Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jun 18.
Published in final edited form as: Lancet HIV. 2026 Feb 13;13(3):e185–e193. doi: 10.1016/S2352-3018(25)00302-9

Racial/Ethnic and Regional Disparities in HIV Testing Amidst the COVID-19 Pandemic: A Nationwide Retrospective Analysis Using National Clinical Cohort Collaborative (N3C) Data

Haishan Yang 1, Duolan Han 1, Eric Hurwitz 1, Jerrod Anzalone 1, Cara D Varley 1, Sandra E Safo 1, Jing Sun 1, Dongmei Li 1, Marlene Camacho-Rivera 1, Nada Fadul 1, Zachary Butzin-Dozier 1, Bohdan Nosyk 1, Rena C Patel 1,*, Xiao Zang 1,*, on behalf of the National Clinical Cohort Collaborative (N3C) Consortium
PMCID: PMC12948494  NIHMSID: NIHMS2149990  PMID: 41698379

Summary

Background:

The COVID-19 pandemic caused major disruptions to healthcare, affecting the delivery of HIV care and prevention services. This study aimed to determine the impact of the pandemic on HIV testing rates in medical settings across regions and racial/ethnic populations in the U.S.

Methods:

We used data from the National Clinical Cohort Collaborative (N3C) from January 2018 to December 2022 of individuals aged 15 to 64 with active healthcare records. In this dataset, COVID-19 cases were matched in a 1:2 ratio to COVID-negative controls on age, sex, race/ethnicity within each site. We used interrupted time-series analyses to estimate changes in the monthly HIV testing rate, measured as the number of individuals tested per 10,000 people in four geographic regions (Midwest, Northeast, South, and West) among four racial/ethnic groups (White [non-Hispanic], Black [non-Hispanic], Hispanic, Others). We estimated these changes in three time periods: pre-pandemic (January 2018 to February 2020), early pandemic (March 2020 to August 2020), and late pandemic (September 2020 to December 2022).

Findings:

Our dataset included 9.7 million patients from 58 clinical sites. During the pre-pandemic, HIV testing rates were relatively stable, though levels varied by region and racial/ethnic group. The impact on testing rates differed substantially during the early and late pandemic, with the West experiencing the sharpest and most prolonged decline. Among racial/ethnic groups, the Black population saw the greatest reduction in March 2020 across all four regions: specifically, testing rate in the Black population in the Northeast dropped by 53.1 individuals tested per 10,000 people (95% confidence interval: −73.4 to −32.8). In contrast, the decline was more modest among White individuals and not significant in the South. Testing rates rebounded to varying degrees during the late pandemic.

Conclusions:

The COVID-19 pandemic, particularly early on, caused substantial disruptions in HIV testing rates across the U.S., with pronounced racial/ethnic and regional disparities. These findings highlight the need for targeted interventions to mitigate the impact of public health emergencies on communities disproportionately affected by pre-existing health inequities, including expanding mobile testing units, supporting community-based outreach, and investing in culturally responsive care to improve testing access during future public health disruptions.

Introduction

Over the past two decades, substantial efforts were made to increase human immunodeficiency virus (HIV) testing across the United States (U.S.).1 Increasing HIV testing and detection is one of four pillars of the Ending the HIV Epidemic initiative, which launched in 2019 with a goal to reduce new HIV infections in the U.S. by 90% by 2030.2 The U.S. National HIV/AIDS Strategy promotes expanding HIV testing, particularly in populations at higher risk for HIV infection.3 The U.S. Centers for Disease Control and Prevention (CDC) also implemented a range of programs and campaigns to promote HIV testing, such as the Act Against AIDS initiative and the Let’s Stop HIV Together campaign.4

A key challenge in monitoring the expansion of HIV testing and studying the impact of these programs is the scarcity of robust and nationally sampled HIV testing data, largely due to a decentralized healthcare system and fragmented surveillance data.5 Although some regional data or sample-based testing information are available, as well as testing data from CDC-funded programs,6 more representative nationwide or statewide population-level databases for HIV testing do not exist. Based on data from the 2022 Behavioral Risk Factor Surveillance System, 36% of participants were tested for HIV at least once in their lifetime, higher among non-Hispanic Black Americans (57%) than non-Hispanic White (32%) and Hispanic Americans (44%).7 HIV testing and incidence in the US are linked to broader social determinants, including poverty, income inequality, residential segregation, lower education, unemployment, and social instability.8

At the onset of the coronavirus disease 2019 (COVID-19) pandemic, concerns were raised regarding the implications for potential declines in HIV testing.9 Many called for focused efforts to promote HIV testing to prevent COVID-19 from undoing the progress made over the past two decades.10 The impacts of the pandemic were greater in communities of color,11,12 partly due to disruptions to access and continuity of care for chronic conditions. HIV testing and other diagnostic services were particularly susceptible because of lockdowns inhibiting travel, the transition to virtual care visits, and diversion of laboratory services to COVID-19 testing.13 Two recent studies revealed a significant decline in HIV testing rates during the early months of the COVID-19 pandemic in New York state and Chicago.14,15 Another CDC study reported 32% reduction in the number of HIV tests from the first to the second quarter of 2020 based on commercial laboratory data.16 Although testing rates rebounded after the initial decline, they remained lower than the pre-pandemic levels or projected levels that would have been expected in the absence of the pandemic. However, these studies did not include national-level analysis, limiting the ability to assess differential impacts across populations.

