Coronavirus disease 2019 vaccine effectiveness among a population-based cohort of people living with HIV : AIDS

Secondary Logo

Journal Logo

FAST TRACK

Coronavirus disease 2019 vaccine effectiveness among a population-based cohort of people living with HIV

Chambers, Catharinea,b; Samji, Hasinac,d; Cooper, Curtis L.e; Costiniuk, Cecilia T.f; Janjua, Naveed Z.c,g,h; Kroch, Abigail E.a,i,j; Arbess, Gordonb,k; Benoit, Anita C.a,l,m; Buchan, Sarah A.a,j,n,o; Chung, Hannahn; Kendall, Claire E.n,p,q; Kwong, Jeffrey C.a,j,k,n,o,r; Langlois, Marc-Andrés; Lee, Samantha M.n; Mbuagbaw, Lawrencet,u,v,w,x,y; McCullagh, Johnz; Moineddin, Rahimk; Nambiar, Devanaa; Walmsley, Sharonr; Anis, Aslam H.g,h; Burchell, Ann N.a,b,k,n

Author Information
doi: 10.1097/QAD.0000000000003405

Abstract

Introduction

People with HIV may be at higher risk for coronavirus disease 2019 (COVID-19), particularly those who have advanced or untreated HIV infections [1]. HIV-associated immune suppression may increase the risk of COVID-19-related hospitalization compared with the general population [2,3], but its effect on incident disease and mortality is less clear [2–6]. Moreover, many people with HIV face multiple intersecting vulnerabilities, such as belonging to low socio-economic or racialized groups, which may increase their risk of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) exposure. They may also have other risk factors, such as older biological age and chronic comorbidities [3].

Although people with HIV who participated in early immunological studies do appear to mount a sufficient antibody response to COVID-19 vaccines [7,8], this population was underrepresented in pre-licensure clinical trials of vaccine efficacy, which showed conflicting results [9–13]. The small number of participants with HIV in these trials represented <1–6% of the primary efficacy population; furthermore, the few who were included had normal absolute CD4+ T-cell numbers (>500 cells/mm3) and few comorbidities [9,10,12]. Observational studies of vaccinated cohorts suggest that people with HIV may be at higher risk of breakthrough infections compared with their non-HIV infected counterparts [14,15].

Prior studies of COVID-19 vaccine effectiveness (VE) in the general population using administrative data in Ontario, Canada, suggest that two doses of mRNA vaccines are highly effective at preventing symptomatic infection and severe COVID-19 outcomes, with VE estimates ranging from 91% to 98% during the initial waves of the pandemic [16]. The extent to which these findings apply to people with HIV is unclear. Our primary objective was to estimate COVID-19 VE against RT-PCR-confirmed SARS-CoV-2 infection for the BNT162b2, mRNA-1273, and ChAdOx1 vaccines during the pre-Omicron period among a population-based cohort of people with HIV in Ontario, Canada. A secondary objective was to estimate VE against COVID-19-related severe outcomes, including hospitalization and death.

Methods

Setting

In Canada, COVID-19 vaccines first became available on December 14, 2020, with four formulations authorized by the end of 2021: Pfizer-BioNTech (BNT162b2), Moderna (mRNA-1273), AstraZeneca (ChAdOx1), and Janssen (AD26.COV2.S). mRNA vaccines (BNT162b2 or mRNA-1273) comprised the majority of distributed product, either as mixed or homologous schedules, with <1% receiving the AD26.COV2.S vaccine. Vaccines were initially prioritized for residents and staff of senior living facilities, older adults aged ≥70 years, healthcare workers, and people in Indigenous communities, with gradual roll out to younger ages and communities with higher COVID-19 incidence. People with HIV were not initially prioritized for the two-dose primary series, but those with advanced or untreated HIV infection were prioritized to receive a third dose starting in September 2021. Publicly-funded testing was available for individuals who were symptomatic, had close contact with a confirmed case, were part of an outbreak investigation, were hospitalized, or were considered at greater risk due to their health condition or employment. Rapid or point-of-care antigen tests were not widely available during the study period.

Study population and design

We conducted a retrospective, population-based study among people with HIV using linked administrative health datasets available at ICES (formerly known as the Institute for Clinical Evaluative Sciences). As of 2020, an estimated 23 380 people were living with HIV in Ontario (range = 20 000–26 500), more than one-third of people with HIV in Canada [17]. The majority (88%) were accessing HIV care and 85% were on antiretroviral therapy, of whom 97% were virally suppressed [18].

We identified adults living with HIV as of December 14, 2020, using a validated case-finding algorithm based on three physician billing claims with an HIV diagnosis (three-digit ICD-9 codes) over a 3-year period [19]. This algorithm has 96.2% sensitivity and 99.6% specificity against medical charts [19]. Additional inclusion criteria were: aged ≥19 years; residing in Ontario; eligible for provincial health insurance; and tested for SARS-CoV-2 by RT-PCR between December 14, 2020 (first availability of COVID-19 vaccines in Ontario) and November 21, 2021 (before first detection of Omicron variant). We excluded individuals who were residents of long-term care facilities, tested positive for SARS-CoV-2 before December 14, 2020, received ≥1 dose of the AD26.COV2.S vaccine or a vaccine not authorized by Health Canada, or received ≥1 dose out of province. Due to the small number of three-dose recipients, we also excluded individuals who received ≥3 doses. Datasets were linked using unique encoded identifiers and analyzed at ICES.

