. 2021 Feb 10;35(4):F1–F10. doi: 10.1097/QAD.0000000000002836

Table 1.

Details of all included studies with summary of mortality findings and quality appraisal.

Study details	Location	Study type and study population	Characteristics of PLWH	Mortality	Quality appraisal^a
Studies including all COVID-19 patients (hospitalized and in the community)
Boulle et al.[20]	Multi-centre: Western Cape, South Africa	Prospective cohort study of 540 552 PLWH (3978 confirmed COVID-19) & 2920 380 HIV-negative adults (18 330 confirmed COVID-19)	34% male, median age 20–39 years. No data on proportions by ethnicity, ART, or any co-morbidity. 7.6% with VL > 1000 copies/ml or CD4⁺ cell count <200 cells/μl, 34.7% with VL unknown in past 15 months. No data on corticosteroid use	115 (2.1/10 000) COVID-19 deaths in PLWH vs. 510 (1.7/10 000) in adults without HIV. Adjusted for age and sex, aHR: 1.97; 95% CI: 1.59–2.45. Adjusted for age, sex and co-morbidities: aHR: 2.14; 95% CI: 1.70–2.70^b	JBI C = 11. Risk of confounding by socio-economic status and BMI
Bhaskaran et al.[22]	Multi-centre, United Kingdom	Retrospective population-based cohort analysis of primary care data of 27 480 PLWH and 17 282 905 adults without HIV. Numbers diagnosed with COVID-19 not presented	65% male, median age 40–55 years, 46.2% White, 5.1% Mixed, 4.3% South Asian, 26% Black, 2.3% Other. No data on proportions with any co-morbidity, or by ART, CD4⁺ cell count, viral suppression. No data on systemic corticosteroid use	25 (9.1/10 000) COVID-19 deaths in PLWH vs. 14 857 (8.6/10 000) in adults without HIV. Adjusted for age and sex, aHR: 2.90, 95% CI: 1.96–4.30. Adjusted for age, sex, ethnicity, deprivation and co-morbidities, aHR 2.30; 95% CI: 1.55–3.41^c	JBI C = 11. Small number of outcomes among PLWH therefore possible overfitting of model
Hadi et al.[12]	Multi-centre, United States	Retrospective cohort study among people with SARS-CoV-2 infection comparing 404 PLWH with 49 763 adults without HIV (including matched cohort of 404 adults^d)	71% male, mean age 48.1, 50% Black or African-American, 34% White, 13% Latinx, 3% Asian. No data on proportions with any co-morbidity. 70% on ART, no data on CD4⁺ cell count or VL. No data on corticosteroid use	20 (4.95%) COVID-19 deaths in PLWH vs. 1585 (3.19%, RR 1.55, 95% CI: 1.01–2.39) in unmatched cohort, and 15 (3.71%, RR 1.33, 95% CI: 0.69–2.57) in matched cohort	JBI C = 10. Unclear how COVID-19 diagnosed. 1 : 1 matching may result in underpowered analysis
del Amo et al.[13]	Multi-centre, 60 Spanish HIV clinics	Cohort study of 236 PLWH receiving ART with COVID-19, recruited from 77 590 patients of HIV clinics	75% male, median age 50–59 years. 100% receiving ART. No data on ethnicity, CD4⁺ cell count, viral load or other co-morbidities. 64% hospitalized, 6% in ICU	20 (8%) PLWH died, giving age and sex standardized risk of death of 3.7/10 000 vs. 2.1/10 000 in general Spanish population	JBI C = 7. Comparison group from a different population. Confounding factors not accounted for
Miyashita et al.[23]	Multi-centre: New York, US	Cohort study of people with SARS-CoV-2 infection comparing 161 PLWH to 8751 people without HIV	78% male, median age 51–65 years. No data on ethnicity, ART, CD4⁺, VL, or proportion with any co-morbidity	23 (14%) PLWH died vs. 1235 (14%) without HIV (but in age <50 years), PLWH had higher risk of death (RR 4.36, 95% CI 1.43–13.3)	JBI C = 7. Confounding factors not accounted for. Insufficient follow-up time for some patients
Ho et al.[24]	Multi-centre, New York, US	Case series of 93 PLWH who presented to ED with positive SARS-CoV-2 RT-PCR test	72% male, median age 58 years, 40.9% Black, 31.2% Hispanic/Latinx. 96% on ART, 84% VL <50 copies/ml, median CD4⁺ cell count 554 cells/μl	19 (20.4%) PLWH died, no mortality comparison given	JBI CS = 8. Deaths among those not admitted not recorded
Di Biagio et al.[25]	Multi-centre, Italy	Case series of 69 PLWH who were diagnoses with SARS-CoV-2 infection	72% male, median age 50–55, 86% white, 100% on ART, 88% VL < 50 copies/ml, median CD4⁺ cell count 580–600 cells/μl	7 (10.8% of those with known outcomes) PLWH died, no mortality comparison given	JBI CS = 5. Not clear if participating centres identified SARS-CoV-2 cases systematically. Statistics unclear
Maggiolo et al.[26]	Single-centre, Bergamo, Italy	Cohort study of 55 PLWH with suspected/confirmed COVID-19 vs. 69 asymptomatic PLWH who tested negative for SARS-CoV-2 (RT-PCR or serology)	80% male, median age 54 (IQR: 49–58), 100% receiving ART, 98% VL < 50 copies/ml, median last CD4⁺ cell count 904 cells/μl, 51% with any co-morbidity	4 (7.2%) PLWH died, no mortality comparison given	JBI C = 6. Follow-up unclear. Confounding factors not accounted for. Sample size not large enough for multi-variable analysis
Etienne et al.[27]	Single centre, Paris, France	Case series of 54 PLWH with symptoms or were hospitalized with COVID-19	61.1% male, median age 54, 100% on ART, 96.2% VL < 40 copies/ml, median CD4⁺ cell count 583 cells/μl, 44.6% sub-Saharan African origin	1 (1.9%) PLWH died, no mortality comparison given	JBI CS = 3. COVID-19 case definition and disease severity not defined. Short follow-up time
