The ongoing digitization of healthcare, which has been much accelerated by the widespread adoption of electronic health records, generates unprecedented amounts of clinical data in a readily computable form. This, in turn, affords great opportunities for making meaningful secondary use of clinical data in the endeavor to improve healthcare, as well as to support epidemiology and medical research. To that end, there is a need for techniques capable of effectively and efficiently analyzing large amounts of clinical data. While machine learning provides the necessary tools, learning effective predictive models from electronic health records comes with many challenges due to the complexity of the data. Electronic health records contain heterogeneous and longitudinal data that jointly provides a rich perspective of patient trajectories in the healthcare process. The diverse characteristics of the data need to be properly accounted for when learning predictive models from clinical data. However, how best to represent healthcare data for predictive modeling has been insufficiently studied. This thesis addresses several of the technical challenges involved in learning effective predictive models from electronic health records. Methods are developed to address the challenges of (i) representing heterogeneous types of data, (ii) leveraging the concept hierarchy of clinical codes, and (iii) modeling the temporality of clinical events. The proposed methods are evaluated empirically in the context of detecting adverse drug events in electronic health records. Various representations of each type of data that account for its unique characteristics are investigated and it is shown that combining multiple representations yields improved predictive performance. It is also demonstrated how the information embedded in the concept hierarchy of clinical codes can be exploited, both for creating enriched feature spaces and for decomposing the predictive task. Moreover, incorporating temporal information leads to more effective predictive models by distinguishing between event occurrences in the patient history. Both single-point representations, using pre-assigned or learned temporal weights, and multivariate time series representations are shown to be more informative than representations in which temporality is ignored. Effective methods for representing heterogeneous and longitudinal data are key for enhancing and truly enabling meaningful secondary use of electronic health records through large-scale analysis of clinical data.

Följande kandidatuppsats undersöker hur en elektronisk hälsoapplikation kan utformasför att öka medvetenheten om den personliga dagliga hälsan, fokuserat på huden och atthålla den frisk. Det sker genom att en prototyp har utformats där användaren genombilder samt text får instruktioner för att utföra en självkontroll. Prototypen innehållerockså en bildregistrering där användaren kan ladda upp bilder på en leverfläck ochgenom registreringen kan om leverfläcken utvecklas över tid. Syftet med prototypen äratt undersöka om en elektronisk hälsoapplikation utformad med fokus på återkopplingkan göra det möjligt för en person som inte är utbildad inom ämnet att undersöka sinhud samt se och följa förändring i en leverfläck genom bildanalys. Detta undersöktesgenom två tester. Ett frågeformulär där svarspersonen fick försöka avgöra om enleverfläck var frisk eller inte. Ett användartest där testpersonen fick testa bildanalysen,det som testades var om personen kunde ladda upp en bild samt se skillnad i bildernagenom bildregistreringen.Resultatet visar att majoriteten av enkätsvaren var korrekta svar, närmare bestämt 76,9procent. Personerna kunde identifiera om leverfläcken var frisk eller sjuk med hjälp avinformationen som tilldelades. Resultatet av användartesten visade att personerna kundeladda upp en bild på webbplatsen och sedan förstå bildanalysen. / This bachelor thesis examins how a electronic health application can be made toincrease awareness about personal daily health, with focus on individual skin andkeeping it healthy. This was made by creating a prototype where the user get usefulinformation on how to perform a self examination through text and images. Theprototype also contains functionality that allows the user to upload images of a moleand by an image registration follow possible changes over long periods of time. Thepurpose of the prototype is to investigate if an electronic health application with focuson feedback can make it possible for a person who is not educated in this area to examinthe skin and also discover possible changes over time with the image registration. Twotests were made to examin if this was possible. The first test was an questionnaire wherethe respondet tried to determine if a mole was healthy or not. The second test was a usertest where the test person got to try out the functionality with image registration in theprototype. The person got to upload an image and see the differences through the imageregistration.The results showed that the majority of the collected repsonses from the quetsionnairewas correct, with a percentage of 76,9 correct answers. The respondent could identify ifthe mole was healthy or not through the information that was presented. The results ofthe user test showed that all test persons could perform and understand the functionalityof uploading an image and then analyze the results throgh image registration.