To estimate VE, we used a test-negative design, which is a modified case-control study whereby vaccination status is compared between SARS-CoV-2 test-positive ‘cases’ and test-negative ‘controls’ [20]. It adjusts for bias arising from differences in access to healthcare between cases and controls by its restriction to those who present for testing [21,22]. For cases, we selected the first positive SARS-CoV-2 test episode. For controls, we randomly selected a negative test episode stratified by pandemic wave (wave 2: December 14, 2020 to March 6, 2021; wave 3: March 7, 2021 to June 26, 2021; wave 4: June 27, 2021 to November 21, 2021) to control for changes in access to testing and vaccine eligibility; individuals who had multiple test-negative episodes could contribute up to three controls (one per wave).

Outcomes

Outcomes included: any RT-PCR-confirmed SARS-CoV-2 infection, RT-PCR-confirmed SARS-CoV-2 infection associated with symptomatic illness, and COVID-19-related hospitalizations and/or death [23]. For symptomatic infections, we restricted the analysis to individuals who had ≥1 symptom indicated on their laboratory requisition form. We defined a COVID-19-related hospitalization as any all-cause hospital admission associated with a RT-PCR-confirmed SARS-CoV-2 infection occurring within 14 days before or 3 days after admission recorded in the Canadian Institute for Health Information Discharge Abstract Database (CIHI-DAD) or any hospitalization associated with a reported COVID-19 case recorded in the COVID-19 reportable disease database (Case and Contact Management Solution [CCM]). We defined a COVID-19-related death as any all-cause death associated with a RT-PCR-confirmed SARS-CoV-2 infection occurring within 30 days before death or 7 days postmortem recorded in the provincial health insurance registry or any death associated with a reported COVID-19 case recorded in CCM. For severe outcomes (hospitalization and/or death), we restricted test-negative controls to symptomatic testers (Figure S1, Supplemental Digital Content, https://links.lww.com/QAD/C657). We used specimen collection date as the index date for all outcomes.

Coronavirus disease 2019 vaccination

We obtained data on vaccination status from the provincial COVID-19 vaccine registry, including number of doses, date(s) of administration, and vaccine product type(s). For the one-dose analysis, we considered participants as vaccinated if they received one dose ≥14 days before their index date; those who were vaccinated within 14 days or received two doses were excluded. For the two-dose analysis, we considered participants as vaccinated if they received two doses ≥7 days before their index date; those who were vaccinated within 7 days or received one dose were excluded. Participants without a vaccination recorded were considered unvaccinated for all analyses.

Covariates

We identified a participants’ age, sex, and public health region as of December 14, 2020, using the provincial health insurance registry. We considered the number of SARS-CoV-2 RT-PCR tests during the 3 months prior to December 14, 2020, as a proxy for routine screening (e.g. healthcare workers who undergo frequent testing and who may be more likely to be vaccinated). We obtained influenza vaccination during the 2019–2020 and/or 2020–2021 seasons as a proxy for health behaviors using physician or pharmacist billing codes in the Ontario Health Insurance Program or the Ontario Drug Benefits databases, respectively [16]. We identified comorbidities associated with an increased risk of COVID-19 and/or severity using validated algorithms and diagnostic codes for chronic diseases, as previously described [16]. We obtained neighborhood-level variables (based on census dissemination area) for the proportion of the working population employed in sales/trades/manufacturing/agriculture (i.e., non-health essential workers), average number of people per dwelling, proportion of the population who self-identified as a visible minority, and median household income from the 2016 Canadian census.

Analysis

We compared test-positive cases and test-negative controls as well as vaccinated and unvaccinated participants using standardized differences (SDs), which were calculated as the absolute difference in proportions in units of the pooled standard deviation. SDs ≤0.1 were taken to indicate negligible differences [24]. We used multivariable logistic regression to estimate odds ratios (OR) and 95% confidence intervals (CIs) comparing the odds of COVID-19 vaccination between test-positive cases and test-negative controls. Models were adjusted for 10-year age group, sex, region, calendar time, SARS-CoV-2 test histories, influenza vaccination histories, number of comorbidities, and neighborhood-level variables for socio-economic status (SES). Calendar time was modeled as a restricted cubic spline with 3 equally-spaced knots based on week of specimen collection. VE was derived as (1 – OR) × 100%. Participants with missing data for ≥1 covariate were excluded.

We created separate models for each outcome and number of doses received. Due to supply shortages, Canada's National Advisory Committee on Immunization recommended extending the dosing interval up to 16 weeks to optimize early vaccine distribution starting in March 2021, and allowed for the interchangeability of COVID-19 vaccines for the two-dose primary series starting in June 2021. For this reason, we stratified analyses according to extended dosing intervals and vaccine product. Due to suspected waning over time [25,26], we also analyzed VE by time since last dose. All analyses were performed using SAS, version 9.4 (SAS Institute Inc., Cary, North Carolina, USA).

Ethics approval

Research Ethics Board approval was obtained from the University of Toronto (protocol #28456).

Results

Participant characteristics

Among 21 023 adults living with HIV, 7461 unique individuals were tested for SARS-CoV-2 between December 14, 2020, and November 21, 2021, comprising 9962 test episodes. After exclusions (Figure S1, Supplemental Digital Content, https://links.lww.com/QAD/C657), there were 9680 (97.2%) test episodes, including 801 (8.3%) test-positive cases and 8879 (91.7%) test-negative controls (representing 6465 unique individuals). Overall, 131 (16.4%) test-positive cases and 985 (11.1%) test-negative controls (representing 877 unique individuals) had recorded COVID-19 symptoms. Compared to those without symptoms recorded, symptomatic test-negative controls were younger, more likely to be vaccinated, tested during the fourth wave (June 27, 2021 to November 21, 2021), reside in Northern, Peel, or South West regions, have <2 SARS-CoV-2 tests prior to the study period, and reside in lower SES neighborhoods (Table S2, Supplemental Digital Content, https://links.lww.com/QAD/C657); among test-positive cases, only public health region and the SES indicators reached statistical significance.

