The healthcare sector is facing a significant challenge: delivering quality clinical care in a costly and intricate environment. There is a general consensus that a solution for many aspects of this problem lies in establishing a framework for effective and efficient clinical decision support. The key to good decision support is offering clinicians just-in-time accessibility to relevant patient specific knowledge. However, at the present time, management of clinical knowledge and patient records is significantly inadequate resulting in sometimes uninformed, erroneous, and costly clinical decisions. One of the contributing factors is that the field of healthcare is characterized by large volumes of highly complex medical knowledge and patient information that must be captured, processed, interpreted, stored, analyzed, and exchanged. Moreover, different clinical information systems are typically not interoperable. This thesis introduces an approach for realizing a clinical decision support framework that manages complex clinical knowledge in a form of evidence-based clinical practice guidelines. The focus of presented work is directed on the interoperability of knowledge, information, and processes in a heterogeneous distributed environment. The main contributions of this thesis include definition of requirements, conceptual architecture, and approach for an interoperable clinical decision support system that is stand-alone, independent, and based on open source standards.

Medical practitioners are often unable to access medical and health information at the point-of-care, thus preventing them from providing quality healthcare. Family Health International 360 (FHI) provided medical practitioners with a locally relevant, reliable, and accurate comprehensive library of medical information on mobile computing devices (MCDs), at the point-of-care, as part of a project in collaboration with the Department of Health in the Eastern Cape Province. As part of the latter project, Ricks (2012:7) conducted an investigation into the impact that accessing health information at the point-of-care, via MCDs, had on the clinical decision-making practice of medical practitioners and professional nurses in public hospitals and primary healthcare settings in the Eastern Cape Province. The researcher identified a gap in the aforementioned study and was thus motivated to conduct this study to explore and describe the experiences of medical practitioners at public hospitals in further detail by conducting a qualitative study, as the previous study was quantitative. The purpose of this study was therefore to explore and describe the experiences of medical practitioners regarding the accessing of information at the point-of-care, via mobile technology, for clinical decision making at public hospitals. To achieve the purpose of the study, a qualitative, explorative, descriptive and contextual research design was used. The research population comprised medical practitioners who were using MCDs to access information at the point-of-care for clinical decision making. Purposive sampling was used to select the research sample. Semi-structured interviews were used to collect the necessary research data. Tesch’s steps were used to analyse the data. The principles for ensuring trustworthiness and ethical considerations were adhered to throughout the study. Two main themes and six sub-themes emerged in relation to the experiences of medical practitioners regarding the accessing of information at the point-of-care, for clinical decision making, via mobile technology. The main findings of the research highlighted the benefits and challenges that were experienced by the medical practitioners when using the MCDs for accessing information at the point-of-care for clinical decision making. The study concludes with recommendations pertaining to the areas of practise, education and research.

An attempt was made to investigate the decision processes of subjects in a bivariate decision making task, similar to that facing a medical specialist who is required to classify a patient as belonging to one of a number of possible disease populations on the basis of the patient's scores of two predictor cues. It was felt that such tasks had been largely neglected in experimental psychology, where the tendency has been towards requiring subjects to learn relationships between continuous predictor variables and a continuous criterion, rather than between continuous predictor variables and a categorical criterion. When the relationship between the predictor variables is the same in both the populations to be discriminated, the best decision function is based on a linear combination of the cues (Fisher’s Linear Discriminant Function). It was found that the decisions of those subjects who learned to use the cues in a way which was at all valid in such situations, could be well approximated by a model which weighted the two cues equally in a linear combination and based it’s decisions on the result. When the relationship between the predictor variables differs from one population to the other, however, the best decision function becomes more complex, including terms in the squares and cross-products of the cues. It was felt that such situations are particularly relevant to medical decision making where clinicians have frequently claimed that the "pattern" of scores of a patient is important, not Just the individual scores on each cue. It was found that if differences in cue intercorrelation were large, then subjects seemed to inolude in their iii decision processes, some nonlinear term to take account of this fact. If, however, differences in cue intercorrelation were only moderate, or if the correlations involved were large hut negative, this seemed to go unnoticed by the subjects and did not lead to any reliance on nonlinear terms. The results show that previous findings in "real life" tasks, that decision making processes could be adequately represented as linear combinations of cues, may be due more to the linear nature of the tasks than to any predisposition towards linear processes on the part of human decision makers, and that the statistical properties of "real life" tasks must be more thoroughly investigated before it is assumed that they require nonlinear decision processes.