The National Clinical Cohort Collaborative (N3C) data is one of the largest repositories of electronic health record (EHR) data available for research in the U.S., created in response to the COVID-19 pandemic. This dataset includes harmonized data from 71 clinical sites across 48 states, including approximately 18.1 million total patients (based on data collected in March 2025). It features clinical, demographic, and historical health information from January 1, 2018 onward, enabling robust longitudinal analyses.17 This comprehensive database facilitates population-level studies while ensuring data quality through rigorous harmonization and institutional review processes.5 In this study, we aimed to leverage the N3C dataset to quantify changes in HIV testing rates over time and determine the differential impacts of the COVID-19 pandemic on HIV testing rates across the U.S., focusing on disparities by region and racial/ethnic groups. The nationally-sampled data offers novel insights into systemic disruptions in HIV testing, investigating inequities in healthcare access and outcomes during different stages of the pandemic.

Methods

Study design

Using N3C data between January 2018 to December 2022, we conducted a retrospective analysis, applying an interrupted time-series (ITS) methods using segmented linear regression — a technique in which the outcome (i.e., monthly HIV testing rates) is modeled as a series of straight-line segments joined at prespecified “breakpoints” (i.e., March 2020 and September 2020). Each segment is allowed its own intercept (capturing any immediate shift in testing rates at the start of a period) and slope (capturing the underlying month-to-month trend), enabling us to estimate both the “level change” and “slope change” associated with the onset and progression of the pandemic. Our analysis examined changes in HIV testing rates across three distinct time periods: pre-pandemic (January 2018 – February 2020), early pandemic (March 2020 – August 2020), and late pandemic (September 2020 – December 2022), consistent with prior research showing the most pronounced disruptions occurred during the early months of the pandemic.18 We further compared these changes across U.S. regions and racial/ethnic groups.

Direct patient consent was not obtained for this repository of deidentified data per N3C policies. The N3C received a waiver of consent from the National Institutes of Health (NIH) Institutional Review Board (IRB), and NIH takes care to ensure the highest privacy and security requirements are met and adhered to for housing and protecting these data in the NIH-managed N3C Enclave. More details can be found in N3C resources. The N3C Enclave is approved through the NIH IRB. Each individual data partner site maintains its own IRB-approved data transfer agreement or joins under a Johns Hopkins University Reliance Protocol (IRB00249128). Each investigator accessing the N3C Enclave receives local IRB approval from their respective institutions. The N3C Data Access Committee approved this project (RP-CA3365).

Data

The N3C dataset was established with broad inclusion criteria encompassing both COVID-19 cases and non-COVID-19 controls, including outpatient and inpatient visits from participating sites. The dataset includes both individuals with confirmed, suspected, or possible COVID-19 and a demographically matched control group of COVID-negative individuals, matched on age, sex, race, and ethnicity at a case-to-control ratio of 1:2 at the same partner site. Historical patient EHR data dating back to January 1, 2018 are also included.

Based on our three periods of analysis, we defined the study period as January 2018 to December 2022 to balance the number of time points before and after the onset of the COVID-19 pandemic (March 2020). We used the N3C dataset’s “observation period” variable to identify the study cohort. In the N3C dataset, the observation period refers to the span of time during which a patient is enrolled in a health benefit plan and whose healthcare events are recorded.17 This observation period is delineated by a start and end date for each patient, representing the duration when a healthcare provider or data source actively captures a person’s medical information. According to CDC guidelines on HIV testing, we included individuals aged 15 to 64 years in 2022 in the study cohort whose observation period overlapped with the study period, i.e., having at least one month of active observation time between January 2018 and December 2022. Within the study period, individuals who have never been diagnosed for HIV were included in the analysis only for the months in which their observation period was active. For example, an individual with an observation period from January 2019 to March 2019 is only counted during these months, which contributes data in the pre-pandemic period only. As such, the composition of the study population varies over time. Additionally, we excluded or censored individuals from subsequent time points after their first detection of HIV or confirmation of an HIV-positive status (see details about our approach to HIV phenotyping elsewhere).19 We collected data in March 2025.

