Contexte : Les patients âgés atteints de néphropathie diabétique ne reçoivent souvent pas un traitement pharmacologique optimal. Les directives de pratique clinique actuelles n'intègrent pas le concept de soins personnalisés. Les algorithmes d'aide à la décision clinique (ADC) qui tiennent compte à la fois des preuves et des soins personnalisés pour améliorer les résultats des patients peuvent améliorer les soins aux personnes âgées. L'objectif de cette recherche est de concevoir et de valider un algorithme ADC pour la prescription d'inhibiteurs du système rénineangiotensine- aldostérone (ISRAA) pour les patients âgés atteints de diabète. Méthodes : La conception de l'algorithme ADC comprenait trois phases principales. Dans la première phase, nous avons recherché, examiné et évalué les preuves actuelles sur plusieurs sujets liés aux décisions de prescription pour les patients âgés et à l'adhésion des cliniciens aux directives de pratique. Nous avons également procédé à un examen systématique et à une méta-analyse d'essais cliniques randomisés afin de déterminer les valeurs du nombre de patients à traiter (NPT) et du délai d'obtention d'un avantage (DOA) applicables à notre population cible en vue de leur utilisation dans l'algorithme. Dans la deuxième phase, nous avons exploré les points de vue des patients et des prestataires de soins de santé sur les outils ADC en menant des entretiens avec les patients et une enquête transversale auprès des prestataires de soins de santé. Dans la troisième et dernière phase, les résultats des études réalisées dans les phases un et deux ont été utilisés pour informer le développement de l'algorithme ADC qui a ensuite été validé dans une étude Delphi. Résultats : Nous avons créé un algorithme ADC qui couvrait 16 scénarios possibles. Neuf scénarios correspondaient à des recommandations de méta-analyses, tandis que cinq scénarios correspondaient à des directives de pratique clinique. Pour les neuf cas, nous avons généré 36 recommandations personnalisées et neuf recommandations générales sur la base des valeurs NPT et DOA calculées et appariées. En outre, nous avons pris en compte l'espérance de vie et la capacité fonctionnelle du patient. L'algorithme a été validé lors de trois tours d'une étude Delphi. Conclusion : Nous avons conçu un algorithme de CDS fondé sur des preuves qui intègre des considérations souvent négligées dans les directives de pratique clinique, notamment l'espérance de vie restante, la charge médicamenteuse et l'état fonctionnel. Les prochaines étapes consistent à le tester dans le cadre d'un essai clinique afin d'étudier s'il est capable d'atteindre des objectifs cliniques prévisibles et réalistes, de maintenir la qualité de vie des personnes âgées et de réduire l'utilisation et le coût du système de santé. / Background: Older patients with diabetic nephropathy often do not receive optimal pharmacological treatment. Current clinical practice guidelines do not incorporate the concept of personalized care. Clinical decision support (CDS) algorithms that consider both evidence and personalized care to improve patient outcomes can improve the care of older adults. The aim of this research is to design and validate a CDS algorithm for prescribing renin-angiotensin aldosterone system inhibitors (RAASi) for older patients with diabetes. Methods: The design of the CDS algorithm included three main phases. In phase one, we searched, reviewed, and evaluated current evidence on several topics related to prescribing decisions for older patients and clinicians' adherence to practice guidelines. We also conducted a systematic review and a meta-analysis of randomized clinical trials to determine the number needed to treat (NNT) and time-to-benefit (TTB) values applicable to our target population for use in the algorithm. In phase two, we explored the views of patients and healthcare providers on CDS tools through conducting patient interviews and a cross-sectional survey of healthcare providers. In the third and final phase, findings from studies completed in phases one and two were used to inform the development of the CDS algorithm which was then validated using modified Delphi methodology. Results: We have created a CDS algorithm that covered 16 possible scenarios. There were nine scenarios matched to meta-analysis recommendations, while five scenarios were matched to clinical practice guidelines. For the nine cases, we have generated 36 personalized and nine general recommendations based on the calculated and matched NNT and TTB values. In addition, we have considered the patient’s life expectancy and functional capacity. The algorithm was validated in three rounds of a modified Delphi study. Conclusion: We designed an evidenceinformed CDS algorithm that integrates considerations often overlooked in clinical practice guidelines, including remaining life expectancy, medication burden, and functional status. The next steps include testing in a clinical trial to study if it is able to achieve predictable and realistic clinical goals, maintaining quality of life in older adults, and reducing healthcare system use and cost.