Compared to test-negative controls, test-positive cases were younger, more likely to be tested during the third wave (March 7, 2021 to June 26, 2021), reside in Toronto or Peel regions, have no SARS-CoV-2 tests prior to the study period, be unvaccinated against influenza, have fewer comorbidities, and reside in lower SES neighborhoods (Table 1); they were also less likely to be vaccinated against COVID-19 (20.1% vs. 47.8%; SD = 0.61). Comparisons between test-positive cases and test-negative controls were similar among the subset of symptomatic testers (Table S1, Supplemental Digital Content, https://links.lww.com/QAD/C657). Among 801 test-positive cases, 101 (12.6%) had a severe outcome, including 99 COVID-19-related hospitalizations and 18 deaths.

Table 1 - Participant characteristics by case status among people with HIV tested for SARS-CoV-2 infection between December 14, 2020 and November 21, 2021, Ontario, Canada.
Test-positive cases Test-negative controls
N % N % SD
Total 801 100.0 8879 100.0
Number of COVID-19 vaccine doses (at time of test)
 0 640 79.9 4635 52.2 0.61
 1 105 13.1 1409 15.9 0.08
 2 56 7.0 2835 31.9 0.66
Age group (years)
 19–29 74 9.2 656 7.4 0.07
 30–39 193 24.1 1699 19.1 0.12
 40–49 206 25.7 2036 22.9 0.07
 50–59 217 27.1 2640 29.7 0.06
 ≥60 111 13.9 1848 20.8 0.18
Sex
 Male 586 73.2 6656 75.0 0.04
 Female 215 26.8 2223 25.0 0.04
Year-month of specimen collection
 2020-12 63 7.9 624 7.0 0.03
 2021-01 127 15.9 1036 11.7 0.12
 2021-02 72 9.0 892 10.0 0.04
 2021-03 119 14.9 1097 12.4 0.07
 2021-04 196 24.5 966 10.9 0.36
 2021-05 108 13.5 789 8.9 0.15
 2021-06 23 2.9 568 6.4 0.17
 2021-07 10 1.2 577 6.5 0.27
 2021-08 28 3.5 633 7.1 0.16
 2021-09 21 2.6 643 7.2 0.21
 2021-10 22 2.7 650 7.3 0.21
 2021-11 12 1.5 404 4.6 0.18
Pandemic wave
 Wave 2: Dec. 14, 2020 to March 6, 2021 284 35.5 2799 31.5 0.08
 Wave 3: March 7, 2021 to June 26, 2021 423 52.8 3099 34.9 0.37
 Wave 4: June 27, 2021 to November 21, 2021 94 11.7 2981 33.6 0.54
Public health region
 Toronto 446 55.7 4457 50.2 0.11
 Central East 7 0.9 295 3.3 0.17
 Central West 99 12.4 1190 13.4 0.03
 Durham 23 2.9 231 2.6 0.02
 Eastern 11 1.4 288 3.2 0.12
 Northern 23 2.9 354 4.0 0.06
 Ottawa 45 5.6 663 7.5 0.07
 Peel 72 9.0 449 5.1 0.15
 South West 44 5.5 698 7.9 0.09
 York 31 3.9 254 2.9 0.06
Number of tests in 3 months before December 14, 2020
 0 600 74.9 5995 67.5 0.16
 1 132 16.5 1697 19.1 0.07
 2+ 69 8.6 1187 13.4 0.15
Receipt of 2019/20 or 2020/21 flu vaccine
 Neither season 471 58.8 4630 52.1 0.13
 Either season 181 22.6 2310 26.0 0.08
 Both seasons 149 18.6 1939 21.8 0.08
Number of comorbidities
 0 442 55.2 4191 47.2 0.16
 1 221 27.6 2566 28.9 0.03
 2+ 138 17.2 2122 23.9 0.17
Comorbidities
 Chronic respiratory disease 155 19.4 2338 26.3 0.17
 Hypertension 151 18.9 2085 23.5 0.11
 Diabetes 103 12.9 1173 13.2 0.01
 Frailty/dementia 73 9.1 838 9.4 0.01
 Chronic heart disease 53 6.6 780 8.8 0.08
 Chronic kidney disease/dialysis 54 6.7 622 7.0 0.01
 Advanced liver disease 34 4.2 374 4.2 0.00
 Autoimmune disorders 24 3.0 426 4.8 0.09
 History of stroke/TIA 13 1.6 162 1.8 0.02
Neighborhood-level essential workers quintile
 1 (0–32.5%) 193 24.1 2921 32.9 0.20
 2 (32.5–42.3%) 134 16.7 1586 17.9 0.03
 3 (42.3–49.8%) 143 17.9 1439 16.2 0.04
 4 (50.0–57.5%) 163 20.3 1480 16.7 0.09
 5 (57.5–100%) 168 21.0 1453 16.4 0.12
Neighborhood-level average number of persons per dwelling quintile
 1 (0–2.1) 314 39.2 4046 45.6 0.13
 2 (2.2–2.4) 119 14.9 1492 16.8 0.05
 3 (2.5–2.6) 87 10.9 861 9.7 0.04
 4 (2.7–3.0) 133 16.6 1415 15.9 0.02
 5 (3.1–5.7) 148 18.5 1065 12.0 0.18
Neighborhood-level proportion of persons self-identifying as visible minority quintile
 1 (0.0–2.2%) 26 3.2 687 7.7 0.20
 2 (2.2–7.5%) 49 6.1 849 9.6 0.13
 3 (7.5–18.7%) 125 15.6 1532 17.3 0.04
 4 (18.7–43.5%) 236 29.5 2886 32.5 0.07
 5 (43.5–100%) 365 45.6 2925 32.9 0.26
Neighborhood-level median income quintile
 1 (lowest) 359 44.8 3161 35.6 0.19
 2 166 20.7 1885 21.2 0.01
 3 123 15.4 1582 17.8 0.07
 4 90 11.2 1087 12.2 0.03
 5 (highest) 63 7.9 1164 13.1 0.17
COVID-19, coronavirus disease 2019; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; SD, standardized difference.