Objective: The researcher replicated 12 questions from a previous user satisfaction study for the C-HOBIC assessment tool at St. Boniface Hospital in Winnipeg Manitoba. The research questions were: (a) what is user satisfaction regarding the C-HOBIC assessment tool 20 months after implementation, (b) has user satisfaction changed since the first evaluation, and (c) do user demographics correlate with user satisfaction and the C-HOBIC assessment tool? Method: 20 months after the previous study (Canadian Nurses Association, 2015), a convenience sample of 71 participants from a pool of approximately 700 clinicians completed an online survey comprised of 12 questions taken from the previous study. Results: The data were analyzed using Shapiro-Wilk, descriptive statistics, chi-square test for independence, and Spearman’s correlation. The Likert-style survey produced discrete, ranked data that did not follow a normal distribution. Overall user satisfaction with the C-HOBIC assessment tool was rated higher in the previous group (n=59) as compared to user satisfaction in the current group (n=71). There was a significant but weak correlation with gender and C-HOBIC patient outcomes positively influencing patient care directions, and improving patient care planning. A significant but weak correlation existed between the years of a participant's clinical experience and the ease of integrating C-HOBIC into practice. Conclusions: There were more participants not satisfied with the use of the C-HOBIC assessment tool than were satisfied. Participants in this study had less user satisfaction with the C-HOBIC assessment tool and associated outcomes than participants from a similar study 20 months before. Gender and years of clinical experience are correlated with user satisfaction. The small sample size, the non-normally distributed data, and convenient sampling method do not support generalization of the results beyond the data set. / Graduate / 0569 / alshunt@gmail.com

Empiric antimicrobial therapy often consists of the combination of Gram-positive coverage with vancomycin (VAN) and Gram-negative coverage, specifically an anti-pseudomonal beta-lactam, such as piperacillin-tazobactam (PTZ). Nephrotoxicity is commonly associated with VAN therapy; however, recent reports demonstrate increasing nephrotoxicity rates among patients treated with the combination of VAN and PTZ. This study evaluated the effect of the VAN/PTZ combination on acute kidney injury (AKI), as defined by the RIFLE criteria, compared to VAN and PTZ monotherapies. Overall, 11,650 patients were analyzed, with 1,647 (14.1%) AKI cases occurring. AKI was significantly more frequent in the VAN/PTZ group (21%) compared to either monotherapy group (VAN 8.3%, PTZ 7.8%, p<0.001 for both). Combination therapy was independently associated with higher AKI odds compared to monotherapy with either agent (aOR=2.03; 95% CI 1.74-2.39; aOR=2.31; 95% CI 1.97-2.71, for VAN and PTZ, respectively). Receipt of concomitant nephrotoxic drugs were independently associated with increased AKI rates, as were increased duration of therapy, length of hospital stay, increasing severity of illness, and increasing baseline renal function. VAN combined with PTZ was associated with twice the odds of AKI development compared to either agent as monotherapy. This demonstrates the need for judicious use of combination empiric therapy.

Abstract PATTERNS IN CHAPLAIN DOCUMENTATION OF ASSESSMENTS AND INTERVENTIONS, A DESCRIPTIVE STUDY by Kevin Eugene Adams, MDiv A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at Virginia Commonwealth University Virginia Commonwealth University, 2015 Diane Dodd-McCue, D.B.A, Department of Patient Counseling There is increasing emphasis on the importance of evidence-based care provided by all disciplines in healthcare. The Electronic Health Record (EHR) is becoming the standard for communicating assessments, plans of care, interventions, and outcomes of patient care. The spiritual care literature demonstrates the importance of assessing religious/spiritual needs and resources and developing plans of care to address the results of such assessment (Anandarajah & Hight, 2001; Borneman, Ferrell, & Puchalski, 2010; Fitchett, 1999; Fitchett & Risk, 2009; H. G. Koenig, 2007). This literature also suggests that addressing religious/spiritual needs of patients and families in the healthcare context can affect healthcare and adherence outcomes. The purpose of this study was to identify patterns of chaplain assessment and patterns of chaplain provision of services. This descriptive study was an exploratory retrospective analysis of categorical data recorded by clinical staff chaplains in the EHR at a single all pediatric healthcare institution, using contingency tables and frequency tables. The study examined chaplain use of assessment and service descriptors and the patterns of these descriptors when documenting chaplain visits. The results indicate chaplain preference for communicating in the EHR using general themes and concepts. This reveals an opportunity for chaplains to develop and implement a model of professional identity and articulation of care that is broad enough to accommodate the diversity of religion/spirituality chaplains encounter, yet able to articulate the specifics of patient and family religion/spirituality. The results found no consistent patterns among assessments or services provided. Further, the results found no indication of patterns between assessments made and the services provided. This presents an opportunity for chaplains to develop and implement a theory-driven, construct-based model of care that will connect the different facets of spiritual care. The assessments made will lead to plans of care that involve specific interventions resulting in appropriate outcomes related to overall patient and family care.

