Abstract
BACKGROUND: Long COVID has become a central public health concern. This study characterized the effectiveness of BNT162b2 BA.4/5 bivalent COVID-19 vaccine (bivalent) against long COVID symptoms. METHODS: Symptomatic US adult outpatients testing positive for SARS-CoV-2 were recruited between 2 March and 18 May 2023. Symptoms were assessed longitudinally using a CDC-based symptom questionnaire at Week 4, Month 3, and Month 6 following infection. The odds ratio (OR) of long COVID between vaccination groups was assessed by using mixed-effects logistic models, adjusting for multiple covariates. RESULTS: At Week 4, among 505 participants, 260 (51%) were vaccinated with bivalent and 245 (49%) were unvaccinated. Mean age was 46.3 years, 70.7% were female, 25.1% had ≥1 comorbidity, 43.0% prior infection, 23.0% reported Nirmatrelvir/Ritonavir use. At Month 6, the bivalent cohort had 41% lower risk of long COVID with ≥3 symptoms (OR: 0.59, 95% CI, 0.36-0.96, p = 0.034) and 37% lower risk of ≥2 symptoms (OR: 0.63, 95% CI, 0.41-0.96, p = 0.030). The bivalent cohort reported fewer and less durable symptoms throughout the six-month follow-up, driven by neurologic and general symptoms, especially fatigue. CONCLUSIONS: Compared with unvaccinated participants, participants vaccinated with the bivalent were associated with approximately 40% lower risk of long COVID and less symptom burden over the six-month study duration.