Inciarte et al.[28]	Single centre, Barcelona, Spain	Cohort study of 5683 PLWH in of whom 53 PLWH had confirmed or suspected COVID-19	85% male, median age 44, median last CD4⁺ cell count 618 cells/μl (IQR: 449–834), 96% on ART	2 (4%) PLWH died, no mortality comparison given	JBI C = 7. COVID-19 case definition unclear. Duration of follow-up unclear. Confounding not accounted for
Gervasoni et al.[29]	Single centre: Milan, Italy	Case series of 47 PLWH with suspected/confirmed COVID-19	74% male, median age 52 years, no ethnicity data. 100% receiving ART, median CD4⁺ cell count 636 cells/μl, 93.6% with undetectable VL. 64% with ≥1 co-morbidity	2 PLWH died (4.2% of cohort, 15.4% of those hospitalized with COVID-19), vs. 17% died among 502 HIV-negative patients admitted with COVID-19 at same hospital	JBI CS = 7. Comparison group from a different population. Length of follow-up unclear. Confounding factors not accounted for
Huang et al.[30]	Multi-centre, Wuhan City, China	Cohort study 6001 PLWH of whom 35 were diagnosed with COVID-19	90% male, median age 37 years, no data on ethnicity. 92% on ART, median CD4⁺ cell count 200–499 cells/μl, 66% VL < 20 copies/ml	2 (5.7%) PLWH with COVID-19 died vs. 3869/50 333 (7.69%) of general population of Wuhan with COVID-19	JBI C = 5. Comparison group from a different population. Follow-up not clear. Confounding not accounted for
Härter et al.[14]	Multi-centre: 12 German HIV centres	Case series of 33 PLWH with confirmed COVID-19	91% male, median age 48 years, no ethnicity data. 100% receiving ART, overall median CD4⁺ cell count 670 cells/μl, 94% virally suppressed, 60%. with ≥1 co-morbidity	3 PLWH died (9.1%) vs. 3.7% mortality in general COVID-19 positive population in Germany	JBI CS = 7. Comparison group from a different population and confounding not accounted for. Not all patients with an outcome
Studies including only hospitalized COVID-19 patients
Geretti et al.[21]	Multi-centre: 207 UK hospitals	Prospective cohort study of people hospitalized with suspected/confirmed COVID-19. 122 PLWH vs. 47 470 HIV-negative adults	66% male, median age 56 (IQR: 49–62) years. White (45.5%), Black (42.9%), Asian (0.9%), Other (10.7%). PLWH had fewer co-morbidities overall, 74.6% ≥ 1 co-morbidity. 91.8% had a record ART, No data on proportions by CD4⁺ or VL. No data on corticosteroid use	By day 28, 30 PLWH died (24.6%) vs. 13 969 (29.4%) in adults without HIV. Adjusted for age and sex, aHR: 1.45; 95% CI: 1.00–2.12 (P = 0.05). Adjusted for age, sex and co-morbidities, aHR: 1.69; 95% CI: 1.15–2.48, P = 0.008^e	JBI C = 11. Risk of confounding by socioeconomic status
Sigel et al.[15]	Multi-centre: 5 hospitals New York, US	Cohort study of 88 PLWH hospitalized with laboratory confirmed COVID-19 compared with matched cohort of 405 hospitalized HIV-negative adults^f	78% male, median age 45–67 years. Ethnicity: White (19%), Black, (40%), Hispanic (30%), Other (11%). No data on proportions with any co-morbidity. 100% on ART, 58% CD4⁺ cell count >200 cells/μl, 81% VL < 50 copies/μl. No data on corticosteroid use	18 (21%) COVID-19 deaths in PLWH vs. 81 (20%) in adults without HIV	JBI C = 6. Follow-up not complete for all patients. Matched on limited number of confounders. Potential overadjustment
Vizcarra et al.[16]	Single centre: Madrid	Case series of 51 PLWH with suspected/confirmed COVID-19	84% male mean age 53.3 years, 88% white ethnicity. 100% receiving ART, median CD4⁺ cell count 565 cells/μl, 98% virally suppressed, 63% with ≥1 co-morbidity. 15 (38%) received corticosteroids	2 PLWH died (4%) vs. 20% among general population admitted to nearby hospital with COVID-19	JBI CS = 7. Comparison group from different population. COVID-19 case definition unclear. Not all cases completed follow-up
Shalev et al.[17]	Single centre: New York, US	Case series of 31 PLWH hospitalized for COVID-19 at a large tertiary care medical centre in New York City	77% male, mean age 60.7 years, 52% black, 29% Hispanic, 9% white. 100% receiving ART, median CD4⁺ cell count 396 cells/μl, 100% virally suppressed, 71% with ≥1 co-morbidity. 8 (25.8%) received corticosteroids	8 PLWH died (27.6%), no mortality comparison given	JBI CS = 8. Not all cases completed follow-up
Karmen-Tuohy et al.[18]	Multi-centre: 4 hospitals in New York, US	Cohort study of 21 PLWH hospitalized with confirmed COVID-19 compared with a matched cohort^g of 42 HIV-negative patients (selected from 2617 non-HIV patients with COVID-19 at same centres)	90.5% male, mean age 60 years, 24% African American, 38% White, 38% Other. 100% receiving ART, median CD4⁺ cell count 298 cells/μl, 71% VL < 50 copies/ml. 4 (19%) PLWH received corticosteroids vs. 0 in the HIV-negative matched cohort	6 PLWH died (28.6%) vs. 10 (23.8%) in HIV-negative cohort (P = 0.682)	JBI C = 11
Childs et al.[19]	Single centre: London, UK	Case series of 18 PLWH hospitalized with confirmed COVID-19	67% male, median age 52, 94% black ethnicity. 100% on ART, median CD4⁺ cell count 395 cells/μl, 94% virally suppressed. No data on corticosteroid use	5 PLWH died (29%), no mortality comparison given	JBI CS = 6. Time period of enrolment and follow-up time not defined

