Real world analysis of Omicron outbreak in South Africa including vaccine effectiveness - December 2021
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Real world analysis of Omicron outbreak in South Africa including vaccine effectiveness December 2021
Preface SA scientists from Network for Discovery Health has continuously The National DoH has created Genomic Surveillance first to announce shared data throughout the pandemic. considerable capacity to vaccinate the the identification of Omicron variant. South African population at scale. The insights herein support the priority of Discovery Health is in an unusual vaccination and the NDoH’s existing SA is first to experience Omicron- position to avail early insights given the approval of third dose Pfizer-BioNTech driven outbreak commencing about 3 dominance of the Omicron variant in boosters. weeks ago. SA, and the availability of member data spanning demographic details, clinical and pathology records and vaccination Discovery Health’s insights have been records. shared with the SA NDoH, the SA National Institute for Communicable Diseases (NICD), the US CDC, leading SA and UK scientists and others Dataset is derived for the early period of the Omicron outbreak. Consequently these preliminary insights may change as this Covid-19 wave extends
Executive summary | Omicron Insights 1. Epidemiological tracking shows steep trajectory of new Covid-19 infections indicating rapid Omicron spread, but with a flatter trajectory of hospital admissions, indicating likely lower severity 2. Anecdotal observations demonstrate nuanced differences in the clinical features of Omicron both out of hospital and in hospital 3. Data indicates that the severity of Omicron is 29% lower than D614G (first) wave of Covid-19 infections in South Africa 4. Vaccine effectiveness of the double dose Pfizer-BioNTech regime: • Has reduced from 80% in Delta wave to 33% in Omicron wave against Covid-19 infection; and • Has reduced from 93% in Delta wave to 70% in Omicron wave against severe complications of Covid-19 (hospital admission) 5. The protective effect of prior infection has reduced over time, and Omicron has eroded that protective effect further 6. Children experiencing very low test-positivity rate relative to adults, and low Covid-19 admissions in absolute terms, but appear to be at 20% greater risk of hospitalisation during Omicron wave relative to D614G wave 4
Background Clinical and Real-world insights epidemiological into vaccine observations effectiveness 5
Discovery Health (DH) is uniquely positioned to derive powerful COVID-19 insights • Pathology results DH Electronic • Hospital admissions Health Record • ICU and ventilation Unique, real-world data • Underlying disease burden Covid-19 insights >3.7 million health Epidemiological insurance lives tracking & administered by modelling,, Discovery Health resilience against Vitality Member severe disease, Wellness vaccination Vaccine Data data effectiveness safety and side effects • Underlying wellness • Vaccination status • Physical activity • Dose adherence • Mobility data • Vaccine type
Discovery Health beneficiaries are diverse, allowing for relevant real world insights to be derived 26% of all individuals are living with chronic Data distributed widely across age groups illness Other 9,6% 9,0% Epilepsy 9,0% Conduction disorder 2,1 8,2% 7,9% Bipolar mood disorder Hypertension 2, % 7,6% 2% 7,4% 3,1% 29,5% 6,6% HIV 6,1% 6,2% 4,6% 5,9% 5,1% Ischaemic heart disease 4,8% 4,7% 4,4% 3,4% Hypothyroidism 5,6% 2,8% 2,0% 9,2% 1,2% 1,2% Diabetes 19,2% 0,8% Hypercholesterolaemia 10,7%
Discovery Health has extensive South Africa-specific, real-world data 10 Dec 2021 We have been tracking and codifying unique member data since the start of the pandemic Tests Positive Active Repeat Recoveries Members Deaths Vaccines conducted cases cases cases admitted administered 2,604,385 467,530 28,415 15,580 424,717 61,579 14,398 2,713,122 DH SA 19,912,714 3,071,064 113,060 2,867,966 90,038 26,781,642 Testing rate COVID proportion % Active Proportion of Recovery rate Proportion Case % Adult of DH lives repeat positive admitted fatality rate vaccinated Testing rate Unique Proportion Hospital Proportion DH entities readmitted mortality rate administered national Pfizer vaccine 14 days – 99 days after second dose. Excludes data on members vaccinated in the public sector post September 2021. Source: Discovery Health Insights https://discv.co/DiscoveryHealthInsights 8
