Leveraging electronic medical records for HIV testing, care, and treatment programming in Kenya-the national data warehouse project.

BACKGROUND: Aggregate electronic data repositories and population-level cross-sectional surveys play a critical role in HIV programme monitoring and surveillance for data-driven decision-making. However, these data sources have inherent limitations including inability to respond to public health priorities in real-time and to longitudinally follow up clients for ascertainment of long-term outcomes. Electronic medical records (EMRs) have tremendous potential to bridge these gaps when harnessed into a centralised data repository. We describe the evolution of EMRs and the development of a centralised national data warehouse (NDW) repository. Further, we describe the distribution and representativeness of data from the NDW and explore its potential for population-level surveillance of HIV testing, care and treatment in Kenya. MAIN BODY: Health information systems in Kenya have evolved from simple paper records to web-based EMRs with features that support data transmission to the NDW. The NDW design includes four layers: data warehouse application programming interface (DWAPI), central staging, integration service, and data visualization application. The number of health facilities uploading individual-level data to the NDW increased from 666 in 2016 to 1,516 in 2020, covering 41 of 47 counties in Kenya. By the end of 2020, the NDW hosted longitudinal data from 1,928,458 individuals ever started on antiretroviral therapy (ART). In 2020, there were 936,869 individuals who were active on ART in the NDW, compared to 1,219,276 individuals on ART reported in the aggregate-level Kenya Health Information System (KHIS), suggesting 77% coverage. The proportional distribution of individuals on ART by counties in the NDW was consistent with that from KHIS, suggesting representativeness and generalizability at the population level. CONCLUSION: The NDW presents opportunities for individual-level HIV programme monitoring and surveillance because of its longitudinal design and its ability to respond to public health priorities in real-time. A comparison with estimates from KHIS demonstrates that the NDW has high coverage and that the data maybe representative and generalizable at the population-level. The NDW is therefore a unique and complementary resource for HIV programme monitoring and surveillance with potential to strengthen timely data driven decision-making towards HIV epidemic control in Kenya. DATABASE LINK: ( https://dwh.nascop.org/ ).

Lessons learned from early implementation of the Growing Expertise in E-health Knowledge and Skills (GEEKS) program in Nigeria, 2019 - 2021.

Introduction: the Growing Expertise in E-health Knowledge and Skills (GEEKS) program is an applied apprenticeship program that aims to improve informatics capacity at various levels of the national health system and create a sustainable informatics workforce. Nigeria adapted the GEEKS model in 2019 as a mechanism to strengthen data quality and use of routine immunization (RI) and vaccine-preventable disease (VPD) surveillance data among Expanded Programme on Immunization (EPI) staff. Since the start of the GEEKS-EPI program, there has not been a formal assessment conducted to measure the extent to which GEEKS-EPI has been able to build local informatics workforce capacity and strengthen RI and VPD surveillance (VPDS) data quality and use in Nigeria. Methods: we conducted a qualitative assessment to inform the extent to which GEEKS-EPI has been able to build informatics skillsets to enhance local workforce capacity, foster collaboration across government agencies, and create a sustainable informatics workforce in Nigeria. In-Depth Interviews (IDIs) and Focus Group Discussions (FGDs) were held with GEEKS-EPI supervisors, mentors, and mentees from previous GEEKS-EPI cohorts. Results: while there were challenges reported during early implementation of the GEEKS-EPI program in Nigeria, particularly early on in the COVID-19 pandemic, participants and supervisors reported that the fellowship provided a framework for building a sustainable RI and VPDS informatics workforce through regular mentorship, peer-to-peer exchanges and Subject Matter Expert (SME)-led trainings. Conclusion: lessons learned from early implementation of GEEKS-EPI in Nigeria will help to inform its implementation in other countries, where strengthened national RI and VPDS informatics capacity is the primary objective.

Strengthening eHealth Systems to Support Universal Health Coverage in sub-Saharan Africa.

The aim of universal health coverage (UHC) is to ensure that all individuals in a country have access to quality healthcare services and do not suffer financial hardship in using these services. However, progress toward attaining UHC has been slow, particularly in sub-Saharan Africa. The use of information and communication technologies for healthcare, known as eHealth, can facilitate access to quality healthcare at minimal cost. eHealth systems also provide the information needed to monitor progress toward UHC. However, in most countries, eHealth systems are sometimes non-functional and do not serve programmatic purposes. Therefore, it is crucial to implement strategies to strengthen eHealth systems to support UHC. This perspective piece proposes a conceptual framework for strengthening eHealth systems to attain UHC goals and to help guide UHC and eHealth strategy development.

Use of unstructured event-based reports for global infectious disease surveillance.

