Use of trained scent dogs for detection of COVID-19 and evidence of cost-saving.

Background: One of the lessons learned from the coronavirus disease 2019 (COVID-19) pandemic is the importance of early, flexible, and rapidly deployable disease detection methods. Currently, diagnosis of COVID-19 requires the collection of oro/nasopharyngal swabs, nasal turbinate, anterior nares and saliva but as the pandemic continues, disease detection methods that can identify infected individuals earlier and more quickly will be crucial for slowing the spread of the virus. Previous studies have indicated that dogs can be trained to identify volatile organic compounds (VOCs) produced during respiratory infections. We sought to determine whether this approach could be applied for detection of COVID-19 in Rwanda and measured its cost-saving. Methods: Over a period of 5 months, four dogs were trained to detect VOCs in sweat samples collected from human subjects confirmed positive or negative for COVID-19 by reverse transcription polymerase chain reaction (RT-PCR) testing. Dogs were trained using a detection dog training system (DDTS) and in vivo diagnosis. Samples were collected from 5,253 participants using a cotton pad swiped in the underarm to collect sweat samples. Statistical analysis was conducted using R statistical software. Findings: From August to September 2021 during the Delta wave, the sensitivity of the dogs' COVID-19 detection ranged from 75.0 to 89.9% for the lowest- and highest-performing dogs, respectively. Specificity ranged from 96.1 to 98.4%, respectively. In the second phase coinciding with the Omicron wave (January-March 2022), the sensitivity decreased substantially from 36.6 to 41.5%, while specificity remained above 95% for all four dogs. The sensitivity and specificity by any positive sample detected by at least one dog was 83.9, 95% CI: 75.8-90.2 and 94.9%; 95% CI: 93.9-95.8, respectively. The use of scent detection dogs was also found to be cost-saving compared to antigen rapid diagnostic tests, based on a marginal cost of approximately $14,000 USD for testing of the 5,253 samples which makes 2.67 USD per sample. Testing turnaround time was also faster with the scent detection dogs, at 3 h compared to 11 h with routine diagnostic testing. Conclusion: The findings from this study indicate that trained dogs can accurately identify respiratory secretion samples from asymptomatic and symptomatic COVID-19 patients timely and cost-effectively. Our findings recommend further uptake of this approach for COVID-19 detection.

SARS-CoV-2-positive patients display considerable differences in proteome diversity in urine, nasopharyngeal, gargle solution and bronchoalveolar lavage fluid samples.

Proteome profile changes post-severe acute respiratory syndrome coronavirus 2 (post-SARS-CoV-2) infection in different body sites of humans remains an active scientific investigation whose solutions stand a chance of providing more information on what constitutes SARS-CoV-2 pathogenesis. While proteomics has been used to understand SARS-CoV-2 pathogenesis, there are limited data about the status of proteome profile in different human body sites infected by the SARS-CoV-2 virus. To bridge this gap, our study aims to characterize the proteins secreted in urine, bronchoalveolar lavage fluid (BALF), gargle solution, and nasopharyngeal samples and assess the proteome differences in these body samples collected from SARS-CoV-2-positive patients. We downloaded publicly available proteomic data from (https://www.ebi.ac.uk/pride/). The data we downloaded had the following identifiers: (i) PXD019423, n = 3 from Charles Tanford Protein Center in Germany. (ii) IPX0002166000, n = 15 from Beijing Proteome Research Centre, China. (iii) IPX0002429000, n = 5 from Huazhong University of Science and Technology, China, and (iv) PXD022889, n = 18 from Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN 55905 USA. MaxQuant was used for the human peptide spectral matching using human and SARS-CoV-2 proteome database which we downloaded from the UniProt database (access date 13th October 2021). The individuals infected with SARS-CoV-2 viruses displayed a different proteome diversity from the different body sites we investigated. Overally, we identified 1809 proteins across the four sample types we compared. Urine and BALF samples had significantly more abundant SARS-CoV-2 proteins than the other body sites we compared. Urine samples had 257(33.7%) unique proteins, followed by nasopharyngeal with 250(32.8%) unique proteins. Gargle solution and BALF had 38(5%) and 73(9.6%) unique proteins respectively. Urine, gargle solution, nasopharyngeal, and bronchoalveolar lavage fluid samples have different protein diversity in individuals infected with SARS-CoV-2. Moreover, our data also demonstrated that a given body site is characterized by a unique set of proteins in SARS-CoV-2 seropositive individuals.

Leveraging artificial intelligence and data science techniques in harmonizing, sharing, accessing and analyzing SARS-COV-2/COVID-19 data in Rwanda (LAISDAR Project): study design and rationale.