SA has experienced 3 prior waves of COVID-19 and now entering a fourth wave with rapid rise in cases driven by Omicron variant National new cases (7-day moving average) DH new cases (7-day moving average) 20 000 3500 15 043 3000 15 000 2500 2000 10 000 1 519 Gauteng Province accounts for 1500 63% of current COVID-19 cases 5 000 1000 nationally, and 67% of Discovery 500 Health cases in fourth wave - 0 Sep-20 Sep-21 May-20 May-21 Mar-20 Nov-20 Mar-21 Nov-21 Sep-20 Sep-21 Jan-21 May-20 May-21 Nov-20 Nov-21 Jan-21 Mar-20 Mar-21 Jul-20 Jul-21 Jul-20 Jul-21 26% of the national population Average weekly new DH cases per 100,000 lives for Gauteng (7-day moving average) resides in Gauteng Province. Upcoming mobility due to high 120 migrant worker base and holiday 100 period pose a risk of imminent 80 spread to other provinces across 60 South Africa 40 20 0 Week ending 20 Nov Week ending 27 Nov Week ending 4 Dec Source: NICD, Discovery Health Analysis 9
Background Clinical and Real-world insights epidemiological into vaccine observations effectiveness 10
Omicron-driven fourth wave is developing at a steeper trajectory of new infections relative to prior waves National data Discovery Health data SA daily cases (7-day average) by wave DH daily cases (7-day average) by wave 25 000 4 000 3 500 20 000 3 000 15 000 2 500 2 000 Steeper increase in 10 000 1 500 new infections and 1 000 5 000 500 test positivity rate - - during the first three weeks of the 1 7 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 121 127 133 139 103 109 115 121 127 133 139 Day from start of wave Day from start of wave fourth wave indicating highly SA weekly test positivity rate by wave DH weekly test positivity rate by wave transmissible 40,0% 35,0% 35,0% variant with rapid 30,0% 30,0% community spread 25,0% 25,0% 20,0% 20,0% 15,0% 15,0% 10,0% 10,0% 5,0% 5,0% D614G wave 0,0% 0,0% Beta wave 1 7 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 121 127 133 139 103 109 115 121 127 133 139 Delta wave Day from start of wave Day from start of wave Omicron wave Source: NICD, Discovery Health Analysis
Omicron variant has displaced Delta and now dominates 04 Dec 2021 new infections in SA 100 Nextclade clades 19B 20A 75 20B 20C 20D Alpha (10I, V1) percentage 50 Beta (20H, V2) Delta (21A) Delta (21I) 25 Delta (21J) Eta (21D) Kappa (21B) 0 Omicron (21K) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 C,1.2 (20D) epiweeks in 2021 • Network of virology and genomic laboratories, scientists and academic institutions across South Africa • Genomic data produced at five sequencing facilities under the guidance of more than 50 investigators and scientists • Launched in June 2020 with support of the Department of Science and Innovation and South African Medical Research Council Source: Network for Genomic Surveillance in South Africa 12
Anecdotal evidence demonstrates nuanced differences in the clinical presentation Out-of-hospital acute care Admissions • Higher reinfections and breakthrough infections • Most hospitalised patients for Covid-19 related disease than other waves, including vaccinated are unvaccinated • Shorter incubation period of 3-4 days • High number of hospitalisations in Gauteng for non-Covid care, present with Covid as an • Milder illness with reported recoveries within 3 days incidental finding on admission • Scratchy/sore throat most common early symptom, • Less respiratory distress on presentation like other waves • Proportion of High care and ICU admissions lower • Typical features include nasal congestion, dry cough compared to previous waves and myalgia, especially lower back pain • Significantly lower proportion of admitted patients requiring oxygen support • Most hypoxic patients requiring oxygenation are unvaccinated • 16% of ICU admissions are vaccinated (raw data) Source: Dr Gerrit De Villiers, Chief Clinical Officer of Mediclinic Southern Africa
