Decreased severity of disease during the first global omicron variant COVID-19 outbreak in a large hospital in Tshwane, South Africa


The coronavirus disease 2019 (COVID-19) first reported in Wuhan China in December 2019, is a global pandemic that is threatening the health and wellbeing of people worldwide. To date there have been more than 274 million reported cases and 5.3 million deaths (World Health Organisation 2021). South Africa has borne the brunt of COVID-19 on the African continent, registering in excess of 3 million cases and 90 000 officially reported deaths (National Department of Health 2021). The number of deaths could be as high as 275,976 (Bradshaw et al., 2021), putting this country's death toll among the highest in the world with a cumulative excess death rate of 464 per 100,000. South Africa is currently experiencing its fourth COVID-19 wave, being driven by the recently identified Omicron variant. Previous waves were associated with the Ancestral, Beta and Delta variants.

The City of Tshwane (incorporating Pretoria and surrounding areas just north of Johannesburg), with its population of 3.31 million people (Office of the Executive Mayor 2021) has had 241,794 cases of SARS-CoV-2, 35 090 hospital admissions and 7,086 deaths (National Institute for Communicable Diseases 2021) since the first COVID-19 admission at the Steve Biko Academic Hospital on 4 May 2020.

The first case of Omicron was documented in the City of Tshwane on 9 November 2021 (World Health Organization 2021). This was followed by a rapid rise in SARS-CoV-2 infections and COVID-19 associated hospitalisations from 14 November 2021, heralding the onset of the fourth wave in South Africa. Omicron rapidly displaced the Delta variant in the City of Tshwane and the Gauteng Province of South Africa (Network for Genomics Surveillance in South Africa (NGS-SA) 2021).

During the resurgence in Tshwane, we observed a difference in the clinical picture of COVID-19 ward patients compared with prior COVID-19 waves. We report from the first global epicentre of Omicron-driven resurgence on the patient profile of admissions to the Steve Biko Academic Hospital in Pretoria, the heart of the Tshwane District. In the current study, we compare the clinical profile of 466 COVID-19 patients admitted in the first 33 days since the commencement of the Omicron driven fourth wave with that of 3962 patients admitted during the 3 pandemic waves over the previous 18 months since May 2020 and provide a description of the clinical profile of 98 patients in the hospital at the peak of the Omicron wave on 14-15 December 2021.


The Steve Biko Academic Hospital (SBAH) is an 800 bed tertiary academic hospital to which is attached the 240-bed Tshwane District Hospital and the University of Pretoria's Health Sciences Faculty. Sections of the hospital, including ICU, high care and general wards were repurposed for the management of adult and paediatric COVID-19 patients. This included areas in the Emergency Units, labour wards and theatres. All clinical departments provided both staff and services to the COVID-19 areas as required.

At the beginning of the pandemic in March 2020, a national hospital admissions surveillance system (DATCOV) was established by the National Institute of Communicable Diseases (NICD). Hospital level data were extracted from the COVID-19 hospital surveillance system for patients admitted to the Steve Biko Academic Hospital (SBAH) from 4 May 2020 to 16 December 2021. These hospital records were reviewed for a comparison between patients admitted during the Omicron wave and previous waves. All patients were included in the sample.

466 records from DATCOV of patients admitted during the Omicron wave were compared to all 3962 records of patients admitted during three previous waves over a period of 18 months. In addition, a snapshot analysis of 98 records of patients occupying COVID-19 beds in the hospital at peak bed occupancy were reviewed for severity of illness, primary indication for admission, oxygen supplementation level, self-reported vaccination and prior COVID-19 infection status. These data were entered into the internal hospital information system. Oxygen supplementation levels for 588 patients admitted to the hospital during the first wave were reviewed (Boswell et al., 09 December 2021).

The record files of 21 deceased patients for the period 14 November through 16 December 2021 were requested from the hospital registries and reviewed for cause of death.

Hospital COVID-19 bed occupancy was obtained from daily statistics captured by the Nursing Services Manager responsible for bed management at the facility.

