Carlos Otero

Health Informatics in Developing Countries: Going beyond Pilot Practices to Sustainable Implementations: A Review of the Current Challenges

Objectives Information technology is an essential tool to improve patient safety and the quality of care, and to reduce healthcare costs. There is a scarcity of large sustainable implementations in developing countries. The objective of this paper is to review the challenges faced by developing countries to achieve sustainable implementations in health informatics and possible ways to address them. Methods In this non-systematic review of the literature, articles were searched using the keywords medical informatics, developing countries, implementation, and challenges in PubMed, LILACS, CINAHL, Scopus, and EMBASE. The authors, after reading the literature, reached a consensus to classify the challenges into six broad categories. Results The authors describe the problems faced by developing countries arising from the lack of adequate infrastructure and the ways these can be bypassed; the fundamental need to develop nationwide e-Health agendas to achieve sustainable implementations; ways to overcome public uncertainty with respect to privacy and security; the difficulties shared with developed countries in achieving interoperability; the need for a trained workforce in health informatics and existing initiatives for its development; and strategies to achieve regional integration. Conclusions Central to the success of any implementation in health informatics is knowledge of the challenges to be faced. This is even more important in developing countries, where uncertainty and instability are common. The authors hope this article will assist policy makers, healthcare managers, and project leaders to successfully plan their implementations and make them sustainable, avoiding unexpected barriers and making better use of their resources.

Why Patient Centered Care Coordination Is Important in Developing Countries? Contribution of the IMIA Health Informatics for Development Working Group.

UNLABELLED Patient Centered Care Coordination (PCCC) focuses on the patient health care needs. PCCC involves the organization, the patients and their families, that must coordinate resources in order to accomplish the goals of PCCC. In developing countries, where disparities are frequent, PCCC could improve clinical outcomes, costs and patients satisfaction. OBJECTIVE the IMIA working group Health Informatics for Development analyzes the benefits, identifies the barriers and proposes strategies to reach PCCC. METHODS Discussions about PCCC emerged from a brief guide that posed questions about what is PCCC, why consider PCCC important, barriers to grow in this direction and ask about resources considered relevant in the topic. RESULTS PCCC encompasses a broad definition, includes physical, mental, socio-environmental and self care. Even benefits are proved, in developing countries the lack of a comprehensive and integrated healthcare network is one of the main barriers to reach this objective. Working hard to reach strong health policies, focus on patients, and optimizing the use of resources could improve the performance in the devolvement of PCCC programs. International collaboration could bring benefits. We believe information IT, and education in this field will play an important role in PCCC. CONCLUSION PCCC in developing countries has the potential to improve quality of care. Education, IT, policies and cultural issues must be addressed in an international collaborative context in order to reach this goal.

Implementation of a clinical decision support system using a service model: results of a feasibility study.

Numerous studies have shown that the quality of health care is inadequate, and healthcare organizations are increasingly turning to clinical decision support systems (CDSS) to address this problem. In implementing CDSS, a highly promising architectural approach is the use of decision support services. However, there are few reported examples of successful implementations of operational CDSS using this approach. Here, we describe how Hospital Italiano de Buenos Aires evaluated the feasibility of using the SEBASTIAN clinical decision support Web service to implement a CDSS integrated with its electronic medical record system. The feasibility study consisted of three stages: first, end-user acceptability testing of the proposed CDSS through focus groups; second, the design and implementation of the system through integration of SEBASTIAN and the authoring of new rules; and finally, validation of system performance and accuracy. Through this study, we found that it is feasible to implement CDSS using a service-based approach. The CDSS is now under evaluation in a randomized controlled trial. The processes and lessons learned from this initiative are discussed.

User-centered design improves the usability of drug-drug interaction alerts

Clinical Decision Support Systems can alert health professionals about drug interactions when they prescribe medications. The Hospital Italiano de Buenos Aires in Argentina developed an electronic health record with drug-drug interaction alerts, using traditional software engineering techniques and requirements. Despite enhancing the drug-drug interaction knowledge database, the alert override rate of this system was very high. We redesigned the alert system using user-centered design (UCD) and participatory design techniques to enhance the drug-drug interaction alert interface. This paper describes the methodology of our UCD. We used crossover method with realistic, clinical vignettes to compare usability of the standard and new software versions in terms of efficiency, effectiveness, and user satisfaction. Our study showed that, compared to the traditional alert system, the UCD alert system was more efficient (alerts faster resolution), more effective (tasks completed with fewer errors), and more satisfying. These results indicate that UCD techniques that follow ISO 9241-210 can generate more usable alerts than traditional design.

