This article detailing a postgraduate medical education research agenda highlights two very distinct realms: upper-income-country (United States, Canada, etc.) international/global emergency medicine (EM) fellowship programs, and low- and middle-income-country postgraduate training. While there is some overlap between these two realms (and therefore of the pressing research questions outlined here), research questions 1, 2, and 3 mostly relate to international/global EM fellowship programs, and research questions 4, 5, and 6 primarily speak to postgraduate training in low- and middle-income countries.
Global emergency medicine (EM) is a rapidly growing field within EM, as evidenced by the increasing number of trainees and clinicians pursuing additional experiences in global health and emergency care. In particular, many trainees now desire opportunities at the postgraduate level by way of global EM fellowship programs. Despite this growing popularity, little is known of the effects of postgraduate training in global health and emergency care on learners and patients in the United States and abroad. During the 2013 Academic Emergency Medicine consensus conference on global health and emergency care, a group of leading educators at the postgraduate medical education level convened to generate a research agenda of pressing questions to be answered in this area. The consensus-based research agenda is presented in this article.
La medicina de urgencias y emergencias (MUE) global es un campo con un rápido crecimiento dentro de la medicina de urgencias y emergencias, como pone de manifiesto el aumento del número de estudiantes y clínicos que buscan experiencias adicionales de atención urgente y salud global. Particularmente, muchos residentes desean ahora oportunidades durante su formación de postgrado a través de los programas de postgrado en MUE global. A pesar de esta creciente popularidad, poco se sabe del impacto de la formación de postgrado en salud global y atención urgente en los residentes y los pacientes en Estados Unidos y en el extranjero. Durante la Conferencia Consenso de la Academic Emergency Medicine de 2013, un grupo de los principales profesores de medicina de postgrado se reunieron para generar un programa de investigación con preguntas prioritarias para ser contestadas en esta área. Este programa de investigación basado en el consenso se presenta en este artículo.
History of and Current Offerings in Global Health Education at Postgraduate Medical Education Level
In 1994, EM was the first specialty to formally create a global health-related fellowship program in the United States. The first international/global EM fellowship program was started at Loma Linda University in 1994, followed by the University of Illinois at Chicago in 1995.[2, 3] In 2005, Baylor College of Medicine began the first pediatric EM global health fellowship program. In 2006, an international pediatric EM fellowship program was started in Canada. Currently, there are at least 34 active international/global EM fellowship programs. These fellowship programs have varying curricula and differ in length and composition. Some offer advanced degrees, some focus on systems development, and still others include extensive field work. In addition, several other global health fellowships have been created by a variety of medical specialties. In total, there are now at least 83 global health-related fellowship programs, including 34 in EM, 14 in family medicine, 11 in internal medicine, 10 in pediatrics, eight interdisciplinary, three in surgery, and three in women's health.
In 2012, in an effort to create a greater sense of community and collaboration among the existing international/global EM fellowship programs, the International Emergency Medicine Fellowship Consortium (IEMFC) was created. The IEMFC maintains a website that offers a uniform application system to facilitate an easier application process for fellowship applicants; the website also serves as a venue for research and educational collaboration across fellowship programs.
Currently, the Accreditation Council for Graduate Medical Education (ACGME) does not provide accreditation for any of these international/global EM fellowship programs. This affords global EM fellows the opportunity to function as independent practitioners of EM, and as such, they can bill for patient care. This has become an important revenue stream for global EM fellowships. While this has advantages, the lack of external accreditation has meant that fellows spend considerable time in U.S. clinical practice—which reduces the time available for education and to do research.
Proposed Society for Academic Emergency Medicine (SAEM) Credentialing Process for Global EM Fellowship Programs
In an effort to raise training standards for global EM, SAEM is currently developing a voluntary credentialing process for interested fellowship programs. To be credentialed, programs will have to meet certain predetermined and standardized criteria. These criteria include academic requirements for fellows and faculty, as well as specific reporting requirements. Fellows in SAEM-credentialed global EM fellowship programs will also have to meet specific milestones. The SAEM Fellowship Credentialing Task Force's global EM work group is in the process of adapting these from the recently launched EM milestones.
Rationale for a Research Agenda
Given the popularity of global health and global EM educational endeavors, it is crucial to outline a research agenda in these areas. We assume, for example, that global health educational experiences result in positive outcomes for global EM fellows and patients (both at the local medical school hospital and at the overseas host site). This, however, may not prove to be the case, and only through a well-laid-out and executed research agenda can this be determined.
