Journal:Development and national scale implementation of an open-source electronic laboratory information system (OpenELIS) in Côte d’Ivoire: Sustainability lessons from the first 13 years

Full article title	Development and national scale implementation of an open-source electronic laboratory information system (OpenELIS) in Côte d’Ivoire: Sustainability lessons from the first 13 years
Journal	International Journal of Medical Informatics
Author(s)	He, Yao; Liams-Hauser, Casey; Assoa, Paul H.; Kouabenan, Yves-Rolland; Komena, Pascal; Pongathie, Adama; Kouakou, Alain; Kirn, Mary; Antilla, Jennifer; Rogosin, Carli; Ngatchou, Patricia S.; Kohemun, Natacha; Koffi, Jean B.; Flowers, Jan; Abiola, Nadine; Adjé-Touré, Christiane; Puttkammer, Nancy; Perrone, Luca A.
Author affiliation(s)	University of Washington, I-TECH Côte d’Ivoire, Ministry of Health and Public Hygiene, U.S. Centers for Disease Control and Prevention
Primary contact	lucy dot perrone at ubc dot ca
Year published	2023
Volume and issue	170
Article #	104977
DOI	10.1016/j.ijmedinf.2022.104977
ISSN	1872-8243
Distribution license	Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Website	https://www.sciencedirect.com/science/article/pii/S138650562200291X
Download	https://www.sciencedirect.com/science/article/pii/S138650562200291X/pdfft

Abstract

Purpose: Côte d'Ivoire has a tiered public health laboratory system of nine reference laboratories, 77 laboratories at regional and general hospitals, and 100 laboratories among 1,486 district health centers. Prior to 2009, nearly all of these laboratories used paper registers and reports to collect and report laboratory data to clinicians and national disease monitoring programs.

Project: Since 2009 the Ministry of Health (MOH) in Côte d'Ivoire has sought to implement a comprehensive set of activities aimed at strengthening the laboratory system. One of these activities is the sustainable development, expansion, and technical support of an open-source electronic laboratory information system (LIS) called OpenELIS, with the long-term goal of Ivorian technical support and managerial sustainment of the system. This project has addressed the need for a comprehensive, customizable, low- to no-cost, open-source LIS to serve the public health systems, with initial attention to HIV clients and later expansion to cover the general population. This descriptive case study presents the first published summary of original work which has been ongoing since 2009 in Côte d’Ivoire to transform the information management systems and processes in laboratories nationally.

Impact: OpenELIS is now in use at 106 laboratories across Côte d’Ivoire. This article describes the iterative planning, design, and implementation process of OpenELIS in Côte d'Ivoire, and the evolving leadership, ownership, and capacity of the Ivorian MOH in sustaining the system. This original work synthesizes lessons learned from this 13-year experience towards strengthening LISs in other low-resource settings.

Highlights of this work:

• Factors for scaling and sustaining an electronic LIS (eLIS) in low- and middle-income countries (LMICs) have not been previously described in the literature.
• Successful adoption, scaling, and sustainment of OpenELIS in Côte d'Ivoire relied on early collaboration with partners, regulatory agencies, and technical experts into every step of the design, implementation, and evaluation phases.
• Entities planning to nationally scale an eLIS should plan for 1) workforce development in both end users and system administrators; 2) financial sustainability; and 3) institutionalization of government ownership and technical leadership.
• An open-source eLIS can increase accuracy and timeliness of clinical laboratory data, supporting diagnostic testing and monitoring in a resource-limited setting such as Cote d’Ivoire.

Keywords: electronic information systems, laboratory information systems, quality, low-resource country

Introduction

In low- and middle-income countries (LMICs), inadequate infrastructure—including limited availability and use of electronic laboratory information systems (LISs)—has hindered quality laboratory service delivery for communities. [1] International development efforts such as the U.S. President's Emergency Fund for AIDS Relief (PEPFAR) have aimed to strengthen national laboratory systems by modernizing infrastructure and diagnostic testing methods, including automated analyzers that require computer connectivity for data management. The increase in the quantity and complexity of data generated from high-throughput analyzers and sophisticated diagnostics necessitates the transition from paper records to LIS.

