Factors Associated With Severity of COVID-19 Disease in a Multicenter Cohort of People With HIV in the United States, March–December 2020

Adrienne E. Shapiro, MD, PhD; Rachel A. Bender Ignacio, MD, MPH; Bridget M. Whitney, PhD; Joseph A. Delaney, PhD; Robin M. Nance, MS; Laura Bamford, MD, MSCE; Darcy Wooten, MD, MS; Jeanne C. Keruly, MS, CRNP; Greer Burkholder, MD, MSPH; Sonia Napravnik, PhD; Kenneth H. Mayer, MD; Allison R. Webel, RN, PhD; H. Nina Kim, MD, MSc; Stephen E. Van Rompaey, PhD; Katerina Christopoulos, MD, MPH; Jeffrey Jacobson, MD; Maile Karris, MD; Davey Smith, MD, MAS; Mallory O. Johnson, PhD; Amanda Willig, RD, PhD; Joseph J. Eron, MD, MPH; Peter Hunt, MD; Richard D. Moore, MD, MHS; Michael S. Saag, MD; W. Christopher Mathews, MD, MSPH; Heidi M. Crane, MD, MPH; Edward R. Cachay, MD, MAS; Mari M. Kitahata, MD, MPH


J Acquir Immune Defic Syndr. 2022;90(4):369-376. 

In This Article


Among 16,056 PWH in care during the study period, 20.9% were female; the median age was 52 years (IQR 40–59) and 44.5% were non-Hispanic Black, 37.9% non-Hispanic White, 12.5% Hispanic, and 5.2% mixed, other, or unreported race/ethnicity. Most of them (95.5%) were on ART, and 85.6% had an undetectable HIV VL. Between March and December 2020, a total of 649 PWH had COVID-19 disease, of whom 28% were female, 52% aged 50 years or older, 51% non-Hispanic Black, 25% non-Hispanic White, and 20% Hispanic of any racial identity. Most of the COVID-19 cases (77%) had a documented positive SARS-CoV-2 reverse transcriptase-polymerase chain reaction or antigen test result within the CNICS system. The remaining 23% of cases were verified from a variety of sources, including polymerase chain reaction and antigen testing performed in community settings, public health testing, or non–CNICS-affiliated medical sites. As in the overall cohort, most of the PWH with COVID-19 were on ART (95%) and had an undetectable VL (86%). While 45% of PWH with COVID-19 had a lowest CD4 count <200 cells/mm3, more than half (66%) had a current CD4 count ≥500 cells/mm3 (Table 1).

One hundred-six (16%) PWH with COVID-19 disease were hospitalized. In univariate analyses, demographic characteristics associated with hospitalization included female sex, Black race, and older age. Compared with nonhospitalized PWH with COVID-19, lowest CD4 count, current CD4 count, elevated FIB-4 and ASCVD risk scores, diabetes, hypertension, CKD (eGFR <60), obesity (BMI ≥30 kg/m2), and COPD, were each associated with the risk of COVID-19 hospitalization (all P < 0.006). Notably, 60% of hospitalized cases had a lowest CD4 count <200 cells/mm3 compared with 42% of nonhospitalized cases, and 50% of hospitalized cases had a current CD4 count ≥500 cells/mm3 compared with 70% of nonhospitalized cases. Having an FIB-4 score >3.25 (highly predictive of hepatic fibrosis) was strongly associated with hospitalization (7% vs. 1%, P < 0.001). PWH with an FIB-4 score >1.45 were also more likely to be hospitalized with COVID-19 (38% vs. 19%, P < 0.001). HCV coinfection, smoking, injection drug use (IDU) as the reported risk of HIV, and being unstably housed/homeless were not associated with hospitalization among cases with available data.

In adjusted analysis, low current CD4 count was more strongly predictive of hospitalization than age or any comorbid condition. Both current CD4 count <350 cells/mm3 [aRR: 2.68; 95% confidence interval (CI): 1.93 to 3.71; P < 0.001] and lowest CD4 count <200 cells/mm3 (aRR 1.67; 95% CI: 1.18 to 2.36; P < 0.005) were associated with hospitalization (Figure 1). There was a trend toward protection against hospitalization with higher current CD4/CD8 ratio, that is, aRR 0.88 for each 1 SD increase in ratio (95% CI: 0.75 to 1.03; P = 0.08). There was no association between risk of hospitalization and detectable VL. Although there were insufficient PWH not receiving ART to evaluate an adjusted risk, the proportion of hospitalized cases on ART before COVID-19 diagnosis (97%) did not differ significantly from nonhospitalized cases (94%).

