Permanent URI for this collection
All papers that are published / presented (or in the process of being published / or to be presented) at a scientific conference by SAMRC staff.
Browse
Recent Submissions
Publication Point of care antibody tests for covid-19: Field-based performance, South Africa(2022) Goga, A; Mayne, E.S; Woeber, K; Takuva, S; Nsibande, D; Lekalakala, M; Jaumdally, S; Mutevedzi, P; Vreede, H; Daniels, B.B; Kufe, C.N; Dheda, K; Chetty, K; Gray, G.E; Nyuyki, C; Madhi, S; SA COVID-19 POC study team; Goga A, Woeber K, Nsibande D, Daniels B, Gray GE. South African Medical Research CouncilBackground: SARS-CoV2 antibody testing is an important auxillary test especially for retrospective diagnosis or in patients with long COVID-19 or multisystem inflammatory syndrome of childhood. Epidemiological serology studies may also assist public health planning. Access to formal laboratory testing is not universal in many low-and middle-income (LMIC) countries and rapid lateral flow antibody tests are an attractive alternative. Performance of these tests has been inconsistent. A large-scale study was undertaken in South Africa, during the beta and delta waves, to assess the field-based performance of rapid point of care (POC) COVID-19 antibody tests. Methods: Symptomatic, ambulatory persons under investigation (PUIs) aged 18 years and older, presenting for SARS-CoV-2 diagnosis at public health facilities in three provinces, South Africa were enrolled at baseline. All patients completed a questionnaire regarding symptoms. Nasopharyngeal swabs were taken and processed for SARS-CoV-2 PCR testing using a GeneXpert (Cepheid, USA), or manual assay (ThermoFisher TaqPath assay or Seegene Allplex assay) on a real-time platform at routine accredited National Health Laboratory Service laboratories as per routine national protocols. Concomitantly, trained study staff performed three facility-based POC lateral flow antibody tests on a on a fingerstick sample and blood was collected for formal serology. POC tests were selected following a rapid in-laboratory evaluation. Asymptomatic contacts of people with confirmed COVID-19 were recruited into the asymptomatic study arm and rapid tests and PCR were performed. PCR and rapid positive patients and 500 negative controls were followed up at 5-14 days. Antibody tests were compared with formal serology performed on 2 platforms-Euroimmun (Euroimmun, Lubeck) IgA and IgG anti-S antibodies and Abbott Architect IgG test. Results: The sensitivity (S), specificity (Sp), positive (PPV) and negative predictive (NPV) values of tests for PUIs and contacts were calculated (Table 1)∗. Analyses using serology as a reference are forthcoming. Conclusion: Compared with PCR, performance of rapid POC COVID-19 antibody tests was poor with low sensitivity. This may reflect the patient cohort tested as humoral responses typically develop from day 7-14. The tests are unlikely to be useful for acute diagnosis but sensitivity may improve at later timepoints and further follow up data will be analysed by duration of symptom onset, severity of symptoms and wave (beta versus delta).Publication Utility of Covid-19 point of care antigen tests in low-middle income settings(Databases of international organizations, 2022-02-16) Goga, A; Mayne, E.S; Woeber, K; Takuva, S; Nsibande, D; Lekalakala, M; Jaumdally, S; Mutevedzi, P; Vreede, H; Daniels, B.B; Kufe, C.N; Dheda, K; Chetty, K; Gray, G.E; Madhi, S; Goga A, Woeber K, Nsibande D, Daniels B, Gray GE. South African Medical Research CouncilBackground: Access to SARS-CoV-2 polymerase chain reaction (PCR) testing is a bottleneck globally, especially in low-and middle-income countries (LMICs). Reliable point-of-care (POC) diagnostics for coronavirus disease 2019 (COVID-19) are cheaper and easier to scale-up than PCR especially in LMICs, and will facilitate interruption of transmission. We report the field-based effectiveness of rapid point-of-care (POC) antigen COVID-19 tests during the beta and delta waves, in South Africa. Methods: We enrolled symptomatic, ambulatory persons under investigation (PUIs) aged 18 years and older, presenting for SARS-CoV-2 diagnosis at public health facilities in three provinces, South Africa. All patients completed a questionnaire regarding symptoms. Nasopharyngeal swabs were taken and processed for SARS-CoV-2 PCR testing using either GeneXpert (Cepheid, USA), or with a manual assay (ThermoFisher TaqPath assay or Seegene Allplex assay) on a real-time PCR platform at routine, accredited National Health Laboratory Service laboratories, as per routine national protocols. Concomitantly, trained study staff performed three facility-based POC