Effect of Coronavirus Disease 2019 Lockdowns on Identification of Advanced Human Immunodeficiency Virus Disease in Outpatient Clinics in Uganda

Abstract

Using data from 67 Ugandan human immunodeficiency virus (HIV) clinics (July 2019–January 2022), we report a 40% (1005/1662) reduction in the number of people with HIV presenting to care after August 2021 compared to prepandemic levels, with a greater proportion presenting with advanced HIV disease (20% vs 16% in the pre–coronavirus disease 2019 period).

Despite widespread availability of antiretroviral therapy (ART), advanced human immunodeficiency virus (HIV) disease persists globally, particularly in low- and middle-income country settings [1–4]. This population with advanced HIV disease, defined as persons having a CD4 count of <200 cells/µL, contributes to a high proportion of HIV-associated morbidity and mortality [5]. Due to severe immunosuppression, persons with advanced HIV disease often present with opportunistic infections including tuberculosis (TB), cryptococcal meningitis, and severe bacterial infections [6–8]. Patients with advanced HIV disease are therefore at high risk for hospitalization and death if they do not receive timely screening, prophylaxis, and ART. Therefore, for people with advanced HIV disease, the World Health Organization recommends screening and prophylaxis for TB and cryptococcal meningitis, in conjunction with rapid ART initiation and adherence counseling [5].

In the context of ongoing challenges including later HIV diagnoses and poor retention in care leading to advanced HIV disease, the COVID-19 pandemic has further disrupted healthcare delivery. With national lockdowns and interruption of public transport and business, patients are faced with challenges in accessing healthcare services. The effects of the COVID-19 pandemic threaten to reverse the recent progress in reducing the global TB burden [9]. TB detection has fallen as resources are redistributed to COVID-19 diagnosis and treatment. Specifically, staff have been reallocated, and diagnostics have been interrupted as GeneXpert machines have been repurposed for COVID-19 diagnostics in many countries. With priority shifting away from endemic diseases to COVID-19, HIV care systems may suffer.

The purpose of this study is to evaluate the number of people with HIV (PWH) and the number with advanced HIV disease entering HIV care and receiving opportunistic infection screening, before and during the COVID-19 pandemic in Uganda.

METHODS

We retrospectively reviewed clinic data from 1 July 2019 to 31 January 2022 in 67 of 94 government-supported HIV clinics offering services to >150 000 PWH surrounding Kampala, Uganda. These facilities span across all levels of HIV care from semirural clinics to clinics associated with national referral hospitals. As part of routine data collection and reporting, clinics disseminate monthly reports of the number of PWH initiating care. Clinics also report the number of persons presenting with advanced HIV disease, defined as a CD4 count <200 cells/μL. These reports are sent to the Infectious Diseases Institute health system strengthening program quarterly; clinics were selected for this study based on access to these routinely collected data. We summed the monthly number of PWH entering care across all 67 clinics to determine the median and interquartile range of PWH entering care and with advanced HIV disease monthly across predefined periods. We calculated the percentage change using the prepandemic period as the denominator. Additionally, we summarized the number and percentage of persons with advanced HIV disease receiving TB and cryptococcal screening. TB screening was defined as any TB test (lipoarabinomannan or Gene Xpert) performed within 6 months of CD4 testing. Cryptococcal screening was defined as evaluation of blood for cryptococcal antigen. Data were summarized and displayed using Microsoft Excel.

We used the COVID-19 data repository by the Center for Systems Science and Engineering at Johns Hopkins University to summarize new Ugandan COVID-19 cases by month from July 2019 to January 2022 [10]. The COVID-19 lockdown dates were acquired from the Ugandan government COVID-19 information website [11]. Between March and May 2020, there was a period of strict lockdown where all public transportation was halted. Between May and July 2020 there was a less stringent lockdown, where public transportation was available, but schools and many businesses remained closed. Between June and July 2021, there was another a strict lockdown where public transportation was closed and business operations were restricted [12].

RESULTS

Overall during the study period (1 July 2019 to 31 January 2022) 39 903 PWH were newly enrolled into care across the 67 participating clinics, of whom 15% (6171/39 903) presented with advanced HIV disease (Supplementary Table 1). Between July 2019 and February 2020 (pre-COVID period), a median of 1662 PWH newly presented to clinics monthly. During the first lockdown (March 2020 to July 2020), there was a 30% decrease in the median number of PWH presenting to clinic monthly (1158/1662) compared to the pre-COVID period. From August 2020 to May 2021, a period without lockdown restrictions, a median of 1298 PWH newly presented to clinics monthly. During the second lockdown (June to July 2021), there was a 24% reduction in the median number of PWH presenting to clinic monthly (987/1298) compared to August 2020 to May 2021. Thereafter, from August 2021 to January 2022, the median number of PWH presenting to clinic monthly in the absence of a COVID-19 lockdown was 40% lower (1005/1662) compared with the pre-COVID period.

Prior to the COVID-19 pandemic (July 2019 to February 2020), 16% (271/1662) of PWH presented monthly with advanced HIV disease (Figure 1). During the COVID-19 lockdown (March 2020 to July 2020), 8% (87/1158) of PWH presented monthly with advanced HIV disease. However, during the second lockdown (June to July 2021), 19% (185/987) of PWH presented monthly with advanced HIV. During the final reporting period (August 2021 to January 2022), 20% (196/1005) presented monthly with advanced HIV disease in the absence of a COVID-19 lockdown.