Coronavirus disease 2019 vaccination

Among 801 test-positive cases, 161 (20.1%) had received ≥1 dose of COVID-19 vaccine, including 63 (9.0%) who received 1 dose ≥14 days before specimen collection and 53 (7.6%) who received two doses ≥7 days before specimen collection. By comparison, among 8879 test-negative controls, 4244 (47.8%) had received ≥1 dose of COVID-19 vaccine, including 1055 (18.5%) who received 1 dose ≥14 days before specimen collection and 2706 (36.9%) who received two doses ≥7 days before specimen collection. Most vaccinated participants received ≥1 mRNA dose, including 80.3% of single-dose recipients and 93.4% of two-dose recipients (of whom, 88.9% received mRNA vaccines for both doses). Compared to unvaccinated participants, vaccinated participants were more likely to be older, test during the third (single-dose recipients) or fourth waves (two-dose recipients), have ≥2 SARS-CoV-2 tests prior to the study period (two-dose recipients only), be vaccinated against influenza, and have a higher number of comorbidities (single-dose recipients only) (Table 2).

Table 2 - Participant characteristics by COVID-19 vaccination status among people with HIV tested for SARS-CoV-2 infection between December 14, 2020 and November 21, 2021, Ontario, Canada.
Unvaccinated Received one dose ≥14 days before index date Received two doses ≥7 days before index date
N % N % SD N % SD
Total 5275 100.0 1118 100.0 2759 100.0
Lab-confirmed SARS-CoV-2 infection
 Test-positive case 640 12.1 63 5.6 0.23 53 1.9 0.41
 Test-negative control 4635 87.9 1055 94.4 0.23 2706 98.1 0.41
Age group (years)
 19–29 449 8.5 64 5.7 0.11 177 6.4 0.08
 30–39 1085 20.6 174 15.6 0.13 537 19.5 0.03
 40–49 1259 23.9 242 21.6 0.05 620 22.5 0.03
 50–59 1540 29.2 340 30.4 0.03 813 29.5 0.01
 ≥60 942 17.9 298 26.7 0.21 612 22.2 0.11
Sex
 Male 3898 73.9 834 74.6 0.02 2114 76.6 0.06
 Female 1377 26.1 284 25.4 0.02 645 23.4 0.06
Year-month of specimen collection
 2020-12 683 12.9 0 0.0 0.55 0 0.0 0.55
 2021-01 1099 20.8 14 1.3 0.66 ≤5 ≤0.2 -
 2021-02 871 16.5 23 2.1 0.51 40-44 1.4–1.6 -
 2021-03 1000 19.0 52 4.7 0.45 80 2.9 0.53
 2021-04 739 14.0 223 19.9 0.16 80 2.9 0.41
 2021-05 370 7.0 346 30.9 0.64 72 2.6 0.21
 2021-06 117 2.2 269 24.1 0.68 138 5.0 0.15
 2021-07 108 2.0 80 7.2 0.25 363 13.2 0.43
 2021-08 94 1.8 50 4.5 0.16 503 18.2 0.57
 2021-09 84 1.6 25 2.2 0.05 543 19.7 0.61
 2021-10 69 1.3 26 2.3 0.08 571 20.7 0.65
 2021-11 41 0.8 10 0.9 0.01 363 13.2 0.50
Pandemic wave
 Wave 2: December 14, 2020 to March 6, 2021 2893 54.8 39 3.5 1.37 63 2.3 1.43
 Wave 3: March 7, 2021 to June 26, 2021 1979 37.5 861 77.0 0.87 319 11.6 0.63
 Wave 4: June 27, 2021 to November 21, 2021 403 7.6 218 19.5 0.35 2377 86.2 2.55
Public Health Region
 Toronto 2645 50.1 551 49.3 0.02 1431 51.9 0.03
 Central East 149 2.8 42 3.8 0.05 94 3.4 0.03
 Central West 699 13.3 168 15.0 0.05 356 12.9 0.01
 Durham 133 2.5 33 3.0 0.03 72 2.6 0.01
 Eastern 169 3.2 44 3.9 0.04 70 2.5 0.04
 Northern 212 4.0 43 3.8 0.01 95 3.4 0.03
 Ottawa 390 7.4 73 6.5 0.03 207 7.5 0.00
 Peel 292 5.5 58 5.2 0.02 145 5.3 0.01
 South West 427 8.1 79 7.1 0.04 207 7.5 0.02
 York 159 3.0 27 2.4 0.04 82 3.0 0.00
Number of tests in 3 months before December 14, 2020
 0 3682 69.8 762 68.2 0.04 1820 66.0 0.08
 1 1001 19.0 204 18.2 0.02 516 18.7 0.01
 2+ 592 11.2 152 13.6 0.07 423 15.3 0.12
Receipt of 2019/20 or 2020/21 flu vaccine
 Neither season 2966 56.2 565 50.5 0.11 1313 47.6 0.17
 Either season 1307 24.8 272 24.3 0.01 771 27.9 0.07
 Both seasons 1002 19.0 281 25.1 0.15 675 24.5 0.13
Number of comorbidities
 0 2526 48.6 481 43.0 0.11 1334 48.4 0.00
 1 1534 29.1 324 29.0 0.00 783 28.4 0.02
 2+ 1179 22.4 313 28.0 0.13 642 23.3 0.02
Comorbidities
 Chronic respiratory disease 1359 25.8 312 27.9 0.05 684 24.8 0.02
 Hypertension 1135 21.5 294 26.3 0.11 688 24.9 0.08
 Diabetes 654 12.4 182 16.3 0.11 367 13.3 0.03
 Frailty/dementia 530 10.0 133 11.9 0.06 196 7.1 0.11
 Chronic heart disease 427 8.1 120 10.7 0.09 228 8.3 0.01
 Chronic kidney disease/dialysis 329 6.2 99 8.9 0.10 205 7.4 0.05
 Advanced liver disease 223 4.2 65 5.8 0.07 96 3.5 0.04
 Autoimmune disorders 232 4.4 54 4.8 0.02 147 5.3 0.04
 History of stroke/TIA 87 1.6 31 2.8 0.08 52 1.9 0.02
Neighborhood-level essential workers quintile
 1 (0–32.5%) 1594 30.2 358 32.0 0.04 997 36.1 0.13
 2 (32.5–42.3%) 933 17.7 196 17.5 0.00 482 17.5 0.01
 3 (42.3–49.8%) 888 16.8 187 16.7 0.00 421 15.3 0.04
 4 (50.0–57.5%) 907 17.2 208 18.6 0.04 431 15.6 0.04
 5 (57.5–100%) 953 18.1 169 15.1 0.08 428 15.5 0.07
Neighborhood-level average number of persons per dwelling quintile
 1 (0–2.1) 2335 44.3 513 45.9 0.03 1273 46.1 0.04
 2 (2.2–2.4) 905 17.2 192 17.2 0.00 434 15.7 0.04
 3 (2.5–2.6) 508 9.6 101 9.0 0.02 293 10.6 0.03
 4 (2.7–3.0) 855 16.2 187 16.7 0.01 418 15.2 0.03
 5 (3.1–5.7) 672 12.7 125 11.2 0.05 341 12.4 0.01
Neighborhood-level proportion of persons self-identifying as visible minority quintile
 1 (0.0–2.2%) 372 7.1 91 8.1 0.04 210 7.6 0.02
 2 (2.2–7.5%) 490 9.3 109 9.7 0.02 247 9.0 0.01
 3 (7.5–18.7%) 902 17.1 198 17.7 0.02 468 17.0 0.00
 4 (18.7–43.5%) 1660 31.5 376 33.6 0.05 935 33.9 0.05
 5 (43.5–100%) 1851 35.1 344 30.8 0.09 899 32.6 0.05
Neighborhood-level median income quintile
 1 (lowest) 1989 37.7 405 36.2 0.03 934 33.9 0.08
 2 1152 21.8 219 19.6 0.06 570 20.7 0.03
 3 881 16.7 208 18.6 0.05 526 19.1 0.06
 4 619 11.7 132 11.8 0.00 360 13.0 0.04
 5 (highest) 634 12.0 154 13.8 0.05 369 13.4 0.04
COVID-19, coronavirus disease 2019; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; SD, standardized difference..