Research has focused heavily on the study of Clinical Decision Support Systems. However, CDS systems have generally had little impact on clinical practice. One of the most important reasons is the lack of human-computer interaction (HCI) considerations in designing these systems. Although physicians play an essential role in healthcare decision-making, there is little literature describing physicians' expectations and preferences prior to the development of these systems, which is an essential phase in user-centered design.This study aims to answer the following research question: What do physicians expect of interacting with future clinical decision support systems? An exploratory qualitative study was conducted, and data were collected by interviewing 9 physicians practicing in Sweden. A thematic analysis was used for data analysis, and the findings are four themes: 1) physicians' Expectations related to clinical practice; 2) physicians' expectations related to physician-patient relationship; 3) physicians' expectations related to the physician's role 4) physicians' expectations related to CDS governance.The research findings contribute to the knowledge of Anticipated UX in the context of healthcare and CDS systems. The empirical findings on potential user expectations are valuable for understanding the diversity of user experience and user expectations as phenomena in the specific domain of CDS systems. Service designers can utilize and build on the empirical findings to develop positive user experiences of future CDS systems

<p>Chronic low back pain is the second most prevalent chronic condition in Canadian primary care settings. The treatment and diagnosis of chronic pain is challenging for primary care clinicians. Their main challenges are lack of knowledge and their approach toward assessing and treating pain. Evidence based guidelines have been developed for neuropathic pain and low back pain.</p> <p>CDSSs for chronic diseases are becoming popular in primary care settings as a mean to implement CPGs. A CDSS prototype for diagnosis and treatment of chronic, non-cancer pain in primary care was developed at McMaster University. It is evident that poor usability can hinder the uptake of health information technologies.</p> <p>The objective of this study was to test the usability of Pain Assistant using think aloud protocols with SUS scores in 2 iterations. In this study 13 primary care providers including family physicians, nurse practitioners and residents used Pain Assistant to complete 3 different patient case scenarios. Participants were asked to comment on both barriers and facilitators of usability of Pain Assistant. Additionally time to complete patient case scenarios was calculated for each participant. A comparison questionnaire gathered user preference between introducing CPGs in paper format and computerized decision support system.</p> <p>This study showed that iterative usability testing of the Pain Assistant with participation of real-end users has the potential to uncover usability issues of the Pain Assistant. Problems of user interface were the main usability barrier in first testing iteration following by problems of content. Changes were made to system design for second round based on the issues came up in the first iteration. However, because of time constrains not all the changes were implemented for second round of testing. Most of the refinements were to resolve user interface issues. In the second iteration, the problems with the content of Pain Assistant were the major barrier. The changes to the system design were successful in resolving user interface problems since the changed issues did not come up again in second round. Pain Assistant had an above the average usability score however no significant changes seen in SUS score. The time needed to complete tasks remained identical in both iterations. In addition, participants preferred to have CPGs in electronic formats than paper. Further study after implementing all the system changes needed to determine the effectiveness of system refinements.</p> / Master of Science (MSc)

New diagnostic methods and novel therapeutic agents spawn additional and heterogeneous in-formation, leading to an increasingly complex decision-making process for optimal treatment of cancer. A great amount of information is collected in organ-specific multidisciplinary tumor boards (MDTBs). By considering the patient’s tumor properties, molecular pathological test re-sults, and comorbidities, the MDTB has to consent an evidence-based treatment decision. Im-munotherapies are increasingly important in today’s cancer treatment, resulting in detailed in-formation that influences the decision-making process. Clinical decision support systems can fa-cilitate a better understanding via processing of multiple datasets of oncological cases and mo-lecular genetic information, potentially fostering transparency and comprehensibility of available information, eventually leading to an optimum treatment decision for the individual patient. We constructed a digital patient model based on Bayesian networks to combine the relevant pa-tient-specific and molecular data with depended probabilities derived from pertinent studies and clinical guidelines to calculate treatment decisions in head and neck squamous cell carcinoma (HNSCC). In a validation analysis, the model can provide guidance within the growing subject of immunotherapy in HNSCC and, based on its ability to calculate reliable probabilities, facilitates estimation of suitable therapy options. We compared actual treatment decisions of 25 patients with the calculated recommendations of our model and found significant concordance (Cohen’s κ=0.505, p=0.009) and 84% accuracy.