Personal Health Records (PHRs) allow patients to access and in some cases manage their own health records. Their potential benefits include access to health information, enhanced asynchronous communication between patients and clinicians, and convenience of online appointment scheduling and prescription refills. Potential barriers to PHR use include lack of computer and internet access, poor computer or health literacy, security concerns, and provider disengagement. PHRs may help those living in rural areas and those with chronic conditions such as heart failure, monitor and manage their disease, communicate with their health care team and adhere to clinical recommendations. To provide some much needed actual research, a descriptive mixed methods study of the usability, usefulness, and disease management potential of PHRs for rural heart failure patients was conducted. Fifteen participants were enrolled. Usability issues fell into three categories: screen layout; applying consistent, standard formatting; and providing concise, clear instructions. Participants used PHR features that were more convenient than other methods or that had some additional benefit to them. There was no difference between rural and urban participants. A heart failure nurse promoted recording daily heart failure symptoms in the PHR. Most participants did so at least once, but many found it cumbersome. Reasons for recording included the comfort of having clinical staff monitor the data. Participants who were stable did not find recording as useful as did those who were newly diagnosed or unstable. Participants used asynchronous communication to send messages to the heart failure nurse that they would not otherwise have communicated. The study expands the knowledge of PHR use by addressing useful functionality and disease management tools among rural patients with heart failure. The patients were able to complete tasks they found useful. The increased communication and disease management tools were useful to some.

This master thesis contributes to a growing body of literature analysing electronic health records (EHRs) and their importance to potentially optimization of the work flow in small medical practices. Additionally, general information about different health care systems is given, especially about the Canadian health care system. It gives a theoretical overview of information technologies used through health care providers. This thesis addresses the new technologies for communication and describes them. Problem: The "lack of clear channels of communication in patient transfer between care facilities leads to fragmentation in care." To prevent this, new communication channels need to be more effective in improving communication. Before trying to enhance communication channels used between care facilities, locally and nationally, the current focus needs to be on communication channels used within care facilities. This is of great importance especially for patients and other stakeholders who deem it as the most critical health care item. The slow implementation of EHRs in small medical practices makes it difficult to improve and evaluate performance and to ensure the confidence of patients in new technologies. On the other hand, the lack of implementation in this area has a negative effect on other health care providers, e.g. hospitals which already implemented the system. Through the existing gap the delivery of health care information is not complete possible and does not help to make the system safer for users. Research Question: The main goal is to obtain an understanding of individual physicians' attitudes and barriers to EHR. The thesis will show the advantages and obstacles as well as the pros and cons for small- and medium-sized practices to adopt EHRs. Result: It is not necessary anymore to discuss if EHRs should be implemented. It is more important to discuss how they should implement. The main problem for implementation is the financial barrier for small medical practices as well as personal attitudes connected with the age distribution of physicians and patients.