Figure 1.

Relative risk of hospitalization with COVID-19 among PWH with COVID-19 by key characteristics. FIB-4, Fibrosis-4 scoring system for liver fibrosis; RR, relative risk. Reference range for age per decade is 18–29 years, the subsequent ordinal categories being 30–39, 40–49, 50–59, and 60+, as in Table 1. aRelative risk regression models adjusted for demographic and clinical characteristics using disease risk scores, except for the ASCVD risk score analysis, which is unadjusted. Disease risk scores were constructed independently for each exposure variable of interest using all nonduplicative covariates.

In adjusted analyses, PWH with CKD (eGFR <60) had a more than 2-fold risk of hospitalization (aRR 2.31; 95% CI: 1.63 to 3.28, P < 0.001), and those with diabetes, hypertension, obesity, or COPD also had a higher risk of hospitalization (Figure 1). In addition, we found a strong association between clinically validated risk scores for cardiovascular and liver disease (ASCVD and FIB-4) and a higher risk of hospitalization. For each 10% increase in ASCVD risk score, the aRR for hospitalization was 1.37 (95% CI: 1.25 to 1.51; P < 0.001). PWH who had an FIB-4 score above the 1.45 cutoff were 2.33 times more likely to be hospitalized with COVID-19 compared with those below this cutoff (95% CI: 1.66 to 3.28; P < 0.001).

Although women were overrepresented among those hospitalized, after adjusting for other factors, sex was not associated with hospitalization [aRR 1.30; 95% CI: 0.92 to 1.85, P = 0.14). As seen in the general population, an age limit of 50 years or older was associated with a more than 2-fold risk of hospitalization (aRR 2.13; 95% CI: 1.48, 3.07) P < 0.001]. Adjusting for other covariates, the risk of hospitalization due to COVID-19 increased 42%, on average, for every additional 10 years of age. PWH identifying as non-Hispanic Black were 51% more likely to be hospitalized with COVID-19 compared with other racial/ethnic identities (aRR 1.51; 95% CI: 1.04 to 2.19, P = 0.03). Hispanic ethnicity was not significantly associated with hospitalization (aRR = 0.71 95% CI: 0.41 to 1.24; P = 0.23).

To further evaluate whether demographic characteristics associated with COVID-19 hospitalization were partially attributable to the differential distribution of comorbidities, we performed sensitivity analyses that also adjusted for comorbidities including diabetes, hypertension, CKD, obesity, COPD, and liver disease (FIB-4) and found similar associations between COVID-19 hospitalization and racial/ethnic identity; the associations between age and hospitalization were attenuated but overall similar.

Twelve (2%) PWH with COVID-19 died; 3 deaths were in persons never hospitalized for COVID-19, and 9 died in hospital or within 30 days of discharge. Thirty-one (29%) of the 106 hospitalized persons were admitted to the ICU, and 16 (15%) were intubated (Table 2). Owing to small numbers who were intubated or died, we have reported proportions but not adjusted risk estimates for critical illness outcomes. Older PWH, those with CD4 count <350 cells/mm3, persons not on ART, and those with diabetes and CKD were overrepresented among those who required intubation or died of COVID-19. No one younger than 30 years was admitted to the ICU or died, whereas 25 (81%) of those admitted to the ICU and 9 (75%) of the deaths were among PWH aged 50 years and older. As with most COVID-19 hospitalizations in the United States, most of the PWH (65%, Table 3) received some degree of oxygen or respiratory support, although presentations with gastrointestinal symptoms, sepsis physiology, acute kidney injury, or COVID-19 pneumonia without significant hypoxemia were observed. Thirty-four percentage of hospitalized PWH required only nasal canula as the maximal level of support, 10% required a high-flow nasal canula, and 15% were intubated; none received extracorporeal life support (Table 3). While the case fatality rate among PWH with COVID-19 was 12 (2%) of 649, 9% of hospitalized COVID-19 cases died, including nearly one-third (31%) of those intubated. Most of the nonhospitalized COVID-19 cases had no record of receiving any approved or experimental COVID-19–specific treatment (94%), whereas 64% of hospitalized COVID-19 cases and 77% of critically ill COVID-19 cases received COVID-19 pharmacotherapy. Consistent with changes in clinical care during 2020, the most common medications administered to hospitalized PWH with COVID-19 were dexamethasone (45%) and remdesivir (42%), with fewer receiving convalescent plasma or hydroxychloroquine and anti-SARS-CoV-2 monoclonal antibodies or participating in any interventional trial (see Table 1, Supplemental Digital Content, https://links.lww.com/QAI/B840).