The purpose of the study was to describe the effects of an EMR on patient management in selected HIV clinics in Johannesburg. A quantitative, descriptive, cross-sectional study was undertaken in four HIV clinics in Johannesburg. The subjects (N=44) were the healthcare workers selected by stratified random sampling. Consent was requested from each subject and from the clinics in Johannesburg. Data was collected using structured questionnaires. Median age of subjects was 36, 82% were female. 86% had tertiary qualifications. 55% were clinicians. 52% had 2-3 years work experience. 80% had computer experience, 86% had over one year EMR experience. 90% used the EMR daily, 93% preferred EMR to paper. 93% had EMR training, 17% used EMR to capture clinical data. 87% perceived EMR to have more benefits; most felt doctor-patient relationship was not interfered with. 89% were satisfied with the EMR’s overall performance. The effects of EMR benefit HIV patient management. / Health Studies / MA (Public Health)

Opportunistic infections (OIs) have been identified as a leading cause of poor outcomes in the ARV therapy (ART) programme. In order to reduce OIs, the Malawi, MoH introduced routine prescription of cotrimoxazole preventive therapy (CPT) in 2005. The MoH also started scaling up a point-of-care electronic medical record (EMR) system in 2007 to improve monitoring and evaluation. This study had the following objectives: i) to quantify prescription of CPT before and after implementing EMR; ii) to compare the difference in CPT prescription before and after implementing EMR. A historically controlled study design was used to compare CPT prescriptions one year before, and one year after implementation of the EMR at two health facilities. The data indicated that there was a significant (P <0.001) decrease in CPT prescribing at one health facility and a significant increase in CPT prescription at another. / Health Studies / M.A. (Public Health)

The purpose of the study was to describe the effects of an EMR on patient management in selected HIV clinics in Johannesburg. A quantitative, descriptive, cross-sectional study was undertaken in four HIV clinics in Johannesburg. The subjects (N=44) were the healthcare workers selected by stratified random sampling. Consent was requested from each subject and from the clinics in Johannesburg. Data was collected using structured questionnaires. Median age of subjects was 36, 82% were female. 86% had tertiary qualifications. 55% were clinicians. 52% had 2-3 years work experience. 80% had computer experience, 86% had over one year EMR experience. 90% used the EMR daily, 93% preferred EMR to paper. 93% had EMR training, 17% used EMR to capture clinical data. 87% perceived EMR to have more benefits; most felt doctor-patient relationship was not interfered with. 89% were satisfied with the EMR’s overall performance. The effects of EMR benefit HIV patient management. / Health Studies / MA (Public Health)

Für viele Krankheitsbilder und Indikationen ist ein breites Spektrum an Arzneimitteln und Arzneimittelkombinationen verfügbar. Darüber hinaus stellen Therapieziele oft Kompromisse zwischen medizinischen Zielstellungen und Präferenzen und Erwartungen von Patienten dar, um Zufriedenheit und Adhärenz zu gewährleisten. Die Auswahl der optimalen Therapieoption kann daher eine große Herausforderung für den behandelnden Arzt darstellen. Klinische Entscheidungsunterstützungssysteme, die Wirksamkeit oder Risiken unerwünschter Arzneimittelwirkung für Behandlungsoptionen vorhersagen, können diesen Entscheidungsprozess unterstützen und \linebreak Leitlinien-basierte Empfehlungen ergänzen, wenn Leitlinien oder wissenschaftliche Literatur fehlen oder ungeeignet sind. Bis heute sind keine derartigen Systeme verfügbar. Im Rahmen dieser Arbeit wird die Anwendung von Methoden aus der Domäne der Recommender Systems (RS) und des Maschinellen Lernens (ML) in solchen Unterstützungssystemen untersucht. Aufgrund ihres erfolgreichen Einsatzes in anderen Empfehlungssystemen und der einfachen Interpretierbarkeit werden zum einen Nachbarschafts-basierte Collaborative Filter (CF) an die besonderen Anforderungen und Herausforderungen der Therapieempfehlung angepasst. Zum anderen werden ein Modell-basierter CF-Ansatz (SLIM) und ein ML Algorithmus (GBM) erprobt. Alle genannten Ansätze werden anhand eines