Compared to paper-based systems, LIS are more efficient and enable better quality control (QC) in collecting, processing, managing, synthesizing, and reporting large amounts of data (Table 1). [1–2] The COVID-19 pandemic has highlighted the acute importance of digital health, including LIS, in providing data and enabling rapid data exchange and sharing to facilitate public health surveillance and data-driven decision-making. [3]

Table 1. Strengths of laboratory information systems (LISs) compared to paper-based processes. Lab aspect LIS feature(s) Actions Impact Reception Batch entry; referral workflows Accelerate the processing and recording of incoming work Reduce turnaround time (TAT) Results generation Work plans Help managers monitor workload and supply management Ensure efficient operation of laboratories Analyzer interfaces Relay results quickly and accurately Reduce TAT Reporting Interface with electronic medical record (EMR) Communicate client laboratory history rapidly to clinicians in referral networks Enable faster clinical decision-making regarding diagnosis, treatment, and prognosis; reduce TAT Interface with client notification systems via Short Message Service (SMS) or email Communicate test results to clients rapidly Prompt immediate behavior augmentation and achieve better health outcomes (e.g., initiate self-quarantine in the case of COVID-19). Interface with routine health information systems at the district, regional, and national levels Enable faster aggregate data reporting Facilitate data-driven policy making; reduce TAT Reports of quality indicators such as turnaround time Help managers monitor quality performance Facilitate continuous quality improvement; increase testing quality Overarching Graphic widgets (e.g., drop-down menus, checkboxes), active logic and value checks, flags and reminders for potential issues, and automation Reduce errors Facilitate quality control; increase testing quality Data storage and data query tools Allow secure, long-term data retainment and easy data retrieval based on specific needs Improve client privacy, long-term data availability, and data usage

However, LIS uptake and routine usage in LMICs remains low due to several barriers. LIS ownership, development, and maintenance can be perceived as bearing high costs of financial and human resources. Numerous proprietary LIS exist but may not be ideally suited for LMICs, where financial resources are limited and, most often, ongoing software support relies on long-term service contracts with private companies. Although some LMICs have implemented proprietary systems [4–6], these systems are vulnerable to market changes and can leave users unsupported if a company suspends its operations or modifies its service terms. Health information systems (HIS), including LIS, that use open-source code and are supported by communities of practice (CoPs) offer an attractive alternative in LMICs. Open-source CoPs have a vested interest in capacity building of software developers and users, including those from LMICs. [7] Implementation time and effort aside, there are no licensing fees, and all software code created is in public ownership, which enables others to customize and improve functionality. [8] Several open-source LISs are currently in use in LMICs in Africa and Asia, with varied scale. [9]

Another barrier to sustained, wide-scale usage of LIS is that LIS initiatives are often international-donor-driven and may lack close collaborations with ministries of health (MOHs) in LMICs. Engaging the MOH in the initial design and feature prioritization stage, building local capacity to own and maintain the software, and securing resource commitment after phaseout of donor support are time- and labor-intensive. [10] Additionally, power asymmetries between donors and LMIC stakeholders may discount local priorities, requirements, and innovations. [11] However, successful experiences of implementing an LIS in LMICs at the national scale have shown that strengthening MOH leadership and cultivating mutually beneficial partnerships are essential to impactful and sustainable laboratory system strengthening. [4][12]

Since 2009, the University of Washington’s (UW) International Training and Education Center for Health (I-TECH) has worked in partnership with the Côte d'Ivoire Ministry of Health and Public Hygiene (Ministère de la Santé et de l’Hygiène Publique, or MSHP) to implement a comprehensive set of activities aimed at strengthening the laboratory system. PEPFAR and the U.S. Centers for Disease Control and Prevention (CDC) have funded the partnership. One of the key activities is the sustainable development, expansion, and technical support of an open-source enterprise-level LIS, the OpenELIS system, with the long-term goal of Ivorian technical support and managerial sustainment of the system. The software serves as both an effective laboratory information management solution and a business operations framework for the laboratory service units (Fig. 1). The collaboration has addressed the need for a comprehensive, customizable, low- to no-cost, open-source LIS to serve the public health systems, with initial attention to HIV clients and later expansion to cover the general population.

Figure 1. OpenELIS laboratory software solution and business process framework. (Color should be used in print).

Côte d'Ivoire has a tiered public health laboratory system of nine reference laboratories, 77 laboratories at regional and general hospitals, and 100 laboratories among 1,486 district health centers. Prior to 2009, most laboratories used paper registers and reports to collect and report laboratory data to clinicians and national disease monitoring programs. The only LIS was a bespoke database located at the national HIV reference laboratory (called rETRO-CI), and this system was paired with commercial software.

