With exponential growth on a daily basis, there is potentially valuable information hidden in complex electronic medical records (EMR) systems. In this thesis, several efficient data mining algorithms were explored to discover hidden knowledge in insurance claims data. The first aim was to cluster three levels of information overload(IO) groups among chronic rheumatic disease (CRD) patient groups based on their clinical events extracted from insurance claims data. The second aim was to discover hidden patterns using three renowned pattern mining algorithms: Apriori, frequent pattern growth(FP-Growth), and sequential pattern discovery using equivalence classes(SPADE). The SPADE algorithm was found to be the most efficient method for the dataset used. Finally, a prototype system named myDietPHIL was developed to manage clinical events for CRD patients’ and visualize the relationships of frequent clinical events. The system has been tested and visualization of relationships could facilitate patient education.

Hydropathic molecular modeling is a computer-aided molecular design technique for obtaining, representing, and understanding the properties and interactions of biomacromolecular complexes in the biological environment. Hydropathic INTeraction (HINT) is a novel empirical force field to calculate the free energy of intermolecular interaction based on experimentally determined partition coefficients (log Po/w). It includes all the expected interactions between molecules such as hydrogen bonding, hydrophobic, electrostatic, acid-base, and Coulombic interactions, entropy, solvation and others. HINT tools were used to determine, evaluate, and analyze protein-ligand interactions in different research projects: 1) We used these tools to discover small molecule inhibitors of PsaA, a potential target for Streptococcus pneumoniae. We screened and scored potential molecules to obtain hits. After the growth conditions for both the wild type and PsaA mutant of S. pneumoniae were optimized, we then tested our hits. A few compounds passed through the three-stage assay protocol and confirmed the inhibition of PsaA with MICs between 125-250 μM. 2) The SAR of C-3 and C-5 pyrrole-based antitubulin agents at the colchicine-binding site with explicitly solvated models was performed. After docking with GOLD at the colchicine site, post-docking scoring and evaluation were performed with HINT. The total HINT score correlates with binding and activity; similarly, the significance of individual functional groups, protein residues and interactions amongst a collection of compounds can be quantitated. The possibility of water-mediated interactions in a way solvent accessible part of the pocket was considered by subjecting molecular models to MD simulations. Several water molecules were identified to be contributing to the binding and were confirmed by HINT scoring. Finally, using hydropathic molecular modeling tools helped us to understand, evaluate, analyze, and improve protein-ligand interactions in different biological systems.

The aim of this study is to examine the impact of a cloud-based CDS alerting system for SIRS, a precursor to sepsis, and sepsis itself, on adult patient and process outcomes at VCU Health System. The two main hypotheses are: 1) the implementation of cloud-based SIRS and sepsis alerts will lead to lower sepsis-related mortality and lower average length of stay, and 2) the implementation of cloud-based SIRS and sepsis alerts will lead to more frequent ordering of the Sepsis PowerPlan and more recording of sepsis diagnoses. To measure these outcomes, a pre-post study was conducted. A pre-implementation group diagnosed with sepsis within the year leading up to the alert intervention consisted of 1,551 unique inpatient visits, and the three-year post-implementation sample size was 9,711 visits, for a total cohort of 11,262 visits. Logistic regression and multiple linear regression were used to test the hypotheses. Study results showed that sepsis-related mortality was slightly higher after the implementation of SIRS alerts, but the presence of sepsis alerts did not have a significant relationship to mortality. The average length of stay and the total number of recorded sepsis diagnoses were higher after the implementation of both SIRS and sepsis alerts, while ordering of the Sepsis Initial Resuscitation PowerPlan was lower. There is preliminary evidence from this study that more sepsis diagnoses are made as a result of alert adoption, suggesting that clinicians can consider the implementation of these alerts in order to capture a higher number of sepsis diagnoses.

The adoption rate of health information technology (HIT) remains low in developing countries, where healthcare institutions experience high operating costs and loss of revenue, which are related to systems and processes inefficiency. The purpose of this case study was to explore strategies leaders in Zimbabwe used to implement HIT. The conceptual framework of the study was Davis's technology acceptance model (TAM). Data were gathered through observations, review of organizational documents (i.e., policies, procedures, and guidelines), and in-depth interviews with a purposive sample of 10 healthcare leaders and end-users from hospitals in Zimbabwe who had successfully implemented HIT. Transcribed interview data were coded and analyzed for emerging themes. Implementation strategies, overcoming barriers to adoption, and user acceptance emerged as the themes most healthcare leaders associated with successful HIT projects. Several subthemes also emerged, including: (a) the importance of stakeholder involvement, (b) the importance of management buy-in, and (c) the low level of IT literacy among healthcare workers. The strategies identified in this study may provide a foundation on which healthcare leaders in developing countries can successfully adopt and implement HIT. The recommendations from this study could lead to positive social change by providing leaders with knowledge and skills to use information technology strategies to deliver better healthcare at lower costs while creating employment for local communities.