Flatter trajectory of hospital admissions, indicating lower severity Hospital admissions per 1,000 infections 44 131 101 38 D614G Beta Delta Omicron Hospital admissions by wave at 3 week snapshot 500 DH daily hospital admissions (7-day average) by wave Caveats: 450 • Underlying Covid-19 400 antibody seroprevalence in SA population high, 350 especially following Delta 300 D614G wave wave Beta wave • Covid-19 treatment has 250 Delta wave improved since D614G 200 Omicron wave wave; especially with the 150 introduction of 100 dexamethasone • Data represents the first 3 50 weeks of Omicron wave, 0 caution in interpretation of hospital admissions 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 118 121 124 127 130 133 136 139 Day from start of wave therefore required Source: Discovery Health Analysis 14
Apparent de-linking of infection curve from admission curve Much steeper increase in new cases during the first three weeks of the Omicron wave compared to the Delta variant driven third wave. However, admissions and deaths are not increasing as rapidly. Cases, admissions and deaths indexed to the Delta wave, each as a share of Delta peak (%) 100% 90% 80% 70% Cases 60% 50% 40% 30% Admissions 20% 10% Deaths 0% Apr May Jun Jul Aug Sep Oct Nov Dec Source: Discovery Health Analysis
Paediatric cases and admissions appear more prevalent than prior waves, but typically present as mild cases Paediatric out-of-hospital cases Paediatric admissions • Higher number of paediatric cases testing positive • Primary diagnoses in children on admission for Covid-19 relative to prior waves related disease in Omicron wave are bronchiolitis and • Children present with a sore throat, nasal congestion pneumonia and fever for 2-3 days, and tend to complain of a • Often with associated diarrhoea and vomiting, and headache dehydration • Seems to resolve quickly with recovery after 3 days • Incidental Covid-19 diagnosis for multiple paediatric admissions, exceeding Covid-19 specific admissions 1,80 DH cases and admissions for under 18 year olds indexed to Delta Wave 1,60 Cases 1,40 1,20 Admissions 1,00 0,80 0,60 0,40 0,20 - Apr May Jun Jul Aug Sep Oct Nov Dec Source: Dr Gerrit De Villiers, Chief Clinical Officer of Mediclinic Southern Africa; Discovery Health Analysis
Increase in paediatric admissions, with high number of incidental Covid-19 diagnoses for unrelated admissions Admission proportion by age band Admission rate by age band Admission risk by age band Increase in proportion of paediatric To date, children under age 5 have admissions in the Omicron wave, possibly 60,0% experienced higher admission risk than driven by incidental Covid-19 diagnosis on adults admission 50,0% COVID19 admissions per 1,000 person years 35,0% 35,00 40,0% 30,0% 30,00 25,0% 30,0% 25,00 20,0% 20,00 15,0% 20,0% 15,00 10,0% 10,00 10,0% 5,0% 5,00 0,0% - 0,0% 0-9 10-19 20-34 35-49 50-59 60-79 80+ 0-4 5-12 13-17 Adults to 04 to 09 to 14 to 19 to 24 to 29 to 34 to 39 to 44 to 49 to 54 to 59 to 64 to 69 to 74 to 79 to 84 85+
Risk-adjusted, Omicron may pose risk of increased severity to children under 18 years of age; very early data which should be carefully followed; low risk in absolute terms Risk of admission relative to South Africa's first wave (fully risk-adjusted) Adult population Child population 1,20 1,2 1,2 1,1 1,1 1,0 1 1,0 1 29% average lower 0,9 0,9 admission risk 0,8 0,8 relative to D614G 0,71 wave. Children, to 0,7 0,7 date, have had a 0,6 0,6 20% higher risk of 0,5 0,5 admission 0,4 0,4 0,3 0,3 0,2 0,2 0,1 0,1 0,0 0,0 D614G wave Beta Wave Delta Wave Omicron Wave Assessed using a Cox proportional hazard model allowing for days since PCR collection date, age, sex, number of documented risk factors, vaccination status and documented prior infection* To be submitted for peer review and publication Source: Discovery Health Insights https://discv.co/DiscoveryHealthInsights