Outcome measures

Given the changing underlying population, we used the monthly HIV testing rate as the primary outcome, defined as the number of individuals who underwent an HIV test in a given month per 10,000 people, allowing for comparisons across periods and populations. Individuals who underwent multiple tests in a month were counted only once (i.e., a binary indicator for HIV testing). We summarized all unique individuals who were tested in a given month and divided by the total number of individuals included in the study cohort for each subgroup (defined by region and race/ethnicity) to calculate the corresponding HIV testing rate. HIV testing in the N3C data is identified through 45 classifications that encompass a range of laboratory measures, including HIV antigen/antibody immunoassays, rapid immunoassays, enzyme-linked immunosorbent assays, and viral load measurements (see the full list of classifications in the appendix pp32–33). Geographic locations were mapped to one of the four U.S. Census regions using patients’ zip codes and their corresponding state (see details in the appendix p2): Midwest, Northeast, South, and West. Individuals were further classified into four mutually exclusive racial/ethnic groups based on their recorded race and ethnicity:20 non-Hispanic White (White), non-Hispanic Black (Black), Hispanic (Hispanic ethnicity, regardless of reported race) and Others (Asian, American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, and those reporting two or more races or with unknown race/ethnicity). We did not further stratify the Others group due to small sample size and resulting high variability.

Statistical analysis

We used a three-phase single-arm (i.e., no control) ITS, corresponding to the three distinctive periods of the pandemic: pre-pandemic (January 2018 to February 2020), early pandemic (March 2020 to August 2020), and late pandemic (September 2020 to December 2022).21 We applied segmented linear regression models to analyze changes in HIV testing rates during the three periods: (1) the underline trend (slope) during the pre-pandemic period, (2) the immediate shift (level change) at the onset of the pandemic in March 2020, (3) the trend (slope) during the late pandemic period, and (4) the change in trend (slope change) comparing the late pandemic to the pre-pandemic period, although the slope and slope change during the early pandemic period, as well as the level change in the late pandemic period were also analyzed but not presented (see details about the ITS model in the appendix p3). In addition, we estimated the level of recovery during the late pandemic period by projecting a counterfactual trend—extending the estimated pre-pandemic trend from January 2018 to February 2020 through December 2022— and compared it to the estimated testing rates at the end of the study period from the ITS models. This assumes that the pre-pandemic trend would have continued beyond February 2020 in the absence of the pandemic. The ITS model was fitted separately for each subgroup. We began by estimating national-level trends. Next, we examined regional differences by aggregating racial/ethnic groups within each region, followed by an analysis of racial/ethnic disparities by aggregating regions within each racial/ethnic group. Finally, we assessed the intersection of region and race/ethnicity by applying the ITS model for each region–race/ethnicity combination.

To evaluate possible seasonal variation, we used the Autocorrelation Function and Partial Autocorrelation Function to check the potential seasonality of HIV testing rates, where the results showed no evident seasonal trends. Therefore, we did not include seasonal adjustments in the final models. In sensitivity analysis, we varied the length of the early pandemic period from six to seven and eight months. In light of evidence showing differential HIV testing rates and potentially greater disruptions among the older population,14 we conducted an additional sensitivity analysis focused on individuals aged 50–64. All statistical analyses were executed using R version 4.4.1 software.22 See details about these analyses in the appendix.

Role of the funding source

The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the article.

Results

A total of 9,704,278 individuals aged 15–64 were included from 58 data partner sites spanning January 1, 2018, to December 31, 2022 (Table 1). Individuals in the Northeast had the highest average age, while those in the Midwest had the lowest. The majority of individuals were White, followed by Others, Black, and Hispanic, with notable geographic differences in racial/ethnic composition. The overall cohort was 58.4% female, though sex distribution varied slightly across regions. Among the study cohort, 18.5% (1,797,541/9,704,278) of individuals underwent at least one HIV test during the study period, with the highest percentage tested in the West (25.6%, 525,997/2,055,066) and the lowest in the Midwest (13.1%, 452,842/3,451,476). Before the pandemic, the Northeast exhibited the highest average monthly testing rate (80.6 individuals tested per 10,000 people), while the Midwest had the lowest rate, averaging 46.5 per 10,000. Among racial/ethnic groups, Black individuals had the highest pre-pandemic monthly testing rate (93.3 per 10,000), followed by Hispanic populations (87.5 per 10,000). White and Other groups had the lowest rate (42.2 and 52.4 per 10,000 respectively).

