kind support for all aspects of building, running and supporting the COVID Symptom Study app and service to all users worldwide. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: Yorkshire & Humber NHS Research Ethics Committee gave ethical approval for the COVID Symptom Study Biobank, Ref: 20/YH/0298. All waves of TwinsUK have received ethical approval associated with TwinsUK Biobank (19/NW/0187), TwinsUK (EC04/015) or Healthy Ageing Twin Study (H.A.T.S) (07/H0802/84) studies from HRA/NHS Research Ethics Committees. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes For the purposes of open access, the author has applied a Creative Commons Attribution (CC BY) licence to any Accepted Author Manuscript version arising from this submission. Access to data in the CSS Biobank is available to bona fide health researchers on application to the CSS Biobank Management Group. Further details are available online at https://cssbiobank.com/information-for-researchers including application forms and contact information. Analysis code used in this study is available openly on GitHub at https://github.com/nathan-cheetham/CSSBiobank_COVIDRecovery. Anonymised COVID Symptom Study data are available to researchers to be shared with researchers according to their protocols in the public interest through Health Data Research UK (HDRUK) and Secure Anonymised Information Linkage consortium, housed in the UK Secure Research Platform (Swansea, UK) at https://web.www.healthdatagateway.org/dataset/fddcb382-3051-4394- 8436-b92295f14259.