Risk-adjusted data correlates with anecdotal evidence demonstrating lower severity in current Omicron wave than previous waves Adults are experiencing 29% lower admission risk of Covid-19 admissions relative to the D614G wave. Those who are hospitalised also have a lower admission acuity and a lower propensity to be admitted to ICU, relative to prior waves Admission risk indexed to the first wave Case mix of Covid 19 admissions indexed to the Proportion of admissions in High care and ICU first wave by wave 1,12 1,18 1,08 5% 1,00 18% 8% 0,94 1,00 21% 22% 0,79 17% 14% 10% 0,71 87% 65% 65% 68% D614G Beta Delta Omicron D614G Beta Delta Omicron D614G Beta Delta Omicron ICU High Care Other wards Assessed using a Cox proportional hazard model allowing for days since PCR collection date, age, sex, number of documented risk factors, vaccination status and documented prior infection* To be submitted for peer review and publication Source: Discovery Health Data 19
Although national excess natural deaths increased in the last week of Nov, excess natural deaths are still significantly lower than previous waves South Africa weekly natural cause excess deaths National weekly excess deaths Oct – Nov 2021 2 076 16 000 1 504 14 000 1 407 1 376 1 134 1 133 1 187 1 091 12 000 804 10 000 3-Oct 10-Oct 17-Oct 24-Oct 31-Oct 7-Nov 14-Nov 21-Nov 28-Nov 8 000 Provincial weekly excess deaths Oct – Nov 2021 600 6 000 500 4 000 400 300 2 000 200 100 0 - Sep-20 Sep-21 Oct-20 Oct-21 May-21 Nov-20 Nov-21 Mar-21 Dec-20 Apr-21 Jan-21 Jul-20 Aug-20 Jul-21 Aug-21 Feb-21 Jun-21 3-Oct 17-Oct 31-Oct 14-Nov 28-Nov EC FS GT KZN LM MP NC NW WC Source: South African Medical Research Council excess natural deaths report 20
Background Clinical and Real-world insights epidemiological into vaccine observations effectiveness 21
Rapid timeline of Omicron research following identification by SA scientists; accelerated severity & vaccine effectiveness insights through DH collaboration with SAMRC 24 November First report of B.1.1.529 from South African scientists 26 November WHO designates Omicron a variant of concern 27 November South African clinicians note mildness of cases relative to prior variants Moderna and Pfizer announce plans for Omicron-specific vaccine Less than 3 weeks 2 December 250% increase in re-infection for Omicron relative to prior variants in South Africa (Juliet Pulliam et al) 7 December 4 lab neutralisation studies pre-print published with ~40% reduction in neutralization (Sigal et al, Ciesek et al, Sheward et al, Pfizer-BioNTech study) 10 December First report of clinical vaccine effectiveness for 2 shots and boosters, 3-fold household transmission contact transmission vs Delta (UKHSA) Today First real world study at scale on Pfizer-BioNTech vaccine effectiveness. Pre-print published by Discovery Health and the South African Medical Research Council Source: Adapted from Eric Topol twitter, with addition in blue
Omicron vaccine effectiveness study parameters study duration Covid-19 PCR test results for adult population Pre-Omicron Period Proxy Omicron Period 1 September 15 November to to 31 October 2021 7 December 2021 63% 37% of sample of sample SGTF positive test Gauteng PCR tests PCR test results results from Lancet amongst symptomatic Laboratories admitted patients 45% 55% 8% of sample of sample of sample 23
Omicron vaccine effectiveness study methodology: Test negative design Method used globally on pathology surveillance data to assess annual influenza vaccine effectiveness • Method is generalisable to other specific respiratory conditions provided vaccination does not have clinical efficacy for other respiratory conditions • Methodology implicitly adjusts for biases due to health-seeking behaviour • Efficacy estimates and associated confidence intervals are consistent with randomized control studies • A number of publications have used a test-negative design for real world Covid-19 vaccine effectiveness studies De Serres G, Skowronski DM, Wu XW, Ambrose CS. The test-negative design: validity, accuracy and precision of vaccine efficacy estimates compared to the gold standard of randomised placebo-controlled clinical trials . Euro Surveill. 2013;18(37):pii=2058 24
