Antiretroviral therapy (ART) has played a crucial role in substantially decreasing Human Immunodeficiency Virus (HIV) viral load (VL) among people living with HIV (PLHIV), leading to reductions in HIV-associated morbidity and mortality.^1,2 PLHIV who have VL below the level of 200 HIV RNA copies/milliliter (c/mL) do not transmit HIV to sexual partners, and there is further evidence of almost zero risk of sexual transmission of HIV with VL less than 1000 c/mL.³ In 2016, the Prevention Access Campaign launched the Undetectable equals Untransmittable (U=U) message to end the HIV pandemic and HIV-related stigma.⁴ U=U established that PLHIV who adhere to their ART regimen and achieve an undetectable VL of <200 c/mL and have risk of transmitting HIV to their sexual partners.⁵ The World Health Organization (WHO) initially defined undetectable as VL<200 (c/mL) in 2016 and later revised the definition to VL ≤50 c/mL in 2021. The actual undetectable VL value depends on the assay platform used and reflects no virus detection, as indicated in the 2023 WHO guidance.⁵

The process and validity when measuring an individual’s HIV VL can vary depending on the method and platform used. In settings with high laboratory capacity, plasma VL testing is the gold standard, and instruments can have a lower limit of detection of 20–200 c/mL of HIV-1 RNA.⁶ Despite the benefit of having a low limit of detection, plasma VL testing may not be possible or feasible in all settings as it requires venous blood draw, plasma processing, and stringent storage and transport conditions.⁷ An alternative to plasma VL testing is using dried blood spots (DBS). The use of DBS eliminates the need for phlebotomy, is cost-effective for remote areas, and facilitates easy transport to central laboratories.^8,9 Additionally, high concordance exists between DBS and plasma results for HIV viral load, indicating the viability of using DBS as an alternative to plasma.^10-12 However, the lower limit of VL detection for DBS is 550–1000 c/mL, depending on the platform being used.^8,9,13-15 Due to this limitation, using DBS as a diagnostic tool for concluding that an individual has VL<200 c/mL and, therefore, is considered to have an undetectable VL is not appropriate. Further, the use of DBS in VL testing can lead to reluctance among providers to advise patients that they cannot transmit HIV to sexual partners, as providers can only determine that a patient’s VL is <1000 c/mL rather than undetectable. Understanding the population-based distribution of VL results among plasma samples can provide a proxy measure and counterfactual for the proportion of DBS samples that are likely to be undetectable. We describe the proportion of PLHIV with an undetectable HIV VL (<200 c/mL and ≤50 c/mL) among those with VL <1000 c/mL, using plasma VL data. We also assess how our findings can inform the interpretation of DBS VL testing results in resource-limited settings, where DBS is more commonly available, using data from 13 countries in sub-Saharan Africa.

To calculate the proportion of virally suppressed individuals who had an undetectable VL, we analyzed results from plasma-based VL testing completed for Population-based HIV Impact Assessment (PHIA) surveys conducted from 2015 to 2019. The PHIA survey’s primary objectives were to measure national HIV incidence and viral load suppression (VLS) at the subnational level. The methods have been previously described.^16,17 VL testing was conducted for participants with positive HIV tests, primarily using plasma samples. In rare instances, DBS was used to obtain VL results. Plasma VL testing was conducted using the Roche COBAS AmpliPrep/COBAS TaqMan (CAP/CTM) HIV-1 test, version 2.0 (Roche Diagnostics, Pleasanton, CA) on a Roche CAP/CTM with either a 48 or a 96 analyzer or the Abbott RealTime HIV-1 assay (Abbott Molecular, Des Plaines, IL) on the Abbott fully automated m2000 platform (m2000sp and m20000rt). All seroreactive samples were tested for the presence of antiretroviral (ARV) medications used in first- and second-line treatment regimens in the respective countries.^14,16 ART use was defined by either self-report or qualitative detection of ARV. We included PHIA surveys from Cameroon (CAMPHIA, 2017–18), Côte d'Ivoire (CIPHIA 2017–18), Eswatini (SHIMS2, 2016–17), Ethiopia (EPHIA, 2017), Kenya (KENPHIA 2018–19), Lesotho (LePHIA, 2016–17), Malawi (MPHIA, 2015–16), Namibia (NAMPHIA, 2016–17), Rwanda (RPHIA, 2018) Tanzania (THIS, 2016–17), Uganda (UPHIA, 2016–17), Zambia (ZAMPHIA, 2016), and Zimbabwe (ZIMPHIA, 2015–16).¹⁸ The Ethiopia survey was limited to urban areas.

We conducted both pooled and individual country-specific data analyses from the 13 PHIA surveys. Survey weights were applied to all analyses to account for differences in sampling probabilities and to adjust for survey non-response.¹⁹ Data from participants aged 15–59 years whose plasma samples tested HIV positive and had plasma VL results were included in the analysis. Data from participants who only had a DBS VL result were excluded.

We assessed the proportion of participants with VL <1000 c/mL who met the 2016 and 2021 WHO definitions of undetectable VL, specifically VL<200 c/mL and ≤50 c/mL, respectively ⁹. We also assessed these proportions based on current ART use of any duration and current ART use for ≥12 months. Additionally, we stratified all analyses by sex (male/female), place of residence (urban/rural), and age group separately to identify any subgroup differences. All analyses were completed in SAS (Version 9.4; SAS Institute, Cary, NC). The PHIA surveys were approved by Institutional Review Boards at the CDC, Columbia University, Westat, and in the respective countries (See 45 C.F.R. part 46; 21 C.F.R. part 56). All participants provided oral consent.

