Large HIV clusters among people who inject drugs (PWID) have been highly visible. However, the extent to which large clusters occur among other groups has not been described. We described characteristics and growth of large HIV molecular clusters in the United States to guide public health response planning.
We used data reported through March 2021 to characterize clusters first detected during 2018–2019. To detect clusters, we analyzed HIV-1 pol sequences reported to the National HIV Surveillance System for persons with HIV diagnosed in the 3 years prior. We conducted quarterly transmission network analysis, inferring clusters using a pairwise threshold of 0.005 substitutions/site, and identified clusters of rapid transmission (those with ? 5 diagnoses during the most recent 12 months). We determined the primary (>50%) risk group (based on transmission category) for each cluster. For large clusters (>25 persons) among men who have sex with men (MSM), we described size at detection and used node ages inferred by molecular clock analysis to estimate HIV transmission rates. We also calculated size and annual growth rate as of March 2021.
During 2018–2019, we identified 144 clusters of rapid transmission; 118 (82%) were primarily MSM, 10 (7%) were primarily PWID, and 16 (9%) had no primary risk group. Of 25 large clusters, 17 (68%) were primarily MSM, 6 (24%) were primarily PWID, and 2 (8%) had no primary risk group. Among large MSM clusters, median size at detection was 11; transmission rates at detection ranged from 11–139 transmission events/100 person-years (median: 23/100 py). Large MSM clusters were primarily found in the West (n=8) and the South (n=7), with 2 in the Northeast; 16 of 17 clusters involved >1 state. Median growth rate was 9 persons/year; median size as of March 2021 was 31 persons. Among 528 persons in large MSM clusters, 33% were White, 32% Hispanic/Latino and 26% Black/African American (Table); 4% identified as transgender.
Most large, rapidly growing HIV molecular clusters in the United States were primarily MSM. High transmission rates and small cluster sizes at the time of detection, coupled with fast growth rates, indicate that swift response when clusters are first detected has the potential for meaningful public health impact. Demographic data show diversity of race/ethnicity and gender identity among persons in these clusters. Interventions accounting for the unique needs of these networks are needed to halt transmission.