Clinical Research during Internal Medicine Residency: A Practical Guide
Article Outline
To the Editor:
The article by Hamann et al in the The American Journal of Medicine1 stimulated our interest. The authors suggested that residents should take advantage of pre-existing research tools (such as clinical databases) rather than starting from scratch. One of the basic requirements for any research is the creation of a reliable and searchable database. Although many hospitals in the United States have an established electronic medical record (EMR) system, the main purpose of these systems seems to be documentation and diagnosis capture. For residents, it is often difficult, using EMRs, to carry out targeted data collection on a prospective basis. On the other hand, residents create daily “sign-out” documents to be used by the team assuming care of their patients. These are usually in the form of Microsoft Word (Microsoft Inc., Redmond, Wash) templates that are normally filled out and printed. A huge amount of data is generated in this way but, because patients’ information is deleted with their discharge, no research benefit is obtained from this effort. Our experience indicates that one easy way to obtain data for resident research is to standardize and store the sign-out information in a permanent retrievable form.
The St. Agnes Residents’ Research Tracking System (STARRTRACKS) is a computer software program that co-ordinates effective patient handover between residents. Designed and developed by the residents in-house, STARRTRACKS automates certain features of patients’ data collection and presents a user-friendly interface that reminds operators of important pieces of information required for inclusion in each sign-out document. Apart from the usual clinical data (chief complaints, code status, diagnosis, allergies), STARRTRACKS also collects information on all devices used on the patient, intravenous fluid type and rates, pain score, applicable prophylaxis, and any other special issues. Each day, STARRTRACKS asks a user on discharge plan for a patient and gives a predischarge checklist of issues to be taken care of 24 hours before an imminent discharge. STARRTRACKS uses the information gathered from the handover activity to generate a sign-out document for the residents and also a nursing communication sheet that comprehensively provides summary of the patient’s medical history, nursing concerns, and discharge planning. Because STARRTRACKS keeps track of this information, it immediately alerts you when a patient is readmitted and imports his/her information from the last sign-out. This substantially saves the resident’s time, ensures consistency in patients’ data, and provides a mechanism for incremental updates on patients’ records during each admission.
Over time, STARRTRACKS has generated a formidable data repository for research, giving residents an easy and effective tool that allows the generation and testing of hypotheses in real time. Queries can be created on specific questions concerning existing data. Additionally, a data collection module can be quickly designed and added to the sign-out routines so that specific information is prospectively collected on each patient.
Although STARRTRACKS was created using Visual Basic (Microsoft Inc, Redmond, Wash) codes, the same objective can be achieved using a spreadsheet program. Database programs offer the additional functionality of interactive forms and data storage.
Within its first year, STARRTRACKS has helped residents submit several scientific abstracts to the regional ACP meeting with several more research projects ongoing. Perhaps the most exciting feature of STARRTRACKS is its ability to track all patients seen by each resident and permit a direct correlation between cases seen by a resident and his/her performance on specific aspects of in-service examination. We eagerly await these results. We would be happy to share our experience with any residency program that is interested in exploring this methodology.
Reference
PII: S0002-9343(06)00389-5
doi:10.1016/j.amjmed.2006.03.020
© 2007 Elsevier Inc. All rights reserved.