Vaccine effectiveness

After adjustment for potential confounders, VE for receipt of two doses ≥7 days before specimen collection was 82% (95%CI = 74–87%) against any infection, increasing to 94% (95% CI = 82–98%) against symptomatic outcomes, and 97% (95% CI = 85–100%) against severe outcomes (Table 3). VE was lower for receipt of 1 dose ≥14 days before specimen collection against all three outcomes. We did not observe waning protection from time since second dose for symptomatic infection or severe outcomes. However, against any infection, VE declined from 86% (95% CI = 77–92%) within 7–59 days after the second dose to 66% (95% CI = −15–90%) ≥180 days after the second dose. VE was slightly lower among participants who received only the ChAdOx1 vaccine (either as a single dose or two doses), but differences were not statistically significant. We did not observe heterogeneity in two-dose VE by age, sex, region, number of comorbidities, or pandemic wave (Table S3, Supplemental Digital Content, https://links.lww.com/QAD/C657).

Table 3 - Vaccine effectiveness estimates among people with HIV tested for SARS-CoV-2 infection between December 14, 2020 and November 21, 2021, Ontario, Canada.
Any infection Symptomatic infection Severe outcomes (hospitalization/death)
Unadjusted Adjusteda Unadjusted Adjusteda Unadjusted Adjusteda
VE (95% CI) VE (95% CI) VE (95% CI) VE (95% CI) VE (95% CI) VE (95% CI)
Received one dose ≥14 days before index date 57 (43, 67) 64 (52, 74) 51 (5, 75) 55 (1, 80) 38 (−19, 67) 67 (23, 86)
 Time since first dose (days)
  14–29 41 (13, 60) 51 (27, 68) 46 (-54, 81) 38 (-105, 81) NE NE
  30–59 66 (47, 78) 74 (58, 83) 56 (-27, 84) 58 (-36, 87) NE NE
  ≥60 62 (34, 79) 66 (39, 82) 49 (-70, 85) 66 (-33, 91) NE NE
 Vaccine product type
  ChAdOx1 28 (-16, 55) 39 (-1, 63) 24 (-163, 78) 22 (-228, 81) NE NE
  BNT162b2 or mRNA-1273(mRNA vaccines) 64 (50, 73) 70 (57, 79) 57 (7, 80) 61 (6, 84) NE NE
Received 2 doses ≥7 days before index date 86 (81, 89) 82 (74, 87) 91 (81, 96) 94 (82, 98) 97 (89, 99) 97 (85, 100)
 Time since second dose (days)
  7–59 88 (80, 93) 86 (77, 92) 87 (58, 96) 92 (69, 98) NE NE
  60–119 86 (78, 91) 78 (62, 87) 89 (70, 96) 93 (74, 98) NE NE
  120–179 84 (72, 91) 77 (53, 89) 96 (69, 99) 98 (73, 100) NE NE
  ≥180 79 (34, 93) 66 (-15, 90) NE NE NE NE
 Vaccine product type
  ChAdOx1 only 84 (56, 94) 77 (34, 92) NE NE NE NE
  ≥1 mRNA vaccine dose 86 (81, 90) 82 (74, 88) 90 (79, 95) 93 (80, 98) 97 (88, 99) 97 (83, 100)
   mRNA only 86 (81, 90) 82 (74, 88) 89 (77, 95) 93 (79, 98) 97 (87, 99) 97 (80, 99)
   Mixed ChAdOx1 and mRNA doses 87 (69, 95) 80 (49, 92) NE NE NE
NE, not estimated; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; VE, vaccine effectiveness.
aAdjusted for age group, sex, public health region, calendar time (restricted cubic spline for week of specimen collection), number of SARS-CoV-2 test episodes in 3 months prior to December 14, 2020, received influenza vaccine in 2019/20 or 2020/21 seasons, number of comorbidities, and neighborhood-level SES indicators. Severe outcomes models adjusted for 4-level region (Greater Toronto Area, East, West, North); any or symptomatic infection models adjusted for public health region.