Table 1:

Demographic characteristics of included individuals aged 15 to 64 stratified by region in the National Clinical Cohort Collaborative (N3C) dataset, January 2018 to December 2022

Overall
(n = 9,704,278)		 Midwest
(n = 3,451,476)		 Northeast
(n = 873,343)		 South
(n = 3,324,393)		 West
(n = 2,055,066)
Age: average (standard deviation) 41.2 (13.5) 40.7 (13.5) 41.9 (13.5) 41.5 (13.7) 41.2 (13.2)
Sex
Male 41.6% 42.9% 41.8% 40.1% 41.6%
Female 58.4% 57.1% 58.2% 59.9% 58.4%
Race/Ethnicity
White 56.6% 65.8% 45.9% 56.3% 46.4%
Black 15.6% 13.8% 15.4% 23.8% 5.4%
Hispanic 11.9% 8.9% 15.9% 6.7% 23.8%
Others 15.8% 11.4% 22.8% 13.2% 24.4%
HIV Testing
Individuals with at least one HIV test during the study period 18.5% 13.1% 22.7% 18.7% 25.6%
Average monthly HIV testing rate during the pre-pandemic period* All: 57.1 All: 46.5 All: 80.6 All: 52.0 All: 72.1
White: 42.2 White: 33.4 White: 59.3 White: 41.9 White: 58.2
Black: 93.3 Black: 95.7 Black: 129.9 Black: 80.9 Black: 115.7
H/L: 87.5 H/L: 68.5 H/L: 134.4 H/L: 65.9 H/L: 92.5
Others: 52.4 Others: 54.1 Others: 46.8 Others: 35.4 Others: 68.6
Open in a new tab
*

Calculated by averaging the monthly HIV testing rates during the pre-pandemic period, measured as the number of individuals tested for HIV per 100,000 population.

Based on ITS analysis, during the pre-pandemic period, HIV testing rates were generally stable (Figure 1, Figure 2), with slight shifts in some regions and racial/ethnic groups. Among regions, the West showed the fastest pre-pandemic growth, increasing by 1.46 individual per 10,000 per month (95% Confidence interval [CI]: 1.10 to 1.82). In comparison, trends in other regions were not statistically significant. Among racial/ethnic groups, the Hispanic population experienced the largest, statistically significant growth in HIV testing rate, at an estimate of 0.95 (95% CI: 0.57 to 1.32), followed by Others (0.45, 95% CI: 0.19 to 0.71) and White (0.39, 95% CI: 0.17 to 0.59) individuals. On the contrary, there was a decline in testing rate that was not statistically significant among Black individuals. At the intersection of region and racial/ethnic group, HIV testing rates among Black individuals in the Midwest had the largest decline at a monthly rate of −0.64 individual tested per 10,000 people (95% CI: −1.16 to −0.12). In contrast, the Black population in the West showed a significant increase (2.31, 95% CI: 1.82 to 2.80).

Figure 1: Interrupted time-series analysis on monthly HIV testing rates by region and race/ethnicity (2018–2022).

Figure 1:

Open in a new tab

The dots represent the observed rates of HIV testing. The solid lines represent the predicted HIV testing rates based on the interrupted time-series models during each of the 3 periods, while the dotted lines represent the predicted HIV testing rates if the pre-pandemic trend continued.

Figure 2: Changes in monthly HIV testing rate by region and race/ethnicity (2018–2022).

Figure 2:

Open in a new tab

The slope change* refers to the change in slopes between the late pandemic and the pre-pandemic periods.

At the onset of the pandemic, all regions and racial/ethnic groups experienced immediate declines in HIV testing rate, although a few of these decreases were not statistically significant (Figure 1, Figure 2). Among regions, the Northeast and West regions showed major declines at −27.68 individuals tested per 10,000 people (95% CI: −40.22 to −15.13) and −30.94 (95% CI: −42.29 to −19.59), respectively. In contrast, the South region experienced the smallest and statistically non-significant decline of −6.86 (95% CI: −15.68 to 1.95). Among racial/ethnic groups, the Black population experienced the steepest decline, with an estimate of −24.54 (95% CI: −37.71 to −11.36), followed by the Hispanic population (−23.94, 95% CI: −35.88 to −11.99), while the White population observed the smallest decline (−13.55, 95% CI: −20.27 to −6.84). At the intersection of region and racial/ethnic group, the Black population experienced the greatest decline in testing rate across all regions, despite not statistically significant in the South. In particular, the Black population in the Northeast saw the largest immediate decline of −53.09 (95% CI: −73.43 to −32.77). We found the smallest statistically significant decline among the White population in the Midwest (−11.73, 95% CI: −17.26 to −6.21). In contrast, no racial/ethnic groups in the South exhibited any statistically significant declines. Following the initial immediate decline, all subgroups showed substantial rebounds in testing rates between March 2020 and August 2020.