The efforts addressed in this thesis refer to applying nonlinear risk predictive techniques based on logistic regression to medical diagnostic test data. This study is motivated and pursued to address the following: 1. To extend logistic regression model of biostatistics to medical informatics 2. Computational preemptive and predictive testing to determine the probability of occurrence (p) of an event by fitting a data set to a (logit function) logistic curve: Finding upper and lower bounds on p based on stochastical considerations 3. Using the model developed on available (clinical) data to illustrate the bounds-limited performance of the prediction. Relevant analytical methods, computational efforts and simulated results are presented. Using the results compiled, the risk evaluation in medical diagnostics is discussed with real-world examples. Conclusions are enumerated and inferences are made with directions for future studies. / by Alice Horn Dupont. / Thesis (M.S.C.S.)--Florida Atlantic University, 2011. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2011. Mode of access: World Wide Web.

Background: Cancer treatment and management require complicated clinical decision making to provide the highest quality of care for an individual patient. This is facilitated in part with ever-increasing availability of medications and treatments but hindered due to barriers such as access to care, cost of medications, clinician knowledge, and patient preferences or clinical factors. Although guidelines for cancer treatment and many symptoms have been developed to inform clinical practice, implementation of these guidelines into practice is often delayed or does not occur. Informatics-based approaches, such as clinical decision support, may be an effective tool to improve guideline implementation by delivering patient-specific and evidence-based knowledge to the clinician at the point of care to allow shared decision making with a patient and their family. The large amount of data in the electronic health record can be utilized to develop, evaluate, and implement automated approaches; however, the quality of the data must first be examined and evaluated. Methods: This dissertation addresses gaps the literature about clinical decision making for cancer care delivery. Specifically, following an introduction and review of the literature for relevant topics to this dissertation, the researcher presents three studies. In Study One, the researcher explores the use of clinical decision support in cancer therapeutic decision making by conducting a systematic review of the literature. In Study Two, the researcher conducts a quantitative study to describe the rate of guideline concordant care provided for prevention of acute chemotherapy-induced nausea and vomiting (CINV) and to identify predictors of receiving guideline concordant care. In Study Three, the researcher conducts a mixed-methods study to evaluate the completeness, concordance, and heterogeneity of clinician documentation of CINV. The final chapter of this dissertation is comprised of key findings of each study, the strengths and limitations, clinical and research implications, and future research. Results: In Study One, the systematic review, the researcher identified ten studies that prospectively studied clinical decision support systems or tools in a cancer setting to guide therapeutic decision making. There was variability in these studies, including study design, outcomes measured, and results. There was a trend toward benefit, both in process and patient-specific outcomes. Importantly, few studies were integrated into the electronic health record. In Study Two, of 180 patients age 26 years or less, 36% received guideline concordant care as defined by pediatric or adult guidelines, as appropriate. Factors associated with receiving guideline concordant care included receiving a cisplatin-based regimen, being treated in adult oncology compared to pediatric oncology, and solid tumor diagnosis. In Study Three, of the 127 patient records reviewed for the documentation of chemotherapy-induced nausea and vomiting, 75% had prescriber assessment documented and 58% had nursing assessment documented. Of those who had documented assessments by both prescriber and nurse, 72% were in agreement of the presence/absence of chemotherapy-induced nausea and vomiting. After mapping the concept through the United Medical Language System and developing a post-coordinated expression to identify chemotherapy-induced nausea and vomiting in the text, 85% of prescriber documentation and 100% of nurse documentation could be correctly categorized as present/absent. Further descriptors of the symptoms, such as severity or temporality, however, were infrequently reported. Conclusion: In summary, this dissertation provides new knowledge about decision making in cancer care delivery. Specifically, in Study One the researcher describes that clinical decision support, one potential implementation strategy to improve guideline concordant care, is understudied or under published but a promising potential intervention. In Study Two, I identified factors that were associated with receipt of guideline concordant care for CINV, and these should be further explored to develop interventions. Finally, in Study Three, I report on the limitations of the data quality of CINV documentation in the electronic health record. Future work should focus on validating these results on a multi-institutional level.