Across the 13 countries, 303,400 participants aged 15-59 years were eligible for participation, and 281,188 (93.1%) completed the individual interview; of these, 266,187 (95.0%) consented to biomarker testing. Of the 16,989 participants with HIV-positive test results, HIV VL data from plasma samples were available for 16,533 (97.3%). Overall, 87.3% of PLHIV on ART had VL <1000 c/mL (Table). Of these, 95.4% had VL <200 c/mL. Restricting the analysis to participants who had VL <1000 c/mL and reported current ART ≥12 months, 96.6% had VL <200 c/mL. Using the updated 2021 WHO guidance threshold of ≤50 c/mL, 84.9% of PHIA participants with VL <1000 c/mL and 87.4% of those on ART for ≥12 months had VL ≤50 c/mL.

The proportion of PLHIV on ART with VL <1000 c/mL and ≤50 c/mL varied across countries. Eswatini, Kenya, Malawi, Namibia, and Rwanda had > 90% of PLHIV on ART with VL <1000 c/mL. Conversely, Cameroon and Côte d’Ivoire had 79.6% and 74.8%, of PLHIV on ART with VL <1000 c/mL, respectively. Among PLHIV on ART with VL <1000 c/mL, most countries reported between 82%–90% also having VL ≤50 c/mL. The exceptions were Côte d’Ivoire (72.3%), Uganda (75.9%), and Malawi (92.2%).

By demographic characteristics, a lower proportion of males compared to females had VL ≤50 c/mL, both among individuals on ART and those who reported being on ART for ≥12 months. A lower proportion of younger adults aged 15–24 years compared to older adults aged 25–59 years had VL <200 c/mL and VL ≤50 c/mL among individuals on ART. Among those who reported being on ART for ≥12 months, younger adults aged 15–24 years had the lowest proportion (94.9%), and older adults aged 45-59 years had the highest proportion (97.3%) with VL <200 c/mL. No differences in all VL outcomes were seen by place of residence.

Although there was variation across countries in the percentage of individuals achieving VLS of <1000 c/mL and being undetectable at VL <200 c/mL, the vast majority of PLHIV who had VL <1000 c/mL and were on ART exhibited an undetectable VL <200 c/mL. Narrowing the study to individuals who reported taking ART for ≥12 months led to a modest increase in the proportion of undetectable individuals. This finding reinforced that among PLHIV who had been on ART for a period of ≥12 months, a small percentage of PLHIV (3.4%) were not undetectable at the level required for zero risk of sexual transmission, according to available scientific evidence.³ Lower VLS among the younger age group (15–24 years) may be due to new diagnosis and recent initiation of ART. By providing quantitative data on the very low frequency of having VL that exceeds 200 c/mL when the VL is confirmed as <1000 c/mL, patients, healthcare providers, and other stakeholders possess tangible evidence to adopt U=U messaging and initiatives, even in settings where only DBS VL testing is available. The recommendation to use DBS VL results is further supported by evidence of almost zero risk of sexual transmission of HIV with VL <1000 c/mL.^3,5

Undetectable viral load was observed in the majority (95.4%) of PLHIV on treatment with viral load <1000 c/mL, providing convincing evidence that onward HIV transmission is effectively eliminated even in the most challenging Sub-Saharan African settings. For PLHIV who are recently exposed, newly diagnosed, or with known infection but not yet on treatment, U=U has the potential to reduce self-stigma, create optimism to seek HIV testing, initiate ART as soon as possible, and stay on ART.^20-22 To fully endorse U=U at both individual and population levels in resource-limited settings, PLHIV and clinicians seek assurance that an undetectable VL under currently accepted definitions has been both obtained and sustained over time.^22,23 Accessibility and scale-up of VL testing and monitoring are therefore critical for PLHIV and HIV care providers, including clinicians, peer educators, HIV counselors, and allied health professionals. The scale-up of plasma-based VL testing is not only beneficial for ensuring no onward sexual transmission of HIV but also for treatment management for PLHIV who have VLS but have not attained VL ≤50 c/mL.

Several limitations exist in our analysis. First, these cross-sectional study results cannot provide longitudinal evidence of the durability of undetectable VL among PLHIV. However, these data are nationally representative across multiple countries and were collected using the same methods, thereby strengthening the inferences drawn from cross-sectional data. Second, since we restricted our analysis to participants aged 15–59 years, the results are not generalizable to PLHIV of all ages. Third, we relied on self-reported ART duration to determine whether participants were on treatment for 12 months and did not ascertain these data from clinical records. Because we observed a small but increased percentage of undetectable results in this self-reported sample, we posited that the duration of ART use was largely accurate. Despite these limitations, the results of this multinational PHIA analysis offer compelling support for the widespread adoption of the U=U message as a public health intervention across all countries, even in resource-limited settings that do not have universal access to plasma-based VL testing.^3,5,24

Nearly all PLHIV with VL <1000 c/mL from plasma samples in the PHIA surveys attained undetectable VL <200 c/mL and ≤50 c/mL, indicating no risk of transmitting HIV to their sexual partners. The finding provides indirect evidence that in regions with limited access to plasma VL testing, the majority of PLHIV with a DBS VL test result <1000 c/mL likely fall below 200 c/mL or are undetectable based on the assay’s lower limit of detection. U=U messaging in resource-limited settings can motivate those who are unsure of their HIV-positive status to seek HIV testing promptly, enabling PLHIV to initiate treatment and effectively suppress their HIV VL.²¹ Further efforts must consider local context to ensure that U=U messaging is effectively communicated by health care providers and public health practitioners, particularly in low- and middle-income countries.²⁴ U=U applies to those with a suppressed VL <1000 c/mL, highlighting the importance of enhancing accessibility to VL testing and monitoring, even in remote areas using DBS testing.