HIV and COVID-19: Intersecting Epidemics With Many Unknowns

Catherine R. Lesko; Angela M. Bengtson


Am J Epidemiol. 2021;190(1):10-16. 

In This Article

Incidence and Prevalence of SARS-CoV-2 in PLWH

It is unclear whether or not PLWH are at higher risk for infection with SARS-CoV-2 or for poor clinical outcomes subsequent to infection. There are reasons to hypothesize that PLWH are a high-risk group: Antibody responses to an immune system challenge are impaired in PLWH, and PLWH have high prevalence of risk factors for severe SARS-CoV-2 infection, including hypertension, diabetes, cardiovascular disease, obesity, lung disease and smoking, male sex, and older age.[1,2] Alternatively, worse COVID-19 outcomes might be due to immune (over)activation, and thus PLWH might actually be at lower risk for poor outcomes following SARS-CoV-2 infection due to their reduced immune response.[3] However, there are not yet sufficient data to support or refute either of these hypotheses.

Early in the course of an epidemic of a novel pathogen, evidence is scarce, and the most practical or indeed the only epidemiologic study design available to us is the case report or case series.[4–8] Early case reports and case series of COVID-19 in PLWH told of an occasionally atypical, but not more severe, disease course relative to people living without HIV.[7–13] Some case series suggested that PLWH with COVID-19 might be younger than persons with COVID-19 in the general population.[11,12] However, incidence and mortality rates for COVID-19 will be a function of the age structure of the underlying populations of people with versus without HIV; thus, it is difficult to compare rates without age standardization.

Data on the incidence of COVID-19 in PLWH is slowly amassing from population- (i.e., surveillance) and clinic-based cohorts of PLWH. While important, absolute risk estimates will reflect: 1) SARS-CoV-2 infection dynamics such as the force of infection in a community and duration of follow-up (cumulative risk infection increases monotonically); 2) demographic and clinical characteristics of PLWH in the population (presumably, higher prevalence of comorbidities would be associated with higher risk of COVID-19, independent any direct causal effect of HIV infection on COVID-19); and 3) excess risk of COVID-19 attributable to HIV infection. Among 1,174 PLWH living in the Wuchang or Qinshan districts of Wuhan, China, 8 had confirmed COVID-19 (0.7%, as of the end of February or beginning of March 2020), which was comparable to the risk in the general population in Wuhan (0.5%).[14] Among 1,339 PLWH engaged in regular care in Madrid, Spain, 51 (3.8%) were diagnosed with COVID-19 as of April 30, 2020. The risk of COVID-19 in Madrid for the same period (4.0%) was comparable.[15] Finally, the SARS-CoV-2 positivity rate among PLWH tested in a medical center in Chicago, Illinois (15%), was comparable to the positivity rate among people without HIV (19%).[16] In contrast to these cohorts suggesting similar infection rates in people with and without HIV, unpublished surveillance data from South Africa's Western Cape province through June 9, 2020, suggest that PLWH were 2.3 times as likely to die from COVID-19 as people without HIV, after age and sex standardization.[17]

Certainly, more information is needed. Surveillance data, such as those available from South Africa or Wuhan, will provide the most complete picture of COVID-19 risk among PLWH (e.g., by not restricting to PLWH who are in care and who are more likely to have well-controlled HIV disease); however, clinical data, such as those from Madrid, might provide the most depth (e.g., by allowing examination of the role of comorbidities, medications, and COVID-19 treatments) as long as potential selection bias is considered. Perhaps the most fruitful investigation would be one that merged clinical and surveillance data.