Currently, C Hospital lacks a standardized nursing informatics competency program to validate nurses' skills and knowledge in using electronic medical records (EMRs). At the study locale, the organization is about to embark on the implementation of a new, more comprehensive EMR system. All departments will be required to use the new EMR, unlike the current policy that allows some areas to still document on paper. The Institute of Medicine, National League of Nursing, and American Association of Colleges of Nursing support and recommend that information technology be an essential core competency for nurses. Evidence of the need for nursing informatic competencies was found through a literature search using CINHAL, Proquest Nursing, Medline, and Pubmed search lines. Concepts searched were competencies, nursing informatics, health information technology, electronic health record, information technology literacy, nursing education, information technology training, and curriculum. The Staggers Nursing Computer Experience Questionnaire was distributed to 300 nurses practicing within the hospital setting to obtain baseline data on current nursing computer knowledge and skill level. This validated tool was created by Nancy Staggers in 1994 and used in other process improvement efforts similar to this one. The assumption was that nursing competency levels with computers were varied through the hospital. The data obtained from the questionnaire, through Zoho Survey tool, confirmed this assumption and were used to help create the education, support, and competency plan for the future. Data was analyzed through the built-in reports and interactive charts that the Zoho survey tool provides. The new EMR and all the new processes that come with it will be the framework of nursing care. Having competent nurses in the use of the EMR will optimize the quality of patient care delivered.

Despite defibrillation as the only effective treatment for sudden cardiac arrest (SCA), less than 15% of homes and public facilities have access to an automated external defibrillator (AED). In the United States, ineffective response to SCA cases occurring each year classifies it as a business problem for medical device manufacturing leaders, emergency responders, and bystanders. The purpose of this multicase study was to explore the marketing strategies AED manufacturing leaders use to expand their consumer customer base. Data were collected via in-depth interviews with a purposive sample of participants from 2 U.S. AED manufacturers on the east coast, 2 AED distributors, and 2 healthcare corporations in Texas, as well as a review of company materials. The framework for this study was product life cycle theory. Initial findings for expanding the U.S. AED market indicated that the market was not led by its manufacturers but by its distributors. This finding became an important theme noted from AED manufacturers in considering the consumer segment, an aftermarket from commercial marketing strategies. A common concern for the security of strategic marketing was evident across the AED manufacturer participants with reluctance to discuss business models and marketing plans. A congruent theme was the curtailment of open discussions regarding AED marketing strategies because of security and confidentiality risk. Also, limited number of approved AED manufacturers by the Federal Drug Administration minimizes AED access. Residual outcomes include improving the quality of life for the aging population while reducing the loss of the lives and costs of healthcare. Social implications include preventing sudden cardiac death by providing more accessibility of AEDs to baby boomers.

Over 700,000 adverse drug events (ADEs) result in emergency hospital visits annually, and many of these ADEs are preventable through the use of health information technology in hospitals. However, only 12.6% of U.S. hospitals have developed the capacity to adopt closed-loop electronic medical records (EMR). Organizational complexity may be a major factor influencing hospitals' adoption of closed-loop EMR. This quantitative study explored how organizational complexity influenced hospitals' adoption of closed-loop EMR. Diffusion of innovation theory was the foundation for this study. Logistic regression was used to establish possible relationships between organizational complexity and hospitals' adoption of EMR for closed-loop medication therapy management. Secondary data from Health Information and Management Systems Society were examined to explore the relationship between organization complexity and hospitals' adoption of EMR for closed-loop medication therapy. The research questions explored whether vendor selection strategy, structural complexity, and management structure influence hospitals' adoption of EMR for closed-loop medication therapy management. The results indicated that all three variables, vendor selection strategy, structural complexity, and management structure, are statistically significant predictors of hospitals' adoption of EMR for closed-loop medication therapy management. Results from this study may promote positive social change by enhancing hospitals' adoption of EMR for closed-loop medication therapy management, which may therefore help improve the quality, efficiency, and safety of health care delivery in U.S. hospitals.