The positive or negative findings of research in global health medical education create opportunities to improve these experiences and to share and implement best practices. With this also comes opportunity to enhance specific outcomes for global EM fellows and for patients and populations (local and abroad) under their care.
Perhaps just as important is a global health education–related research agenda that includes determination of effect on global health and global EM educators. For instance, are global health and global EM academicians promoted with equal parity to their non–global health academic counterparts? This article will highlight six pressing research questions related to global health education at the postgraduate medical education level.
Research Question 1: What Is the Impact of Global EM Fellowship Programs?
Research Question 2: What Is the Long-term Impact to Patients (in the United States and Abroad) Under the Care of Global EM Fellows?
Impact of Global EM Fellowship Programs Abroad
As with all training programs, the challenge is to show the efficacy of the educational effort and meaningful impact to the intended, as well as unintended, targets. The effects of global EM training are wide-ranging and often are extremely difficult to measure, especially because one needs to evaluate the program's effect on the fellow's productivity and success and effects both at home and abroad.
The most obvious (and perhaps most important and most difficult to measure) marker of a fellow's or project's effect is patient outcomes. As an example, the literature includes multiple studies that examine the effectiveness of educational interventions on resuscitation in the developing world.[12-26] However, the collection of follow-up or long-term mortality data can pose immense challenges even in the best of settings and can prove almost impossible in many developing world settings. When collection of mortality data is not feasible, an alternative strategy is to use metrics that are more readily obtainable and correlated with outcomes. This allows a simplified assessment that infers change in outcomes. Possible examples include time to triage, time to clinician assessment, time to intravenous fluid administration, time to antibiotic administration, or administration of aspirin for heart-related chest pain.
More subtle effect markers include evaluation of systems development, educational and research infrastructure, and legislative policy change. It is a natural evolution of health research to move from studying effects at the patient level to studying effects at a systems level. This research, however, requires intimate knowledge of the host country along with an invested local research partner. This type of relationship typically takes a significant amount of time to develop, often over the course of years.
Systems research poses additional challenges in that it can be quite subjective. For example, hospital policies on acute care may change, but the resources allocated to effect that change may not. A national institutional review board may be designated in name, but may not meet or may be dysfunctional. In these cases, it is doubtful that meaningful systems change has occurred or that patient outcomes have improved as a result. Evaluation of educational and research infrastructure can be more objective, as the quantitative and qualitative output can be objectively measured. Given the difficulty of directly measuring patient outcomes (as a result of global EM fellowship programs) in the developing world setting, identification and validation of unique surrogate outcome indicators is of great importance and is a fertile area of future research.
Impact of Global EM Fellowship Programs at Home
Global EM training certainly strives to improve health care abroad, but also serves to improve care at home as well. For example, global EM-trained health care providers who serve immigrant populations at home are often better suited to recognize and understand both the physical and the psychological burdens of disease that these populations can have. Moreover, as the world's population becomes even more mobile, with more than 300 million border crossings per year in the United States alone, knowledge of tropical medicine, parasitology, and epidemiology will be paramount to the early detection and treatment of a myriad of communicable and noncommunicable diseases here at home.
Health care providers who are accustomed to working in resource-limited environments have been shown to order fewer tests[31-34] and feel more comfortable in their clinical judgment. Because this most likely represents an association as opposed to causation, this very topic is worthy of further state-of-the-art research. Although not thoroughly studied, this may result in more cost-effective and efficient health care. This demonstrates an example of how the lessons learned working in a resource-limited overseas locale could positively affect the fields of medicine and public health here at home at a time when health care costs are soaring.
Additionally, global EM training imparts skills in a variety of areas, and global EM fellows can be expected to share their skills and knowledge to educate their peers and treat patients at home. The future success of global EM may depend on demonstrating the benefits to local institutions beyond simple altruism.
Research Question 3: What Is the Long-term Impact to Global EM Fellows?
Evaluations of global EM fellowship programs often focus on patient and systems outcomes. However, at their most fundamental level, global EM fellowship programs are designed specifically toward the education of fellows. From this perspective, program success can be measured by analyzing the fellows' success.
In general, global EM fellowship programs stem from academic departments, and therefore traditional measures of academic success can be used, in part, to judge fellows' progress. The number and quality of publications and successful grant applications can be used as indicators of fellow development and achievement. However, not all global EM fellowship programs or projects lend themselves to prompt or prolific publication, nor do they necessarily fit current granting agendas.