Data for the city and province-wide cases, deaths and hospital admissions were provided by the NDOH (National Department of Health 2021) and the NICD (National Institute for Communicable Diseases 2021). Data analysis was done using Excel and STATA 16. Data smoothing was performed using LOWESS in STATA 16 (Stata/IC 16 2020).


There was an exponential increase in SARS-CoV-2 infections in the City of Tshwane, commencing in the week of 7 November 2021 as shown in Figure 1, reaching 11 010 cases in the week of 28 November 2021 and peaking in the week of 5 December 2021.

Figure 1

Figure 1 Weekly number of SARS-CoV-2 cases in Tshwane District, 24 October through 11 December 2021 (NICD)
View Large Image | Download Hi-res image | Download (PPT)

The highest single day occupancy of COVID beds during the Omicron wave was 108 on 13 December 2021, much lower than the highest level of COVID bed occupancy over previous waves, which was 213 beds occupied on 13 July 2021 at the peak of the Delta wave.

There were 466 admissions to the COVID-19 wards between 14 November and 16 December 2021 compared to 20 admissions in the preceding two weeks, showing the rapid rise in hospital admissions during this period (Figure 2).

Figure 2

Figure 2 Daily number of COVID-19 hospital admissions for Steve Biko Academic Hospital, 31 October to 16 December 2021
View Large Image | Download Hi-res image | Download (PPT)

Table 1 compares 466 patients admitted during the Omicron wave and 3962 during previous waves, showing significant differences in the age distribution, outcomes, level of care required, and length of hospital stay. Mean age was significantly lower (39 vs 49.8 years), most admissions were in the 30 – 39 year age group, 68% of admissions were for those below age 50 compared to 44.5% previously, and the proportion of admissions in 0-9 year old age group doubled.

Table 1 Description of COVID-19 admissions at Steve Biko Academic Hospital Complex, fourth wave compared to previous waves


14/11/21 – 16/12/21





n(%) or mean(SD)

n(%) or mean(SD)



# Admissions





Mean age



t= - 10.2

p< 0.00001

Proportions in age groups








z= 5.1

p< 0.00001




z = 2

P = 0.044





p< 0.00001





p< 0.00001




z= - 1.9

p = 0.15





p< 0.00001





p< 0.00001




z = -5.5

p< 0.00001




z= -2.7

P = 0.0061




z= -5.1

p< 0.00001


0 (0)

30 (0.8)



Length of stay



t = - 5.4

p< 0.00001




z= -3.4

P 0.0007




z= - 8.7

P< 0.00001

There were 21 (4.5%) hospital deaths compared to 847 (21.3%) and 5 (1%) ICU admissions compared to 172 (4.3%) in the Omicron wave compared to previous waves. Length of hospital stay was significantly shorter (4.0 days) in the Omicron wave compared to 8.8 days in previous waves.

A cause of death analysis for 21 deceased patients in the Omicron wave at the hospital showed COVID-19 with a confirmed COVID pneumonia as the cause of death in 10 (48%) patients, another cause exacerbated by COVID pneumonia in 4 (19%), and a cause unrelated to COVID pneumonia in 7 (33%).

Figure 3 and Table 2 show clinical severity of 98 patients in the COVID wards on 14-15 December 2021.

Figure 3

Figure 3 Tree diagram showing COVID-19 disease severity in patients at the Steve Biko Academic Hospital on 14-15 December 2021
View Large Image | Download Hi-res image | Download (PPT)

Table 2 Levels of oxygen supplementation for COVID pneumonia patients at Steve Biko Academic Hospital Complex 14-15 December 2021

Oxygen Supplementation Modality

Room Air

Nasal Prongs Oxygen

Face Mask Oxygen

Double Oxygen Nasal Prongs+ Face Mask Oxygen

High Flow Nasal Oxygen

Non Invasive Ventilation

Mechanical Ventilation


Confirmed COVID Pneumonia

5 (14%)

13 (36%)

8 (22%)


1 (3%)

4 (11%)

4 (11%)

36 (100%)

Thirty-six patients (37%) had a confirmed diagnosis of COVID pneumonia, of which 31 (86%) required oxygen supplementation.