The purpose of this project was to evaluate the effectiveness of a practice-based dermoscopy training program for dermatology healthcare providers in order to improve their technique of performing clinical skin exams for the early detection of melanomas. The overall incidence of melanoma continues to rise. More than 75% of all skin cancer deaths are from melanoma. Advanced melanoma spreads to lymph nodes and internal organs and can result in death. One American dies from melanoma almost every hour (American Cancer Society [ACS], 2009). Early diagnosis and excision are essential to reduce morbidity and to improve patient survival. This one-group before-and-after study design utilized a convenience sample of three dermatology healthcare providers (DHPs). The primary investigator conducted a retrospective review of the pathology logs for each provider. The time frame for the review was a three-month period in 2010, which represented the same time frame that the study was conducted in 2011. The DHPs participated in a four-hour training workshop that included pattern analysis recognition using dermoscopy. Following the workshop, each DHP was given a DermLite 3Gen DL100 to use in practice when performing clinical skin examinations. All DHPs completed a data collection sheet to document their pattern of decision making with and without a DermLite. The outcome of interest was the use of dermoscopy by DHPs to demonstrate an increased detection of melanoma when compared to naked-eye examination. The outcome was evaluated 12 weeks postworkshop training. There were 120 evaluations made with the DermLite as compared to the naked eye. The overall agreement was 0.52, AC1 coefficient (95% CI) was 0.36 (0.30, 0.42), p < .001, and kappa coefficient (95% CI) was 0.27 (0.20, 0.43), p < .001. Overall, the risk of lesion under exam being suspicion for skin cancer was higher on 27.5% (33 out of 120) of the evaluations and lower on 20.8% (25 out of 120) evaluations. The risk of lesion was evaluated the same on 51.7% (62 out of 120) of the evaluations. This is an indication of “Poor” agreement between the two methods. The diagnosis and disposition made using DermLite compared to naked-eye results for both coefficients provided an “Intermediate to Good” agreement between the two methods in assigning diagnosis and disposition. This indicates that there is no difference between DermLite and naked-eye evaluations. More studies are needed in order to provide better evidence on the value of dermoscopy in clinical practice at the Dermatology and Laser Center. Future projects should be more explicit regarding the methods used and lesion selection in order to better understand the benefits of dermoscopy.

abstract: Clinical Decision Support (CDS) is primarily associated with alerts, reminders, order entry, rule-based invocation, diagnostic aids, and on-demand information retrieval. While valuable, these foci have been in production use for decades, and do not provide a broader, interoperable means of plugging structured clinical knowledge into live electronic health record (EHR) ecosystems for purposes of orchestrating the user experiences of patients and clinicians. To date, the gap between knowledge representation and user-facing EHR integration has been considered an “implementation concern” requiring unscalable manual human efforts and governance coordination. Drafting a questionnaire engineered to meet the specifications of the HL7 CDS Knowledge Artifact specification, for example, carries no reasonable expectation that it may be imported and deployed into a live system without significant burdens. Dramatic reduction of the time and effort gap in the research and application cycle could be revolutionary. Doing so, however, requires both a floor-to-ceiling precoordination of functional boundaries in the knowledge management lifecycle, as well as formalization of the human processes by which this occurs. This research introduces ARTAKA: Architecture for Real-Time Application of Knowledge Artifacts, as a concrete floor-to-ceiling technological blueprint for both provider heath IT (HIT) and vendor organizations to incrementally introduce value into existing systems dynamically. This is made possible by service-ization of curated knowledge artifacts, then injected into a highly scalable backend infrastructure by automated orchestration through public marketplaces. Supplementary examples of client app integration are also provided. Compilation of knowledge into platform-specific form has been left flexible, in so far as implementations comply with ARTAKA’s Context Event Service (CES) communication and Health Services Platform (HSP) Marketplace service packaging standards. Towards the goal of interoperable human processes, ARTAKA’s treatment of knowledge artifacts as a specialized form of software allows knowledge engineers to operate as a type of software engineering practice. Thus, nearly a century of software development processes, tools, policies, and lessons offer immediate benefit: in some cases, with remarkable parity. Analyses of experimentation is provided with guidelines in how choice aspects of software development life cycles (SDLCs) apply to knowledge artifact development in an ARTAKA environment. Portions of this culminating document have been further initiated with Standards Developing Organizations (SDOs) intended to ultimately produce normative standards, as have active relationships with other bodies. / Dissertation/Thesis / Doctoral Dissertation Biomedical Informatics 2018

Nurses are often the first members of the health care team with whom patients interact. The initial impression of the nurses’ receptiveness to the patients’ needs influences the patients’ views of their overall care. Researchers have suggested that understanding communication between individuals can provide the human link, or social element, to the successful implementation and use of electronic health records, including documentation (Lanham, Leykum, & McDaniel, 2012). Zadvinskis, Chipps, and Yen (2014) identified that the helpful features of bedside documentation systems were offset by the mismatch between the system and nurse’s workflow. The purpose of this micro-ethnography study was to explore the culture of nurse-patient interaction associated with electronic documentation at the bedside. Data were collected through passive participant observation, audio-taping of the nurse-patient interactions, and informal and semi-structured interviews with the nurses. A total of twenty-six observations were conducted on three nursing units at an urban healthcare facility in New England. These three units were occupied by similar patient populations and all patients required cardiac monitoring. Three themes consistently emerged from qualitative data analysis: the nurses paused during verbal communication, the nurses played a game of tag between the patient and the computer, and the nurses performed automatic or machine-like actions. The participants described these themes in the informal and semi-structured interviews. The nurses’ actions were observed during passive participant observation, and the audio-taped interactions supported these themes. Understanding the adaptation of caregiving necessitated by bedside electronic documentation will have a positive impact on developing systems that interface seamlessly with the nurses’ workflow and encourage patients’ active participation in their care.