Discussion

In a population-based study of adults living with HIV in Ontario, Canada, two doses of a COVID-19 vaccine offered significant protection against SARS-CoV-2 infection, symptomatic illness, and severe outcomes when Alpha and Delta variants, along with wild-type viruses, were predominately circulating. Although some previous reports suggest lesser effectiveness in immunocompromised individuals [27,28], our findings do not support a broad conclusion that COVID-19 VE is lower among people with HIV who, for the most part, are attending HIV care, taking antiretroviral medication, and are virally suppressed [18].

Our VE estimates against symptomatic infection and severe outcomes (94–97%) are comparable to those measured in the general population (82–98%) during similar time periods and against non-Omicron variants [16,29]. Consistent with immunological evidence [7,8], these findings provide reassurance that HIV-mediated immune suppression does not meaningfully interfere with real-world protection from (predominantly) mRNA COVID-19 vaccines against clinically-relevant outcomes for populations in which HIV is well managed. Nonetheless, we anticipate that VE will be lower against the Omicron variant and with increasing time since the last dose [30]. We also anticipate lower VE in people with HIV who have AIDS-defining illness, low CD4+ cell count (e.g., <200 cells/μl), or unsuppressed HIV viral loads and in people with other immunocompromised conditions [7,8].

Few observational studies have evaluated the real-world effectiveness of COVID-19 vaccines in people with HIV, and none specifically examined VE for mRNA vaccines. These studies may also be prone to confounding bias due to healthcare seeking behavior. In a retrospective cohort study among people with HIV on antiretroviral therapy in Russia from January to July 2021, VE of the Sputnik V viral vector vaccine was 79% (95% CI = 70–81%) against any COVID-19 infection and 90% (95% CI not reported) against hospitalization [31]. That study found slightly lower VE in participants with low CD4 counts (<350 cells/μl) but differences were not statistically significant [31]. Among healthcare workers living with HIV in South Africa from February to May 2021, VE for a single dose of the AD26.COV2.S viral vector vaccine was 73% (95% CI = 58–85%) against COVID-19-related hospitalization and 65% (95% CI = 13–93%) against death [32]. Among a small, prospective cohort of people seeking HIV care at a hospital in Taiwan between March and September 2021, the incidence rate of SARS-CoV-2 infection was 6.4 per 100 000 person-days in the unvaccinated group, 2.9 per 100 000 in the partially-vaccinated group, and 0 per 100 000 in the fully-vaccinated group, corresponding to VE of 53% and 100% for partial and full vaccination, respectively [33]. In that study, most participants (>70%) had received the ChAdOx1 vaccine, with less than one-quarter receiving mRNA vaccines [33]. Differences in study design, vaccine products and dosing, healthcare setting, and HIV populations preclude direct comparison of these findings.

To our knowledge, ours is among the first to evaluate the real-world effectiveness of mRNA COVID-19 vaccines in people with HIV using a test-negative design. This study design, via restriction to those who present for SARS-CoV-2 testing, mitigates potential bias arising from differences in access to healthcare. As in the general population, we found lower protection against any infection, with evidence of waning immunity with time since last dose [25,26]. However, this analysis is more prone to bias since it includes asymptomatic testers undergoing routine screening or those in close contact with a suspected case for whom the likelihood of test positivity is lower than that in symptomatic testers [21]. Individuals with recorded symptoms systematically differed from those without symptom information, although this was largely driven by regional variability as not all laboratories had the ability to enter symptom information into the provincial system. As such, symptom information was missing for >85% of test episodes, limiting our sample size for the more conventional test-negative design analysis restricted to symptomatic testers [20]. Nonetheless, we expect that most testers likely had symptoms consistent with COVID-19, as per provincial test guidelines during this period.

There are other limitations. We used specimen collection date as a proxy for date of symptom onset as the latter was inconsistently reported. Some severe outcomes may be misclassified due to delays in capturing hospital discharges in CIHI-DAD and deaths in the provincial insurance registry or incomplete reporting of severe outcomes in CCM. Although we attempted to control for confounding due to healthcare-seeking behavior, neighborhood-level SES and other demographic factors, residual confounding may remain. We used the number of SARS-CoV-2 tests in the 3 months prior to the study period as a proxy for routine screening but we were otherwise unable to standardize the testing indication for SARS-CoV-2 [22]. Data on HIV clinical indicators were unavailable at the time of analysis.

