Physicians use of a computerized inpatient sign-out tool has been shown to reduce the risk of preventable adverse events. The researcher evaluated sign-out software usage at Vanderbilt University Medical Center in order to understand user behavior and identify software enhancements. To accomplish these goals, the researcher created software to record sign-out data, determined descriptive statistics of software utilization, collected feedback from users regarding new software enhancements, and analyzed the content produced by sign-out tool users. Results included the identification of unanticipated software usage by non-providers, different use patterns across hospital units/services, and a variety of discipline-specific sign-out note styles. These results combined with a comparison to the literatures recommendations guided the design specification for new sign-out software. Further study is required to determine relevant outcome measures related to both the sign-out process and the impact of sign-out software.

This Masters Thesis project had as its objectives: (1) to optimize algorithms for solvent-accessible surface area (SASA) approximation to develop an environment free energy knowledge-based potential; and, (2) to assess the knowledge-based environment free energy potentials for de novo protein structure prediction. This project achieved its goals by developing, implementing, optimizing, and evaluating four different algorithms for approximating the SASA of a given protein model and generating knowledge-based potentials for de novo protein structure prediction. The algorithms are entitled Neighbor Count, Neighbor Vector, Artificial Neural Network, and Overlapping Spheres.

Dealing with a cancer diagnosis and cancer treatment involves communication among clinicians, patients, families, friends and others affected by the illness. The hypothesis for this research is that an informatics system can effectively support the communication needs of cancer patients and their formal and informal caregivers. How can an online communication system be made accessible and desirable for all of the different players involved in the patients care and support? The initial design of a novel informatics-based system for cancer communication was conducted in three phases (The Discovering Phase, The Developing Phase and The Testing Phase). Five types of clinical and supportive relationships were identified and supported by in-depth interviews with cancer patients and their informal caregivers. A prototype web-based communication system was created with an emphasis on the interpersonal relationships between patients, families, and clinicians. The system was used over the course of two months by a group of clinicians and head and neck cancer patients in the Vanderbilt Ingram Cancer Center. A case study is presented of one patients detailed feedback and use of the system with the clinicians and with family and friends. A larger study of the systems effect on the clinical and social outcomes will be conducted during the PhD phase of this research. The current phase of this research concludes with the initial user feedback and iterative design of the system prototype.

Tandem mass spectrometry-based shotgun proteomics has become a widespread technology for analyzing complex protein mixtures. A number of database searching algorithms have been developed to assign peptide sequences to tandem mass spectra. Assembling the peptide identifications to proteins, however, is a challenging issue because many peptides are shared among multiple proteins. IDPicker is an open-source protein assembly tool that derives a minimum protein list from peptide identifications filtered to a specified False Discovery Rate. Here, we update IDPicker to increase confident peptide identifications by combining multiple scores produced by database search tools. By segregating peptide identifications for thresholding using both the precursor charge state and the number of tryptic termini, IDPicker retrieves more peptides for protein assembly. The new version is more robust against false positive proteins, especially in searches using multispecies databases, by requiring additional novel peptides in the parsimony process. IDPicker has been designed for incorporation in many identification workflows by the addition of a graphical user interface and the ability to read identifications from the pepXML format. These advances position IDPicker for high peptide discrimination and reliable protein assembly in large-scale proteomics studies. The source code and binaries for the latest version of IDPicker are available from http://fenchurch.mc.vanderbilt.edu/.

Healthcare organizations are often stymied in their efforts to prevent insider attacks that violate patient privacy. Numerous high-profile privacy breaches involving celebrities have brought this deficiency to the public's attention. In response, recent legislation aims to improve this situation by means of regulations and sanctions. While the public and government may demand more privacy safeguards, the current state-of-the-art tools in healthcare security, such as access control and auditing, will still be limited in their ability to solve the issue technically. These technologies are theoretically sound and tested in other industries, yet are suboptimal because no feasible methods exist for generating the policies these systems must act upon, due to the inherent complexities of modern healthcare organizations. To address this shortcoming, we present a novel open-source framework, which mines low-level statistics of how users interact within the organization from the access logs of the organization's information systems. Our framework is scalablecapable of handling real world data integrity issues. We demonstrate the use of our tool by modeling the Vanderbilt University Medical Center. Additionally, we compare our framework's model to traditional experts who would attempt to manually generate a similar model.