The objectives of this case study are to: 1) present the iterative implementation process of OpenELIS in Côte d'Ivoire; 2) describe the evolving leadership, ownership, and capacity of the Côte d'Ivoire MSHP in sustaining routine use of OpenELIS; and 3) synthesize lessons learned for strengthening LIS in other LMICs.

Materials and methods

We used qualitative methods in this case study to describe the implementation and collaboration around OpenELIS and its supporting activities. [12–13] We reviewed monthly and annual project reports from the implementers; activity and trip reports from technical experts, as well as from government and donor representatives; and software development roadmap and technical documentation from 2009 to 2020. This project was determined to be non-research by the University of Washington and has been approved by Côte d'Ivoire Comité National d'Ethique des Sciences de la Vie et de la Santé (CNESVS, Ivorian Institutional Review Board; reference number 006–21/MSHP/CNESVS-km) and the U.S. CDC.

Information abstracted from the project documents was initially summarized in chronological order and later coded deductively and inductively. Next, we coded the summaries deductively using the components of the Stages of Continuous Improvement (SOCI) Framework for Health Information Systems. [14] The SOCI Framework has previously been used in Uganda and Cameroon to assess the maturity of digital health tools.[14–16] In addition to deductive codes, we also created new codes inductively where the SOCI Framework did not apply. Inductive codes captured the different steps of OpenELIS implementation and evolving collaboration among stakeholders. When coding was complete, codes and their corresponding excerpts from the summaries were organized into themes. YH coded the summaries. The other co-authors reviewed and verified the coding and analysis to ensure accuracy. The coding and thematic analysis took place in ATLAS.ti 8 Windows.

To ensure that LIS implementation translates into high-quality health services and population health benefits [17], it is important to monitor how data systems mature over time. As such, the SOCI Framework outlines 13 components and 39 subcomponents within five domains, i.e., HIS leadership and governance, management and workforce, infrastructure, standards and interoperability, and data quality and use.[18–19]

Results

Design and development

OpenELIS was initially designed and implemented at two U.S. state public health laboratories in Minnesota and Iowa. [20] Its implementation in global health began as part of a collaborative effort between the U.S. CDC and the Government of Vietnam to strengthen HIV service delivery and laboratory systems in 2005. [12] As part of the open-source medical record system (OpenMRS) consortium of partners [21], UW I-TECH received the OpenELIS source code and joined the development of the OpenELIS codebase in 2009.

An agile software design and development methodology was adopted by UW I-TECH for OpenELIS to enable an inclusive, iterative approach to software creation together with the local stakeholders in Côte d'Ivoire.[22] Since its introduction and adaptation in Côte d'Ivoire in 2009, OpenELIS has evolved through multiple iterations of improving functionality and flexibility, and through new releases twice per year. The first versions of OpenELIS were limited in scope and focused primarily on HIV care. They provided pre-defined forms and reports for common HIV-related tests such as early infant diagnosis of HIV (EID), CD4 counts, and HIV genotyping.

In 2011, UW I-TECH merged all versions of OpenELIS for different laboratories into one core, i.e., OpenELIS Global 1.4, greatly reducing the burden of maintenance and update. Before 2011, we created a customized copy of OpenELIS for each laboratory. This approach was manageable when OpenELIS was implemented in four national laboratories in Côte d'Ivoire and Haiti. However, as OpenELIS expanded to new laboratories, the time burden of maintaining and updating the software across different versions became excessive. The shift to a common code base shortened the software testing cycle for each new version from approximately 200 hours to 40 hours and permitted a more thorough software testing process and a higher-quality final release.

The core software improvements also enabled greater customizability and made it possible for a lab to install and configure OpenELIS without engaging a software developer to customize elements. We added test catalog management features, allowing laboratories to modify catalogs between version releases. Laboratories could make changes to existing tests or add new tests or panels without a software developer adding them manually to the system. New analyzer interfaces were added so that laboratories could import data from analyzers beyond those related to HIV. The core software could be deployed without impacting previous local customizations and further adapted to any setting without code re-write or maintenance. This allowed for countries and laboratories to have more control over what features to include in the software.

In 2018, UW I-TECH began an extensive review of the OpenELIS application and identified where the older technology had become obsolete and presented data security risks. OpenELIS Global 2.0, completed in 2019, upgraded the core framework to mitigate security risks (Table 2).

