ABSTRACT
Background: The pandemic response measures have had significant global economic and health impacts with transient reductions in HIV clinic attendance and self-reported anti-retroviral therapy (ART) adherence reported in prior studies. Since viral suppression (VS) is an indication of ART adherence and effective service delivery, we assessed VS in the context of the COVID-19 pandemic in 3 African countries Methods: Since 2013, the African Cohort Study (AFRICOS) has enrolled individuals 18 years or older with and without HIV, in an approximate 5:1 ratio, at 12 clinics across 5 HIV care programs in Tanzania Uganda, Kenya, and Nigeria. For people living with HIV (PLWH), ART history was extracted from medical records and viral load was assessed at each visit. This assesses VS (< 1000 c/ml) before and during the COVID-19 pandemic (categorized into 4 surges and a consolidated non-surge period;defined in Table 1) among PLWH. Tanzania was excluded due to inadequate pandemic data. Logistic regression with generalized estimating equations, clustered by participant, was used to estimate odds ratios (ORs) and 95% confidence intervals (CI) comparing VS before and during COVID-19. Models are adjusted for age, sex, and program. Result(s): Of the 1741 study participants, 368 are from Uganda, 1156 are from Kenya, and 217 are from Nigeria;730 are males, 1011 are females, and 147 are under the age of 30. PLWH were less likely to be virally suppressed during the first surge period (OR 0.85, CI 0.46-1.56), but VS significantly increased during the second surge period (OR 1.95, CI 1.23-3.04) compared to the pre-COVID period. The third and fourth surge periods also saw a higher VS (table 1). Females are more likely to be virally suppressed than males (OR 1.58, CI 1.09- 2.29) and PLWH ages 40-49 have higher VS (OR 2.43, CI 1.32-4.48) compared to PLWH under. PLWH at the AFRICOS sites in Kenya and Nigeria show lower VS than the Ugandan cohort (ORs 0.46, CI 0.26-0.79 and OR 0.32, CI 0.17-0.60 respectively). Conclusion(s): The initial drop in VS may be attributed to reduced clinic access due to lockdowns. Many HIV programs supported by the President's Emergency Plan for AIDS Relief (PEPFAR) adapted their strategies to serve PLWH by scaling up community ART dispensing and multi-month dispensing (MMD) of ART for stable clients, which could have led to increased VS during the other surge periods.
ABSTRACT
Background. Multi-month dispensing (MMD) of antiretroviral therapy (ART) decreases logistical burdens on HIV clinics and patients, which is especially important during the COVID-19 pandemic. HIV programs are scaling-up 6-month dispensing (6MD), but the impact on viral suppression (VS) has not been well-documented in programmatic settings. Methods. The African Cohort Study (AFRICOS) is an international observational study of people living with HIV (PLWH) receiving HIV care. In Nigeria and Kenya, this includes 6MD. Participants undergo semiannual viral load quantification and were included in analysis if they had complete data, documentation of MMD (self-reported) and at least two follow-up visits after initiating MMD. In stratified analyses for each country, we used multivariable logistic regression with generalized estimating equations to estimate adjusted odds ratios (aOR) and 95% confidence intervals (95%CI) comparing VS < 50 copies/mL among those who received 6MD to those given 3-5 months dispensing (3-5MD) of ART. Analyses were adjusted for age, education, employment, distance to clinic, duration on ART, ART regimen (TLD vs other) and self-reported ART adherence (>= 1 missed dose in past 30 days). Results. Between January 30, 2020, when MMD data was first collected, and September 1, 2021, 1176 PLWH in Kenya and 272 in Nigeria had at least one visit;285 participants from Kenya, totaling 442 visits, and 177 participants from Nigeria, totaling 382 visits, met criteria and were included in analysis. At most recent visit, VS < 50 copies/mL was documented in 266 (93.3%) participants from Kenya and 125 (70.6%) participants from Nigeria. Among the 35 participants given 6MD in Kenya, compared to 250 participants given 3-5MD, the aOR for VS was 0.42 (95% CI: 0.13-1.37);among the 91 participants given 6MD in Nigeria, compared to the 86 participants given 3-5MD, the aOR was 3.01 (95%CI: 1.70-5.31). Conclusion. The positive association between 6MD and VLS in Nigeria, as compared to 3-5MD, should prompt more aggressive scale-up of 6MD. The lack of an association in Kenya merits further investigation, but likely relates to high overall VS and few participants on 6MD. Updated data from specific geographic and demographic sub-populations is needed to inform programming as 6MD is scaled.