Abstract
BACKGROUND: Long COVID is defined as otherwise unexplained symptoms following a SARS-CoV-2 infection. OBJECTIVE: To examine the prevalence of preexisting symptoms compatible with long COVID in individuals with a diagnosis of long COVID. METHODS: This retrospective, observational study included the adult population (aged 18 years and older) in Region Västra Götaland, with at least one recorded healthcare contact between January 1, 2020, and November 30, 2023, from a regional database comprising all levels of healthcare contacts. Data on long COVID, relevant symptoms before and after the pandemic started (2016-2023), and SARS-CoV-2 infection status were extracted using the International Classification of Diseases version 10 (ICD-10) codes. Individuals who had been hospitalized due to a SARS-CoV-2 infection were considered separately. RESULTS: Out of 1,415,885 individuals, 9202 (0.6%) had been diagnosed with long COVID. Among the non-hospitalized individuals, the record of at least one of the relevant symptoms was more common in those with long COVID compared to those without it (57.6% vs. 36.3% for men and 71.6% vs. 50.4% for women), already before January 1, 2020. Among individuals with any relevant symptom, the odds ratios (ORs) of having long COVID were OR = 2.28 (95% confidence interval [CI] = 2.10-2.48) for men and OR = 2.32 (95% CI = 2.18-2.48 for women) after adjusting for age group, obesity, asthma, and anxiety, compared with individuals without any relevant symptom. CONCLUSIONS: Individuals diagnosed with long COVID had more healthcare contacts for relevant symptoms even before the pandemic compared to individuals without long COVID.