BACKGROUND: Since the outbreak of COVID-19 pandemic in Rwanda, a vast amount of SARS-COV-2/COVID-19-related data have been collected including COVID-19 testing and hospital routine care data. Unfortunately, those data are fragmented in silos with different data structures or formats and cannot be used to improve understanding of the disease, monitor its progress, and generate evidence to guide prevention measures. The objective of this project is to leverage the artificial intelligence (AI) and data science techniques in harmonizing datasets to support Rwandan government needs in monitoring and predicting the COVID-19 burden, including the hospital admissions and overall infection rates. METHODS: The project will gather the existing data including hospital electronic health records (EHRs), the COVID-19 testing data and will link with longitudinal data from community surveys. The open-source tools from Observational Health Data Sciences and Informatics (OHDSI) will be used to harmonize hospital EHRs through the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). The project will also leverage other OHDSI tools for data analytics and network integration, as well as R Studio and Python. The network will include up to 15 health facilities in Rwanda, whose EHR data will be harmonized to OMOP CDM. EXPECTED RESULTS: This study will yield a technical infrastructure where the 15 participating hospitals and health centres will have EHR data in OMOP CDM format on a local Mac Mini ("data node"), together with a set of OHDSI open-source tools. A central server, or portal, will contain a data catalogue of participating sites, as well as the OHDSI tools that are used to define and manage distributed studies. The central server will also integrate the information from the national Covid-19 registry, as well as the results of the community surveys. The ultimate project outcome is the dynamic prediction modelling for COVID-19 pandemic in Rwanda. DISCUSSION: The project is the first on the African continent leveraging AI and implementation of an OMOP CDM based federated data network for data harmonization. Such infrastructure is scalable for other pandemics monitoring, outcomes predictions, and tailored response planning.

Effect of unmanned aerial vehicle (drone) delivery on blood product delivery time and wastage in Rwanda: a retrospective, cross-sectional study and time series analysis.

BACKGROUND: The accessibility of blood and blood products remains challenging in many countries because of the complex supply chain of short lifetime products, timely access, and demand fluctuation at the hospital level. In an effort to improve availability and delivery times, Rwanda launched the use of drones to deliver blood products to remote health facilities. We evaluated the effect of this intervention on blood product delivery times and wastage. METHODS: We studied data from 20 health facilities between Jan 1, 2015, and Dec 31, 2019, in Rwanda. First, we did a cross-sectional comparison of data on emergency delivery times from the drone operator collected between March 17, 2017, and Dec 31, 2019, with two sources of estimated driving times (Regional Centre for Blood Transfusion estimates and Google Maps). Second, we used interrupted time series analysis and monthly administrative data to assess changes in blood product expirations after the commencement of drone deliveries. FINDINGS: Between March 17, 2017, and Dec 31, 2019, 12 733 blood product orders were delivered by drones. 5517 (43%) of 12 733 were emergency orders. The mean drone delivery time was 49·6 min (95% CI 49·1 to 50·2), which was 79 min faster than existing road delivery methods based on estimated driving times (p<0·0001) and 98 min faster based on Google Maps estimates (p<0·0001). The decrease in mean delivery time ranged from 3 min to 211 min depending on the distance to the facility and road quality. We also found a decrease of 7·1 blood unit expirations per month after the start of drone delivery (95% CI -11·8 to -2·4), which translated to a 67% reduction at 12 months. INTERPRETATION: We found that drone delivery led to faster delivery times and less blood component wastage in health facilities. Future studies should investigate if these improvements are cost-effective, and whether drone delivery might be effective for other pharmaceutical and health supplies that cannot be easily stored at remote facilities. FUNDING: Canadian Institutes for Health Research.

The impact of Covid-19 on malaria services in three high endemic districts in Rwanda: a mixed-method study.