Figure 1.

Number of PWH presenting to HIV care by month. Blue bars represent PWH with CD4>200 cells/µL. Orange bars represent PWH with advanced HIV disease (CD4≤200 cells/µL). Covid-19 cases in Uganda plotted in dashed black line. There has been a marked decline in overall number of PWH presenting to care since onset of the Covid-19 pandemic, though the absolute number with advanced HIV disease remains stable.

The percentage of persons with advanced HIV disease who were screened for TB rose from 32% in the pre-COVID period (July 2019 to February 2020) to 52% after the eased restrictions (August 2021 to January 2022). Similarly, among persons with advanced HIV disease, 77% received cryptococcal screening in the pre-COVID period, and 81% received cryptococcal screening after eased restrictions.

DISCUSSION

We identified a 40% reduction in the median number of PWH presenting to care since the onset of COVID-19 in Uganda; this reduction has persisted even in the absence of COVID-19 lockdowns. While only 9% of persons presented with advanced HIV disease during the first March 2020 lockdown, in the final reporting period (August 2021 and beyond), 20% of PWH presented monthly with advanced HIV disease. This reduction in PWH presenting to care has not rebounded to prepandemic levels. Overall, fewer PWH are presenting to care, and a greater proportion are presenting with advanced HIV disease. The percentage of persons receiving TB and cryptococcal screening was unaffected by COVID-19 lockdowns.

The reduction of new PWH registering for care that we describe, if generalizable to other centers, threatens the achievement of the Joint United Nations Programme on HIV/AIDS 95-95-95 targets. A study of 11 African countries identified a decrease in the number (1) presenting for HIV testing, (2) testing HIV positive, and (3) initiating ART [13]. However, these reductions quickly recovered upon lifting of national restrictions. Conversely, Uganda had the longest period of national COVID-19 restrictions, including extended closure of public transport and schools. These restrictions were implemented with increasing level of stringency over time; this explains why there was a marked decline in PWH presenting to care in the 2021 lockdown compared to 2020. In another study evaluating effects of the COVID-19 pandemic on 44 President’s Emergency Plan for AIDS Relief–supported countries, there was a 19% decline in the number of PWH testing positive and enrolling in care [14]. A South African study revealed a more dramatic 48% decrease in HIV testing, and a 46% reduction in ART initiation after COVID-19 lockdowns [15]. Our study identifies a growing proportion of PWH presenting to care with advanced HIV disease, demonstrating the need to ensure continued delivery of HIV services despite the ongoing COVID-19 pandemic. With respect to the effect of COVID-19 lockdowns on opportunistic infection screening, a Kenyan study observed a 31% decrease in presumptive TB diagnoses and similar reduction in patients registered in TB care [16]. A national South African surveillance study identified a decline in cryptococcal screening, with a greater proportion testing positive [17].

Limitations of our analysis are related to selection bias with respect to clinics included in this study, as we selected a convenience sample where aggregate data could easily be extracted. Furthermore, we could not control the number of months of the COVID-19 lockdowns, and we did not adjust for seasonality. Data availability was inconsistent prior to July 2019; therefore, the prepandemic reporting period could not be extended across the 67 participating clinics. In a post hoc analysis, we examined available data from 49 clinics during April 2018 to 2019 and found that 11% of PWH presented with advanced HIV disease monthly, which is notably lower than 20% presenting with advanced HIV disease in the post-lockdown period.

The coexistence of the HIV and COVID-19 epidemics has threatened gains previously achieved in the HIV pandemic. Facility-based HIV care, which has historically been the mainstay of delivery of HIV care services in sub-Saharan Africa, will continue to be challenged; a transition to community-based patient care may be a more sustainable service delivery model [18].

In this study of >39 000 PWH entering HIV care, we observed a 30% decline in median number of PWH presenting to HIV care monthly during the COVID-19 lockdown restrictions in Uganda. Even in the absence of COVID-19 restrictions, a 40% reduction in the number of PWH presenting to care has persisted. Persons who finally present to care were more likely to have advanced HIV disease. Innovation and optimization of HIV service delivery is urgently needed.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Acknowledgments. Special thanks to the government of Uganda, the Centers for Disease Control and Prevention, Kampala, and other stakeholders for their commitment to ensuring that HIV prevention, care, and treatment services are made readily available and accessible to all Ugandans at no cost.

Financial support. R. R. and D. R. B. are supported by the National Institute of Allergy and Infectious Diseases (award numbers K23AI138851 and U01AI125003). C. P. S. is supported by the National Institutes of Health National Center for Advancing Translational Sciences (award numbers KL2TR002492 and UL1TR002494). P. S. is supported by a combined National Institute of Neurologic Disorders and Stroke and Fogarty International Center award (award number D43TW009345) via the Northern Pacific Global Health Fellows Program. J. E. is supported by a Wellcome Trust Clinical PhD Fellowship (grant number 203905/Z/16/Z). K. H. H. is supported by the University of Minnesota, Division of Biostatistics.