antigen tests on a nasal/nasopharyngeal swab, as recommended by the manufacturer. Asymptomatic contacts of people with confirmed COVID-19 were recruited into the asymptomatic study arm and rapid tests and PCR were performed. The sensitivity (S), specificity (Sp), positive (PPV) and negative predictive (NPV) values of tests for PUIs and contacts were calculated using PCR as the reference standard. Results: Between Oct 2020-2021 1816 participants were enrolled;472 (26%) tested PCR or rapid test positive;235 positives (49.8%) and 532 negatives were followed up at 5-14 days;574 asymptomatic contacts were enrolled, of which 21 (3.7%) were PCR positive. Performance of the three antigen tests are shown in Table 1∗. Conclusion: In a real world setting, during the beta and delta waves, compared with PCR the sensitivity of rapid antigen tests ranged from 35-68%. This may reflect low viral loads at diagnosis. Further work will compare antigen test performance in patients with high versus lower cycle threshold (Ct) values. Meanwhile, PCR testing capacity needs urgent scale-up in LMICs and improved POC diagnostics are needed to facilitate COVID-19 diagnosis in LMICs.Publication Are people living with HIV (PLWH) more prone to relapse?(2022-11-08) Loveday, M; Loveday M, HIV and other Infectious Diseases Research Unit, South African Medical Research CouncilNo Abstract Available.Publication The genetics of PD in sub-Saharan Africa: What do we know?(2023-01-23) Bardien, S; Bardien S, Centre of Tuberculosis. South African Medical Research CouncilNo abstract.Publication A national household SARS CoV-2 seroprevalence survey in South Africa, 2020-2021(2022-03) Simbayi, L; Moyo, S; Zuma, K; Puren, A; Zungu, M; Jooste, S; Mabaso, M; Reddy, T; Parker, W; Naidoo, I; Morris, L; Moore, P; Manda, S; Goga, A; Ngandu, N; Cawood, C; Reddy T, Manda S, Ngandu N. Goga A. South African Medical Research CouncilSouth Africa is one of the African countries most affected by the COVID-19 pandemic. SARS-CoV-2 seroprevalence surveys provide valuable epidemiological information given the existence of asymptomatic cases. We report the findings of the first nationwide household-based population estimates of SARS-CoV-2 seroprevalence among people aged 12 years and older in South Africa. The survey used a cross-sectional multi-stage stratified cluster design undertaken over two separate time periods (November 2020 – February 2021 and April – June 2021) which coincided with the second and third waves of the pandemic in South Africa. The Abbott® and Euroimmun® ani-SARS CoV-2 antibody assays were used to test for SARS-CoV-2 antibodies, the latter being the final result. The survey data was weighted with final individual weights benchmarked against 2020 mid-year population estimates by age, race, sex, and province. Frequencies were used to describe characteristics of the study population and SARS-CoV-2 seroprevalence. Bivariate and multivariate logistics regression analysis were used to identify factors associated with SARS-CoV-2 seropositivity. 13640 participants gave a blood sample. The SARS-CoV-2 seroprevalence using the Euroimmun assay was 19.6% (95% CI 17.9–21.3) over the study period, translating to an estimated 8 675 265 (95% CI 7 508 393 – 9 842 137) estimated infections among people aged 12 years and older across South Africa by June 2021. Seroprevalence was higher in the Free State (26.8%), and Eastern Cape (26.0%) provinces (Figure). Increased odds of seropositivity were associated with prior PCR testing [aOR=1.29 (95% CI: 0.99-1.66)], being female [aOR=1.28 (95% CI 1.00-1.64), p=0.048] and hypertension, [aOR=1.28 (95% CI 1.00-1.640, p=0.048]. These findings highlight the burden of infection in South Africa by June 2021, and support testing strategies that focus on individuals with known exposure or symptoms since universal testing is not feasible. Females and younger people were more likely to be infected suggesting need for additional strategies targeting these populations. The estimated number of infections was 6.5 times higher than the number of SARS-CoV-2 cases reported nationally, suggesting that the country’s testing strategy and capacity partly explain the dynamics of the pandemic. It is therefore essential to bolster testing capacity and to rapidly scale up vaccinations in order to contain the spread of the virus in the country.Publication Performance of three SARS-CoV-2 serology assays in a household serosurvey in South Africa(2022-03) Moyo, S; Puren, A; Simbayi, L; Zuma, K; Zungu, N; Marinda, E; Jooste, S; Fortuin, M; Brukwe, Z; Singh, B; Mabaso, M; Reddy, T; Parker, W; Naidoo, I; Manda, S; Goga, A; Ngandu, N; Cawood, C; Morris, L; Moore, P; Rehle, T; Reddy