Physicians Perceptions of an Educational Support System Integrated into an Electronic Health Record

The purpose of this study is to determine the perceptions by physicians of an educational system integrated into an electronic health record (EHR). Traditional approaches to continuous medical education (CME) have not shown improvement in patient health care outcomes. Hospital Italiano de Buenos Aires (HIBA) has implemented a system that embeds information pearls into the EHR, providing learning opportunities that are integrated into the patient care process. This study explores the acceptability and general perceptions of the system by physicians when they are in the consulting room. We interviewed 12 physicians after one or two weeks of using this CME system and we performed a thematic analysis of these interviews. The themes that emerged were use and ease of use of the system; value physicians gave to the system; educational impact on physicians; respect for the individual learning styles; content available in the system; and barriers that were present or absent for using the CME system. We found that the integrated CME system developed at HIBA was well accepted and perceived as useful and easy to use. Future work will involve modifications to the system interface, expansion of the content offered and further evaluation.

Health Informatics in Developing Countries: A Review of Unintended Consequences of IT Implementations, as They Affect Patient Safety and Recommendations on How to Address Them

It is a great honor for me to write the president’s statement for 2017 IMIA Year Book on ‘Learning from experience: Secondary use of patient data’ as a special topic. Healthcare providers around the world record millions of patients’ health information in electronic health records (EHRs) every day. Researchers are using data from EHRs to answer their research questions on health services and public health. As the researchers extract knowledge from data collected during routine patient care and turn that knowledge into guidance for healthcare providers to use at the point of care, they help create what Lowrance called “learning from experience”, [1] and the Institute of Medicine called “learning health care systems”[2]. Learning from experience in healthcare is a process with a series of continuous and iterative cycles patient data is collected from clinical encounters, data is aggregated and analyzed, outcomes are evaluated, evidence is used to inform changes in care delivery, and practice is adjusted as needed for continual improvement [2, 3]. Since learning from experience studies real world patient care experiences, it can provide feedback to improve real world patient care experiences. It can enrich the findings of randomized control trials by allowing the study of subgroups of patients that can’t be studied in a controlled trial. It can also help to find patterns of diagnostic accuracy, and assess conformity of practice with guidance by allowing working back from outcomes. Government agencies, health care organizations, and private companies are using data from EHRs and administrative claims to predict the risk of certain disease or adverse events, to conduct comparative effectiveness research, and to develop clinical practice guidelines and decision support tools for practicing clinicians. Research funding agencies are beginning to fund researches, which mine the data drawn from EHRs to advance medical knowledge. With the widespread adoption of health IT, the digital capture of health data, and big data analytics, the opportunities for learning from experience will only increase. To foster learning from experience, a paradigm shift in healthcare is needed with continuous feedback loops between research and practice, and evidence and learning flowing in both directions. Whether these kinds of research involve a clinician using electronic databases to find an answer to his research problem or a data scientist using automated rules to identify patterns and trends in clinical big data, it requires informatics and data science for real time capture of the patient care experience, generation of knowledge, and access to knowledge.Background: Patient safety concerns every healthcare organization. Adoption of Health information technology (HIT) appears to have the potential to address this issue, however unanticipated and undesirable consequences from implementing HIT could lead to new and more complex hazards. This could be particularly problematic in developing countries, where regulations, policies and implementations are few, less standandarized and in some cases almost non-existing. Methods: Based on the available information and our own experience, we conducted a review of unintended consequences of HIT implementations, as they affect patient safety in developing countries. Results: We found that user dependency on the system, alert fatigue, less communications among healthcare actors and workarounds topics should be prioritize. Institution should consider existing knowledge, learn from other experiences and model their implementations to avoid known consequences. We also recommend that they monitor and communicate their own efforts to expand knowledge in the region.BACKGROUND Patient safety concerns every healthcare organization. Adoption of Health information technology (HIT) appears to have the potential to address this issue, however unanticipated and undesirable consequences from implementing HIT could lead to new and more complex hazards. This could be particularly problematic in developing countries, where regulations, policies and implementations are few, less standandarized and in some cases almost non-existing. METHODS Based on the available information and our own experience, we conducted a review of unintended consequences of HIT implementations, as they affect patient safety in developing countries. RESULTS We found that user dependency on the system, alert fatigue, less communications among healthcare actors and workarounds topics should be prioritize. Institution should consider existing knowledge, learn from other experiences and model their implementations to avoid known consequences. We also recommend that they monitor and communicate their own efforts to expand knowledge in the region.