In conclusion, we measured high VE against symptomatic infection and severe illness among adults living with HIV in Ontario prior to the emergence of the Omicron variant. These findings may provide reassurance for people with HIV and their healthcare providers, particularly for those whose HIV is well controlled, and has important implications for COVID-19 vaccination guidelines, including the potential need for booster doses. Further research is required to estimate the effectiveness of COVID-19 vaccines among people with HIV who have moderate to severe immunodeficiencies, and among all people with HIV against Omicron or future variants.

Acknowledgements

Additional members of the COVAXHIV Study Team are Mark Brockman, Zabrina Brumme, Troy Grennan, Mark Hull, Mohammad-Ali Jenabian, Mel Krajden, Enrico Mandarino, Shari Margolese, Emily McDonald, Nasheed Moqueet, Mario Ostrowski, Caroline Quach-Thanh, Ron Rosenes, Sergio Rueda, Joel Singer, Darian Taylor, Vanessa Tran, and Jason Wong.

The authors would like to thank staff at Unity Health Toronto (Anna Yeung, Ashley Mah) and at the Canadian HIV Trials Network (Judy Needham, Erin Cherban) for their administrative support. We are also grateful for guidance from members of our Community Advisory Board, co-chaired by Muluba Habanyama and Devan Nambiar.

Sources of funding: This study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). This study also received funding from a Project Grant (# MM1-174900) by the Canadian Institutes of Health Research to ANB. Parts of this material are based on data and/or information compiled and provided by CIHI. However, the analyses, conclusions, opinions and statements expressed in the material are those of the author(s), and not necessarily those of CIHI. We thank IQVIA Solutions Canada Inc. for use of their Drug Information File. This study was carried out in collaboration with the COVAXHIV Study which is supported by funding to AA from the Public Health Agency of Canada, through the Vaccine Surveillance Reference Group and the COVID-19 Immunity Task Force (# 2122-HQ-000075). The analyses, conclusions, opinions, and statements expressed herein are solely those of the authors and do not reflect those of the funding or data sources; no endorsement is intended or should be inferred.

C.C. is a Vanier scholar (CIHR # 415141) and Canadian Immunization Research Network trainee. A.N.B. is a Canada Research Chair in Sexually Transmitted Infection Prevention. A.N.B. and J.C.K. are recipients of a University of Toronto Department of Family and Community Medicine non-clinician scientist and clinician-scientist award, respectively. C.T.C. is supported by a Fonds de recherche du Québec-Santé (FRQS) Junior 2 career award. C.E.K. is supported by a Clinical Chair Award from the Faculty of Medicine at the University of Ottawa.

Conflicts of interest

N.Z.J. declares speaking and advisory board participation for Abbvie. All other authors have no conflicts of interest.

