OpenMRS aims to improve patient-care by offering quality, shelf-ready OpenMRS solutions that provide health care workers with a better user experience and are easy for implementers to re-use and deploy. By fostering collaboration among highly experienced community members from multiple organizations and other global goods communities, this investment will be used to iteratively create a standards-based, OpenMRS Version 3 (O3) ANC MVP package that incorporates the Antenatal Care (ANC) Digital Adaptation Kit (DAK) and can be used with the new OpenMRS 3 Framework.
OpenMRS is a scalable, modular, open source electronic medical records platform (EMR) used by institutions and nations across the globe to build customized medical record systems for use in resource-constrained settings where structured patient record keeping systems can improve health outcomes. Since 2004, OpenMRS has grown to become the point-of-service system used to support care for more than 14.6 million patients in over 6,500 health facilities in 40+ countries to date. It is increasingly recognized as a de-facto EMR standard, supported by the OpenMRS community.
This funding will be used to create a standards-based OpenMRS ANC MVP package using new OpenMRS frontend technology and based on the ANC DAK. Our consortium will a) adapt the ANC DAK for OpenMRS end-users through a specifications analysis and user experience assessment, b) create a base O3 ANC package incorporating the ANC DAK’s core data dictionary, functional, and non-functional requirements, c) extend the O3 ANC package to include the ANC DAK’s decision support component, and d) incorporate ANC DAK-based program indicators and reporting into the ANC package.
Since 2020, the OpenMRS Community’s FHIR Squad and Quality Assurance Team have collaborated with OpenHIE and other partners to develop testable, standards-based data exchange solutions that leverage the OpenHIE architectural specification and testing framework, supporting alignment with the InstantHIE project. During this period, the OpenMRS community has also collaborated with OpenMRS implementers and Open Concept Lab (OCL) to improve OpenMRS dictionary management products. Building on these experiences, our consortium will closely collaborate with the Frontend Squad and the broader OpenMRS community, specifically FHIR Squad, Dictionary Management Squad, QA Support Team, and OpenMRS implementations. We will also foster our connection with the OpenHIE community, OCL, and InstantHIE representatives in order to advance this work.
OpenMRS, Inc. In addition to being an open source EMR, OpenMRS is an open source community that functions as a consortium, with many organizations working in LMICs supporting the work of individual OpenMRS contributors. As such, the community seeks to engage and motivate both volunteers and supporting organizations to actively contribute to all aspects of the software development process.
The OpenMRS Community currently provides community and product management support to the Frontend Squad, which consists of contributors from OpenMRS implementers such as AMPATH, Mekom Solutions, and PIH. In this role, the OpenMRS community will facilitate collaboration between the Frontend Squad, the FHIR Squad, the Dictionary Manager Squad, Quality Assurance Support Team, OCL, and the OpenHIE community.
AMPATH The Academic Model Providing Access to Healthcare (AMPATH) is a 30-year partnership between Moi Teaching and Referral Hospital and Moi University College of Health Sciences in Eldoret, Kenya and a consortium of academic medical centers and universities led by Indiana University. To support the expansion of HIV care and services, AMPATH was one of the first programs in a low-income country to develop an electronic medical records system. ThS, an open-source medical records system developed by medical informatics experts and scientists from the Indiana University School of Medicine and Regenstrief Institute that is now used in over 70 countries worldwide. AMPATH uses OpenMRS to support real-time EMR access to clinicians at over 50 clinic sites, and has developed and piloted additional modules to support non-HIV related care in non-communicable diseases. AMPATH brings its expertise in OpenMRS 3 product leadership, UX design, and development to all three work packages.
Mekom Solutions, an OpenMRS service provider and technology partner of large international humanitarian organisations, supports EMR and HIS projects globally, with a special focus on Bahrain, Cambodia, Haiti, Honduras, Laos and Dominican Republic. Mekom's mission is to deliver digital health solutions virtually anywhere while contributing to open-source software and medical informatics standards. Mekom is a core contributing organisation of OpenMRS, co-leading development of the novel OpenMRS 3.0 with AMPATH Kenya and PIH. Mekom Solutions will primarily contribute their expertise in OpenMRS 3 architecture and development activities in all three work packages.
Partners In Health is a nonprofit social justice organization that brings the benefits of modern medical science to the poorest and sickest communities around the world. Since its founding in Haiti in 1987, PIH has been at the forefront of health interventions in the most marginalized settings around the world, including Rwanda, Lesotho, Haiti, Sierra Leone and Liberia. PIH has been a leader in EMR development since 2001, co-founding OpenMRS in 2004. PIH continues to lead development of OpenMRS 2.0 (Reference Application) and currently co-leads development for PIH-supported health facilities and the novel OpenMRS 3.0 with MEKOM and AMPATH Kenya. The PIH EMR is currently deployed in 5 countries and includes ANC and Women’s Health capabilities. PIH’s contributions to the three proposed work packages will focus on business analysis, requirements gathering, UX design, and user acceptance testing.
Background
OpenMRS is an approved global good used to manage 14.6 million patients’ health data in over 6,500 health facilities located in more than 40 low and middle income countries (LMIC). Although the OpenMRS Platform and core modules are widely used by OpenMRS implementers, many have created customized frontend features designed to improve service delivery to patients. These siloed efforts often duplicate development efforts, limit standards-based interoperability with other health systems, and are difficult for other implementers to use.