In the late pandemic period, testing rates generally showed negative slope changes compared with the pre-pandemic trend (growth slowed, reversed into decline, or existing declines became steeper), where the most pronounced negative slope change occurred in the West region (−1.65 individuals tested per 10,000 people, 95% CI: −2.12 to −1.18) (Figure 1, Figure 2). We observed an exception to this in the Northeast and among the Black population, where the slope/trend shifted from negative to positive, though these slope changes were not statistically significant. We observed similar patterns across subgroups defined by the combination of region and racial/ethnic group, with significant negative slope changes identified among all racial/ethnic groups in the West, Hispanic and White populations in the South, and Other in the Northeast—the largest negative slope change occurring among the Black population in the West (−2.41, 95% CI: −3.05 to −1.77). In contrast, we observed statistically significant positive slope changes among Black individuals in both the Northeast and Midwest, indicating a relative increase in testing trends.

We assessed the level of recovery in testing rates by comparing the observed trendlines (based on ITS models) to the counterfactual trendlines projected by extending the pre-pandemic trends through the end of the study period (appendix p34). Among regions, the Northeast and South exhibited higher HIV testing rates than their respective counterfactuals during the late pandemic period, indicating strong recovery. In contrast, the Midwest and the West showed lower observed rates than the counterfactuals, with the gap in the West widening over time. Among racial/ethnic groups, only the Black population demonstrated a strong recovery, with observed testing rates exceeding counterfactual projections in all regions except the West. By comparison, we observed strong recovery only among White individuals in the Northeast and South, Hispanic individuals in the Northeast, and Others in the South.

In the sensitivity analysis focusing on individuals aged 50 to 64, we found that the overall trends and the impacts of the pandemic were largely consistent with those observed in the broader 15 to 64 age group; however, HIV testing rates were significantly lower among older adults (appendix pp30–31).

Discussion

Our analysis, using one of the largest EHR datasets available with HIV testing data, revealed differential impacts of the COVID-19 pandemic on HIV testing rates by region and race/ethnicity. Prior to the pandemic, HIV testing rates were generally stable or increasing, with the West and Hispanic populations experiencing the fastest growth. At the onset of the pandemic, all groups experienced immediate declines in HIV testing rates, with the steepest drop observed among Black individuals. The West region also experienced a sharp immediate decline. In contrast, declines were more modest among White individuals and in the South. Although testing rates rebounded between March and August 2020, longer-term recovery patterns varied. In the late pandemic period, testing trends generally declined compared to pre-pandemic, most sharply in the West and among the Black population in that region. Recovery relative to pre-pandemic trends was strongest in the Northeast and South, and more among Black individuals, except in the West.

Declines in HIV testing, particularly during the early pandemic period, may be attributed to several possible factors, including disruptions to healthcare system, stay-at-home orders and social distancing, deprioritization of routine care, and reduced community outreach.23 The pandemic disproportionately impacted the racial/ethnic minority populations, exacerbating pre-existing inequities and creating additional barriers to healthcare access.24 These communities have historically faced barriers such as limited insurance coverage, fewer healthcare facilities in their neighborhoods, and greater mistrust of the healthcare system due to experiences of discrimination and systemic racism.25,26 The pandemic likely exacerbated these challenges, as many safety-net clinics temporarily closed or reduced services, and public health resources were redirected to the COVID-19 response16. Greater COVID-19 severity and care burden among Black and Hispanic populations may further worsen barriers to accessing routine HIV testing.23 Regional differences in HIV testing recovery following the onset of the COVID-19 pandemic may, in part, reflect variation in state-level policy responses and public health infrastructure. Regions with more robust healthcare systems and infrastructures (including testing centers, laboratories, mobile clinics, etc.), such as the Northeast,27 may have been better equipped to adapt to the challenges posed by the pandemic. As a result, these areas recovered more quickly from the pandemic-induced declines. In contrast, the West region, which had one of the highest HIV testing rates before the pandemic, experienced a declining trend during the late pandemic period and showed the weakest recovery relative to pre-pandemic trajectories. This slower recovery may be attributed to extended COVID-19 restrictions that limited healthcare access and delayed the resumption of routine services. Meanwhile, the Hispanic population showed a worsening trend in HIV testing and weaker recovery across most regions (except in the Northeast) in the late pandemic period. Disruptions to testing may have downstream consequences for delayed diagnosis, which in turn may postpone entry into care and initiation of antiretroviral therapy. These delays can result in worse clinical outcomes for individuals and may increase the risk of onward HIV transmission. These findings highlight the need for targeted, equity-focused interventions that strengthen testing infrastructure and ensure continuity of HIV care services, particularly in underserved communities.

