EVALUATION COMMITTEE

The Faculty Council has appointed the following adjudication committee to evaluate the thesis and the associated lecture: 

• Xavier Moisset, Professor, MD, PhD, Centre Hospitalier Universitaire de Clermont-Ferrand, Neurology department, Inserm U1107 – NeuroDol, France
• Anders Peter Hviid, Professor, Head of Department of Epidemiology Research at Statens Serum Institut, Denmark
• Niels Nygaard Rossing, Associate  Professor, HST, Aalborg University, Denmark (Chariman)

Moderator: Prof. Lars Arendt-Nielsen, HST, Aalborg University

ABSTRACT

Up until May 2024, the number of reported incidences of positive Coronavirus, 2019 (COVID-19) cases reached 775 million, while over 7 million deaths have been reported worldwide. The majority of people survive the acute phase of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2 virus) infection, recovering fully. However, a proportion of COVID-19 survivors develop persistent long-term sequelae, termed long-COVID. More than 100 different sequelae of both physical and psychological characters have been recognized. Of those, pain, termed post-COVID pain, is a major contributor. Upwards to a fifth of all COVID-19 survivors that suffer from long-COVID experience one or more types of pain. This dissertation consisted of a self-reported population-based survey (questionnaire 1) distributed to 593,741 adult Danish residents with a positive reverse transcription-polymerase chain reaction (RT-PCR) test between March 2020 and December 2021. Data was collected on average 21.0 months post-infection. A follow-up survey (questionnaire 2) was distributed to 106,045 participants from the initial questionnaire collecting additional information about previous pain and disease conditions. Registered socio-economic data was retrieved from Statistics Denmark. Study I mapped the prevalence of de novo post-COVID pain in a non-hospitalized cohort (n=130,443). Study II investigated the de novo post-COVID pain prevalence in a previously hospitalized cohort (n=4,712). Study III applied an exploratory prediction model based on 58 variables from the surveys and registered data. Within the non-hospitalized cohort, 5.3% suffered from de novo, multi-type post-COVID pain. The previously hospitalized cohort reported any post-COVID pain prevalence of 38.6%, while 18.0% reported de novo multi-type post-COVID pain. Female sex, higher Body-Mass Index (BMI), history of previous whiplash, and low socio-economic status (lower income and educational level) were identified as risk factors for the experience of de novo post-COVID pain. Unique for Study I risk factors were older age, history of migraine, stress, type-2 Diabetes Mellitus, and neurological disorders, while younger age, anxiety, and intensive care unit (ICU) admission were unique to the previously hospitalized cohort in Study II. Study III applied a predictive model to 58 potential predictors to investigate the predictive profiles of the full cohort (n=65,028, Model 1), a pre-COVID pain subgroup (n=9,090, Model 2), and a non-pre-COVID pain group (n=55,938, Model 3). Model 1 calculated a cross-validated area under the curve (cvAUC) of 0.68. Models 2 and 3 provided cvAUCs of 0.69 and 0.65, respectively, suggesting a moderate predictive ability. The most prominent predictors contributing to the predictive value of Model 1 were the use of pain medication, stress, higher income, age [40,60], female gender, and weight. Model 2 highlighted the use of pain medicine, breathing pain, stress, height, physical activity, and weight. Model 3 identified stress, female gender, weight, higher education, age [40,60], higher income, and physical activity.

In conclusion, long-term post-COVID pain is widely present in the general population of COVID-19 survivors. A more severe SARS-CoV-2 infection leading to hospitalization more often leads to persistent pain symptoms reflecting widespread chronic pain conditions affecting quality of life. Thus, this adds to the existing burden of managing chronic pain patients in the clinics. By further exploring the pre-COVID data variables in the prediction model, these results serve to provide a better understanding of what pre-COVID conditions to direct attention to, when exploring what might predict the pain/no-pain outcome post-COVID.