Effectiveness, Comparative Effectiveness, and Harms of COVID-19 Vaccines in Adults Who Are Not Pregnant or Immunocompromised: A Rapid Review for the American College of Physicians

Andreea Dobrescu; Larisa Pinte; Amin Sharifan; Arianna Gadinger; Isabel Moser; Christopher Cooper; Gerald Gartlehner

doi:10.7326/ANNALS-25-05044

Effectiveness, Comparative Effectiveness, and Harms of COVID-19 Vaccines in Adults Who Are Not Pregnant or Immunocompromised: A Rapid Review for the American College of Physicians

Ann Intern Med. 2026 Feb 24. doi: 10.7326/ANNALS-25-05044. Online ahead of print.

Authors

Andreea Dobrescu¹, Larisa Pinte², Amin Sharifan¹, Arianna Gadinger¹, Isabel Moser¹, Christopher Cooper¹, Gerald Gartlehner³

Affiliations

¹ Cochrane Austria, Department for Evidence-based Medicine and Evaluation, University for Continuing Education Krems, Krems, Austria (A.D., A.S., A.G., I.M., C.C.).
² Department of Internal Medicine Clinical Hospital Colentina, University of Medicine and Pharmacy Carol Davila, Bucharest, Romania (L.P.).
³ Cochrane Austria, Department for Evidence-based Medicine and Evaluation, University for Continuing Education Krems, Krems, Austria, and RTI International, Center for Advanced Public Health Methods, Research Triangle Park, North Carolina (G.G.).

PMID: 41730209
DOI: 10.7326/ANNALS-25-05044

Abstract

Background: The SARS-CoV-2 Omicron variant continues to pose a global health burden.

Purpose: To assess the effectiveness, comparative effectiveness, and harms of COVID-19 vaccines in nonpregnant, nonimmunocompromised adults.

Data sources: Medline via Ovid and DynaMedex from January 2022 to September 2025.

Study selection: Two reviewers independently selected English-language randomized controlled trials (RCTs) and nonrandomized studies of interventions (NRSIs).

Data extraction: One reviewer extracted data and assessed the certainty of evidence (CoE), and a second reviewer verified; 2 reviewers independently assessed risk of bias.

Data synthesis: Five RCTs and 18 NRSIs were included. In adults of all ages, Omicron-adapted vaccination probably reduces all-cause mortality (vaccine effectiveness [VE] ranged from 26.6% [95% CI, 5.5% to 42.3%] to 75.2% [CI, 70.6% to 79.9%]; moderate CoE) and COVID-19-related hospitalization (VE ranged from 16.6% [CI, 6.5% to 25.8%] to 67.8% [CI, 63.1% to 72.5%]; moderate CoE) compared with no Omicron-adapted vaccination. When administered more than 365 days after the prior vaccine, it probably reduces all-cause mortality (VE, 36.1% [CI, 14.8% to 54.1%]; moderate CoE) and COVID-19-related hospitalization (VE, 22.2% [CI, 11.4% to 32.0%]; moderate CoE). When administered earlier, it may result in no difference in COVID-19-related hospitalization. Omicron-adapted vaccination may increase myocarditis (incidence rate ratio, 2.7 [CI, 1.0 to 7.0]; low CoE) in adults aged 50 years or older. The mRNA-1283.222 bivalent vaccine probably results in no difference in all-cause mortality or serious adverse events compared with mRNA-1273.222 in adults of all ages.

Limitations: No RCTs assessed the effectiveness of Omicron-adapted versus no Omicron-adapted vaccination. Evidence on harms was limited.

Conclusion: Omicron-adapted vaccines improve protection compared with no Omicron-adapted vaccines, particularly when administered more than 365 days after the prior vaccination.

Primary funding source: American College of Physicians. (PROSPERO: CRD420251136017).

Publication types

Review