Overall, our findings align with prior research. Similar to a previous observational study that found significant declines in HIV testing in medical settings across four U.S. cities during stay-at-home orders,9 our study identified substantial reductions in HIV testing rates nationally, particularly among Black and Hispanic populations. If left unaddressed, prolonged disparities in HIV testing could undermine progress toward national EHE goals to reduce new HIV infections by 90% by 2030 as well as hinder progress toward Healthy People 2030 objectives, which prioritize reducing new HIV infections and ensuring timely access to diagnosis and treatment for all populations. These findings are also consistent with studies showing widespread declines in HIV testing globally, with HIV testing declining by approximately 45% in Latin America, 39–47% in Asia-Pacific, 35% in Africa, and 22–26% in Europe during the pandemic, highlighting the global scale of this challenge.13,28 To address these challenges, community-based and decentralized testing options, such as self-sampling kits, telehealth, and outreach programs, may effectively mitigate access barriers when traditional healthcare settings face capacity constraints.13 Strengthening partnerships with trusted community organizations and expanding culturally responsive care are also critical. Meantime, future emergency preparedness efforts should incorporate strategies to ensure sustainment of HIV testing and prevention services even during public health emergencies.

Our study has several limitations. First, due to the absence of individual-level risk data in the N3C dataset, this study did not account for individual-level risk factors that influence HIV testing recommendations and behaviors. Despite that, analyzing overall testing rates across populations and regions remains valuable, as it reflects system-wide disparities in access to and uptake of HIV testing services. Second, while we included approximately 10 million patients from the N3C dataset, N3C is largely a hospital-encounter-based dataset from many academic medical centers. Thus, our findings may not be generalizable to all medical settings, community-based testing sites, or to all populations undergoing HIV testing. Third, we did not capture HIV tests administered in non-medical settings, including self-testing programs, which have gained popularity in recent years.29 As a result, our analysis may have underestimated HIV testing rates, particularly among individuals who rely on community-based or at-home testing options. Fourth, we assume that the pre-pandemic trends in HIV testing rates would have continued in the absence of the pandemic. While testing rates will eventually reach a natural ceiling, it is unlikely that this ceiling was close to being reached during our study period.30 Fifth, we did not account for specific policies or restrictions implemented at the county or state levels, or for other simultaneous public health interventions or disruptions beyond the COVID-19 pandemic. Future studies should consider incorporating local policy variations and contextual factors to better isolate the causal impact of the pandemic on HIV testing behavior. For example, local jurisdictions that implemented expanded telehealth services or targeted HIV outreach programs during the pandemic may have mitigated declines in testing. In addition, changes in HIV testing rates may have resulted not only from formal policy interventions such as stay-at-home orders, but also from limited access to testing sites or study hospitals, health system capacity constraints, and reductions in routine care utilization, including routine health exams, during the pandemic period.12 Finally, while we conducted region-stratified analysis, potential heterogeneities in HIV testing trends within each region may have been masked by broader regional patterns.

In conclusion, the COVID-19 pandemic, particularly early on, caused substantial disruptions in HIV testing rates across the U.S., with pronounced racial/ethnic and regional disparities. Despite some recovery in the late pandemic period, testing rates had still not fully recovered to pre-pandemic trajectories in many subgroups as of December 2022. The persistent racial/ethnic and regional gaps highlight the need for targeted interventions to ensure equitable healthcare access. The disproportionate impact on communities historically experiencing higher rates of HIV infection—such as Black and Hispanic populations—and those facing structural barriers to healthcare access underscores the need for emergency response plans that explicitly incorporate strategies to sustain equitable access to preventive health services in the face of large-scale disruptions. Strengthening healthcare infrastructure, addressing structural inequities, and implementing effective public health strategies are crucial to mitigating the impact of future pandemics and improving HIV care for all populations.

Supplementary Material

1

Research in context.

Evidence before this study

The COVID-19 pandemic caused significant disruptions to global healthcare systems, including HIV testing, prevention, and care services. Declines in HIV testing were reported during the pandemic across specific U.S. regions. A PubMed search using terms including “HIV testing,” “COVID-19,” “racial/ethnic disparities,” and “United States” (published through December 2024) identified four studies that focused on state or community-level impacts or testing in commercial laboratories, offering limited analysis on racial/ethnic disparities in HIV testing rates, and none used nationally sampled data to examine these patterns across geographic regions during the course of the COVID-19 pandemic.

Added value of this study

By leveraging one of the largest harmonized electronic health record repositories in the US, this study provides the first nationally sampled evidence of how HIV testing trends evolved across multiple phases of the pandemic, with a focus on racial/ethnic and regional inequities. Beyond documenting disruptions, our findings highlight heterogeneity in recovery trajectories, showing that some populations and regions rebounded while others experienced persistent declines, thereby addressing an important evidence gap about long-term impacts. These findings deepen understanding of how structural inequities intersect with public health crises, and offers critical evidence to guide equity-focused HIV testing policies and emergency preparedness planning.