exemplarischen Therapieempfehlungssystems evaluiert, das auf die Behandlung der Autoimmunkrankheit Psoriasis abzielt. Um das Risiko der Empfehlung kontraindizierter oder gar gesundheitsgefährdender Medikamente zu reduzieren, werden Regeln aus evidenzbasierten Leitlinien und Expertenempfehlungen implementiert, um solche Therapieoptionen aus den Empfehlungslisten herauszufiltern. Insbesondere die Nachbarschafts-basierten CF-Algorithmen zeigen insgesamt kleine durchschnittliche Abweichungen zwischen geschätztem und tatsächlichem Therapie-Outcome. Auch die aus den Outcome-Schätzungen abgeleiteten Empfehlungen zeigen eine hohe Übereinstimmung mit der tatsächlich angewandten Behandlung. Die Modell-basierten Ansätze sind den Nachbarschafts-basierten Ansätzen insgesamt unterlegen, was auf den begrenzten Umfang der verfügbaren Trainingsdaten zurückzuführen ist und die Generalisierungsfähigkeit der Modelle erschwert. Im Vergleich mit menschlichen Experten sind alle untersuchten Algorithmen jedoch hinsichtlich Übereinstimmung mit der tatsächlich angewandten Therapie unterlegen. Eine objektive und effiziente Bewertung des Behandlungserfolgs kann als Voraussetzung für ein erfolgreiches ``Krankheitsmanagement'' angesehen werden. Daher wird in weiteren Untersuchungen für ausgwählten klinische Anwendungen der Einsatz von ML Methoden zur automatischen Quantifizierung von Gesunheitszustand und Therapie-Outcome erprobt. Zusätzlich, als weitere Quelle für Informationen über Therapiewirksamkeiten, wird der Einsatz von Sentiment Analysis Methoden zur Extraktion solcher Informationen aus Medikamenten-Bewertungen untersucht. / Under most medical conditions and indications, a great variety of pharmaceutical drugs and drug combinations are available. Beyond that, trade-offs need to be found between the medical requirements and the patients' preferences and expectations in order to support patients’ satisfaction and adherence to treatments. As a consequence, the selection of an optimal therapy option for an individual patient poses a challenging task to prescribers. Clinical Decision Support Systems (CDSSs), which predict outcome as effectiveness and risk of adverse effects for available treatment options, can support this decision-making process and complement guideline-based decision-making where evidence from scientific literature is missing or inappropriate. To date, no such systems are available. Within this work, the application of methods from the Recommender Systems (RS) domain and Machine Learning (ML) in such decision support systems is studied. Due to their successful application in other recommender systems and good interpretability, neighborhood-based CF algorithms are transferred to the medical domain and are adapted to meet the requirements and challenges of the therapy recommendation task. Moreover, a model-based CF method (SLIM) and a state of the art ML algorithm (GBM) are employed. All algorithms are evaluated in an exemplary therapy recommender system, targeting the treatment of the autoimmune skin disease Psoriasis. In order to reduce the risk of recommending contraindicated or even health-endangering drugs, rules derived from evidence-based guidelines and expert recommendations are implemented to filter such options from the recommendation lists. Especially the neighborhood-based CF algorithms show small average errors between estimated and observed outcome. Also, the recommendations derived from outcome estimates show high agreement with the ground truth. The performance of both model-based approaches is inferior to the neighborhood-based recommender. This is primarily assumed to be due to the limited training data sizes, which renders generalizability of the learned models difficult. Compared with recommendations provided by various experts, all proposed approaches are, however, inferior in terms of agreement with the ground truth. An objective and efficient assessment of treatment response can be regarded a prerequisite for successful ``disease management''. Therefore, the use of ML methods for the automatic quantification of health status and therapy outcome for selected clinical applications is investigated in further experiments. Moreover, as additional source of information about drug effectiveness, the use of Sentiment Analysis, in order to extract such information from drug reviews, is investigated.