Healthcare Information Systems to Assess Influenza Outbreaks: An analysis of the 2009 H1N1 Epidemic in Buenos Aires

OBJECTIVE To determine whether a private HIS could have detected the influenza epidemic outbreaks earlier through changes in morbidity and mortality patterns. METHODS Data Source included a health information system (HIS) from an academic tertiary health care center integrating administrative and clinical applications. It used a local interface terminology server which provides support through data autocoding of clinical documentation. Specific data subsets were created to compare the burden of influenza during the epidemiological week (EW) 21 to 26 for years 2007 to 2009 among 150,000 Health Maintenance Organization members in Argentina. The threshold for identifying an epidemic was considered met when the weekly influenza-like illness (ILI) rate exceeded 200 per 100,000 visits. Case fatality rates and mortality rates of severe acute respiratory infection (SARI) from 2007 to 2009 were retrospectively compared. Case fatality rates and mortality rates for A/H1N1 influenza 2009 also were estimated. RESULTS The HIS detected the outbreak in EW 23 while the government Ministry of Health (MoH) gave a national epidemic alert during EW 25. The number of visits for ILI increased more than fourfold when comparing 2009 to the period 2007-2008. The SARI mortality rate in 2009 was higher than in 2008 (RR 2.8; 95%CI 1.18-6.63) and similar to that of 2007 (RR 1.05; 95%CI 0.56-1.49). 2009 was the first year with mortalities younger than 65 years attributable to SARI. The estimated A/H1N1 case fatality rate for SARI was 6.2% (95%CI 2.5 to 15.5) and A/H1N1 mortality rate was 6 per 100,000 (95%CI 0 to 11.6). CONCLUSION Our HIS detected the outbreak two weeks before than the MoH gave a national alert. The information system was useful in assessing morbidity and mortality during the 2009 influenza epidemic H1N1 outbreak suggesting that with a private-public integration a more real-time outbreak and disease surveillance system could be implemented.

A Machine Translation Approach for Medical Terms.

We describe the task of translating clinical term descriptions from Spanish to Brazilian Portuguese. We build a statistical machine translation system (SMT) using in-domain parallel corpora and available machine learning tools. The performance of this SMT was compared with general purpose machine translation systems available online. We used different techniques to validate the result of the different systems, using reference domain terminology and the occurrence of translated descriptions in a corpus of medical scientific literature and in domain specific web pages. We also use two sets of 1000 description terms that were revised and checked by a Portuguese speaker. The performance of the SMT we built had very good preliminary results.

Terminology Services: Standard Terminologies to Control Health Vocabulary: Experience at the Hospital Italiano de Buenos Aires

Summary Healthcare Information Systems should capture clinical data in a structured and preferably coded format. This is crucial for data exchange between health information systems, epidemiological analysis, quality and research, clinical decision support systems, administrative functions, among others. Structured data entry is an obstacle for the usability of electronic health record (EHR) applications and their acceptance by physicians who prefer to document patient EHRs using “free text”. Natural language allows for rich expressiveness but at the same time is ambiguous; it has great dependence on context and uses jargon and acronyms. Although much progress has been made in knowledge and natural language processing techniques, the result is not yet satisfactory enough for the use of free text in all dimensions of clinical documentation. In order to address the trade-off between capturing data with free text and at the same time coding data for computer processing, numerous terminological systems for the systematic recording of clinical data have been developed. The purpose of terminology services consists of representing facts that happen in the real world through database management in order to allow for semantic interoperability and computerized applications. These systems interrelate concepts of a particular domain and provide references to related terms with standards codes. In this way, standard terminologies allow the creation of a controlled medical vocabulary, making terminology services a fundamental component for health data management in the healthcare environment. The Hospital Italiano de Buenos Aires has been working in the development of its own terminology server. This work describes its experience in the field.

P1-423 Assessment of influenza outbreaks using a private healthcare information system: an analysis of the 2009 H1N1 epidemic in Buenos Aires

Introduction This study aims to determine if the A/H1N1 influenza outbreak could have been earlier detected through changes in morbidity and mortality patterns analysed from a health information system (HIS). Methods Specific data subsets were created to compare the burden of influenza during the epidemiological week (EW) 21 to 26 for years 2007 to 2009 among 150 000 Health Maintenance Organization members in Buenos Aires. The threshold for identifying an epidemic was considered met when the weekly influenza-like illness (ILI) rate exceeded 200 per 100 visits. Mortality rates of severe acute respiratory infection (SARI) from 2007 to 2009 were compared. Case fatality and mortality rates for A/H1N1 influenza 2009 also were estimated. Results The HIS detected the outbreak in EW 23 while the government Ministry of Health (MoH) gave a national epidemic alert during EW 25. The number of visits for ILI increased more than fourfold when comparing 2009 to the 2007–2008. SARI mortality rate in 2009 was higher than in 2008 (RR 2.8; 95% CI 1.18 to 6.63) and similar to that of 2007 (RR 1.05; 95% CI 0.56 to 1.49). 2009 was the first year with mortalities younger than 65 years attributable to SARI. The estimated A/H1N1 case fatality rate for SARI was 6.2% (95% CI 2.5 to 15.5). The estimated A/H1N1 mortality rate was 6 per 100 000 (95% CI 0 to 11.6). Conclusions the outbreak was detected 2 weeks before than the MoH gave a national alert suggesting that with a private-public integration a more real-time outbreak and disease surveillance system could be implemented.