aHR, adjusted hazard ratio; ART, antiretroviral therapy; CI, confidence interval; COVID-19, coronavirus disease 2019; ED, emergency department; IQR, interquartile range; PLWH, people living with HIV; RR, risk ratio; RT-PCR, real-time-polymerase chain reaction; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; VL, viral load. Bold indicates number of PLWH diagnosed with COVID.

Joanna Briggs Institute (JBI) Cohort Study (C) or Case Series (CS) checklist items completed, with comment. See Tables S2 and S3 for full scores.

Adjusted for age, sex, diabetes, tuberculosis history and other co-morbidities (hypertension, kidney disease, lung disease).

Adjusted for age, sex, deprivation, ethnicity, obesity, smoking and presence of co-morbidities (hypertension, chronic respiratory disease, chronic cardiac disease, diabetes, non-haematological cancer, haematological cancer, chronic liver disease, stroke, dementia, other neurological disease, reduced kidney function, organ transplant, asplenia, rheumatoid arthritis, lupus, psoriasis or other immunosuppresive disorders).

Matched on age, sex, ethnicity, smoking, BMI, diabetes, hypertension and chronic lung diseases.

Adjusted for age, sex, ethnicity, baseline date, indeterminate/probable hospital acquisition of COVID-19, 10 co-morbidities and hypoxia/receiving oxygen at presentation.

Matched on age, sex, race/ethnicity and week of SARS-CoV-2 infection.

Matched on age, sex, BMI, smoking, co-morbidities (chronic kidney disease, hypertension, asthma, chronic obstructive pulmonary disease, heart failure) and admission date, all from same centre.