Pfizer-BioNTech vaccine 33% effective in reducing Omicron-related infection Real-world Pfizer vaccine Waning effectiveness against infection effectiveness against infection over time 100 100 90 80% 90 Omicron has 80 80 materially reduced 70 70 vaccine effectiveness 60 60 against new 50 50 infections, potentially 40 40 compounded by 30 30 waning durability 56% 20 20 37% 25% 10 10 0 0 Unvaccinated Vaccinated 2 to 4 weeks 1 to 2 months 3 to 4 months Pre-Omicron Omicron wave post 2nd dose post 2nd dose 15 Novpost to 2nd 3 Dec dose2021 *To be submitted for peer review and publication Authors have no conflicts of interest direct or in kind Source: Discovery Health Insights https://discv.co/DiscoveryHealthInsights 25
Pfizer-BioNTech vaccine 70% effective in reducing Omicron-related hospital admissions Real-world Pfizer-BioNTech vaccine effectiveness against hospital admission 100 93% 90 80 Real world effectiveness of the Pfizer-BioNTech vaccine 70 against hospital admission has 60 reduced from 93% in the Delta 50 (pre-Omicron) wave to 70% in the Omicron wave, continuing 40 to provide substantial 30 protection against hospital 20 admission 10 0 Unvaccinated Vaccinated Pre-Omicron Omicron wave * Submitted for peer review and publication with the NEJM Authors have no conflicts of interest direct or in kind Source: Discovery Health analysis of Pfizer-BioNTech effectiveness 26
Vaccine effectiveness retracts slightly with increasing age but is maintained across various comorbidities – durability impact confounding Vaccine effectiveness by age during Omicron Vaccine effectiveness by comorbidity during period Omicron period 1,00 1,00 92% 0,90 0,90 82% 0,80 75% 0,80 74% 68% 69% 0,70 67% 0,70 63% 64% 59% 0,60 0,60 0,50 0,50 0,40 0,40 0,30 0,30 0,20 0,20 0,10 0,10 - - Hyper- Ischemic Heart 18-29 30-39 40-49 50-59 60-69 70-79 Diabetes Hypertension cholesterolamia Disease *To be submitted for peer review and publication Authors have no conflicts of interest direct or in kind Source: Discovery Health analysis of Pfizer-BioNTech effectiveness
Although prior infection confers reduced risk of re-infection, this is diminished against Omicron re-infection D614G Beta Delta 0,73 The protective effect of prior 0,59 infection has reduced over time, 0,48 and Omicron has 0,41 eroded that protective effect 0,29 further 0,27 Odds of reinfection- Omicron period Odds of reinfection Sept/Oct 2021 *To be submitted for peer review and publication, Pre-Omicron period refers to September - October Source: Discovery Health Insights https://discv.co/DiscoveryHealthInsights
Acknowledgements Shirley Collie, Jared Champion, Lizelle Steenkamp, Sameera Haneef, Matt Zylstra, Tommy Chen, Michael Cohen, Ismail Rasool, Thishen Govender Milani Wolmarans Mary Anne Davies Andrew Boulle Petro Rossouw Reshma Kassanjee Prof Glenda Gray Prof Linda Gail Bekker Dr Harry Moultrie Dr Richard Lessels Prof Cheryl Cohen Tulio de Oliveira Hendrik van de Venter John Sole 29
Conclusion| Omicron Insights 1. Apparent 29% lower severity based on early data (in high seroprevalence population) and supported by anecdotal clinical feedback 2. Vaccination remains single most important intervention to mitigate against severe Covid-19, with double dose of Pfizer-BioNTech vaccine showing 70% effectiveness in reducing risk of hospitalisation 3. Vaccine effectiveness against infection is materially reduced, with high numbers of breakthrough infections in vaccinated individuals 4. The protective effect of prior infection has reduced over time, and Omicron has eroded that protective effect further, with high re-infection rates in previously Covid-19 positive individuals 5. Notwithstanding the lower severity, health systems could still be over-run by the sheer volume of cases, considering Omicron’s rapid community spread 6. Children experiencing very low test positivity rate relative to adults, and low Covid-19 admissions in absolute terms, but appear to be at 20% greater risk of hospitalisation during Omicron wave relative to D614G wave 30
Real world analysis of omicron outbreak in South Africa including vaccine effectiveness December 2021
You can also read