Health care today harms too frequently and routinely fails to deliver its potential benefits. Significant evidence suggests that high quality primary care can positively affect health outcomes. I explored three related topics mentioned frequently in current United States health reform €“ Electronic Health Records (EHR), physician profiling and Medical Home. An investment in these areas is expected to significantly improve quality of care and efficiency; however, there is only a patchwork of evidence supporting such claims. To achieve EHR promises, my research employed a standardization lens to study the dynamics between EHR embedded structures and primary care processes. Using grounded theory, a standardization dynamics model was created describing the influencers, conditions and consequences of the process state. A matrix of two conditions, information exchange and patient complexity, identified four distinct pathways that require a different balance between standardization and flexibility. The value of such pathways is that they frame choices about how to use embedded IT structures to support effective delivery processes. Physician profiling is an emerging methodology used in health care quality improvement programs. Efforts to measure performance at the individual physician level face a number of challenges, including the need for sufficient sample size to support reliable measurement. A process for creating a physician profiling model was developed, and a model designed for a case study site. Results indicate that reliable physician profiling is possible across care domains using a hierarchical composite model. Patient-Centered Medical Home (PCMH) is a new care delivery approach for providing comprehensive primary care that seeks to strengthen the physician-patient relationship. This exploratory study utilizes Pearson correlation coefficients to test four hypotheses about relationships between two sources of data: (1) PPC-PCMH Survey results that measure adoption of PCMH structures and (2) patient experience data from Massachusetts Health Quality Partners (MHQP). The results showed that the PPC-PCMH structures of access and communication were negatively correlated with the related patient experience measure. This study contributes to the literature by addressing deficiencies in how EHR-enabled processes, physician profiling models and Medical Home constructs are measured, to support improved outcomes.

The United States health-care system is considered to be unsustainable due to its unbearably high cost. Many of the resources are spent on acute conditions rather than aiming at preventing them. Preventive medicine methods, therefore, are viewed as a potential remedy since they can help reduce the occurrence of acute health episodes. The work in this dissertation tackles two distinct problems related to the prevention of acute disease. Specifically, we consider: (1) early detection of incorrect or abnormal postures of the human body and (2) the prediction of hospitalization due to heart related diseases. The solution to the former problem could be used to prevent people from unexpected injuries or alert caregivers in the event of a fall. The latter study could possibly help improve health outcomes and save considerable costs due to preventable hospitalizations. For body posture detection, we place wireless sensor nodes on different parts of the human body and use the pairwise measurements of signal strength corresponding to all sensor transmitter/receiver pairs to estimate body posture. We develop a composite hypothesis testing approach which uses a Generalized Likelihood Test (GLT) as the decision rule. The GLT distinguishes between a set of probability density function (pdf) families constructed using a custom pdf interpolation technique. The GLT is compared with the simple Likelihood Test and Multiple Support Vector Machines. The measurements from the wireless sensor nodes are highly variable and these methods have different degrees of adaptability to this variability. Besides, these methods also handle multiple observations differently. Our analysis and experimental results suggest that GLT is more accurate and suitable for the problem. For hospitalization prediction, our objective is to explore the possibility of effectively predicting heart-related hospitalizations based on the available medical history of the patients. We extensively explored the ways of extracting information from patients' Electronic Health Records (EHRs) and organizing the information in a uniform way across all patients. We applied various machine learning algorithms including Support Vector Machines, AdaBoost with Trees, and Logistic Regression adapted to the problem at hand. We also developed a new classifier based on a variant of the likelihood ratio test. The new classifier has a classification performance competitive with those more complex alternatives, but has the additional advantage of producing results that are more interpretable. Following this direction of increasing interpretability, which is important in the medical setting, we designed a new method that discovers hidden clusters and, at the same time, makes decisions. This new method introduces an alternating clustering and classification approach with guaranteed convergence and explicit performance bounds. Experimental results with actual EHRs from the Boston Medical Center demonstrate prediction rate of 82% under 30% false alarm rate, which could lead to considerable savings when used in practice.