Opportunistic infections (OIs) have been identified as a leading cause of poor outcomes in the ARV therapy (ART) programme. In order to reduce OIs, the Malawi, MoH introduced routine prescription of cotrimoxazole preventive therapy (CPT) in 2005. The MoH also started scaling up a point-of-care electronic medical record (EMR) system in 2007 to improve monitoring and evaluation. This study had the following objectives: i) to quantify prescription of CPT before and after implementing EMR; ii) to compare the difference in CPT prescription before and after implementing EMR. A historically controlled study design was used to compare CPT prescriptions one year before, and one year after implementation of the EMR at two health facilities. The data indicated that there was a significant (P <0.001) decrease in CPT prescribing at one health facility and a significant increase in CPT prescription at another. / Health Studies / M.A. (Public Health)

La prescription électronique, définie comme la saisie et la transmission électronique de diverses données de prescriptions (médicaments, requêtes de laboratoires, imagerie), est une technologie qui promet d’augmenter la productivité de l’exécution d’une prescription, de diminuer les erreurs reliées à l’illisibilité des prescriptions manuscrites et d’améliorer l’usage approprié des médicaments. Toutefois, la réalisation des bénéfices associés à cette technologie dépend grandement du contexte local de l’implantation et la configuration du système, qui doivent être adaptés aux besoins de l’établissement de santé et aux pratiques locales des professionnels. Bien que la prescription électronique soit implantée depuis plus d’une décennie dans plusieurs établissements de santé à travers le monde, il s’agit d’une technologie émergente au Québec et au Canada. Le Centre hospitalier universitaire (CHU) Sainte-Justine est l’un des premiers établissements de santé au Québec qui a implanté un système informatisé d’entrée d’ordonnances (SIEO) en 2019. L’outil, développé par un fournisseur local, a été adapté spécifiquement aux besoins de cet hôpital pédiatrique. Ainsi, l’objectif principal de ce mémoire est d’évaluer les effets de ce SIEO sur la sécurité de l’usage des médicaments. Plus spécifiquement, ce mémoire vise à 1) mesurer et décrire les problèmes liés à l’usage des médicaments avant et après l’implantation du SIEO, 2) identifier les caractéristiques du SIEO qui influencent la sécurité de l’usage des médicaments et 3) formuler des recommandations pour optimiser les bénéfices de l’outil de prescription électronique pour les patients et les utilisateurs. Afin de répondre à ces objectifs, ce travail présente deux études distinctes : 1. Une première analyse heuristique de l’utilisabilité portant spécifiquement sur la vulnérabilité du système a été effectuée en préimplantation du SIEO. Des scénarios visant à identifier les vulnérabilités du système ont été élaborés, puis un score permettant de noter la capacité du système à pallier ces vulnérabilités a été attribué par trois experts indépendants, afin de formuler des recommandations sur le design des fonctionnalités clés de cet outil. 2. Une étude observationnelle pré-post a été menée dans la période précédant l'implantation du système, et suivant l'implantation du système, dans l'unité pilote de pédiatrie générale. L’étude observationnelle est composée de deux volets, soit : a) une analyse des erreurs liées aux prescriptions de médicaments pour un échantillon d’ordonnances rédigées pendant une semaine par une analyse des interventions des pharmaciens et un audit de conformité des prescriptions et b) une analyse pré-post des erreurs liées au circuit du médicament, à partir des rapports d’incidents et accidents déclarés en lien avec le médicament. Les types d'erreurs ont été analysés afin de bien comprendre leur nature, ainsi que le rôle potentiel de la technologie sur la sécurité de l’usage des médicaments. Ces analyses ont été contextualisées par une description des fonctionnalités du SIEO (par l’utilisation d’outils validés pour l’évaluation des SIEO), des flux cliniques (par l’observation directe), et du projet d’implantation (par l’analyse de documents et des discussions avec les parties prenantes) afin de formuler des recommandations visant à optimiser les bénéfices du SIEO. Le premier article rapporte l'analyse de l'utilisabilité (étude 1) et des problèmes liés à la prescription de médicaments (étude 2a). Les résultats suggèrent que le système d’aide à la décision intégré au SIEO ne disposait pas de fonctionnalités recommandées pour limiter les vulnérabilités liées à l’usage de ce type d’outil. Néanmoins, les erreurs de conformité, qui représentaient la majorité des problèmes de prescription avant l’implantation ont été complètement éliminées par le nouveau SIEO. Toutefois, il n’y a pas eu de différence sur les erreurs de dosage et les autres interventions des pharmaciens. Ainsi, les résultats obtenus confirment qu’il est nécessaire de configurer un système d’aide à la décision avancé et adapté aux soins hospitaliers pédiatriques afin de réduire davantage les erreurs cliniques liées aux ordonnances de médicaments. Le deuxième article présente l’analyse des rapports d’incidents et accidents (étude 2b), et vise à estimer les effets du