Table 2. Open-source components in OpenELIS. HL7 = Health Level 7; FHIR = Fast Healthcare Interoperability Resources; ASTM = American Society of Testing and Materials Component OpenELIS Global 1.X series OpenELIS Global 2.X series Timeframe Before 2011 After 2011 Operating system Ubuntu 12 Ubuntu 20 Programming framework Java Struts 1 Java Spring Web application Tomcat Tomcat Database PostgreSQL PostgreSQL Interoperability Mirth, HL7 2.5.1 FHIR R4, HL7 2.5.1, ASTM Customization Hard-coded multiple forks by developers for each use case Core software with flexible customization by users

Many expansions and improvements in later iterations of OpenELIS stemmed from close collaboration between UW I-TECH and the multi-stakeholder OpenELIS technical working group (TWG) in Côte d'Ivoire, convened in 2015. The TWG consists of representatives from the MSHP, I-TECH Côte d'Ivoire, PEPFAR's NGO implementing partners (IPs), and the CDC. UW I-TECH and the design team within the TWG would lead design workshops that start with needs gathering at regional and district hospitals and laboratories using a standardized questionnaire. After reconvening, group members would share the findings on identified needs and gaps in LIS functionality and reporting, prioritizing design features, adding the designs to the roadmap [23], and writing software specifications.

The software development process followed a similar collaborative trajectory as the design process. Before 2014, UW I-TECH conducted all software development, mostly by one senior software developer and joined later by a second developer. All reports, enhancements, and changes relied on the two developers and was dependent on the next release. Starting in 2015, UW I-TECH has provided software development training in Côte d'Ivoire and created a more collaborative workflow with Ivorian developers so that both design and development have become more localized.

Deploy, scale-up, and support

OpenELIS deployment and scale-up took place in several stages since 2009 (Table 3), and the respective roles of the MSHP and I-TECH in the process evolved over time. Before 2015, I-TECH led the deployment process. The deployment at a laboratory began with a team of I-TECH staff from Côte d'Ivoire and Seattle conducting an initial laboratory assessment. Then, during a second visit, the team deployed a customized version. By the end of 2015, OpenELIS had been deployed at 11 laboratories, covering the entire network of laboratories performing viral load testing at the time (Fig. 2).

Table 3. Timeline of the OpenELIS scale-up in Côte d'Ivoire. Year Number of laboratories implementing OpenELIS New Cumulative 2009–2012 3 3 2013–2015 8 11 2016 1 12 2017 31 43 2018 12 55 2019 13 68 2020 28 93 2021 15 111

Figure 2. Laboratories in Côte d'Ivoire that implement OpenELIS before and after 2015. (Color should be used in print).

In 2015, after the MSHP defined and ratified LIS requirements, they determined that OpenELIS satisfied most of these requirements and would be the system deployed at laboratories throughout the country. The MSHP became the owner of OpenELIS in Côte d'Ivoire and started collaborating with I-TECH closely in deployment to roll out OpenELIS. The scale-up first prioritized regional referral laboratories that newly started providing viral load testing as part of a MSHP- and CDC-led initiative to scale-up viral load testing. OpenELIS was then deployed in hospitals at the district and local level to support their collection and management of data on routine laboratory testing for both HIV and non-HIV clients. As of 2021, 111 laboratories in Côte d'Ivoire had implemented OpenELIS (Fig. 2). OpenELIS currently supports testing of hematology, biochemistry, molecular biology, serology, and testing of microbiology, pathology, and immunohistochemistry will be available in January 2023.

The collaborative deployment and scale-up process started with a series of deployment workshops with relevant stakeholders from the OpenELIS TWG in 2015. Participants received hands-on training on setting up servers and workstations, conducting maintenance and updates, and troubleshooting. During the deployment wave each year, starting from 2015, a team of MSHP implementation, I-TECH Côte d'Ivoire staff, and IP representatives would organize a week-long implementation trip to deploy the software and perform the initial training of laboratory staff. The team would follow up with the laboratories by phone weekly and conduct a coaching and review visit pne to three months after the deployment and training.

Starting in 2020, the MSHP began leading the deployment process with minimal support from I-TECH. The MSHP team conducts all follow-up visits, and the training is facilitated by MSHP trainers with co-facilitation by I-TECH Côte d’Ivoire. The MSHP also independently operates the OpenELIS Technical Assistance Unit (Cellule d'Assistance Technique de OpenELIS, or CATOE), the technical support call center for OpenELIS. Established in 2016 within the MSHP, CATOE provides direct, real-time technical support to OpenELIS users. CATOE technicians hold bidirectional call services with laboratories to conduct monthly check-ins that proactively probe for issues and to respond to technical assistance requests. CATOE consists of four MSHP staff members who were recruited from the participants of the first OpenELIS implementers' workshop in 2015. The technicians received targeted training on providing real-time support, equipment, phone credits, and a six-month mentorship with the primary support and development staff at I-TECH. CATOE meets monthly with the OpenELIS TWG to review activities and findings from the support calls, feeding information into the software roadmap.