Clinical decision support in electronic prescribing (e-Rx) systems can improve patient safety and quality of care. Despite the availability of drug information knowledgebases and decision support modules, users often disable this functionality or customize it to minimize irrelevant or insignificant alerts, due to concerns about alert fatigue, i.e., decreasing the "attention cost" of alerts. We postulate that novel user interfaces may decrease the "attention cost" of alerts, as has been shown in inpatient CPOE. This study explores alternative approaches to display alerts, and examines whether and how human factors based interface design can be used to improve signal detection from noisy data (alerts and reminders) in an existing e-Rx system. The issues in presenting multiple drug alerts in an outpatient e-Rx system are described. Several novel drug alert presentation interfaces are introduced. Both expert evaluation and formal usability testing show that the TreeDashboard-View is better perceived than the text-centric ScrollText-View in delivering multiple drug alerts during e-Rx practice. Physician prescribers' perceptions are discussed.

In health care, a widespread crisis of emergency department crowding has arisen from increasing patient demand and diminishing bed capacity. Predictable fluctuations in patient demand suggest that dynamic resource mobilization may allow for efficient, just-in-time allocation of personnel and beds; this strategy, however, would require a method to forecast near-future crowding. The dissertation presents techniques from queuing theory and discrete event simulation that enable accurate forecasting of emergency department operating conditions. A systematic review of the literature described the causes, effects, and solutions of emergency department crowding, revealing that several measures have been proposed to measure crowding, although none have been validated for the purpose of real-time forecasting. An independent, prospective validation of four previously published crowding measures indicated that three of them accurately discriminate present ambulance diversion, but none of them reliably forecast future ambulance diversion. A discrete event simulation, named ForecastED, was developed to mimic the process of patient flow through the emergency department, such that a single model could forecast many different outcome measures. A prospective evaluation of the ForecastED system demonstrated that the model provides reliable, real-time forecasts of seven different measures of crowding up to eight hours into the future. This technology may provide a foundation for health care providers to coordinate and avoid potentially dangerous crowding situations.

The emergency room is a challenging place to practice medicine. Overcrowding, episodic care, and an interrupt driven environment are all factors contributing to medical errors in the ED . We performed a comprehensive literature review to determine the nature of these errors at time of discharge from the ED. We follow through with a retrospective electronic chart review focusing on potentially inappropriately written drugs in the elderly in the ED. We conclude by presenting an integrated web based patient discharge framework that we custom designed and recently implemented in the Vanderbilt Adult Emergency Department.

The concept of health information exchange (HIE) among organizations in the United States has evolved over the last 25 years, from early efforts to build Community Health Information Networks to recent collaborations in the form of Regional Health Information Organizations (RHIOs). Previous HIE research focused on organizational issues, technological challenges, and impact on healthcare utilization. The research study was set in the MidSouth eHealth Alliance (MSeHA), a RHIO in southwestern Tennessee that has exchanged health data since 2006. This research applied qualitative methods to evaluate the impact of the HIE system on workflow and information flow at MSeHA-participating sites. The study was conducted over a 9-month period at six emergency departments and nine ambulatory clinics. Data collection focused on use of the same HIE technology across widely varying clinical contexts and health information technology (HIT) infrastructures. The researcher observed use of HIE technology and healthcare workflow for 187 hours and conducted semi-structured interviews to confirm observation data. Research results include identification of HIE-related workflow patterns, reasons for HIE technology use, and outcomes of HIE use. The research contributes an understanding of real-world HIE technology usage patterns across organizations, proposes context-aware implementation strategies for HIE systems, and identifies the potential of HIE technology to help transform healthcare.

<p>Answering questions such as Which genes are related to breast cancer? usually requires retrieving relevant publications through the PubMed search engine, reading these publications, and manually creating gene lists. This process is both time-consuming and prone to errors. </p> <p>We report GLAD4U (Gene List Automatically Derived For You), a novel, free web-based gene retrieval and prioritization tool. The quality of gene lists created by GLAD4U for three Gene Ontology terms and three disease terms was assessed using gold standard lists curated in public databases. We also compared the performance of GLAD4U with that of another gene prioritization software, EBIMed. </p> <p>GLAD4U has a high overall recall. Although precision is generally low, its prioritization methods successfully rank truly relevant genes at the top of generated lists to facilitate efficient browsing. GLAD4U is simple to use, and its interface can be found at: http://bioinfo.vanderbilt.edu/glad4u.</p>