SIEO sur la sécurité de l'usage des médicaments, ainsi que mieux comprendre les erreurs de médicaments dans l’ensemble du processus des soins. L’article met en évidence le rôle important de la prescription électronique dans la simplification des étapes de la relève, de la transmission et de la transcription de la prescription. De plus, l'amélioration de l’utilisabilité de la feuille d’administration des médicaments électronique (FADMe) pourrait contribuer à réduire davantage le nombre d'erreurs liées au médicament. Ces deux articles permettent d’explorer les liens entre les caractéristiques du SIEO et les effets sur la sécurité de l’usage des médicaments, durant l’étape de prescription spécifiquement ainsi qu’à travers l’entièreté du circuit du médicament. Des recommandations sur l’utilisabilité du système et des stratégies de prévention sont présentées afin de réduire les erreurs liées au médicament. / Computerized provider order entry (CPOE), defined as a system used for entering and transmitting orders (e.g., for drugs, imaging, or lab requests) electronically, is a technology that can increase the productivity of order dispensing, reduce errors related to the illegibility of handwritten prescriptions and increase the appropriate use of medication. However, achieving the benefits associated with this technology depends on the local context of the implementation and configuration of the system, which must be adapted to the needs of the healthcare institution and the local practices of the healthcare professionals. Although CPOEs have been implemented for more than a decade in many healthcare institutions worldwide, it is an emerging technology in Quebec and Canada. The Centre hospitalier universitaire (CHU) Sainte-Justine is one of the first healthcare institutions in Quebec to implement a CPOE system in 2019. The CPOE, which was developed by a local vendor, was tailored specifically to meet the needs of the CHU Sainte-Justine's pediatric inpatient population. Thus, this study aims to evaluate the effects of the CPOE on medication safety. More specifically, this study seeks to 1) measure and describe problems related to medication use before and after the implementation of the CPOE, 2) identify the characteristics of the CPOE that influence medication safety, and 3) provide recommendations to optimize the benefits of the CPOE for patients and users. To address these objectives, two studies were conducted: 1. An expert-based heuristic vulnerability analysis of the system was performed to analyze the usability of the CPOE in the pre-implementation phase. Scenarios to identify system vulnerabilities were developed, and a score to rate the CPOE's ability to address these vulnerabilities was assigned by three independent experts to make recommendations on the design of the CPOE's key features. 2. A pre-post observational study was conducted prior to and following the CPOE implementation in the general pediatrics unit. The observational study included two components: a) An analysis of medication orders problems for a sample of prescriptions ordered for one week through the documentation of pharmacists’ interventions and a prescription conformity audit; b) An analysis of medication-related incident and accident reports throughout the year in pre and post implementation. The types of errors were described to understand their nature, as well as the potential role of technology on the safety of medication use. The analyses were contextualized with descriptions of the CPOE features (through the use of validated tools for CPOE evaluation), clinical workflows (through direct observation) and implementation project (through secondary document analysis and discussions with stakeholders) in order to make recommendations to improve medication safety. The first article covers the vulnerability analysis (study 1) and the medication orders problems at the prescribing step (study 2a). The results show that the clinical decision support system (CDSS) integrated into the CPOE lacked the recommended features to identify pediatric order errors. Conformity errors, which accounted for most prescribing errors, were completely eliminated by the prescriber implementation. However, there was no difference in dosing errors and other pharmacist interventions. Thus, the results obtained from these two components suggest the need to configure an advanced CDSS tailored to pediatric hospital care to further reduce clinical errors. The second article, focused on the analysis of incident and accident reports (study 2b), aims to estimate the impacts of the electronic prescriber on medication safety, as well as to better understand medication errors in the overall care process. The article highlights the importance of simplifying the acknowledgment, transmission, and transcription steps by implementing a CPOE. Improving the usability of the electronic medication administration record (eMAR) could further reduce medication errors. These two articles explore the relationship between the characteristics of the CPOE and their impact on medication safety, specifically at the prescribing step and throughout the entire medication management process. Recommendations on system usability and other prevention strategies are presented to improve medication safety.