Background The landscape of clinical decision support systems (CDSSs) is evolving to include increasingly sophisticated data-driven methods, such as machine learning, to provide clinicians with predictions about patients’ risk for negative outcomes or their likely responses to treatments (predictive CDSSs). However, trust in predictive CDSSs has shown to challenge clinician adoption of these tools, precluding the ability to positively impact patient outcomes. This is particularly salient in the hospital setting where clinician time is scarce, and predictive CDSSs have the potential to decrease preventable mortality. Many have advised that clinicians should be involved in the development, implementation, and evaluation of predictive CDSSs to increase translation from development to adoption. Yet, little is known about the prevalence of clinician involvement or the factors that influence clinicians’ trust in predictive CDSSs for the hospital setting. The specific aims of this dissertation were: (a) to survey the literature on predictive CDSSs for the hospital setting to describe the prevalence and methods of clinician involvement throughout stages of system design, (b) to identify and characterize factors that influence clinicians’ trust in predictive CDSSs for in-hospital deterioration, and (c) to explore the use of a trust conceptual framework for incorporating clinician expertise into machine learning model development for predicting rapid response activation among hospitalized non-ICU patients using electronic health record (EHR) data. Methods To address the first aim (presented in Chapter Two), a scoping review was conducted to summarize the state of the science of clinician (nurse, physician, physician assistant, nurse practitioner) involvement in predictive CDSS design, with a specific focus on systems using machine learning methods with EHR data for in-hospital decision-making. To address the second aim (presented in Chapter Three), semi-structured interviews with nurses and prescribing providers (i.e., physicians, physicians assistants, nurse practitioners) were conducted and analyzed inductively and deductively (using the Human-Computer Trust conceptual framework) to identify factors that influence trust in predictive CDSSs, using an implemented predictive CDSS for in-hospital deterioration as a grounding example. Finally, to address the third aim (presented in Chapter Four), clinician expertise was elicited in the form of model specifications (requirements, insights, preferences) for facilitating factors shown to influence trust in predictive CDSSs, as guided by the Human-Computer Trust conceptual framework. Specifications included: (a) importance ranking of input features, (b) preference for a more sensitive or specific model, (c) acceptable false positive and negative rates, and (d) prediction lead time. Specifications informed development and evaluation of machine learning models predicting rapid response activation using retrospective EHR data. Results The scoping review identified 80 studies. Seventy-six studies described developing a machine learning model for a predictive CDSS, 28% of which described involving clinicians during development. Clinician involvement during development was categorized as: (a) determining clinical relevance/correctness, (b) feature selection, (c) data preprocessing, and (d) serving as a gold standard. Only five studies described implemented predictive CDSSs and no studies described systems in routine use. The qualitative investigation with 17 clinicians (9 prescribing providers, 8 nurses) confirmed that the Human-Computer Trust concepts of perceived understandability and perceived technical competence are factors that influence hospital clinicians’ trust in predictive CDSSs and further characterized these factors (i.e., themes). This study also identified three additional themes influencing trust: (a) actionability, (b) evidence, and (c) equitability, and found that clinicians’ needs for explanations of machine learning models and the impact of discordant predictions may vary according to the extent to which clinicians rely on the predictive CDSS for decision-making. Only two of 28 categories/sub-categories and one theme emerged uniquely to nurses or prescribing providers. Finally, the third study elicited model specifications from fifteen total clinicians. Not all clinicians answered all questions. Vital sign frequency was ranked the most important feature category on average (n = 8 clinicians), the most frequently preferred prediction lead time was shift-change/8-12 hours (n = 9 clinicians), most preferred a more specific than sensitive model (71%; n = 7 clinicians), the average acceptable false positive rate was 42% (n = 9 clinicians), the average acceptable false negative rate was 29% (n = 6 clinicians). These specifications informed development and testing of four machine learning classification models (ridge regression, decision trees, random forest, and XGBoost). 