The ability to critique published manuscripts and to conduct quality research is another milestone that a global EM fellow should achieve. With each planned intervention (whether educational, systems-based, political, etc.), one must concurrently lay out a plan for outcome assessment.
Last, fellowships are intended to foster careers in global EM. Often, the most powerful interventions occur at the educational and systems level, which typically take years of sustained effort to achieve. Does a global EM fellowship program promote a long-term commitment to academics and the field of global EM as well? Do fellows' efforts overseas remain sustained?
Examples of Methodologies for Measuring Fellows' Outputs, and Outcomes to Fellows and Patients (in the Home Country and Abroad)
Methodologies for Measuring Fellow Outputs and Outcomes
To define the value of fellowship training in global EM, the outputs and outcomes of fellow-level learners and the value of these to the learners themselves to their home institutions and to the international host communities must be identified and measured. To promote interoperability and consistency, best practice metrics for measuring these outputs and outcomes should be shared across all those performing this research. There are little published data on community outcomes for fellow-level global health programs and none at all examining the effect of fellow-level training on international host communities, institutions, and patients.
Fellow-facing Outputs and Outcomes
The majority of the literature evaluates either “soft” learner outputs and outcomes, as assessed by self-report of efficacy, satisfaction, or comfort with skills learned, or “hard” learner outputs and outcomes, as measured by cognitive knowledge, operational performance, or psychomotor skills assessment (including simulated team behavior). A limited additional body of literature examines effects on career choices as both an outcome and an output of fellowship training.
Studies that examine local learners' self-efficacy and individual perception of comfort with skills often fail to correlate these with cognitive or skills assessment and tend to use questionnaires and five-point Likert scale evaluations for measuring self-efficacy and satisfaction.[17, 35] When addressing hard learner outputs like cognitive knowledge change, much of the literature uses pre- and posttraining tests. These tests are frequently taken from the training program curriculum or are modified from existing tests with input from the host institution. In addition to pre- and posttraining tests for determining cognitive knowledge outcomes, many studies use checklists (similar to an objective structure clinical examination) and “mock codes” for determining psychomotor skill acquisition and simulated team behavior.[35-40] Qualitative interview-based methodology is often the primary means used to identify barriers to successful training.
Community-facing Outputs and Outcomes
Few fellow-level training studies look at patient or community outcomes, and so, the best practice methodologies for these assessments have not yet been well defined. Taking from the resuscitation literature again, the use of mixed-methods research allows for an assessment of community outcomes as related to learner knowledge and operational performance, as well as self-efficacy and comfort with new skills. One suggested outcome of fellow-level training at the community or patient level is increased collaboration between different medical specialties, as a part of direct clinical care or educational initiatives supporting international work or in support of research. Survey- and interview-based methodologies to investigate the extent of effects deriving from this type of increased cross-departmental and cross-school scholarly work at both fellows' home institutions and at local host institutions are a fertile potential ground for future research. There are multiple reports of the development of either direct cross-departmental/cross-school work through formal scholarly efforts (such as cross-disciplinary courses attracting applicants from both schools of medicine and public health) and the development of multidisciplinary corps of learners as a part of international service work. However, there is little rigorous analysis of the outcomes that derive from these outputs on communities or patients.
The use of mixed-method strategies could be used to assess changes in attitudes and decision-making at host institutions. Complementing quantitative assessments of cognitive knowledge with interview and focus group–based research may allow for more rigorous analysis of the community outcomes. Objective surrogate measures for care may include use of evidence-based emergency medical decision-making, use of clinical decision rules, clinical practice guidelines, and the integration of features and principles from developed acute care systems.
Another methodology is simple descriptive policy analysis, focusing on changes in policies resulting from fellowship training programs. These may be micro (addressing changes in systems/policies at overseas hospital or regional level) or macro (addressing changes at the national or multinational level). Descriptive methods also lend themselves to detailing potential infrastructural changes that may result from a fellowship training program, including enhanced resources and improved medical information systems for delivery of care, improved funding and academic support for research, and improved structure for EM professional development through educational programs. Alternatively, the effects of training programs related to other aspects of acute care, including process changes (such as the implementation of a triage system, or the introduction of clinical guidelines) may also be measured through the analysis of hospital-level data on admissions and mortality before and after implementation of training.
Transitioning to the overlap between fellow- and community-facing outcomes, evaluation of the effects of global EM training programs on individual patients can examine and measure operational performance via observational checklists as well as by pre- and poststudies.
Research Question 4: What Is the Best Way to Train Emergency Care Providers Around the World?