References

1. Centers for Disease Control and Prevention. Underlying medical conditions associated with higher risk for severe COVID-19: Information for healthcare professionals. Atlanta: Centers for Disease Control and Prevention; 2022. Available at: https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-care/underlyingconditions.html [Accessed 19 May 2022].
2. Tesoriero JM, Swain CE, Pierce JL, Zamboni L, Wu M, Holtgrave DR, et al. COVID-19 outcomes among persons living with or without diagnosed HIV infection in New York State. JAMA Netw Open 2021; 4:e2037069.
3. Hadi YB, Naqvi SFZ, Kupec JT, Sarwari AR. Characteristics and outcomes of COVID-19 in patients with HIV: a multicentre research network study. AIDS 2020; 34:F3–8.
4. Del Amo J, Polo R, Moreno S, Díaz A, Martínez E, Arribas JR, et al. Incidence and severity of COVID-19 in HIV-positive persons receiving antiretroviral therapy: A cohort study. Ann Intern Med 2020; 173:536–541.
5. Inciarte A, Gonzalez-Cordon A, Rojas J, Torres B, de Lazzari E, de la Mora L, et al. Clinical characteristics, risk factors, and incidence of symptomatic coronavirus disease 2019 in a large cohort of adults living with HIV: a single-center, prospective observational study. AIDS 2020; 34:1775–1780.
6. Geretti A, Stockdale A, Kelly S, Villa G, Turtle L, Semple M, et al. Outcomes of COVID-19 related hospitalisation among people with HIV in the ISARIC WHO Clinical Characterisation Protocol (UK): a prospective observational study. J Int AIDS Soc 2020; 73:2095–2106.
7. Lee ARYB, Wong SY, Chai LYA, Lee SC, Lee MX, Muthiah MD, et al. Efficacy of COVID-19 vaccines in immunocompromised patients: systematic review and meta-analysis. BMJ 2022; 376:e068632.
8. Galmiche S, Luong Nguyen LB, Tartour E, de Lamballerie X, Wittkop L, Loubet P, et al. Immunological and clinical efficacy of COVID-19 vaccines in immunocompromised populations: a systematic review. Clin Microbiol Infect 2022; 28:163–177.
9. Polack FP, Thomas SJ, Kitchin N, Absalon J, Gurtman A, Lockhart S, et al. Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine. N Engl J Med 2020; 383:2603–2615.
10. Baden LR, El Sahly HM, Essink B, Kotloff K, Frey S, Novak R, et al. Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine. N Engl J Med 2021; 384:403–416.
11. Voysey M, Clemens SAC, Madhi SA, Weckx LY, Folegatti PM, Aley PK, et al. Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil, South Africa, and the UK. Lancet 2021; 397:99–111.
12. Sadoff J, Gray G, Vandebosch A, Cárdenas V, Shukarev G, Grinsztejn B, et al. Final analysis of efficacy and safety of single-dose Ad26.COV2. S N Engl J Med 2022; 386:847–860.
13. Shinde V, Bhikha S, Hoosain Z, Archary M, Bhorat Q, Fairlie L, et al. Efficacy of NVX-CoV2373 Covid-19 vaccine against the B.1.351 variant. N Engl J Med 2021; 384:1899–1909.
14. Coburn SB, Humes E, Lang R, Stewart C, Hogan BC, Gebo KA, et al. COVID-19 infections postvaccination by HIV status in the United States [Preprint]. medRxiv 5 December 2021 [cited 19 May 2022]. Available at: doi:10.1101/2021.12.02.21267182.
15. Sun J, Zheng Q, Madhira V, Olex AL, Anzalone AJ, Vinson A, et al. Association between immune dysfunction and COVID-19 breakthrough infection after SARS-CoV-2 vaccination in the US. JAMA Intern Med 2022; 182:153–162.
16. Chung H, He S, Nasreen S, Sundaram ME, Buchan SA, Wilson SE, et al. Effectiveness of BNT162b2 and mRNA-1273 covid-19 vaccines against symptomatic SARS-CoV-2 infection and severe covid-19 outcomes in Ontario, Canada: test negative design study. BMJ 2021; 374:n1943.
17. Public Health Agency of Canada. HIV in Canada – people living with HIV and new HIV infections, 2020. Ottawa: Government of Canada; 2022. Available at: https://www.canada.ca/en/public-health/services/publications/diseases-conditions/hiv-canada-people-living-with-hiv-new-infections-2020.html [Accessed 19 July 2022].
18. Ontario HIV Epidemiology and Surveillance Initiative. HIV care cascade in Ontario: Linkage to care, in care, on antiretroviral treatment, and virally suppressed, 2018. Toronto: OHESI; 2021. Available at: https://www.ohesi.ca/wp-content/uploads/2021/06/2018-HIV-care-cascade-Ontario-report.pdf [Accessed 19 May 2022].
19. Antoniou T, Zagorski B, Loutfy MR, Strike C, Glazier RH. Validation of case-finding algorithms derived from administrative data for identifying adults living with human immunodeficiency virus infection. PLoS One 2011; 6:e21748.
20. Foppa IM, Haber M, Ferdinands JM, Shay DK. The case test-negative design for studies of the effectiveness of influenza vaccine. Vaccine 2013; 31:3104–3109.
21. Sullivan SG, Tchetgen Tchetgen EJ, Cowling BJ. Theoretical basis of the test-negative study design for assessment of influenza vaccine effectiveness. Am J Epidemiol 2016; 184:345–353.
22. Patel MK, Bergeri I, Bresee JS, Cowling BJ, Crowcroft NS, Fahmy K, et al. Evaluation of postintroduction COVID-19 vaccine effectiveness: summary of interim guidance of the World Health Organization. Vaccine 2021; 39:4013–4024.
23. Grewal R, Kitchen SA, Nguyen L, Buchan SA, Wilson SE, Costa AP, et al. Effectiveness of a fourth dose of COVID-19 vaccine among long-term care residents in Ontario, Canada: test-negative design study. BMJ 2022; 378:e071502.
24. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res 2011; 46:399–424.
25. Tartof SY, Slezak JM, Fischer H, Hong V, Ackerson BK, Ranasinghe ON, et al. Effectiveness of mRNA BNT162b2 COVID-19 vaccine up to 6 months in a large integrated health system in the USA: a retrospective cohort study. Lancet 2021; 398:1407–1416.
26. Andrews N, Tessier E, Stowe J, Gower C, Kirsebom F, Simmons R, et al. Duration of protection against mild and severe disease by COVID-19 vaccines. N Engl J Med 2022; 386:340–350.
27. Whitaker HJ, Tsang RSM, Byford R, et al. Pfizer-BioNTech and Oxford AstraZeneca COVID-19 vaccine effectiveness and immune response amongst individuals in clinical risk groups. J Infect 2022; 84:675–683.
28. Tenforde MW, Patel MM, Ginde AA, Douin DJ, Talbot HK, Casey JD, et al. Effectiveness of SARS-CoV-2 mRNA vaccines for preventing Covid-19 hospitalizations in the United States [Preprint]. medRxiv 8 July 2021 [cited 19 May 2022]. Available at: Doi:10.1101/2021.07.08.21259776.
29. Nasreen S, Chung H, He S, Brown KA, Gubbay JB, Buchan SA, et al. Effectiveness of COVID-19 vaccines against symptomatic SARS-CoV-2 infection and severe outcomes with variants of concern in Ontario. Nat Microbiol 2022; 7:379–385.
30. Buchan SA, Chung H, Brown KA, Austin PC, Fell DB, Gubbay JB, et al. Effectiveness of COVID-19 vaccines against Omicron or Delta symptomatic infection and severe outcomes [Preprint]. medRxiv 28 January 2022 [cited 19 May 2022]. Available at: https://doi.org/10.1101/2021.12.30.21268565.
31. Gushchin VA, Tsyganova EV, Ogarkova DA, Adgamov RR, Shcheblyakov DV, Glukhoedova NV, et al. Sputnik V protection from COVID-19 in people living with HIV under antiretroviral therapy. EClinicalMedicine 2022; 46:101360.
32. Bekker LG, Garrett N, Goga A, Fairall L, Reddy T, Yende-Zuma N, et al. Effectiveness of the Ad26.COV2.S vaccine in health-care workers in South Africa (the Sisonke study): Results from a single-arm, open-label, phase 3B, implementation study. Lancet 2022; 399:1141–1153.
33. Lin KY, Wu PY, Liu WD, Sun HY, Hsieh SM, Sheng WH, et al. Effectiveness of COVID-19 vaccination among people living with HIV during a COVID-19 outbreak. J Microbiol Immunol Infect 2022; 55:535–539.

Names of other study team members are listed in the Acknowledgment section.

Keywords:

coronavirus disease 2019; HIV; observational study; vaccine; vaccine effectiveness

Supplemental Digital Content

Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc.