OpenCAST: Digital solutions for Community AMR Surveillance & Treatment

OpenMRS will be used to capture community and patient-level data on common infections and antimicrobial consumption in Nigeria. Portable real-time DNA/RNA sequencing devices will be used for targeted antimicrobial prescribing. Electronic clinical decision support (eCDS) systems/real-time dashboards will be introduced for better decision-making to improve antimicrobial stewardship and patient outcomes.

Dr Akaninyene Otu. Executive Director Foundation for Healthcare Innovation and Development (FHIND) Nigeria. He has >10 years experience of implementing  electronic health (eHealth) training in low- and middle-income communities (LMICs)

Antimicrobial resistance (AMR) rates are untenably high in Nigeria and AMR awareness is low among communities and healthcare workers (HCW). A survey in Cross River State (CRS) showed that for 56% of children with symptoms of a respiratory tract infection, carers sought treatment in public health facilities, but 37% bought antibiotics over-the-counter without a prescription. The 4 million people in CRS are served by 1,028 publicly-owned health facilities where medical records are currently paper-based. AMR surveillance is hampered by a lack of laboratory capacity, insufficient data collection/analysis.

OpenCAST will introduce OpenMRS in sentinel site health facilities in CRS to capture community and patient-level data on common infections and antimicrobial consumption. Genomics will be used for rapid pathogen identification and focused antimicrobial prescribing. Digital education on AMR will be provided to communities and HCW. 

By demonstrating a viable working solution in CRS we can leverage the already deployed base of OpenMRS in over 1300 health facilities across Nigeria. We have a route to global dissemination through the academic support of University of Leeds (UoL) and Brown University (with technical lead of OpenMRS Ian Bacher, and co-founder Hamish Fraser), and a route to global implementation through open source and the OpenMRS community. 

Our target audience will be communities in CRS which is a coastal state in South Eastern Nigeria spanning 20,156 square kilometres. The state’s healthcare challenges are exacerbated by a shortage of doctors with only 25 doctors employed by the state to work in secondary health facilities (16 general hospitals).  In CRS, ~70% of clients, predominantly peri-urban and rural poor who are often vulnerable, access public facilities, where antibiotics are prescribed by HCW who are inadequately trained  on antimicrobial stewardship (AMS) or AMR. Antibiotics can be purchased without prescriptions from unregulated drug shops and patent medicine vendors following self-diagnosis.

In the OpenCAST project, we aim to provide educational material on AMS for communities and HCW in CRS using digital platforms. This will promote a greater understanding of rational antimicrobial use and ways of reducing AMR. As part of our ongoing THET Partnerships for Antimicrobial Stewardship 2 project (grant ID Cw2 C.12) which is currently in its 7th month in CRS, we have conducted Point Prevalence Surveys on antimicrobial consumption. Additionally, we have conducted community surveys in CRS to better understand the key drivers of antimicrobial mis(use). We are constantly engaging communities in CRS to contribute to AMS strategies to guarantee sustainability.

Impoverished individuals and populations are especially vulnerable to harm from AMR. About 63% of Nigerians (133 million people) are projected to be experiencing poverty based on the 2022 Multidimensional Poverty Index (MPI). In the OpenCAST project, we will develop new features for the OpenMRS system that will be globally accessible under the existing free, open source licences. In line with a key strategy of the WHO’s Global Action Plan to contain AMR, we will prioritise public and HCW enlightenment initiatives using android-facing interfaces. We will introduce OpenMRS in health facilities in CRS to facilitate the longitudinal collection of health information about individual patients and the population including genomics data on AMR. 

OpenMRS has the capacity to improve communication, health worker compliance with best practices, patient safety, quality of care and health outcomes while reducing medical errors. To complement OpenMRS, the android-facing user interface will be made available to the public to provide them with access to their medical records on mobile devices. We project that the benefits of this project will extend beyond AMR for Nigerian communities (since OpenMRS is a full clinic and patient record management system). This feasible and sustainable solution could ultimately be scaled up beyond Nigeria.

OpenCAST will target the underserved/vulnerable populations in CRS who are disproportionately affected by AMR. HCW in CRS use poorly written and grossly inadequate ‘Standing Orders’ for the care of sick children and adults. Antibiotics are prescribed unnecessarily contributing to AMR. We’ll revise the Standing Orders and provide training on rational use of antibiotics/AMR through online continuing professional development (CPD) on smartphones.

 We will collaborate with local stakeholders to co-create, pilot and upscale OpenMRS in all General Hospitals (initially) then primary healthcare centres in CRS. Community-level data on common  infections and antimicrobial consumption will be captured and used to develop a smartphone-based eCDS tool/real-time dashboard to provide evidence-based support to HCW. The prescribing algorithms will initially be guideline-based progressing to machine learning-based once 1+ year of clinical data has been collected. The eCDS tool will provide personalized recommendations for antibiotic prescribing.

FHIND and UoL have previously worked with a Nigerian health informatics company to upscale online CPD training for frontline HCW on Ebola (https://doi-org.ezproxyberklee.flo.org/10.1186/s12960-016-0100-4), COVID (https://doi-org.ezproxyberklee.flo.org/10.1177/20499361211040704), noncommunicable diseases (https://bmjopen-bmj-com.ezproxyberklee.flo.org/content/12/8/e060304.in) and sepsis (https://doi-org.ezproxyberklee.flo.org/10.1177/20499361241233816 ). 

Sustaining impact is an issue in Nigeria and elsewhere in LMICs. FHIND and UoL have a long-term presence through a series of projects in CRS. 

We'll introduce OpenMRS into 2 General Hospitals per district (6 facilities in total) in the first year. Over the next 2 years, OpenMRS will be deployed to the remaining 10 General Hospitals in CRS and some PHCs. 

Five work packages will be aimed at:

1) Producing a tailored version of OpenMRS for CRS and AMR with the promised additional integration (to laboratories, to DNA sequences, to each other and a big database) that will use HL7 FHIR as an integration standard. 

2) Political and health service negotiation to ensure all stakeholders are supportive of a move from paper to OpenMRS and the funding of any additional costs not covered by Trinity Challenge.

3) Deployment of OpenMRS into facilities in CRS - notably travel, training, hardware, software set-up, data take-on and hand-holding.

4) Development of CPD educational material and a smartphone-based eCDS tool/real-time dashboard for HCW. The tool will utilize guideline-based CDSS progressing to machine learning augmented algorithms/CDSS once 1+ year of clinical data has been collected. 

5) Development of a community-facing tool to promote rational use of antibiotics among the public.

A "back office" system will be created to manage role-based access (RBAC) for users, other administration tasks and reporting.

A monitoring/evaluation framework will be used to articulate measurable objectives and define relationships among inputs, activities, outputs, outcomes and impacts

Measurable indicators will include: 

1. Number of health facilities in CRS with OpenMRS.

2. Number of community members in CRS educated on AMR

3. Number of HCW educated on AMS/AMR

4. Number of HCW who have access to the eCDS tool

5. Percentage reduction in antibiotic prescribing in health facilities 

Expected outcomes will be:

1. Strengthened HIS in CRS with improved interoperability and data exchange capabilities. 

2. Enhanced clinical decision-making and AMS practices through use of the eCDS tool.

3. Reduction in unnecessary antibiotic prescribing and AMR rates in CRS. 

4. Empowerment of communities and HCW with knowledge on AMS/AMR via digital platforms.

An observational study will be carried out to measure antibiotic prescribing rates for key diagnoses (1) before implementation of OpenCAST, (2) after the OpenMRS EHR has been implemented and is in regular use, and (3) after implementation of the eCDS component of OpenCAST. The level of prescribing and dispensing will be measured through routine data collected in OpenMRS. As sites will be implementing at different times we'll be able to use difference of difference or step wedge study designs to obtain data to refine our solution. 

Specific barriers to the OpenCAST project include:

1)  Poor digital literacy: We will bring our extensive experience of providing digital training for HCW and communities at-scale to bear for the purpose of tackling AMR in Nigeria.

2)  Lack of commitment by government and stakeholders: Advocacy will be done before and during the OpenCAST project to ensure buy-in from the main actors.

3)  Lack of finances to sustain OpenMRS beyond the Trinity Challenge: We will embed the Open CAST project into the existing realities in the CRS health service while promoting public-private partnerships to promote local ownership and sustainability.

4)  The diverse geography and uneven distribution of resources in CRS could pose obstacles. The Trinity Challenge will facilitate the deployment of digital transformative strategies for AMR to those hardest to reach (the so-called last mile challenge). 

5)  Limited access to stable internet connections and electricity in some parts of CRS. However, OpenMRS provides offline data capture capability, and is widely used in countries with limited electrical power, network connectivity and technical support. 

6)  Lack of regulatory frameworks for EMR: We will advocate for the establishment of clear regulatory guidelines for digital health technologies to foster a conducive environment for innovation

Some barriers to AMS in CRS (and indeed Nigeria) include poor knowledge of AMS/AMR among HCW and communities, indiscriminate and unregulated use of antibiotics and lack of laboratory capacity. OpenMRS is already in use in 1365  facilities in Nigeria and the move to cascade this to PHCs across the country is ongoing. Open MRS can be used in very resource-poor environments and can be modified with the addition of new data items, forms, and reports. Thus extending this to capture community-level data on common infections and antimicrobial consumption, design eCDS systems and train communities and HCW on AMR should be feasible. 

The Trinity Challenge affords us the opportunity to solve the big challenge of AMR in Nigeria by driving innovation in the form of upscaling OpenMRS. This intervention could potentially be replicated across Nigeria and similar contexts thus meaningfully impacting millions of lives.

A major challenge would be to design and deploy simple enough solutions that can be self-sustaining in the long term without donor funding. Also how to get the benefits of advanced technologies without the cost. However, cloud, artificial intelligence, and smartphone integration can all be done very cheaply if the right skillset and approach are adopted.

1. CRS Ministry of Health
2. Federal Ministry of Health Nigeria
3. National Primary Health Care Development Agency (NPHCDA) Nigeria 
4. Nigeria Centre for Disease Control (NCDC) Nigeria
5.  OpenMRS community
6. British Society of Antimicrobial Chemotherapy (BSAC) 
7. Meta

We believe that by working collaboratively with most/all of the above-named organizations, we will successfully initiate, accelerate and scale OpenCAST across the 36 states in Nigeria. Our project in CRS will be a use case to demonstrate that OpenMRS can be established at secondary care facilities in an LMIC setting. We aim to demonstrate the feasibility of obtaining streamlined EMR via OpenMRS and creating a purpose-built inventory management and requirement forecasting for antibiotics and other drugs. We believe that the interoperability in OpenCAST will have major impacts on the delivery and quality of care in CRS and result in gains in efficient use of resources. The learning from this project will enable use refine our approach for scale-up across Nigeria and similar LMIC settings.

 OpenCAST is from the outset designed for scale-up. Therefore the resulting package of care will be feasible within the context and will address local priorities. We have almost 30 years experience of utilising the embedded approach which we will adopt for OpenCAST

https://health-policy-systems....