Strengthening local leadership and ownership

The Côte d'Ivoire MSHP started collaborating with I-TECH in OpenELIS implementation more actively in 2015. The MSHP initiated a reorganization process in 2014 that spurred the creation of the Department of Informatics and Health Information (Direction de l'Informatique et de l'Information Sanitaire, or DIIS) in 2015 to consolidate and coordinate health informatics activities in the country. This designation signaled a leadership commitment to ownership of HIS development, implementation, and management activities nationally. In 2017 the MSHP incorporated the designation of OpenELIS and its scale-up in the National Strategic Plan for the Development of Medical Biology Laboratories 2017–2020. [24]

The continuous strengthening of MSHP leadership demonstrates comprehensive ownership of not only the software and hardware but also the essential knowledge, skills, and processes to sustain routine OpenELIS implementation. Before 2015, the MSHP was only involved in site selection for OpenELIS. After the formation of DIIS, the MSHP joined I-TECH throughout the implementation process from forming the TWG, gathering needs, and training developers and technicians to deploying the software and providing technical support. By 2020, the DIIS was conducting follow-up visits and operating CATOE independently, leading the deployment process, configuration of servers, and training of new and existing users.

Building local capacity

In addition to providing the software and hardware and strengthening local leadership, building local capacity for implementing OpenELIS in Côte d'Ivoire is also critical. Training on various topics along the software development pipeline and in various formats has equipped Ivorian policymakers and technicians with necessary knowledge and skills for routine OpenELIS implementation and maintenance.

Before 2015, UW I-TECH staff initially provided on-the-job training (OJT) on OpenELIS deployment and upgrade to selected staff at the MSHP, national laboratories, and IPs. In 2014, a targeted month-long training on OpenELIS development took place in Seattle for software developers with background in Java from the MSHP, Institut Pasteur Côte d'Ivoire, Institut National Polytechnique Félix Houphouët-Boigny (INPHB; a public polytechnic institute of higher education), and I-TECH Côte d'Ivoire. After participants returned to Côte d'Ivoire, UW I-TECH staff provided remote mentoring and exercises. While the results of the OJT and targeted training were promising, the time and resources for replicating these formats became unsustainably high as the OpenELIS scale-up started in 2015.

In response, UW I-TECH started collaborating with the MSHP in 2015 to develop a full suite of materials for OpenELIS, including a package for trainers and users, and to lead a series of training of trainers (TOT) sessions within Côte d'Ivoire. The week-long TOTs used theoretical and practical activities and took a holistic technical approach covering a range of topics from design, planning, development, and programming to implementation and support. Among the participants at the first TOT, one became the deputy director of DIIS, and most joined the OpenELIS TWG. The TOT participants later conducted cascade training led by the MSHP during the OpenELIS scale-up. Cascade training participants provided largely positive ratings, appreciating the theoretical and practical exercises. The TOTs and cascade trainings have been an integral part of the deployment and scale-up process since 2015, and they have formed a cadre of trained teams in Côte d'Ivoire to ensure sustainable OpenELIS deployment and support.

The collaboration between I-TECH and INPHB incorporated OpenELIS into the computer science curriculum and started the OpenELIS internship at INPHB. In 2016, two INPHB faculty received training on OpenELIS software development from UW I-TECH, and they started using OpenELIS as an example LIS to teach students in computer science. In 2018, the OpenELIS internship program started with three students studying at INPHB and working with I-TECH Côte d'Ivoire simultaneously to practice OpenELIS software development, deployment, and technical support. The training program and assessment process became standardized in 2019, and two interns per year embedded with I-TECH or MSHP teams for one year. Ten interns have graduated from the program, and one was eventually hired as I-TECH Côte d'Ivoire staff.

In addition to periodic training and the internship program, software developer retreats and opportunities to attend and present at international and regional conferences for HIV/AIDS or health informatics have also been part of capacity building.

Notes

This presentation is faithful to the original, with only a few minor changes to presentation. Grammar was cleaned up for smoother reading. In some cases important information was missing from the references, and that information was added. No other changes were made in accordance with the "NoDerivatives" portion of the content license.