T, Biostatistics Unit. Ngandu N, HIV Prevention. Goga A, Health Systems Research Unit. South African Medical Research CouncilIn this paper we present results from the first national house-hold based SARS CoV- 2 sero-prevalence survey conducted in South Africa in 2020-21. The survey used a cross-sectional multi-stage stratified cluster design. The data was collected over two separate time periods (November 2020 - February 2021 and April - June 2021). This periods coincided with the second and third waves of the pandemic in South Africa. In this paper we compare the performance of the The Abbott® and Euroimmun® ani-SARS CoV- 2 antibody assays.Publication Investigating the impact of Covid-19 restrictions on interpersonal violence in South Africa(2022-11-28) Prinsloo, M; Abrahams, N; Mhlongo, S; Dekel, B; Ketelo, A; Jabar, A; Jewkes, R; Parry, C; Matzopoulos, R; Prinsloo M, Jabar A, Matzopoulos R. Burden of Disease Research Unit. Abrahams N, Mhlongo S, Dekel B. Gender and Health Research Unit. Jewkes R, Research Strategy in the SAMRC Office of the President. Parry C, Alcohol, Tobacco and Other Drug Research Unit, South African Medical Research Council.Background South Africa’s (SA’s) response to the Covid-19 pandemic included rapid testing, social distancing, quarantining and a stringent five-phased lockdown approach. The necessitated countrywide lockdown restricted movement to the purchase of essential items, and alcohol and tobacco sales bans. This overlapped with a substantial decrease in injury mortality and hospital trauma cases. It is unclear whether lockdown restrictions decreased overall exposure to violence in public spaces or whether it increased in the domestic setting, where women and children were possibly more vulnerable to some adult males as the main perpetrators of violence. Methods We conducted a retrospective review of post-mortem reports using a multistage random sample, stratified by province and mortuary size. The study period from April 2020 to March 2021 spans lockdown levels and alcohol bans. The sample comprises approximately 30,000 records from 81 mortuaries, which will be weighted to a national estimate. Age-specific and age-standardized mortality rates will be calculated for cause of death by sex for the study period and for combinations of lockdown-related restrictions and alcohol sales bans. Comparative rates for corresponding periods in 2009 and 2017 will be calculated. Results The injury mortality profile for SA and rates during the Covid-19 pandemic will be presented. This will identify any increase or decrease in the causes of injury by sex during the national response to the pandemic, particularly for interpersonal violence. Conclusion As the third nationally representative sample of injury-related mortality, the findings will guide future emergency crisis response in high injury settings such as SA.Publication Antimicrobial susceptibility and molecular epidemiology of Neisseria gonorrhoeae infection in East London, South Africa(2023-04-18) Peters, R.P.H; Bamford, C; Jung, H; Mdingi, M; Gigi, R; Martinez, F.PNo abstract available.Publication Ntombi Vimbela! One-year post-intervention outcomes from a sexual violence prevention intervention pilot feasibility study, in South African higher education institutions(2022) Machisa, M; Mahlangu, P; Chirwa, E; Nunze, N; Sikweyiya, Y; Jewkes, R; Dartnall, EPublication News media coverage of COVID-19 lockdown tobacco sales ban in South Africa(2022-03-18) Egbe, C.O; Ngobese, P; Barca, H; Crosbie, EPublication Exploring mental health as a pathway to adverse pregnancy outcomes: Analysis from A rape cohort(2022-11-06) Abrahams, N; Chirwa, E; Mhlongo, S; Seedat, S; Myers, B; Peer, N; Kenge, A; Lombard, C; Garcia-Morena, C; Jewkes, RPublication Exposure to childhood abuse is associated with hypertension in South African women(2021-04-14) Nguyen, K; Kengne, A.P; Abrahams, N; Jewkes, R; Mhlongo, S; Peer, N; Peer N, Nguyen K, Kengne AP, Non-Communicable Diseases Research Unit, South African Medical Research Council. Jewkes R, Shibe M, Gender and Health Research Unit, South African Medical Research CouncilPublication SASBi-SC/SAGS 2022 Student symposium(2022) Tromp, G; Tromp G, Centre of Tuberculosis, South African Medical Research CouncilPublication Biomimetic nanoparticle-based host-directed therapy for the eradication of mycobacterium tuberculosis(2023) Du Plessis, SNo Abstract AvailablePublication Automatic tuberculosis and COVID-19 cough classification using deep learning(IEEE, 2022-07-02) Pahar, M; Klopper, M; Reeve, B; Warren, R; Theron, G; Diacon, A; Niesler, T; Klopper M, Reeve B, Warren R, Theron G. Cente for Tuberculosis, South