BACKGROUND: Rwanda has achieved impressive reductions in malaria morbidity and mortality over the past two decades. However, the disruption of essential services due to the current Covid-19 pandemic can lead to a reversal of these gains in malaria control unless targeted, evidence-based interventions are implemented to mitigate the impact of the pandemic. The extent to which malaria services have been disrupted has not been fully characterized. This study was conducted to assess the impact of Covid-19 on malaria services in Rwanda. METHODS: A mixed-methods study was conducted in three purposively selected districts in Rwanda. The quantitative data included malaria aggregated data reported at the health facility level and the community level. The data included the number of malaria tests, uncomplicated malaria cases, severe malaria cases, and malaria deaths. The qualitative data were collected using focus group discussions with community members and community health workers, as well as in-depth interviews with health care providers and staff working in the malaria programme. Interrupted time series analysis was conducted to compare changes in malaria presentations between the pre-Covid-19 period (January 2019 to February 2020) and Covid-19 period (from March 2020 to November 2020). The constant comparative method was used in qualitative thematic analysis. RESULTS: Compared to the pre-Covid-19 period, there was a monthly reduction in patients tested in health facilities of 4.32 per 1000 population and a monthly increase in patients tested in the community of 2.38 per 1000 population during the Covid-19 period. There was no change in the overall presentation rate for uncomplicated malaria. The was a monthly reduction in the proportion of severe malaria of 5.47 per 100,000 malaria cases. Additionally, although healthcare providers continued to provide malaria services, they were fearful that this would expose them and their families to Covid-19. Covid-19 mitigation measures limited the availability of transportation options for the community to seek care in health facilities and delayed the implementation of some key malaria interventions. The focus on Covid-19-related communication also reduced the amount of health information for other diseases provided to community members. CONCLUSION: The Covid-19 pandemic resulted in patients increasingly seeking care in the community and poses challenges to maintaining delivery of malaria services in Rwanda. Interventions to mitigate these challenges should focus on strengthening programming for the community and home-based care models and integrating malaria messages into Covid-19-related communication. Additionally, implementation of the interrupted interventions should be timed and overlap with the malaria transmission season to mitigate Covid-19 consequences on malaria.

One hundred thirty-three observed COVID-19 deaths in 10 months: unpacking lower than predicted mortality in Rwanda.

The African region was predicted to have worse COVID-19 infection and death rates due to challenging health systems and social determinants of health. However, in the 10 months after its first case, Rwanda recorded 10316 cases and 133 COVID-19-related deaths translating to a case fatality rate (CFR) of 1.3%, which raised the question: why does Rwanda have a low COVID-19 CFR? Here we analysed COVID-19 data and explored possible explanations to better understand the disease burden in the context of Rwanda's infection control strategies.We investigated whether the age distribution plays a role in the observed low CFR in Rwanda by comparing the expected number of deaths for 10-year age bands based on the CFR reported in other countries with the observed number of deaths for each age group. We found that the age-specific CFRs in Rwanda are similar to or, in some older age groups, slightly higher than those in other countries, suggesting that the lower population level CFR reflects the younger age structure in Rwanda, rather than a lower risk of death conditional on age. We also accounted for Rwanda's comprehensive SARS-CoV-2 testing strategies and reliable documentation of COVID-19-related deaths and deduced that these measures may have allowed them to likely identify more asymptomatic or mild cases than other countries and reduced their reported CFR.Overall, the observed low COVID-19 deaths in Rwanda is likely influenced by the combination of effective infection control strategies, reliable identification of cases and reporting of deaths, and the population's young age structure.

A pooled testing strategy for identifying SARS-CoV-2 at low prevalence.

Suppressing infections of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) will probably require the rapid identification and isolation of individuals infected with the virus on an ongoing basis. Reverse-transcription polymerase chain reaction (RT-PCR) tests are accurate but costly, which makes the regular testing of every individual expensive. These costs are a challenge for all countries around the world, but particularly for low-to-middle-income countries. Cost reductions can be achieved by pooling (or combining) subsamples and testing them in groups1-7. A balance must be struck between increasing the group size and retaining test sensitivity, as sample dilution increases the likelihood of false-negative test results for individuals with a low viral load in the sampled region at the time of the test8. Similarly, minimizing the number of tests to reduce costs must be balanced against minimizing the time that testing takes, to reduce the spread of the infection. Here we propose an algorithm for pooling subsamples based on the geometry of a hypercube that, at low prevalence, accurately identifies individuals infected with SARS-CoV-2 in a small number of tests and few rounds of testing. We discuss the optimal group size and explain why, given the highly infectious nature of the disease, largely parallel searches are preferred. We report proof-of-concept experiments in which a positive subsample was detected even when diluted 100-fold with negative subsamples (compared with 30-48-fold dilutions described in previous studies9-11). We quantify the loss of sensitivity due to dilution and discuss how it may be mitigated by the frequent re-testing of groups, for example. With the use of these methods, the cost of mass testing could be reduced by a large factor. At low prevalence, the costs decrease in rough proportion to the prevalence. Field trials of our approach are under way in Rwanda and South Africa. The use of group testing on a massive scale to monitor infection rates closely and continually in a population, along with the rapid and effective isolation of people with SARS-CoV-2 infections, provides a promising pathway towards the long-term control of coronavirus disease 2019 (COVID-19).