249,676 patient admissions from 2015–2018 at a large northeastern hospital system were modeled to predict whether or not patients would have a rapid response within the 12-hour shift. The random forest classifier met clinician’s average acceptable false positive (27.7%) and negative rates (28.9%) and was marginally more specific (72.2%) than sensitive (71.1%) on a holdout test set. Conclusions Studies do not routinely report clinician involvement in model development of predictive CDSSs for the hospital setting and publications on implementation considerably lag those on development. Nurses and prescribing providers described largely shared experiences of trust in predictive CDSSs. Clinicians’ reliance on the predictive CDSS for decision-making within the target clinical workflow should be considered when aiming to facilitate trust. Incorporating clinician expertise into model development for the purpose of facilitating trust is feasible. Future research is needed on the impact of clinician involvement on trust, clinicians’ personal attributes that influence trust, and explanation design. Increased education for clinicians about predictive CDSSs is recommended.

Clinical risk assessment of chronic illnesses is a challenging and complex task which requires the utilisation of standardised clinical practice guidelines and documentation procedures in order to ensure consistent and efficient patient care. Conventional cardiovascular decision support systems have significant limitations, which include the inflexibility to deal with complex clinical processes, hard-wired rigid architectures based on branching logic and the inability to deal with legacy patient data without significant software engineering work. In light of these challenges, we are proposing a novel ontology and machine learning-driven hybrid clinical decision support framework for cardiovascular preventative care. An ontology-inspired approach provides a foundation for information collection, knowledge acquisition and decision support capabilities and aims to develop context sensitive decision support solutions based on ontology engineering principles. The proposed framework incorporates an ontology-driven clinical risk assessment and recommendation system (ODCRARS) and a Machine Learning Driven Prognostic System (MLDPS), integrated as a complete system to provide a cardiovascular preventative care solution. The proposed clinical decision support framework has been developed under the close supervision of clinical domain experts from both UK and US hospitals and is capable of handling multiple cardiovascular diseases. The proposed framework comprises of two novel key components: (1) ODCRARS (2) MLDPS. The ODCRARS is developed under the close supervision of consultant cardiologists Professor Calum MacRae from Harvard Medical School and Professor Stephen Leslie from Raigmore Hospital in Inverness, UK. The ODCRARS comprises of various components, which include: (a) Ontology-driven intelligent context-aware information collection for conducting patient interviews which are driven through a novel clinical questionnaire ontology. (b) A patient semantic profile, is generated using patient medical records which are collated during patient interviews (conducted through an ontology-driven context aware adaptive information collection component). The semantic transformation of patients’ medical data is carried out through a novel patient semantic profile ontology in order to give patient data an intrinsic meaning and alleviate interoperability issues with third party healthcare systems. (c) Ontology driven clinical decision support comprises of a recommendation ontology and a NICE/Expert driven clinical rules engine. The recommendation ontology is developed using clinical rules provided by the consultant cardiologist from the US hospital. The recommendation ontology utilises the patient semantic profile for lab tests and medication recommendation. A clinical rules engine is developed to implement a cardiac risk assessment mechanism for various cardiovascular conditions. The clinical rules engine is also utilised to control the patient flow within the integrated cardiovascular preventative care solution. The machine learning-driven prognostic system is developed in an iterative manner using state of the art feature selection and machine learning techniques. A prognostic model development process is exploited for the development of MLDPS based on clinical case studies in the cardiovascular domain. An additional clinical case study in the breast cancer domain is also carried out for the development and validation purposes. The prognostic model development process is general enough to handle a variety of healthcare datasets which will enable researchers to develop cost effective and evidence based clinical decision support systems. The proposed clinical decision support framework also provides a learning mechanism based on machine learning techniques. Learning mechanism is provided through exchange of patient data amongst the MLDPS and the ODCRARS. The machine learning-driven prognostic system is validated using Raigmore Hospital's RACPC, heart disease and breast cancer clinical case studies.