Sixty-two (63%) patients were incidental COVID admissions, having been admitted for another serious primary medical, surgical, obstetric or psychiatric diagnosis. These cases have been labelled ‘incidental COVID’ as they were diagnosed as the result of hospital admission procedures, rather than having the typical clinical profile or meeting a case definition for COVID-19. This phenomenon of ‘incidental COVID’ is not a phenomenon observed before in South Africa and most likely reflects high levels of asymptomatic disease in the community with Omicron infection. As all patients being admitted to the hospital are tested for SARS-CoV-2 as per the policy, those testing positive are admitted to the designated COVID wards.

Fifty-four patients (55%) coped on room air without supplemental oxygen. Fewer than half (45%) of patients in COVID-19 wards compared to 99.5% (Boswell et al., 09 December 2021) in the first wave required oxygen supplementation.

Table 2 shows the level of oxygen supplementation as an index of severity among patients with COVID-19 pneumonia of whom 26 (72%) required no or low levels of oxygen supplementation. Ten patients (28%) required high care or ICU admission. Among the 4 ICU admissions, 3 patients exhibited features of a COVID-19 pneumonia, however only 1 patient required invasive mechanical ventilation primarily for COVID-19 associated respiratory failure. One (1) patient required invasive mechanical ventilation for confirmed pneumonia with severe COPD and cardiogenic shock in the Emergency Care Unit. The six paediatric admissions to paediatric high care/ICU were attributed to diagnoses unrelated to COVID-19.

The Emergency Medical Unit in the SBAH complex reported a marked decline in the use of reticulated oxygen volumes compared to previous waves.

Among 55 pregnant women admitted to the COVID-19 labour ward only 2 required face-mask oxygen, mirroring the 'incidental COVID-19' picture seen in the adult and paediatric COVID-19 wards. One of these two patients had a mild COVID-19 pneumonia requiring supplemental oxygen for 3 days. The second required supplemental oxygen for a diagnosis unrelated to COVID-19.

The findings described above for the Steve Biko Academic Hospital were comparable to city-wide trends when cases and admissions from all public and private hospitals reported in the national hospital surveillance system were observed. The data from the NICD DATCOV database showed that there was a total of 33,643 SARS-CoV-2 cases, 3,233 COVID-19 hospitalisations and 130 deaths reported in the City of Tshwane in the same period (14 November to 16 December 2021), reflecting a lower admission per case ratio, lower death rate and lower rates of admission to the ICU compared to previous waves.

Figure 4 further shows an uncoupling of the case and death rates for the Gauteng Province as a whole, confirming the local hospital experience of significantly fewer admissions to the ICU and deaths compared to previous waves.

Figure 4:

Figure 4COVID-19 cases and reported deaths rates for the Gauteng Province (National Department of Health 2021)
View Large Image | Download Hi-res image | Download (PPT)


As it has been demonstrated elsewhere that the Omicron variant rapidly displaced Delta in the region in which this study was conducted (Viana et al., 2021), the assumption that the clinical profile described in this paper represents disease caused by the Omicron variant is reasonable. The Omicron outbreak has spread and declined in the City of Tshwane with unprecedented speed, peaking within 4 weeks of its commencement. Hospital admissions increased rapidly and began to decline within a period of 33 days. This demonstrates a significantly different transmission trajectory and epidemiological profile from that of previous variants of concern and can be expected to be replicated in other parts of the world.

Peak bed occupancy was about half that of the third (Delta) wave suggesting a lower rate of hospital admissions relative to the number of cases in the Omicron wave compared to previous waves. The mean age of hospitalized patients in the Omicron wave was 11 years younger than previous waves and may reflect the higher rate of vaccination in the elderly population. Fewer ICU admissions and deaths and a shorter length of hospital stay indicate decreased severity of disease caused by the Omicron variant. A third of deaths resulted from a cause other than COVID-19, and there were no paediatric deaths related to severe COVID-19 disease. Sixty three percent of COVID-19 patients in the snapshot at peak bed occupancy were in hospital for an alternative primary diagnosis, and were ‘incidental COVID’ patients as they were diagnosed as the result of hospital admission procedures, rather than having the typical clinical profile or meeting a case definition for COVID. This phenomenon has not been observed to this extent before in the Steve Biko Academic Hospital or anywhere in South Africa and most likely reflects high levels of asymptomatic disease in the community with Omicron infection.

