Using Generalized Structured Additive Regression Models to determine factors associated with and clusters for COVID-19 hospital deaths in South Africa (preprint)

Background The first case of COVID-19 in South Africa was reported in March 2020 and the country has since recorded over 3.6 million laboratory-confirmed cases and 100 000 deaths as of March 2022. Transmission and infection of SARS-CoV-2 virus and deaths in general due to COVID-19 have been shown to be spatially associated but spatial patterns in in-hospital deaths have not fully been investigated in South Africa. This study uses national COVID-19 hospitalization data to investigate the spatial effects on hospital deaths after adjusting for known mortality risk factors.Methods COVID-19 hospitalization data and deaths were obtained from the National Institute for Communicable Diseases (NICD). Generalized structured additive logistic regression model was used to assess spatial effects on COVID-19 in-hospital deaths adjusting for demographic and clinical covariates. Continuous covariates were modelled by assuming second-order random walk priors, while spatial autocorrelation was specified with Markov random field prior and fixed effects with vague priors respectively. The inference was fully Bayesian.Results The risk of COVID-19 in-hospital mortality increased with patient age, with admission to intensive care unit (ICU) (aOR = 4.16; 95% Credible Interval: 4.05–4.27), being on oxygen (aOR = 1.49; 95% Credible Interval: 1.46–1.51) and on invasive mechanical ventilation (aOR = 3.74; 95% Credible Interval: 3.61–3.87). Being admitted in a public hospital (aOR = 3.16; 95% Credible Interval: 3.10–3.21) was also significantly associated with mortality. Risk of in-hospital deaths increased in months following a surge in infections and dropped after months of successive low infections highlighting crest and troughs lagging the epidemic curve. After controlling for these factors, districts such as Vhembe, Capricorn and Mopani in Limpopo province, and Buffalo City, O.R. Tambo, Joe Gqabi and Chris Hani in Eastern Cape province remained with significantly higher odds of COVID-19 hospital deaths suggesting possible health systems challenges in those districts.Conclusion The results show substantial COVID-19 in-hospital mortality variation across the 52 districts. Our analysis provides information that can be important for strengthening health policies and the public health system for the benefit of the whole South African population. Understanding differences in in-hospital COVID-19 mortality across space could guide interventions to achieve better health outcomes in affected districts.

Using Generalized Structured Additive Regression Models to Determine Factors Associated with and Clusters for COVID-19 Hospital Deaths in South Africa (preprint)

Background: The first case of COVID-19 in South Africa was reported in March 2020 and the country has since recorded over 3.6 million laboratory-confirmed cases and 100 000 deaths as of March 2022. Transmission and infection of SARS-CoV-2 virus as well as deaths in general due to COVID-19 have been shown to be spatially associated but spatial patterns in hospital deaths have not fully been investigated in South Africa. This study uses national COVID-19 hospitalization data to investigate the spatial effects on hospital deaths after adjusting for known mortality risk factors. Methods: COVID-19 hospitalization data and deaths were obtained from the National Institute for Communicable Diseases (NICD), who together with the South African National Department of Health (SANDoH) collected hospital admissions data through DATCOV, an active electronic hospital surveillance system for COVID-19. We used the generalized structured additive logistic regression model that allows for modelling spatial correlation to realistically estimate risk factors for hospital COVID-19 deaths. The model included patient demographic and clinical factors as well as time in months which accounted for different waves. Continuous covariates were modelled by assuming second-order random walk priors, while spatial autocorrelation was specified with Markov random field prior and fixed effects with vague priors respectively. The inference was fully Bayesian. Results: The risk of COVID-19 in-hospital mortality increased with patient age as well as with admission to intensive care unit (ICU) (aOR=4.16; 95% Credible Interval: 4.05-4.27), being on oxygen (aOR=1.49; 95% Credible Interval: 1.46-1.51) and on invasive mechanical ventilation (aOR=3.74; 95% Credible Interval: 3.61-3.87). Being admitted in a public hospital (aOR= 3.16; 95% Credible Interval: 3.10-3.21) was also a significant risk factor for mortality. Risk of deaths also increased in months following a surge in infections and dropped after months of successive low infections highlighting crest and troughs lagging the epidemic curve. After controlling for these factors, districts such as Vhembe, Capricorn and Mopani in Limpopo province, and Buffalo City, O.R. Tambo, Joe Gqabi and Chris Hani in Eastern Cape province remained with significantly higher odds of COVID-19 hospital deaths suggesting possible health systems challenges in those districts. Conclusion: The results show substantial COVID-19 in-hospital mortality variation across the 52 districts. This highlights the importance of modelling spatial patterns simultaneously with fixed and nonlinear effects of continuous covariates to identify clusters at high risk of health outcome. The flexible approach to modelling data that has spatial patterns helps to account for possible loss of efficiency due to spatial correlation that spatial patterns can induce in data. Our analysis suggests notable COVID-19 hospital deaths clustering in some districts in Limpopo and Eastern Cape provinces and this information can be important for strengthening health policies and the public health system for the benefit of the whole South African population. Understanding differences in in-hospital COVID-19 mortality across space could guide interventions to achieve better health outcomes.

