21

project be reasonable relative to the fellow’s available time and expertise. The fellow should consider the tradeoff between conducting a study with greater methodological quality (e.g., a prospective design versus a retrospective design), and/or power and one that can be completed more rapidly and reliably. A frustrating research experience can adversely affect a trainee’s long-term interest in research, and generating a publication track record is particularly important early in one’s career, as it can establish one’s potential for success as a researcher and improve competitiveness for employment and grant funding. For a fellow participating in a dedicated research fellowship, as opposed to a clinicallyfocused one, it is critical that he or she devote the majority of his or her efforts towards the completion of a project that is primarily his or hers. In other words, the primary focus should be on a project for which he or she is the unequivocal principal investigator and thus responsible for all phases of the research — definition of the study objectives, obtaining institutional and IRB approvals, data collection, data analysis, manuscript preparation, and interaction with the peer review system. For this primary project, it must be feasible for the fellow to complete the project and draft the primary manuscript within the duration of the fellowship. Taking a single project from conception to completion ensures the fellow understands the entire arc of the research process and, just as importantly, provides a tangible example of the fellow’s intellectual work product. Prospective employers may, appropriately, consider the success and quality of the fellow’s primary project as an indicator of the likelihood of success in subsequent research endeavors. PROJECT EXECUTION After a tentative research question is identified, a detailed timeline should be established. The timeline can help provide a final assessment of the project’s feasibility and ensure that the trainee’s clinical and other educational activities will not be compromised. The timeline will also be useful as a benchmark to gauge progress throughout the study. As explained next, the timeline should include time allocated for manuscript preparation at both the beginning and the end. Before any data are collected or analyzed, the trainee should perform a literature search and write as much of the manuscript as possible. A draft of the introduction, methods, a portion of the discussion, and skeletons of tables and figures can often be produced in advance. In addition to increasing the likelihood of eventual publication, this exercise has three other purposes. First, it ensures that the trainee finds the research question to be engaging enough that his or her interest will be maintained throughout the duration of the project. Second, it demonstrates a level of commitment that justifies a reciprocal degree of commitment from the faculty mentor. Third, and most importantly, it will result in a higher-quality study. By thinking critically from introduction to conclusion, the trainee will develop a better understanding of the project in its entirety, make improvements to the study design, and more efficiently execute the project. For example, after writing the methods section, most likely in collaboration with a statistician, the trainee will be better positioned to collect data accurately and efficiently. Not infrequently, during the statistical CHAPTER 6 — TRAINEES AND EMERGENCY CARE RESEARCH 35 analysis phase of the project, long after the data collection has been completed, it becomes apparent that a crucial variable is either missing from the dataset or coded incorrectly. To protect against this undesirable eventuality, some researchers collect data on many more variables than could possibly be employed in the analysis, which is both inefficient and increases the potential for errors. Creating skeleton tables and figures can help ensure all necessary variables are collected, while helping the trainee to focus data collection efforts on only those data elements that are required. Unfortunately, thorough preparation is no guarantee that a study will be successfully completed. Anticipated and unanticipated setbacks are inevitable and often times beyond the control of investigators. Bench experiments may fail for one reason or another, animals may not survive, eligible subjects might not be reliably identified, or a dataset may turn out to be unavailable. In some cases, an oversight committee, such as an IRB for human subjects research or an animal care and use committee (ACUC) for animal research, may request that a protocol be modified or require additional information. Determining how best to negotiate hurdles is part of the learning process, and an experienced mentor can be extremely helpful in this regard. Successful completion of a study requires substantial persistence and an element of luck. A project that does not result in publication, or at least submission, of a manuscript arguably fails to accomplish the primary objectives of a trainee research experience, which are to acquire research-related skills and, similar to any other research project, to improve the emergency care provided to patients. Of all the skills required to be successful as a researcher, the ability to write well is perhaps the most important. All researchers, regardless of seniority, spend a 

1   2    3    4    5    6   7   8   9   10   11   12   13   14   15   16   17   18   19   20   21   22   23   24   25