Global Postgraduate Training in Emergency Care
In most countries, the specialty of EM is new, if it is even recognized as an independent specialty. Many international partnerships have been developed to train emergency care providers, including physicians, nurses, midlevel providers, midwives, and community responders. The training needs among these populations are extremely variable. Some providers are interested in formal postgraduate training programs, such as residency programs, while others have gained hands-on experience working in “emergency department” environments for many years. The latter persons are often not interested in formal, full-time education and training programs, but instead are interested in, and could benefit from, short courses or experiences more appropriate to their educational needs.
The majority of emergency care in some parts of the world is provided by nonphysicians, and many providers in other specialties provide some components of emergency care, such as midwives providing emergency obstetric care. Conducting a formal needs assessment in collaboration with local partners is a valuable tool for defining needs and priorities. It is critical that local partners take an equal role with international partners in priority-setting, making best use of the local capacities and strengths.
Role of a Formal Needs Assessment
A formal needs assessment is an essential early step in development of any program. This process will facilitate understanding of the current state of education and emergency care training, existing infrastructure, and gaps. Needs assessment is a systematic way to study a problem or innovation, enabling researchers to prioritize effectively as well as to make informed decisions about program development.[43, 44] It is crucial that every step of the project, including the needs assessment, is a collaborative effort between local stakeholders and international consultants and researchers. In most cases, the local stakeholders should drive the decision-making and prioritization process. An effective needs assessment will lead to improved program performance in the long run.
There are various strategies that can be employed in conducting a needs assessment, including questionnaires, interviews, focus groups, and environmental scans. Alternative strategies can include data review; a strengths, weaknesses, opportunities, threats (SWOT) analysis; or a Delphi consensus technique to gain different perspectives. While a needs assessment is a first step of any project development, ongoing reevaluation and reassessment is a best practice for long-term programmatic success.
Measuring Effectiveness of Global Emergency Care Training
The Kirkpatrick model for evaluation of educational interventions can be used when assessing educational interventions. By Kirkpatrick's scheme, educational interventions can be evaluated along four levels: level 1 (reaction), level 2 (knowledge), level 3 (behavior), and level 4 (outcomes). It is both more challenging and more desirable to measure at higher levels since the goal of medical education is to improve patient care.
Much of what has been previously described in terms of global EM educational partnerships is the process of partnership establishment and the reaction to those partnerships. More extensive initiatives have gone on to describe program development and the implementation of some evaluation methodology for these programs. For example, an emergency care education and training partnership in Italy was established to train non–EM-trained physicians actively practicing in emergency departments. Evaluation metrics included pre- and posttests, as well as self-assessment.[48, 49] More comprehensive evaluation methodologies were employed in a program developed in postconflict Rwanda. Direct observation of provider-learners was employed as an evaluation technique, in addition to posttesting and self-evaluation. Specific measurable outcomes were noted related to the educational intervention.
Many programs use a “training of trainers” methodology whereby those enrolled in the courses then become the teachers of the course itself or other training programs. This can multiply the effect of one program to reach and educate more practitioners. Clearly, there remains a significant need for further evaluation and measurement of the most effective ways to train emergency care providers globally.
Research Question 5: What Is the Long-term Impact to Non–EM-trained Clinicians Receiving Additional Training in Emergency Care?
Often times, leaders in the global EM community have individual interactions with providers who go on to spearhead the development of the specialty of EM in that country. While it can be difficult to quantify these effects, qualitative methodology could provide a better understanding of this type of effect. Additionally, measurable indicators, such as publications, grant funding, and promotion and tenure decisions, may provide quantitative information as to the effect on specific clinicians as a result of medical education and training initiatives.
Research Question 6: What Is the Long-term Impact to Patients Under the Care of These Clinicians Providing Emergency Care?
Resuscitation training is perhaps the best-studied area in terms of clinical outcomes for emergency care training in the developing world. At least 44 articles have been published on the effects of resuscitation training, including adult, pediatric, newborn, and trauma resuscitation in this part of the world. Fourteen of these studies focused specifically on clinical outcomes, such as mortality, including seven in trauma patients, six in neonates, and two in adult resuscitation studies.
