Estimated Annual Number of HIV Infections — United States, 1981–2019

Karin A. Bosh, PhD; H. Irene Hall, PhD; Laura Eastham, MPH; Demetre C. Daskalakis, MD; Jonathan H. Mermin, MD

Disclosures

Morbidity and Mortality Weekly Report. 2021;70(22):801-806. 

In This Article

Abstract and Introduction

Introduction

The first cases of Pneumocystis carinii (jirovecii) pneumonia among young men, which were subsequently linked to HIV infection, were reported in the MMWR on June 5, 1981.[1] At year-end 2019, an estimated 1.2 million persons in the United States were living with HIV infection.[2] Using data reported to the National HIV Surveillance System, CDC estimated the annual number of new HIV infections (incidence) among persons aged ≥13 years in the United States during 1981–2019. Estimated annual HIV incidence increased from 20,000 infections in 1981 to a peak of 130,400 infections in 1984 and 1985. Incidence was relatively stable during 1991–2007, with approximately 50,000–58,000 infections annually, and then decreased in recent years to 34,800 infections in 2019. The majority of infections continue to be attributable to male-to-male sexual contact (63% in 1981 and 66% in 2019). Over time, the proportion of HIV infections has increased among Black/African American (Black) persons (from 29% in 1981 to 41% in 2019) and among Hispanic/Latino persons (from 16% in 1981 to 29% in 2019). Despite the lack of a cure or a vaccine, today's HIV prevention tools, including HIV testing, prompt and sustained treatment, preexposure prophylaxis, and comprehensive syringe service programs, provide an opportunity to substantially decrease new HIV infections. Intensifying efforts to implement these strategies equitably could accelerate declines in HIV transmission, morbidity, and mortality and reduce disparities.

To estimate annual HIV incidence among persons aged ≥13 years in the United States during 1981–2019, CDC applied mathematical modeling to data reported to the National HIV Surveillance System. Three eras of HIV incidence estimates were used based on changes in methodology and available data*.[3,4] The cumulative number of HIV infections over the period was estimated by summing annual incidence estimates. The distributions of HIV incidence were compared overall and by sex at birth, race/ethnicity, and transmission category for the period examined at the beginning (1981), at the peak number of annual infections (1984–1985), and at the end of the study period (2019). Trends in the annual number of HIV infections over the entire period were assessed for selected racial/ethnic groups and transmission categories. For racial/ethnic groups, only trends among Black, Hispanic/Latino, and White persons were described.§ Increases or decreases in the numbers and proportions are reported for relative changes of ≥5%.

During 1981–2019, there were an estimated 2.2 million new HIV infections among persons aged ≥13 years in the United States. The estimated number of infections increased from 20,000 in 1981 (Figure 1) to 130,400 in 1984 and 1985, then declined rapidly to between 84,200 and 84,800 annually during 1986–1990. HIV incidence remained relatively stable from 1991 to 2007, with about 50,000 to 58,000 infections per year, and declined in recent years to an estimated 34,800 in 2019. HIV incidence decreased by 73% from the highest annual number of infections (130,400 in 1984 and 1985) to 34,800 in 2019.

Figure 1.

Estimated HIV incidence* among persons aged ≥13 years — United States, 1981–2019
Abbreviations: BCA = back-calculation approach; CD4 = CD4+ T-lymphocyte model; SEA = stratified extrapolation approach.
*HIV incidence estimates for 1981–2006 were derived from the extended BCA applied to HIV surveillance data reported to CDC through June 2007. HIV incidence in 2007 was estimated using the SEA applied to HIV surveillance data reported to CDC through June 2011. HIV incidence estimates during 2008–2019 were derived from the CD4 model applied to HIV surveillance data reported to CDC through December 2020.

A larger proportion of infections occurred among females in 2019 (18%) than did in 1981 (8%) or in 1984–1985 (12%). The number of HIV infections among White persons decreased during 1985–2019 (Table) (Figure 2) and the proportion of infections among White persons decreased from 56% in 1981 to 25% in 2019. The number of infections among Black persons increased during 1981–1990 and then decreased through 2019. In 1988, the number of infections among Black persons surpassed the number among White persons and remained higher than in any other racial/ethnic group through 2019. Black persons accounted for 29% of infections in 1981, 30% of infections in 1984–1985, and 41% of infections in 2019. Hispanic/Latino persons represented 16% of infections in 1981, 14% of infections in 1984–1985, and 29% of infections in 2019.

Figure 2.

Estimated HIV incidence* among persons aged ≥13 years, by selected race/ethnicity and transmission category§ — United States, 1981–2019
Abbreviations: BCA = back-calculation approach; CD4 = CD4+ T-lymphocyte model; IDU = injection drug use; MMSC = male-to-male sexual contact; SEA = stratified extrapolation approach.
*HIV incidence estimates for 1981–2006 were derived from the extended BCA applied to HIV surveillance data reported to CDC through June 2007. HIV incidence in 2007 was estimated using the SEA applied to HIV surveillance data reported to CDC through June 2011. HIV incidence estimates during 2008–2019 were derived from the CD4 model applied to HIV surveillance data reported to CDC through December 2020.
Hispanic/Latino persons can be of any race.
§Transmission categories assigned on the basis of sex at birth and include transgender persons. Data by transmission category have been adjusted to account for missing risk-factor information.

Male-to-male sexual contact accounted for more than one half of infections in all years except during 1988–2002, when infections attributed to heterosexual contact increased. The proportion of infections attributed to male-to-male sexual contact or male-to-male sexual contact and injection drug use was 75% in 1981, 67% in 1984–1985, and 70% in 2019. The proportion of infections attributed to heterosexual contact was higher in 2019 (22%) than in 1981 (2%) or in 1984–1985 (6%), whereas the proportion of infections among persons who inject drugs was lower in 2019 (7%) than in 1981 (22%) or in 1984–1985 (25%).

*HIV incidence estimates for 1981–2006 were derived from the extended back-calculation approach applied to HIV surveillance data reported to CDC through June 2007. HIV incidence in 2007 was estimated using the stratified extrapolation approach applied to HIV surveillance data reported to CDC through June 2011 (https://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-supplemental-report-vol-17-4.pdf). HIV incidence estimates during 2008–2019 were derived from the CD4 model applied to HIV surveillance data reported to CDC through December 2020.
Transmission categories were assigned on the basis of sex at birth, regardless of gender identity.
§Trends were not assessed for racial/ethnic groups other than White, Black, and Hispanic/Latino because of changes in data collection that were required in 2003 to align with revised standards for classification of federal data on race and ethnicity for other racial categories, as well as the small number of infections.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....