Implications of all the available evidence

The disruptions in HIV testing during the pandemic highlight critical vulnerabilities in healthcare access for racial/ethnic minority populations. Enhancing the resilience of healthcare systems, particularly in underserved regions and communities, is imperative for protecting vulnerable populations during future public health crises and ensuring equitable access to diagnostic and preventive services. Sustained federal and local efforts, implementation studies to explore barriers and facilitators as well as interventions such as mobile testing units are essential to address these disparities and meet national HIV prevention targets.

Acknowledgements

The analyses described in this manuscript were conducted with data or tools accessed through the NCATS N3C Data Enclave (https://n3c.ncats.nih.gov) and N3C Attribution & Publication Policy v 1.2-2020-08-25b supported by NCATS U24 TR002306. This research was possible because of the patients whose information is included within the data and the organizations (https://ncats.nih.gov/n3c/resources/data-contribution/data-transfer-agreement-signatories) and scientists who have contributed to the on-going development of this community resource (https://doi.org/10.1093/jamia/ocaa196).

Please see detailed N3C acknowledgements in the appendix.

Financial support:

This work was supported by the National Institutes of Health/National Institute of Mental Health (grant number: R01MH131542, prior to a change in scope of award) and the National Institutes of Health/National Institute on Drug Abuse (grant number R01DA041747).

Funding:

National Institutes of Health/National Institute of Mental Health, the National Institutes of Health/National Institute on Drug Abuse

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflicts of interest: We declare no conflict of interest

Data sharing:

The individual-level data for this study cannot be shared due to privacy and confidentiality restrictions. However, aggregate data generated and analyzed are presented in the article and its appendix. All data were collected through the NCATS N3C Data Enclave, which includes a powerful analytics platform and tool set for online discovery, visualization, and collaboration using PySpark built on Palantir’s Foundry platform. Investigators can request access to the N3C Enclave here: https://ncats.nih.gov/n3c/about/applying-for-access. All concept sets in use are available in the N3C Knowledge Store. All data management and analyses were conducted in the N3C Data Enclave using Python. The interrupted series analyses were performed using R. Data analysis code can be made available in GitHub upon request.

Reference:

  • 1.Campbell CK, Lippman SA, Moss N, Lightfoot M. Strategies to Increase HIV Testing Among MSM: A Synthesis of the Literature. AIDS Behav 2018; 22: 2387–412. [DOI] [PubMed] [Google Scholar]
  • 2.Hamilton DT, Hoover KW, Smith DK, et al. Achieving the ‘Ending the HIV Epidemic in the US’ incidence reduction goals among at-risk populations in the South. BMC Public Health 2023; 23. [Google Scholar]
  • 3.Bonacci RA, Holtgrave DR. Evaluating the impact of the US National HIV/AIDS Strategy, 2010–2015. AIDS Behav 2016; 20: 1383–9. [DOI] [PubMed] [Google Scholar]
  • 4.Beer L, McCree DH, Jeffries WL 4th, Lemons A, Sionean C. Recent US Centers for Disease Control and Prevention activities to reduce HIV stigma. J Int Assoc Provid AIDS Care 2019; 18: 2325958218823541. [Google Scholar]
  • 5.Yang X, Sun J, Patel RC, et al. Associations between HIV infection and clinical spectrum of COVID-19: a population level analysis based on US National COVID Cohort Collaborative (N3C) data. Lancet HIV 2021; 8: e690–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Patel D, Williams WO, Wright C, et al. HIV testing services outcomes in CDC-funded health departments during COVID-19. J Acquir Immune Defic Syndr 2022; 91: 117–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Adults Who Report Ever Receiving an HIV Test by Race/Ethnicity. KFF. 2024; published online Oct 21. https://www.kff.org/other/state-indicator/adults-who-report-ever-receiving-an-hiv-test-by-race-ethnicity/ (accessed May 16, 2025). [Google Scholar]
  • 8.Buot M-LG, Docena JP, Ratemo BK, et al. Beyond race and place: distal sociological determinants of HIV disparities. PLoS One 2014; 9: e91711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Moitra E, Tao J, Olsen J, et al. Impact of the COVID-19 pandemic on HIV testing rates across four geographically diverse urban centres in the United States: An observational study. Lancet Reg Health Am 2022; 7: 100159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Dinenno E, Delaney K, Pitasi MA, et al. HIV testing before and during the COVID-19 pandemic - United States, 2019–2020. MMWR Morbidity and mortality weekly report 2022; 71: 820–4. [DOI] [PubMed] [Google Scholar]
  • 11.Tai DBG, Sia IG, Doubeni CA, Wieland ML. Disproportionate impact of COVID-19 on racial and ethnic minority groups in the United States: A 2021 update. J Racial Ethn Health Disparities 2022; 9: 2334–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kantamneni N The impact of the COVID-19 pandemic on marginalized populations in the United States: A research agenda. J Vocat Behav 2020; 119: 103439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Mude W, Mwenyango H, Preston R, O’Mullan C, Vaughan G, Jones G. HIV testing disruptions and service adaptations during the COVID-19 pandemic: A systematic literature review. AIDS Behav 2024; 28: 186–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.O’Grady TJ, Yuan Y, Harris JM, Massaroni RJ, Fuller JA, Tesoriero JM. Impact of COVID-19 on HIV Testing Among AIDS Institute–Funded Providers in New York State–A Time Series Analysis. JAIDS Journal of Acquired Immune Deficiency Syndromes 2023; 92: 27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Stanford KA, McNulty MC, Schmitt JR, et al. Incorporating HIV screening with COVID-19 testing in an urban emergency department during the pandemic. JAMA Intern Med 2021; 181: 1001–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Hoover KW, Zhu W, Gant ZC, et al. HIV services and outcomes during the COVID-19 pandemic - United States, 2019–2021. MMWR Morb Mortal Wkly Rep 2022; 71: 1505–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Haendel MA, Chute CG, Bennett TD, et al. The National COVID Cohort Collaborative (N3C): Rationale, design, infrastructure, and deployment. J Am Med Inform Assoc 2021; 28: 427–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Li Z, Farmanesh P, Kirikkaleli D, Itani R. A comparative analysis of COVID-19 and global financial crises: evidence from US economy. Econ Res-Ekon Istraž 2022; 35: 2427–41. [Google Scholar]
  • 19.Hurwitz E, Varley CD, Anzolone AJ, et al. Identifying People Living With or Those at Risk for HIV in a Nationally Sampled Electronic Health Record Repository Called the National Clinical Cohort Collaborative: Computational Phenotyping Study. JMIR Medical Informatics 2025; 13(1): e68143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Cook L, Espinoza J, Weiskopf NG, et al. Issues with variability in electronic health record data about race and ethnicity: Descriptive analysis of the National COVID Cohort Collaborative Data Enclave. JMIR Med Inform 2022; 10: e39235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Bernal JL, Cummins S, Gasparrini A. Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol 2017; 46: 348–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.R Core Team. R: A Language and Environment for Statistical Computing. 2024. https://www.R-project.org/. [Google Scholar]
  • 23.Mackey K, Ayers CK, Kondo KK, et al. Racial and Ethnic Disparities in COVID-19–Related Infections, Hospitalizations, and Deaths. Ann Intern Med 2021; 174: 362–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Garcia MA, Homan PA, García C, Brown TH. The Color of COVID-19: Structural Racism and the Disproportionate Impact of the Pandemic on Older Black and Latinx Adults. J Gerontol B Psychol Sci Soc Sci 2021; 76: e75–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Chen AM. Barriers to health equity in the United States of America: can they be overcome? International Journal for Equity in Health 2025; 24(1): 39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Yearby R Racial disparities in health status and access to healthcare: The continuation of inequality in the United States due to structural racism. Am J Econ Sociol 2018; 77: 1113–52. [Google Scholar]
  • 27.Lockamy A III. Benchmarking healthcare quality in the United States. Management in Healthcare 2021; 6: 70–100. [Google Scholar]
  • 28.Rick F, Odoke W, van den Hombergh J, Benzaken AS, Avelino-Silva VI. Impact of coronavirus disease (COVID-19) on HIV testing and care provision across four continents. HIV Med 2022; 23: 169–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kozieł A, Domański I, Szymczak A, Dudzik T, Knysz B, Szetela B. HIV self-testing: A discussion on the benefits, limitations, and implications for public health with a focus on Poland. Diagnostics (Basel) 2024; 14. DOI: 10.3390/diagnostics14222475. [DOI] [Google Scholar]
  • 30.Pitasi MA, Beer L, Cha S, et al. Vital signs: HIV infection, diagnosis, treatment, and prevention among gay, bisexual, and other men who have sex with men - United States, 2010–2019. MMWR Morb Mortal Wkly Rep 2021; 70: 1669–75. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2149990-supplement-1.docx (1.1MB, docx)

Data Availability Statement

The individual-level data for this study cannot be shared due to privacy and confidentiality restrictions. However, aggregate data generated and analyzed are presented in the article and its appendix. All data were collected through the NCATS N3C Data Enclave, which includes a powerful analytics platform and tool set for online discovery, visualization, and collaboration using PySpark built on Palantir’s Foundry platform. Investigators can request access to the N3C Enclave here: https://ncats.nih.gov/n3c/about/applying-for-access. All concept sets in use are available in the N3C Knowledge Store. All data management and analyses were conducted in the N3C Data Enclave using Python. The interrupted series analyses were performed using R. Data analysis code can be made available in GitHub upon request.

RESOURCES