Nos últimos anos, a ciência e a tecnologia proporcionaram uma larga gama de ferramentas aos profissionais de saúde. Em especial, as Tecnologias da Informação, pois favorecem o aprimoramento considerável da qualidade dos serviços de saúde prestados à população, quando gerenciadas adequadamente. O objetivo deste estudo foi identificar a relação entre risco de desenvolvimento de lesão por pressão e complexidade assistencial em pacientes críticos internados na unidade de terapia intensiva de um hospital universitário por meio dos registros eletrônicos de saúde. Trata-se de estudo correlacional, longitudinal e descritivo, com abordagem quantitativa. A coleta de dados foi conduzida durante 120 dias, a amostra foi composta por 74 pacientes que atenderam aos critérios de inclusão da pesquisa. Em relação às características sociodemográficas e clínicas, foi observado maioria do sexo masculino (56,8%), brancos (73%), na faixa etária de 60 a 79 anos (40,5%) e o tempo médio de internação nessa unidade correspondeu a 10,5 dias. A maioria dos indivíduos apresentou risco elevado para a lesão por pressão com média de 11,7%, complexidade assistencial média foi de 84,7% e frequência média diária de 5,5% reposicionamentos, registrados no sistema de informação hospitalar. Quanto ao desfecho dos pacientes, 28 (37,8%) apresentaram lesão por pressão notificada no sistema de informação hospitalar, 27 (36,5%) evoluíram para óbito na Unidade de Terapia Intensiva e 15 (20,3%) evoluíram a óbito e desenvolveram lesão por pressão, mostrando uma associação estatisticamente significante (p= 0,017). Foi observado significância estatística (p<0,001) e relação inversa para a complexidade assistencial e risco para desenvolvimento. As variáveis complexidade assistencial, risco para desenvolvimento de lesão por pressão, posições observadas foram registradas e também frequência de reposicionamento foram coletadas 776 vezes e observou-se que 605 (78%) da amostra em relação ao escore de complexidade assistencial foram registradas. Em 50% dos dias que os profissionais de enfermagem foram escalados com um paciente identificou-se que não foi atingida a capacidade máxima de trabalho do mesmo. No entanto, foi possível identificar que a capacidade máxima foi ultrapassada quando os profissionais assumiram o segundo paciente, ocorrendo uma possível sobrecarga de trabalho em 75% dos dias. Foi possível identificar diariamente os registros inexistentes dos escores relacionados à complexidade assistencial, risco para o desenvolvimento de lesão por pressão e reposicionamento. Portanto, esta pesquisa evidencia a relevância dos dados e informações produzidas pela equipe de enfermagem para identificar os pacientes em risco, estabelecer medidas preventivas para os mesmos e consequentemente melhorar os indicadores de qualidade por meio dos registros eletrônicos e, assim, superar os desafios relacionados a segurança, qualidade e efetividade da assistência de enfermagem / In recent years, science and technology have provided a wide range of tools to health professionals. In particular, information technology, because they favor the improvement of quality of considerable health care provided to the population, when properly managed. The aim of this study was to identify the relationship between risk of pressure injury development and complexity care in critically ill patients admitted to the intensive care unit of a university hospital through electronic records. This is a longitudinal and correlational descriptive study with quantitative approach. Data collection was conducted for 120 days; the sample was composed of 74 patients who met the inclusion criteria. In relation to the sociodemographic and clinical characteristics, it was observed mostly male (56.8%), white (73%), aged 60 to 79 years (40.5%) and the average time of staying in this unit was 10.5 days. The majority of individuals presented a high risk for pressure injury with an average of 11.7%, average complexity care was 84.7% and average daily frequency of replacement registered was 5.5%, on the hospital information system. As for the outcome of patients, 28 (37.8%) had notified pressure injury in the hospital information system, 27 (36.5%) evolved to death in the intensive care unit and 15 (20.3%) evolved to death and developed pressure injury, showing a statistically significant association (p=0.017). Statistical significance was observed (p < 0.001) and inverse relationship to the complexity and risk to development assistance. The variables care complexity, risk for pressure injury development, positions observed, recorded and also repositioning frequency were collected 776 times and it was observed that 605 (78%) of the sample in relation to the care complexity scores were recorded. In 50% of the days that the nursing professionals have been scaled with a patient identified that was not achieved the maximum working capacity of the same. However, it was possible to identify the maximum capacity was exceeded when the professionals took the second patient, a possible overload of work in 75% of the days. It was possible to identify daily non-existent records of scores related to complexity, risk for pressure injury development and repositioning. Therefore, this research highlights the importance of data and information produced by the nursing staff to identify patients at risk, establish preventive measures to the same and consequently improve the quality indicators by means of electronic records and thus overcome the challenges related to safety, quality and effectiveness of nursing care