Free or low-cost sources of unstructured information, such as Internet news and online discussion sites, provide detailed local and near real-time data on disease outbreaks, even in countries that lack traditional public health surveillance. To improve public health surveillance and, ultimately, interventions, we examined 3 primary systems that process event-based outbreak information: Global Public Health Intelligence Network, HealthMap, and EpiSPIDER. Despite similarities among them, these systems are highly complementary because they monitor different data types, rely on varying levels of automation and human analysis, and distribute distinct information. Future development should focus on linking these systems more closely to public health practitioners in the field and establishing collaborative networks for alert verification and dissemination. Such development would further establish event-based monitoring as an invaluable public health resource that provides critical context and an alternative to traditional indicator-based outbreak reporting.

Building workforce capacity for effective use of health information systems: Evaluation of a blended eLearning course in Namibia and Tanzania.

BACKGROUND: Electronic health information systems (HIS) are critical components of national health systems, and have been identified as a key element in the development and strengthening of health systems globally. Novel approaches are needed to effectively and efficiently train health care workers on the use of HIS. One such approach is the use of digital eLearning programs, either alone or blended with face-to-face learning activities. METHODS: We developed a novel blended eLearning course based on an in-person HIS training package previously developed by the United States Centers for Disease Control and Prevention. We then conducted a pilot implementation of the eLearning course in Namibia and Tanzania. RESULTS: The blended eLearning pilot program enrolled 131 people, 72 (55%) from Namibia and 59 (45%) from Tanzania. The majority of enrollees were female (n = 88, 67%) and were nurses (n = 66, 50%). Of the 131 people who participated in the in-person orientation, 95 (73%) completed some or all of the eLearning modules. Across all three modules, the mean score on the post-test was significantly greater than on the pre-test (p < 0.001). When comparing results from previous in-person workshops and the blended eLearning course, we found that participants experienced strong learning gains in both, although learning gains were somewhat greater in the in-person course. Blended eLearning course participants reported good to very good satisfaction with the overall content of the course and with the eLearning modules (3.5 and 3.6 out of 5-point Likert scale). We estimate that the total cost per participant is 2.2-3.4 times greater for the in-person course (estimated cost USD $980) than for the blended eLearning course (estimated cost USD $287-$437). CONCLUSION: A blended eLearning course is an effective method with which to train healthcare workers in the basic features of HIS, and the cost is up to 3.4 times less expensive than for an in-person course with similar content.

A UMLS-based spell checker for natural language processing in vaccine safety.

BACKGROUND: The Institute of Medicine has identified patient safety as a key goal for health care in the United States. Detecting vaccine adverse events is an important public health activity that contributes to patient safety. Reports about adverse events following immunization (AEFI) from surveillance systems contain free-text components that can be analyzed using natural language processing. To extract Unified Medical Language System (UMLS) concepts from free text and classify AEFI reports based on concepts they contain, we first needed to clean the text by expanding abbreviations and shortcuts and correcting spelling errors. Our objective in this paper was to create a UMLS-based spelling error correction tool as a first step in the natural language processing (NLP) pipeline for AEFI reports. METHODS: We developed spell checking algorithms using open source tools. We used de-identified AEFI surveillance reports to create free-text data sets for analysis. After expansion of abbreviated clinical terms and shortcuts, we performed spelling correction in four steps: (1) error detection, (2) word list generation, (3) word list disambiguation and (4) error correction. We then measured the performance of the resulting spell checker by comparing it to manual correction. RESULTS: We used 12,056 words to train the spell checker and tested its performance on 8,131 words. During testing, sensitivity, specificity, and positive predictive value (PPV) for the spell checker were 74% (95% CI: 74-75), 100% (95% CI: 100-100), and 47% (95% CI: 46%-48%), respectively. CONCLUSION: We created a prototype spell checker that can be used to process AEFI reports. We used the UMLS Specialist Lexicon as the primary source of dictionary terms and the WordNet lexicon as a secondary source. We used the UMLS as a domain-specific source of dictionary terms to compare potentially misspelled words in the corpus. The prototype sensitivity was comparable to currently available tools, but the specificity was much superior. The slow processing speed may be improved by trimming it down to the most useful component algorithms. Other investigators may find the methods we developed useful for cleaning text using lexicons specific to their area of interest.

Web-based public health geographic information systems for resources-constrained environment using scalable vector graphics technology: a proof of concept applied to the expanded program on immunization data.

BACKGROUND: Geographic Information Systems (GIS) are powerful communication tools for public health. However, using GIS requires considerable skill and, for this reason, is sometimes limited to experts. Web-based GIS has emerged as a solution to allow a wider audience to have access to geospatial information. Unfortunately the cost of implementing proprietary solutions may be a limiting factor in the adoption of a public health GIS in a resource-constrained environment. Scalable Vector Graphics (SVG) is used to define vector-based graphics for the internet using XML (eXtensible Markup Language); it is an open, platform-independent standard maintained by the World Wide Web Consortium (W3C) since 2003. In this paper, we summarize our methodology and demonstrate the potential of this free and open standard to contribute to the dissemination of Expanded Program on Immunization (EPI) information by providing interactive maps to a wider audience through the Internet. RESULTS: We used SVG to develop a database driven web-based GIS applied to EPI data from three countries of WHO AFRO (World Health Organization - African Region). The system generates interactive district-level country immunization coverage maps and graphs. The approach we describe can be expanded to cover other public health GIS demanding activities, including the design of disease atlases in a resources-constrained environment. CONCLUSION: Our system contributes to accumulating evidence demonstrating the potential of SVG technology to develop web-based public health GIS in resources-constrained settings.