A pesquisa buscou avaliar se a ultrassonografia portátil, durante consulta clínica em uma unidade básica de saúde pública do Sistema Único de Saúde brasileiro, foi capaz de fornecer informações adicionais confiáveis para a tomada de decisões terapêuticas e diminuir o tempo de espera para uma eventual avaliação especializada, por meio da comparação entre os achados obtidos na ultrassonografia portátil e na ultrassonografia convencional. A ultrassonografia é o segundo método de avaliação por imagem mais utilizado na prática médica, após o Raio-X. Não emite radiações ionizantes e seu efeito sobre os tecidos biológicos é seguro, desde que em nível de intensidade acústica mínima para obtenção dos resultados desejados. Este trabalho surgiu da observação de quão prolongado é o tempo de espera por exames de ultrassom a que estão sujeitos os pacientes da saúde pública brasileira, chegando até dois anos em alguns casos, contrariando protocolos de recomendação e podendo comprometer a saúde do paciente. Avaliou-se custo-efetividade da tecnologia, suas principais limitações, a necessidade de criação de protocolos para sua utilização e os subsídios fornecidos para a tomada de decisões pelo médico clínico. A tecnologia foi aplicada em exames nas áreas de medicina interna, sistema musculoesquelético e cardiologia. Empregou aparelho de ultrassom portátil aprovado pela Anvisa, modelo Vscan™ Dualprobe. Conforme protocolos do Instituto Americano de Ultrassom em Medicina (AIUM), os sujeitos da pesquisa foram submetidos a exame com aparelho portátil, durante o momento da consulta com o pesquisador, após anamnese e exame físico. Posteriormente, foram submetidos a exame empregando aparelho convencional, com médico possuindo titulação em ultrassonografia. As imagens e os achados foram comparados entre os exames e se calculou um índice de confiabilidade Kappa global nas três áreas estudadas de 91,12%, embora o índice para a área cardíaca tenha sido de 76,92%, o que demonstra a limitação do uso do ultrassom portátil para a área. A pesquisa demonstrou que o emprego da ultrassonografia portátil pode vir a ser um importante aliado do médico clínico, permitindo melhoria da qualidade da saúde prestada aos pacientes do Sistema Único de Saúde brasileiro. / The research aimed to evaluate whether portable ultrasonography, during clinical consultation in a basic public health unit of the Brazilian Unified Health System, was able to provide reliable additional information for therapeutic decision-making and to reduce the waiting time for a specialized evaluation, by means of the comparison between the findings obtained in portable ultrasonography and conventional ultrasonography. Ultrasonography is the second method of image evaluation more used in medical practice after the X-ray. It does not emit ionizing radiation and its effect on biological tissues is safe, provided that at minimum acoustic intensity level to obtain the desired results. This study was based on the observation of the length of time the patient waits for the ultrasound exams to which Brazilian public health patients are subjected, up to two years in some cases, contrary to recommendation protocols and potentially compromising patient´s health. It evaluated the cost-effectiveness of the technology, its main limitations, the need to create protocols for its use and the subsidies provided for the decision-making by the clinician. The technology was applied in examinations in the areas of internal medicine, musculoskeletal system and cardiology. It was used a portable ultrasound device approved by Anvisa, model Vscan™ Dualprobe. According to protocols of the American Institute of Ultrasound in Medicine (AIUM), the subjects were examined with a portable device, during the moment of the consultation with the researcher, after anamnesis and physical examination. Subsequently, they were submitted to an examination using a conventional device, with a doctor having ultrasound titration. The images and findings were compared between the exams and a global Kappa reliability index was calculated in the three studied areas of 91.12%, although the index for the cardiac area was 76.92%, which demonstrates the limitation of the use of portable ultrasound for the area. The research demonstrated that the use of portable ultrasonography can be an important ally of the clinician, allowing an improvement in the quality of health provided to patients of the Brazilian Unified Health System.