In 2019, the OpenMRS Community supported the establishment of an OpenMRS Frontend Squad to address this problem. This squad spearheaded an initiative to update the OpenMRS Reference Application frontend to more modern technologies that promote greater collaboration and re-use, driving down duplication of development effort. For instance, it is highly likely that multiple OpenMRS implementers have ANC solutions that are unique to their OpenMRS distributions. The new OpenMRS 3.0 Framework makes it easier for developers to share content and frontend modules by creating shelf-ready “OpenMRS packages,” such as a package of base EMR features that can be used with packages of program-specific content, functionality, and features (ie: HIV, MCH, MDR-TB).
Current Frontend Squad efforts focus on growing the marketplace of OpenMRS frontend packages, such as base EMR, HIV, COVID-19, mental health, physical rehabilitation and non-communicable diseases (NCD) packages. With a growing number of organizations expressing interest in designing and developing packages to support ante-natal care (ANC), this DAK-based O3 ANC MVP package will be a valuable contribution to this marketplace and the wider OpenMRS implementer community.
Objectives
This investment will allow us to iteratively develop a shelf-ready, O3 ANC MVP package that implements priority components of the ANC DAK, adapted to reflect end-user needs and preferences. By creating an O3 ANC MVP package using designs informed by the ANC DAK, OpenMRS will be able to serve as a leader and advocate for adapting and integrating DAKs into shared OpenMRS 3.x products.
Our consortium will apply a proven design-driven software development process that is highly collaborative, user centered, standards based, and promotes reuse. OpenMRS will leverage its Fellowship Program to expand the capacity of promising mid-level technical professionals to work with modern OpenMRS technologies within the open source community environment and with OpenMRS implementers at the country-level.
Our consortium will achieve our goal through three work packages. The initial work package will create a base OpenMRS ANC MVP package that meets prioritized ANC DAK core data dictionary, functional, and non-functional requirements. Work package 2 will expand on this work to incorporate the decision support component of the ANC DAK. Work Package 3 will similarly build off of the O3 ANC Core Package, addressing the Program Indicator component. While Work Packages 2 and 3 each rely on Work Package 1 deliverables, Work Package 2 and Work Package 3 are not dependent.
Deliverables and Schedule
|
Deliverable |
Timeframe |
|
Work Package 1: O3 ANC MVP Package: Core Dictionary, Functional, and Non-Functional Objective: Coordinate design and development of an O3 ANC MVP package that meets ANC DAK core dictionary, functional, and non-functional requirements. |
|
|
Prioritized core dictionary, functional, and non-functional requirements based on a specifications analysis |
Month 2 |
|
User experience assessment |
Month 1-2 |
|
Agreed upon ANC dictionary concepts and ANC DAK source available in OCL with core data elements mapped to standard terminology codes (e.g. ICD) |
Month 3 |
|
User experience final designs developed |
Month 4 |
|
Developed and tested frontend UI, missing or inadequately implemented |
Months 4-5 |
|
Work Package 2: O3 ANC MVP Package: Decision Support Objective: Coordinate design and development of an O3 ANC MVP package for the decision support component of the ANC DAK |
|
|
Prioritized decision support requirements based on a specifications analysis |
Month 4 |
|
User experience assessment |
Months 4-5 |
|
User experience final designs developed |
Month 6 |
|
Developed and tested frontend UI, missing or inadequately implemented APIs |
Months 7-8 |
|
FHIR 2 module improvements released |
Month 9 |
|
Work Package 3: O3 ANC MVP Package: Program Indicators Objective: Coordinate design and development of an O3 ANC MVP package for prioritized program indicator components of the ANC DAK |
|
|
Prioritized program indicator requirements |
Month 5 |
|
User experience assessment |
Months 6-7 |
|
User experience final designs developed |
Month 8 |
|
Build the capability for reading FHIR Measure resources for prioritized ANC indicators |
Months 9-10 |
|
Developed and tested APIs and frontend UI based on final designs |
Month 11 |
Risk Mitigation
Low risk of community resistance to adoption of a DAK-based O3 ANC package. This is mitigated through a design-driven development process, routine community showcases, and open community meetings to encourage early buy-in.
Comments
Digital Square comments
Thank you for the application and the detailed schedule, in particular. For the full technical application, in addition to general recommendations per the email, please expand more on:
-How the consortium's roles and responsibilities will be broken out.
-Plans for engaging users/the user experience assessments as this relates to adapting DAK components and associated learnings from this.
For the planned decision support and indicator work, it would be helpful to see more connecting the work to the ANC FHIR implementation guide with identification of potential dependencies or gaps. This can include the current level of support for FHIR in OpenMRS and what known FHIR module improvements this work would allow for.
Very good description of the
Very good description of the task. It will be helpful to add:
1. What standards are being proposed to be used
2. Will this be tested in a country, which one?
Machine learning used for clinical decision support in ANC DAK?
Hi,
I'm a postdoc at Columbia University Irving Medical Center working with the OBGYN department here on maternal health.
I have built clinical decision support tools that use machine learning to help clinicians stratify patient risk during preconception visits.
For example, one of the models I have built can accurately predict who is at risk of a severe maternal morbidity event, and this can affect clinican decision-making. An example model I've built that's used at medical centers is https://arxiv.org/abs/1904.05342
Do you have a reference for which CDS tools will be built for ANC DAK?
I'd be happy to give technical input on the development of CDS tools in the long-term, as I believe AI and machine learning are the future of such tools, and can accurately guide clinical decision-making. If this may be of interest, my email is ja3014@cumc.columbia.edu.
Thanks!
Jaan
Jaan Altosaar, PhD
Columbia University Irving Medical Center
NewYork-Presbyterian Hospital
622 W 168th St PH-20 WS-10