African Medical Research CouncilWe present a deep learning based automatic cough classifier which can discriminate tuberculosis (TB) coughs from COVID-19 coughs and healthy coughs. Both TB and COVID-19 are respiratory diseases, contagious, have cough as a predominant symptom and claim thousands of lives each year. The cough audio recordings were collected at both indoor and outdoor settings and also uploaded using smartphones from subjects around the globe, thus containing various levels of noise. This cough data include 1.68 hours of TB coughs, 18.54 minutes of COVID-19 coughs and 1.69 hours of healthy coughs from 47 TB patients, 229 COVID-19 patients and 1498 healthy patients and were used to train and evaluate a CNN, LSTM and Resnet50. These three deep architectures were also pre-trained on 2.14 hours of sneeze, 2.91 hours of speech and 2.79 hours of noise for improved performance. The class-imbalance in our dataset was addressed by using SMOTE data balancing technique and using performance metrics such as F1-score and AUC. Our study shows that the highest F1-scores of 0.9259 and 0.8631 have been achieved from a pre-trained Resnet50 for two-class (TB vs COVID-19) and three-class (TB vs COVID-19 vs healthy) cough classification tasks, respectively. The application of deep transfer learning has improved the classifiers’ performance and makes them more robust as they generalise better over the cross-validation folds. Their performances exceed the TB triage test requirements set by the world health organisation (WHO). The features producing the best performance contain higher order of MFCCs suggesting that the differences between TB and COVID-19 coughs are not perceivable by the human ear. This type of cough audio classification is non-contact, cost-effective and can easily be deployed on a smartphone, thus it can be an excellent tool for both TB and COVID-19 screening.Publication TB or not TB? Acoustic cough analysis for tuberculosis classification(Interspeech, 2022-09) Frost, G; Theron, G; Niesler, T; Theron G. Centre for Tuberculosis Research, South African Medical Research CouncilIn this work, we explore recurrent neural network architectures for tuberculosis (TB) cough classification. In contrast to previous unsuccessful attempts to implement deep architectures in this domain, we show that a basic bidirectional long short-term memory network (BiLSTM) can achieve improved performance. In addition, we show that by performing greedy feature selection in conjunction with a newly-proposed attention-based architecture that learns patient invariant features, substantially better generalisation can be achieved compared to a baseline and other considered architectures. Furthermore, this attention mechanism allows an inspection of the temporal regions of the audio signal considered to be important for classification to be performed. Finally, we develop a neural style transfer technique to infer idealised inputs which can subsequently be analysed. We find distinct differences between the idealised power spectra of TB and non-TB coughs, which provide clues about the origin of the features in the audio signal.Publication National survey in South Africa reveals high tuberculosis prevalence among previously treated people(2022-09) Hesseling, A.C; Marx, F.M; Martinson, N; Theron, G; Cohen, T; Theron G. South African Medical Research Council Centre for Tuberculosis ResearchNo abstract available.Publication Back to the drawing board: Rethinking potential predictors of preschool executive function in low-income South Africa(eScholarship, 2022) Caylee, C; Draper, C; Howard, S; Merkley, R; Scerif, G; Caylee, C Draper C. Health Systems Research. South African Medical Research UnitThis study aimed to explore cross-sectional associations between executive function (EF), and community and household factors (household SES, caregiver education, home learning environment, caregiver/child interaction, caregiver wellbeing, and exposure to community violence) in a sample of children from very low-SES settings in Cape Town, South Africa. Results revealed that children exposed to higher levels of violence perform worse on inhibition tasks. No other associations were significant, highlighting the need to reassess how researchers can better understand these settings and the effects on EF development.Publication Vaccine information, misinformation and hesitancy: what can we learn from the evidence? ‘Research evidence in a post-truth world – tackling the information, misinformation and disinformation affecting healthcare decision making’(2022-11) Cooper, S; Cooper S, Cochrane South Africa. South African Medical Research CouncilNo Abstract Available.