The low percentage of patients in the COVID-19 wards with a confirmed diagnosis of COVID-19 pneumonia has implications for the application of clinical skill and expertise being deployed to the COVID-19 wards, with all specialties required to manage their ‘incidental COVID' patients under COVID-19 infection control standards. It also implies much lower oxygen utilization levels in the COVID-19 wards. The categorization of patients into ‘incidental SARS-CoV-2’ and moderate to severe COVID-19 disease may lead to a radically different internal organization of COVID-19 wards at the hospital.

A similar profile of patients is being seen in COVID-19 wards in the Western Cape Province of South Africa (Mendelsohn et al., 2021) and reports of similar patterns of the clinical profile of patients have been described by one of South Africa's largest private hospital groups (Maslo, 2021r).

The changing clinical presentation of SARS-CoV-2 infection is likely due to high levels of prior infection and vaccination coverage in this setting. The estimated seroprevalence of immunity from prior infection and vaccine induced immunity for the City of Tshwane is 66.7% (95% CI, 54.2 to 69.0) (Madhi Shabir et al., 2021). About 36% of adults aged 18 to 49 and 58% over age 50 in the Gauteng Province are vaccinated. Another plausible cause for the lower number of admissions and decreased severity is a decrease in pathogenicity or virulence of the highly mutated Omicron variant, though more research is required to fully establish this theory.

A similar pattern is likely to emerge in other provinces in South Africa as Omicron spreads rapidly across the country, but may differ in countries where levels of hybrid immunity or the mix of immunity from prior infection and vaccination are different.

Limitations of the study include the inability to compare the Omicron wave to each of the three previous waves separately due to the difficulty of defining the beginning and end dates of previous waves. Another limitation of the study is that it was unable to compare clinical parameters of patients in the COVID-19 wards across waves due to poor electronic record-keeping of these parameters, including clinician evaluations, chest-xray finding and blood biomarkers for COVID-19 disease.


There was decreased severity of disease in the Omicron-driven fourth wave in the City of Tshwane, its first global epicentre, with fewer deaths, ICU admissions and a shorter length of hospital stay. The younger age profile of patients is likely to have been a factor of this clinical profile.

The wave increased at a faster rate than previous waves, completely displacing the Delta variant within weeks and began its decline in both cases and hospital admissions in the fifth week following its commencement.

There are clear signs that case and admission rates in South-Africa may decline further over the next few weeks. If this pattern continues and is repeated globally, we are likely to see a complete decoupling of case and death rates, suggesting that Omicron may be a harbinger of the end of the epidemic phase of the Covid pandemic, ushering in its endemic phase.


F.Abdullah1, 2, 11, J.Myers4, D.Basu3, G.Tintinger2, V.Ueckermann12, M.Mathebula5, R.Ramlall6, S.Spoor6, Villiers6, Z.Van der Walt6, J.Cloete7, P.Soma-Pillay8, P.Rheeder2, F.Paruk9, A.Engelbrecht10, V.Lalloo10, M.Myburg5, J.Kistan3, W.von Hougenhouck-Tulleken2, M.T.Boswell2, G.Gray11, R.Welch12, L.Blumberg12, 13, W.Jassat12

1 Division of Infectious Diseases, Steve Biko Academic Hospital
2 Department of Internal Medicine, Steve Biko Academic Hospital and University of Pretoria
3 Department of Public Health Medicine, University of Pretoria
4 School of Public Health, University of Cape Town
5 Steve Biko Academic Hospital
6 Tshwane District Hospital
7 Department of Paediatrics, Steve Biko Academic Hospital and University of Pretoria
8 Department of Obstetrics, Steve Biko Academic Hospital and University of Pretoria
9 Department of Critical Care, Steve Biko Academic Hospital and University of Pretoria
10 Department of Emergency Medicine, Steve Biko Academic Hospital and University of Pretoria
11 South African Medical Research Council
12 National Institute of Communicable Diseases
13 Right to Care, South Africa

Original article available on International Journal of Infectious Diseases Website