As an example, five of the seven studies on trauma resuscitation training showed significant decreases in patient mortality, with absolute risk reductions in trauma-related mortality ranging from 3% to 33%.[12, 13, 16, 20, 21] These studies included the training of both prehospital and hospital-based providers working in both rural and urban regions of developing countries. Two studies failed to show benefits, although these were limited by questionable training methods in one case and limited data collection in the other.[14, 16]
In attempting to measure the long-term effect to patients under the care of clinicians providing emergency care, several common barriers to effective training exist. These include language barriers, differences in resources and standards of care between developed countries offering the trainings and the developing countries targeted by them, and inadequate time allotted for the courses. Medical educators seeking to measure long-term patient outcomes in low- and middle-income countries should also consider the following in developing emergency care training programs: medical educators researching this topic should tailor individual trainings to the local clinical setting, collaborate with local experts, allot appropriate time in the training for language interpretation and recognition of local cultural norms, and finally, consider equipment needs and maintenance pre- and posttraining. Last, emergency care skills likely wane over time, making refresher trainings (and remeasurement of patient effect) an important component of any emergency care training program.
Examples of Methodologies for Measuring Best Practices for Training Clinicians Providing Emergency Care Internationally
While limited in scope and quality, a number of prior authors have studied the effects of clinician training in emergency care on both learners and patients. One clear point to emerge from this research is that any new emergency care training program should incorporate a system of monitoring and evaluation to ensure that the training has affected practice and outcomes and that the effects are being maintained over time. Research into the effectiveness of emergency care training can also be improved by the establishment of logbooks or Internet-based data collection instruments, to document provider performance and patient outcomes after trainings. In addition, it is important to establish clear definitions of the anticipated outcomes to improve comparisons pre- and posttrainings. Given that knowledge and skills will wane over time, evaluations should continue for at least 3 to 12 months after the training, to better understand exactly when refresher trainings are needed.
Evaluations of emergency care training should also be sure to measure not only improvements in provider knowledge and skills, but also changes in operational practice and patient outcomes (Kirkpatrick level 4). Measuring hard outcomes, such as patient morbidity or mortality, is ideal for demonstrating the effectiveness of an emergency care training program, but this type of data may not be available or appropriate depending on the nature or setting of the training. An alternative strategy is to use quality metrics used by hospitals in the developed world. Examples may include those mentioned above such as time to triage. These process indicators could be used as surrogates for patient outcomes, especially when the process measures have already been closely linked to hard outcomes in prior studies.
In training, and in the research used to study its effectiveness, local government and hospital administration need to be involved from the beginning. This will ensure that the training being conducted matches the resources available and epidemiologic profile of the setting and that the appropriate data can be collected to measure the effects of the training. In addition, local bodies will be better poised to monitor the long-term effects of emergency care trainings well into the future and long after the specific research study has closed.
Finally, when implementing new types of training or an emergency care training in an entirely new setting, an interventional trial that includes a defined control group may be more appropriate than a simple pre- and poststudy, to be certain that the improvements being measured in provider knowledge, skills, practices, or patient outcomes are due to the training itself and not due to other longitudinal factors. The vast majority of educational research in emergency care in the developing world suffers from the time-series bias caused by the fact that patient outcomes are in general improving. While a pre- and poststudy might be appropriate for ensuring that an already well-studied form of emergency care training is delivered appropriately in a specific hospital or community, controlled trials or crossover studies should be used whenever possible to study the effects of a newly developed emergency care training or the implementation of an established emergency care training in a completely new area.
We present a consensus-based research agenda as generated at the 2013 Academic Emergency Medicine consensus conference in Atlanta, Georgia. The research agenda of pressing research questions related to global health education at the postgraduate medical education level includes:
- What is the impact of global EM fellowship programs?
- What is the long-term impact to patients (in the United States and abroad) under the care of global EM fellows?
- What is the long-term impact to global EM fellows?
- What is the best way to train emergency care providers internationally?
- What is the long-term impact to non–EM-trained clinicians receiving additional training in emergency care?
- What is the long-term impact to patients under the care of these clinicians providing emergency care?
The findings of research aimed at answering these and other questions in global health medical education create opportunity to improve these experiences, to share and implement best practices, and to do the ultimate: improve outcomes.
Given the difficulty of directly measuring patient outcomes (as a result of global EM fellowship programs) in the developing world setting, identification and validation of unique surrogate outcome indicators is an important strategy for attempting to answer some of these research questions. Other strategies for answering questions on this consensus-based research agenda include examining and measuring operational performance through observational checklists and pre- and poststudies and employing interventional trials that include a defined control group, allowing for greater certainty that the measured improvements in provider knowledge, skills, practices, and patient outcomes are due to the medical education intervention itself and not due to other longitudinal factors. And finally, no matter the methodology for attempting to answer individual research questions on this agenda (or for any others), conducting a formal needs assessment in collaboration with invested local partners is an invaluable tool for defining priorities and needs.