TRENDS IN CASES, HOSPITALISATION AND MORTALITY RELATED TO THE OMICRON BA.4/BA.5 SUB-VARIANTS IN SOUTH AFRICA (preprint)

Introduction: The Omicron BA.1/BA.2 wave in South Africa had lower hospitalisation and mortality than previous SARS-CoV-2 variants and was followed by an Omicron BA.4/BA.5 wave. This study compared admission incidence risk across waves, and the risk of mortality in the Omicron BA.4/BA.5 wave, to the Omicron BA.1/BA.2 and Delta waves. Methods: Data from South Africa's national hospital surveillance system, SARS-CoV-2 case linelist and Electronic Vaccine Data System were linked and analysed. Wave periods were defined when the country passed a weekly incidence of 30 cases/100,000 people. Mortality rates in the Delta, Omicron BA.1/BA.2 and Omicron BA.4/BA.5 wave periods were compared by post-imputation random effect multivariable logistic regression models. Results: In-hospital deaths declined 6-fold from 37,537 in the Delta wave to 6,074 in the Omicron BA.1/BA.2 wave and a further 7-fold to 837 in the Omicron BA.4/BA.5 wave. The case fatality ratio (CFR) was 25.9% (N=144,798), 10.9% (N=55,966) and 7.1% (N=11,860) in the Delta, Omicron BA.1/BA.2, and Omicron BA.4/BA.5 waves respectively. After adjusting for age, sex, race, comorbidities, health sector and province, compared to the Omicron BA.4/BA.5 wave, patients had higher risk of mortality in the Omicron BA.1/BA.2 wave (adjusted odds ratio [aOR] 1.43; 95% confidence interval [CI] 1.32-1.56) and Delta (aOR 3.22; 95% CI 2.98-3.49) wave. Being partially vaccinated (aOR 0.89, CI 0.86-0.93), fully vaccinated (aOR 0.63, CI 0.60-0.66) and boosted (aOR 0.31, CI 0.24-0.41); and prior laboratory-confirmed infection (aOR 0.38, CI 0.35-0.42) were associated with reduced risks of mortality. Conclusion: Overall, admission incidence risk and in-hospital mortality, which had increased progressively in South Africa's first three waves, decreased in the fourth Omicron BA.1/BA.2 wave and declined even further in the fifth Omicron BA.4/BA.5 wave. Mortality risk was lower in those with natural infection and vaccination, declining further as the number of vaccine doses increased.

Clinical severity of COVID-19 patients admitted to hospitals during the Omicron wave in South Africa (preprint)

Background: Clinical severity of patients hospitalised with SARS-CoV-2 infection during the Omicron (fourth) wave was assessed and compared to trends in the D614G (first), Beta (second), and Delta (third) waves in South Africa. Methods: Weekly incidence of 30 laboratory-confirmed SARS-CoV-2 cases/100,000 population defined the start and end of each wave. Hospital admission data were collected through an active national COVID-19-specific surveillance programme. Disease severity was compared across waves by post-imputation random effect multivariable logistic regression models. Severe disease was defined as one or more of acute respiratory distress, supplemental oxygen, mechanical ventilation, intensive-care admission or death. Results: 335,219 laboratory-confirmed SARS-CoV-2 admissions were analysed, constituting 10.4% of 3,216,179 cases recorded during the 4 waves. In the Omicron wave, 8.3% of cases were admitted to hospital (52,038/629,617) compared to 12.9% (71,411/553,530) in the D614G, 12.6% (91,843/726,772) in the Beta and 10.0% (131,083/1,306,260) in the Delta waves (p<0.001). During the Omicron wave, 33.6% of admissions experienced severe disease compared to 52.3%, 63.4% and 63.0% in the D614G, Beta and Delta waves (p<0.001). The in-hospital case fatality ratio during the Omicron wave was 10.7%, compared to 21.5%, 28.8% and 26.4% in the D614G, Beta and Delta waves (p<0.001). Compared to the Omicron wave, patients had more severe clinical presentations in the D614G (adjusted odds ratio [aOR] 2.07; 95% confidence interval [CI] 2.01-2.13), Beta (aOR 3.59; CI: 3.49-3.70) and Delta (aOR 3.47: CI: 3.38-3.57) waves. Conclusion: The trend of increasing cases and admissions across South Africa's first three waves shifted in Omicron fourth wave, with a higher and quicker peak but fewer admitted patients, who experienced less clinically severe illness and had a lower case-fatality ratio. Omicron marked a change in the SARS-CoV-2 epidemic curve, clinical profile and deaths in South Africa. Extrapolations to other populations should factor in differing vaccination and prior infection levels.