The one laptop per Child (OLPC) computer for health clinics in developing countries.

The 'One Laptop per Child' or XO computer was developed as an education tool for children in developing countries around the world. We wanted to know whether it could also be used to access knowledge sources at the NLM formatted for low bandwidth environments. We report on our experience in using the OLPC computer as a gateway to NLM databases (MEDLINE/PubMed), a virtual slide collection and electronic health records. Based on our experiments, we believe that the OLPC computer can also be a 'One Laptop per Clinic' or a 'One Laptop per Doc' computer.

Assessment of anthrax vaccination data in the Defense Medical Surveillance System, 1998-2004.

PURPOSE: Understanding the completeness and accuracy of U.S. military anthrax vaccination data is important to the design and interpretation of studies to assess the safety of anthrax vaccine. We estimated the agreement between electronically recorded anthrax vaccination data in the Defense Medical Surveillance System (DMSS) versus anthrax vaccination data abstracted from hardcopy medical charts in a representative sample of the U.S. military from 1998 to 2004. METHODS: Medical chart abstractions were conducted at 28 military treatment facilities for 4201 personnel. Abstracted anthrax vaccination data for 1817 personnel, representing 7400 anthrax vaccine doses, were compared with electronically captured data in the DMSS from 1998 to 2004. Sensitivity, positive predictive value (PPV), specificity and negative predictive value (NPV) were calculated using weighted analyses. RESULTS: Weighted person-level analysis revealed DMSS sensitivity = 93.8% (95%CI = 91.1, 95.8), specificity = 87.0% (79.0, 92.3), PPV = 85.6% (77.2, 91.3) and NPV = 94.5% (91.7, 96.4). Report of anthrax vaccination within a +/-7 days window in both medical chart and DMSS electronic data had a sensitivity of 88.3% (85.4, 90.7) and a PPV of 86.6% (84.9, 88.2) in the vaccine dose-level analysis. CONCLUSIONS: These results support that anthrax vaccination data captured by the DMSS are adequate for post-marketing surveillance investigations in the U.S. military and are of comparable quality to data captured by other vaccine safety databases.

Open source Scalable Vector Graphics components for enabling GIS in web-based public health surveillance systems.

Geographic Information Systems (GIS) are useful for visual analysis and sharing of spatial data. Open standards like Scalable Vector Graphics (SVG) provide viable alternatives to overcome traditional GIS limitations in resource-constrained settings (low-bandwidth, cost and manpower barriers). This project describes the design and implementation of reusable SVG components for managing geographic information for web-based public health surveillance systems.

Txt2MEDLINE: text-messaging access to MEDLINE/PubMed.

We developed a text messaging system for processing incoming Short Message Service (SMS) queries, retrieving medical journal citations from MEDLINE/PubMed and sending them back to the user in the text message format. A database of medical terminology abbreviations and acronyms was developed to reduce the size of text in journal citations and abstracts because of the 160-character per message limit of text messages. Queries may be sent as full-length terms or abbreviations. An algorithm transforms the citations into the SMS format. An abbreviated TBL (the bottom-line) summary instead of the full abstract is sent to the mobile device to shorten the resulting text. The system decreases citation size by 77.5+/-7.9%. Txt2MEDLINE provides physicians and healthcare personnel another rapid and convenient method for searching MEDLINE/PubMed through wireless mobile devices. It is accessible from any location worldwide where GSM wireless service is available.

Concept negation in free text components of vaccine safety reports.

Large amounts of information are locked up in free text components of clinical reports. Surveillance systems that monitor adverse events following immunizations (AEFI) can utilize these components after concept extraction using natural language processing (NLP). Specifically, our method for the identification and filtering of negated concepts using the Unified Medical Language System (UMLS) potentially improves the quality of AEFI surveillance systems.

Linking primary care information systems and public health information networks: lessons from the Philippines.

Community-based primary care information systems are one of the building blocks for national health information systems. In the Philippines, after the devolution of health care to local governments, we observed "health information system islands" connected to national vertical programs being implemented in devolved health units. These structures lead to a huge amount of "information work" in the transformation of health information at the community level. This paper describes work done to develop and implement the open-source Community Based Health Information Tracking System (CHITS) Project, which was implemented to address this information management problem and its outcomes. Several lessons learned from the field as well as software development strategies are highlighted in building community level information systems that link to national level health information systems.

Critically Appraised Topics (CAT) peer-to-peer network.

A peer-to-peer network of Critically Appraised Topics or CATs would allow sharing of relevant clinical information regarding specific clinical problems among physicians. This proposed network would enable clinicians to develop and share CATs to other users within the network. This poster describes a proposed implementation of a peer-to-peer internet based sharing of critically appraised topics in the Philippines.