Determinants of pre-vaccination antibody responses to SARS-CoV-2: a population-based longitudinal study (COVIDENCE UK)

Background: Prospective population-based studies investigating multiple determinants of pre-vaccination antibody responses to SARS-CoV-2 are lacking. Methods: We did a prospective population-based study in SARS-CoV-2 vaccine-naive UK adults recruited between May 1 and November 2, 2020, without a positive swab test result for SARS-CoV-2 prior to enrolment. Information on 88 potential sociodemographic, behavioural, nutritional, clinical and pharmacological risk factors was obtained through online questionnaires, and combined IgG/IgA/IgM responses to SARS-CoV-2 spike glycoprotein were determined in dried blood spots obtained between November 6, 2020, and April 18, 2021. We used logistic and linear regression to estimate adjusted odds ratios (aORs) and adjusted geometric mean ratios (aGMRs) for potential determinants of SARS-CoV-2 seropositivity (all participants) and antibody titres (seropositive participants only), respectively. Results: Of 11,130 participants, 1696 (15.2%) were seropositive. Factors independently associated with higher risk of SARS-CoV-2 seropositivity included frontline health/care occupation (aOR 1.86, 95% CI 1.48–2.33), international travel (1.20, 1.07–1.35), number of visits to shops and other indoor public places (≥ 5 vs. 0/week: 1.29, 1.06–1.57, P-trend = 0.01), body mass index (BMI) ≥ 25 vs. < 25 kg/m ² (1.24, 1.11–1.39), South Asian vs. White ethnicity (1.65, 1.10–2.49) and alcohol consumption ≥15 vs. 0 units/week (1.23, 1.04–1.46). Light physical exercise associated with lower risk (0.80, 0.70–0.93, for ≥ 10 vs. 0–4 h/week). Among seropositive participants, higher titres of anti-Spike antibodies associated with factors including BMI ≥ 30 vs. < 25 kg/m ² (aGMR 1.10, 1.02–1.19), South Asian vs. White ethnicity (1.22, 1.04–1.44), frontline health/care occupation (1.24, 95% CI 1.11–1.39), international travel (1.11, 1.05–1.16) and number of visits to shops and other indoor public places (≥ 5 vs. 0/week: 1.12, 1.02–1.23, P-trend = 0.01); these associations were not substantially attenuated by adjustment for COVID-19 disease severity. Conclusions: Higher alcohol consumption and lower light physical exercise represent new modifiable risk factors for SARS-CoV-2 infection. Recognised associations between South Asian ethnic origin and obesity and higher risk of SARS-CoV-2 seropositivity were independent of other sociodemographic, behavioural, nutritional, clinical, and pharmacological factors investigated. Among seropositive participants, higher titres of anti-Spike antibodies in people of South Asian ancestry and in obese people were not explained by greater COVID-19 disease severity in these groups.