Increased mortality among individuals hospitalised with COVID-19 during the second wave in South Africa (preprint)

IntroductionSouth Africa experienced its first wave of COVID-19 peaking in mid-July 2020 and a larger second wave peaking in January 2021, in which the SARS-CoV-2 501Y.V2 lineage predominated. We aimed to compare in-hospital mortality and other patient characteristics between the first and second waves of COVID-19. MethodsWe analysed data from the DATCOV national active surveillance system for COVID-19 hospitalisations. We defined four wave periods using incidence risk for hospitalisation, pre-wave 1, wave 1, pre-wave 2 and wave 2. We compared the characteristics of hospitalised COVID-19 cases in wave 1 and wave 2, and risk factors for in-hospital mortality accounting for wave period using multivariable logistic regression. ResultsPeak rates of COVID-19 cases, admissions and in-hospital deaths in the second wave exceeded the rates in the first wave (138.1 versus 240.1; 16.7 versus 28.9; and 3.3 versus 7.1 respectively per 100,000 persons). The weekly average incidence risk increase in hospitalisation was 22% in wave 1 and 28% in wave 2 [ratio of growth rate in wave two compared to wave one: 1.04, 95% CI 1.04-1.05]. On multivariable analysis, after adjusting for weekly COVID-19 hospital admissions, there was a 20% increased risk of in-hospital mortality in the second wave (adjusted OR 1.2, 95% CI 1.2-1.3). In-hospital case fatality-risk (CFR) increased in weeks of peak hospital occupancy, from 17.9% in weeks of low occupancy (<3,500 admissions) to 29.6% in weeks of very high occupancy (>12,500 admissions) (adjusted OR 1.5, 95% CI 1.4-1.5). Compared to the first wave, individuals hospitalised in the second wave, were more likely to be older, 40-64 years [OR 1.1, 95% CI 1.0-1.1] and [≥]65 years [OR 1.1, 95% CI 1.1-1.1] compared to <40 years; and admitted in the public sector [OR 2.2, 95% CI 1.7-2.8]; and less likely to have comorbidities [OR 0.5, 95% CI 0.5-0.5]. ConclusionsIn South Africa, the second wave was associated with higher incidence and more rapid increase in hospitalisations, and increased in-hospital mortality. While some of this is explained by increasing pressure on the health system, a residual increase in mortality of hospitalised patients beyond this, could be related to the new lineage 501Y.V2. RESEARCH IN CONTEXT O_TEXTBOXEvidence before this studyMost countries have reported higher numbers of COVID-19 cases in the second wave but lower case-fatality risk (CFR), in part due to new therapeutic interventions, increased testing and better prepared health systems. South Africa experienced its second wave which peaked in January 2021, in which the variant of concern, SARS-CoV-2 501Y.V2 predominated. New variants have been shown to be more transmissible and in the United Kingdom, to be associated with increased hospitalisation and mortality rates in people infected with variant B.1.1.7 compared to infection with non-B.1.1.7 viruses. There are currently limited data on the severity of lineage 501Y.V2. Added value of this studyWe analysed data from the DATCOV national active surveillance system for COVID-19 hospitalisations, comparing in-hospital mortality and other patient characteristics between the first and second waves of COVID-19. The study revealed that after adjusting for weekly COVID-19 hospital admissions, there was a 20% increased risk of in-hospital mortality in the second wave. Our study also describes the demographic shift from the first to the second wave of COVID-19 in South Africa, and quantifies the impact of overwhelmed hospital capacity on in-hospital mortality. Implications of all the available evidenceOur data suggest that the new lineage (501Y.V2) in South Africa may be associated with increased in-hospital mortality during the second wave. Our data should be interpreted with caution however as our analysis is based on a comparison of mortality in the first and second wave as a proxy for dominant lineage and we did not have individual-level data on lineage. Individual level studies comparing outcomes of people with and without the new lineage based on sequencing data are needed. To prevent high mortality in a potential third wave, we require a combination of strategies to slow the transmission of SARS-CoV-2, to spread out the peak of the epidemic, which would prevent hospital capacity from being breached. C_TEXTBOX