Tracking Omicron Lineages Across the U.S., January 2022–May 2023

This page describes a CDC report on variants circulating through May 2023. Variant patterns have changed since.

Since December 2020 CDC has used national genomic surveillance to track SARS-CoV-2 variants. This report covers January 2, 2022, to May 13, 2023 — a period when Omicron dominated throughout, while one descendant lineage after another rose to predominance (more than 50% of cases).

How variants were tracked

CDC combined sequences from three sources: its National SARS-CoV-2 Strain Surveillance (NS3) program, CDC-contracted commercial laboratories, and public databases (GISAID and NCBI GenBank). Of 1,697,197 sequences from 56 U.S. jurisdictions, 1% came from NS3, 60% from commercial labs and 38% from repositories. As cases fell, weekly sequenced specimens dropped from about 65,000 in January 2022 to about 4,400 in April 2023. The median time from specimen collection to available data was 16 days.

Sequences were quality-filtered and assigned Pango lineages, and weighted to correct for sampling bias. For the latest 3 weeks, where data were still thin, CDC used nowcasts — model projections from the previous 21 weeks. Cases per variant were estimated by multiplying positive test counts by variant proportions.

Who led, and when

Each takeover coincided with a surge in COVID-19 cases. BA.5, XBB.1.5 and BA.1.1 held predominance longest — 19, 16 and 10 weeks. The most cases were attributed to BA.1.1 (14 million), the original Omicron B.1.1.529 (9.8 million) and BA.5 (8.0 million).

National weekly variant proportions (A) and estimated cases attributed to each variant (B). CDC figure.

Late 2022: a crowd of lineages

In the second half of 2022, many descendants of BA.2, BA.4 and BA.5 — among them BA.2.75, BA.4.6, BF.7, BQ.1, BQ.1.1, BA.5.2.6, BN.1, BF.11 and CH.1.1 — each passed 1% at some point. None dominated alone, but BQ.1 and BQ.1.1 together peaked at 59.3% by December 24, 2022, during a winter surge.

Several of these lineages independently acquired the same spike mutations in the receptor-binding domain — R346T, K444T, N460K and F486S/P. This convergent evolution suggests those sites are under selective pressure; the mutations are associated with escape from neutralizing antibodies, including monoclonal antibody treatments that had been authorized earlier.

The rise of XBB.1.5

XBB — a recombinant of two BA.2 descendants — appeared in the United States in late fall 2022 but stayed under 5% nationally. Its descendant XBB.1.5, with an added S486P mutation that has been observed to increase infectivity, was first reported in New York City in October 2022. It became predominant first in HHS Region 2 (New York, New Jersey, Puerto Rico and the U.S. Virgin Islands) on December 31, 2022, then Region 1 (Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island and Vermont) on January 7, 2023, and spread south and west to national predominance by January 28.

When XBB.1.5 became predominant in each HHS region. CDC figure.

As of May 13, 2023, the leading lineages were XBB.1.5 (61.5%), XBB.1.9.1 (10.0%) and XBB.1.16 (9.4%). The fastest-growing were XBB.1.16 and XBB.1.16.1 (with an extra K478R mutation; doubling times of 15.7 and 16.7 days) and XBB.2.3 (with P521S; 20.3 days). Over the whole period, the fastest doubling times belonged to BA.2.12.1 (5.4 days), BQ.1.1 (6.3), BA.5 (6.8) and XBB.1.5 (7.0), compared with 3.2 days for the original Omicron.

Changes after the emergency ended

From May 13, 2023, after the public health emergency declaration expired and as cases and specimens declined, CDC moved to 2-week reporting periods with nowcasts for the latest 4 weeks, dropped state-level estimates, and switched its weighting to regional data from the National Respiratory and Enteric Virus Surveillance System. Old and new methods gave similar estimates.

Why it matters

Variant data helped guide decisions to revoke emergency use authorizations for monoclonal antibody treatments that lost effectiveness against Omicron lineages, and the Food and Drug Administration used them to recommend including BA.4/BA.5 in the updated bivalent vaccines in June 2022. They were expected to guide future vaccine composition.

Limitations: early estimates can be imprecise because of limited data and reporting lags; falling specimen numbers reduce precision; changes in data sources, lineage definitions and methods complicate comparisons; and case counts for recent lineages are undercounted as more people test at home.

Sources

• Ma KC, Shirk P, Lambrou AS, et al. "Genomic Surveillance for SARS-CoV-2 Variants: Circulation of Omicron Lineages — United States, January 2022–May 2023." Morbidity and Mortality Weekly Report 72(24), Centers for Disease Control and Prevention.