Siboniso - A Community Data Observatory for Antimicrobial Resistance

Siboniso is a digital Community Data Observatory that collects, integrates, and analyses community-based data on antimicrobial resistance (AMR). ‘Siboniso’ - ‘to see’ - captures the design philosophy. Siboniso utilises advanced technologies and innovative approaches to influence patient behaviour and improve public health responses through actionable insights into AMR trends in South Africa.

Dr Isabel Meyer is an Operations Researcher and veterinarian, experienced in systems optimisation, logistics research, ICT4D and multidisciplinary research projects. She is a principal researcher at CSIR's Smart Mobility cluster.

We are solving the problem of limited access to, and use of, community-based information sources to influence AMR in South Africa.

In addition to the global perspective on the AMR burden, South Africa’s high burden of bacterial infectious diseases was found to increase broad spectrum antimicrobial use (Chetty, 2021). 

Effluent from sewage treatment plants perpetuates antimicrobial resistant bacteria, many originating from the regional human population (Larsson et al., 2003). Treatment plants are ineffective in removing antimicrobials, antibiotic-resistant bacteria, and resistant genes, thus impacting aquatic, environmental, and human health. Wastewater sustains horizontal gene transfer and hence acquisition of new antibiotic resistant genes (Hazra and Durso, 2022). 

AMR is perpetuated at the patient level, where AMR awareness and stewardship capacity is limited (e.g., Manderson, 2020; Chetty, 2021), and mechanisms to understand and influence behaviour are underdeveloped (e.g., Farley, 2018). Compliance with the AMR national strategic framework (59.5%) was found to be lowest in community health centers (Chetty, 2021). Monitoring individual patients for AMR is expensive. Community-based approaches, such as wastewater surveillance and behaviour modification, can address these shortcomings and can add to existing efforts and information bases for AMR surveillance and prevention.  

The solution reduces antimicrobial resistance in two ways: first, by providing decision-makers with accurate community-based information to inform policy, strategy, and intervention design and, second, by influencing patient behaviour towards improved antimicrobial stewardship. For the first purpose, government and private sector decision-makers are targeted. Government stakeholders include the National and Provincial Departments of Health, including clinic staff; private sector stakeholders include medical and pharmaceutical professionals and private health insurance companies. The needs of government stakeholders are understood through ongoing engagement via networks, structures, and projects of the project team, while private sector needs will be assessed via targeted key stakeholder interviews. The observatory will provide all stakeholders (public and private sector) with access to surveillance data and risk profiling information, the latter being particularly relevant to clinic staff.

The AMR data observatory supports well-being of the public as follows:

• Publicly accessible, community-specific AMR data and knowledge products for education of the public users

• A platform that can be modularly expanded to reflect AMR use in different communities, thus expanding the footprint to more beneficiaries

• A freely available tool to support effective AMU at patient level, thus improving both patient outcomes and reducing AMR risk.

• Integration across community-level data sources to provide community-specific insights to influence both policy and behaviour, as such enhancing the quality of decision-making

• Postgraduate student support; graduation of postgraduate students

• CPD-accredited training of health practitioners on use of the observatory

• Peer-reviewed journal articles
  

• Technology demonstrator (TRL 6 and above) for submission to Department of Science and Innovation 

In seeking to enhance decision-making and patient outcomes through the collection and provision of AMR-related insights, the observatory will eventually reduce the risk of AMR for all citizens.

The project aims to influence AMR from two perspectives, namely, by enhancing the quality of decision-making with respect to response and intervention design, and by influencing patient behaviour towards responsible use. 

It supports the enablers and strategic objectives of the framework as follows:

Enablers: health systems strengthening, research, education, and communication

An accessible community-based information platform is established to support decision-making in specific communities. The patient-facing app will include awareness creation and education elements to inform end-user AMU competence, and hence influence behaviour and AMR.

Enhanced surveillance

Wastewater surveillance, combined with the platform’s information capability, will enhance the national capacity for context-specific AMR surveillance and efficient responses by enhancing intervention design.

Stewardship

The app will guide patients to improve AMU, and to clinicians on trends in AMU by patients - in both cases with a view on influencing patient-facing engagement with antimicrobials, and hence stewardship. Future developments could include extension of the app to provide information to inform antimicrobial selection options of clinical staff.

Improved patient outcomes

Users will benefit from better health outcomes through improved compliance with prescription guidelines.  

Evidence to support the effectiveness of the various elements were provided in the technology evidence section.

• The platform lends itself to a modular approach (geography, data sources), and is therefore inherently positioned for scaling. Over the project’s 3-year duration, the platform will be modularly expanded to include different communities and integrate more data sources. 

• The data analysis competence of the platform will be modularly scaled as more funding is secured.

• Once established, a communication strategy will be developed to provide training and awareness of the platform amongst the healthcare profession, to ensure their effective contribution to and benefit from the project.

• AMR testing and analysis of wastewater samples will start with the already established sampling system at wastewater treatment plants in Ekurhuleni. Once the effectiveness of the techniques have been established, sampling, testing, and analysis will be extended to other communities.

• The Siboniso lotto, using a smartphone app, will be piloted on a small scale. Challenges will be identified through user feedback and utilised for app enhancement before rolling it out to more communities. Impact will thereafter be upscaled by using various communication channels to enhance adoption and participation.

• Following feedback from the initial project phases, engagements with potential medical insurance partners will be carried out to find ongoing support and further scaling of the app.

Impact of different project aspects will be monitored and evaluated at regular intervals and adjustments made to the project plan where necessary. Some specific impact measures include:

• Platform expansion/growth – track the expansion of the platform for different communities and measure the number of data sources integrated into the data observatory. 

• Decision-making – use a feedback mechanism on the platform to evaluate the impact of the data observatory on decision making and general awareness.

• Platform performance – use metrics that monitor the data processing time, data completeness, and user engagement with the platform. 

• Accuracy - Evaluate the effectiveness of wastewater surveillance in predicting AMR trends; compare the trend analysis of surveillance data with external sources, including the use of mobility data.

• Behavioural change – track the number of participants of the AMR lotto, their technology usage, and the correlation of these with reduced AMR instances.

• Community satisfaction - Conduct surveys amongst AMR lotto users to measure satisfaction and monitor app usage trends over time.

• Quality of research outputs - number and impact of peer-reviewed articles

• Note: the proposed platform is based on application of known approaches or technologies to AMR - pilot performance data is therefore not available.

FINANCIAL

• Initial & sustainable funding 

Public-private partnerships, government funding, and subscription models are considered.

Minimum: Seed funding.

TECHNICAL

Data Integration 

• Integration of diverse data sources; data quality; interoperability.

CSIR’s skills and infrastructure (e.g., Covid Data Observatory); SEPIMOD’s modelling skills. 

Minimum: Robust IT infrastructure, advanced data analytics, interoperability standards.

Wastewater surveillance 

• Routine sample collection disrupted by social unrest and natural disasters; have not been disrupted for more than two weeks over past 3 years. 

Monthly sampling over a 2-3 year period for sufficient data points.

Mobility analysis, trends, risk profiling 

• Capacity of SEPIMOD staff and postgraduate students. 

Minimum: Funding to alleviate constraints.

Behaviour modification

• Appropriate app design and adequate uptake.

CSIR’s ICT4D expertise and private sector partners will design the intervention.

LEGAL

• Privacy and security of patient data for regulatory compliance.

In-house legal expertise and private sector data privacy experts will develop a data governance framework.

Minimum: Compliance with the Protection of Personal Information Act (POPIA).

CULTURAL and EDUCATIONAL

• Barriers to participation by communities and healthcare professionals. 

Minimum: AMR and data sharing awareness campaigns.

INFRASTRUCTURE

• Communication infrastructure may limit data collection and reporting.

The app will be zero-rated to lower entry barriers for clinic patients.

Minimum: Basic digital reporting capabilities.

Funding is sought to expand wastewater-based surveillance into a new area, and to integrate and packaging results for improved decision-making. 

The SEPIMOD team initiated wastewater-based modelling in 2023, with limited funding. Funding will be used to extend sampling and to increase capacity through postgraduate bursaries and appointment of an administrator. 

We believe that participation in the Trinity Challenge will improve visibility of our work, enabling us to collaborators that could assist in enhancing impact. a larger project will enable us to approach private medical insurers to improve impact (through co-development of the app, etc.).

CSIR has experience in observatory development for improved decision-making, and wants to replicate this impact in different areas. The link between information and decision-making is in practice sometimes tenuous, and this project will provide another opportunity to further develop and implement this capability for improved impact.

We believe that, in addition to providing funding, the Trinity Challenge could support this work by connecting the project team to international best practice with respect to a variety of dimensions of this work, as indicated in the section regarding barriers and collaborations, as well as point the team to alternative sources of funding, should this application not be successful.

To initiate & accelerate

Local

• Medical insurer, e.g., Discovery (Vitality programme) - for learning in behavioural modification

• NICD - for collaboration and integration with their existing surveillance efforts

• ERWAT - wastewater sampling

• DoH - awareness creation, use, and future scaling

African

• Africa CDC - for collaboration and integration with existing surveillance efforts

International

• WHO - best practice in international platforms

To scale

• For future development:  Animal-focused AMR information platform initiatives e.g., WOAH Animuse (https://amu.woah.org/amu-system-portal)