Designing eHealth interventions for children with complex care needs requires continuous stakeholder collaboration and co-creation.

Objective: Hospital-to-home (H2H) transitions challenge families of children with medical complexity (CMC) and healthcare professionals (HCP). This study aimed to gain deeper insights into the H2H transition process and to work towards eHealth interventions for its improvement, by applying an iterative methodology involving both CMC families and HCP as end-users. Methods: For 20-weeks, the Dutch Transitional Care Unit consortium collaborated with the Amsterdam University of Applied Sciences, HCP, and CMC families. The agile SCREAM approach was used, merging Design Thinking methods into five iterative sprints to stimulate creativity, ideation, and design. Continuous communication allowed rapid adaptation to new information and the refinement of solutions for subsequent sprints. Results: This iterative process revealed three domains of care - care coordination, social wellbeing, and emotional support - that were important to all stakeholders. These domains informed the development of our final prototype, 'Our Care Team', an application tailored to meet the H2H transition needs for CMC families and HCP. Conclusion: Complex processes like the H2H transition for CMC families require adaptive interventions that empower all stakeholders in their respective roles, to promote transitional care that is anticipatory, rather than reactive. Innovation: A collaborative methodology is needed, that optimizes existing resources and knowledge, fosters innovation through collaboration while using creative digital design principles. This way, we might be able to design eHealth solutions with end-users, not just for them.

Designing a Collaborative Patient-Centered Digital Health Platform for Pediatric Diabetes Care in British Columbia: Formative Needs Assessment by Caregivers of Children and Youths Living With Type 1 Diabetes and Health Care Providers.

BACKGROUND: Digital health apps are becoming increasingly available for people living with diabetes, yet data silos continue to exist. This requires health care providers (HCPs) and patients to use multiple digital platforms to access health data. OBJECTIVE: In this study, we gathered the perspectives of caregivers of children and youths living with type 1 diabetes (T1D) and pediatric diabetes HCPs in the user-centered design of TrustSphere, a secure, single-point-of-access, integrative digital health platform. METHODS: We distributed web-based surveys to caregivers of children and youths living with T1D and pediatric diabetes HCPs in British Columbia, Canada. Surveys were designed using ordinal scales and had free-text questions. Survey items assessed key challenges, perceptions about digital trust and security, and potential desirable features for a digital diabetes platform. RESULTS: Similar challenges were identified between caregivers of children and youths living with T1D (n=99) and HCPs (n=49), including access to mental health support, integration of diabetes technology and device data, and the ability to collaborate on care plans with their diabetes team. Caregivers and HCPs identified potential features that directly addressed their challenges, such as more accessible diabetes data and diabetes care plans. Caregivers had more trust in sharing their child's data digitally than HCPs. Most caregivers and HCPs stated that an integrative platform for T1D would support collaborative patient care. CONCLUSIONS: Caregiver and HCP perspectives gathered in this study will inform the early prototype of an integrative digital health platform. This prototype will be presented and iterated upon through a series of usability testing sessions with caregivers and HCPs to ensure the platform meets end users' needs.

Prediction of Blood Risk Score in Diabetes Using Deep Neural Networks.

Improving the prediction of blood glucose concentration may improve the quality of life of people living with type 1 diabetes by enabling them to better manage their care. Given the anticipated benefits of such a prediction, numerous methods have been proposed. Rather than attempting to predict glucose concentration, a deep learning framework for prediction is proposed in which prediction is performed using a scale for hypo- and hyper-glycemia risk. Using the blood glucose risk score formula proposed by Kovatchev et al., models with different architectures were trained, including, a recurrent neural network (RNN), a gated recurrent unit (GRU), a long short-term memory (LSTM) network, and an encoder-like convolutional neural network (CNN). The models were trained using the OpenAPS Data Commons data set, comprising 139 individuals, each with tens of thousands of continuous glucose monitor (CGM) data points. The training set was composed of 7% of the data set, while the remaining was used for testing. Performance comparisons between the different architectures are presented and discussed. To evaluate these predictions, performance results are compared with the last measurement (LM) prediction, through a sample-and-hold approach continuing the last known measurement forward. The results obtained are competitive when compared to other deep learning methods. A root mean squared error (RMSE) of 16 mg/dL, 24 mg/dL, and 37 mg/dL were obtained for CNN prediction horizons of 15, 30, and 60 min, respectively. However, no significant improvements were found for the deep learning models compared to LM prediction. Performance was found to be highly dependent on architecture and the prediction horizon. Lastly, a metric to assess model performance by weighing each prediction point error with the corresponding blood glucose risk score is proposed. Two main conclusions are drawn. Firstly, going forward, there is a need to benchmark model performance using LM prediction to enable the comparison between results obtained from different data sets. Secondly, model-agnostic data-driven deep learning models may only be meaningful when combined with mechanistic physiological models; here, it is argued that neural ordinary differential equations may combine the best of both approaches. These findings are based on the OpenAPS Data Commons data set and are to be validated in other independent data sets.

A Bridging Opportunities Work-frame to develop mobile applications for clinical decision making.

BACKGROUND: Mobile applications (apps) providing clinical decision support (CDS) may show the greatest promise when created by and for frontline clinicians. Our aim was to create a generic model enabling healthcare providers to direct the development of CDS apps. METHODS: We combined Change Management with a three-tier information technology architecture to stimulate CDS app development. RESULTS: A Bridging Opportunities Work-frame model was developed. A test case was used to successfully develop an app. CONCLUSION: Healthcare providers can re-use this globally applicable model to actively create and manage regional decision support applications to translate evidence-based medicine in the use of emerging medication or novel treatment regimens.

Personalized medicine policy challenges: measuring clinical utility at point of care.

Pharmacogenomics, driven by advances in genomics, helps to explain patients' individual variability in response to therapies. Personalized medicine, the application of the increasing understanding of pharmacogenomics, and information technology are intertwined from discovery to delivery at point of care, through to tracking clinical outcomes. Although exemplary cases of personalized medicine adoption demonstrate patient benefit and cost-effectiveness, a remaining barrier to large-scale real-world uptake of this novel approach in medicine is policy change. At point-of-care implementation, case studies will need to measure personalized medicine application outcomes of relevance to policy-makers and as evidence of clinical utility. Assessments need to be consistent across case studies. Standardizing specifications for case studies will better inform policy-makers performing economic evaluations on the use of personalized medicine.

Improving oncology outcomes through targeted therapeutics will require electronic delivery systems.

Typically, chemotherapy selection takes into account patient demographic data, including disease symptoms, family history, environmental factors and concurrent medications. Although validated and approved genomics tests are available for targeted therapeutics, a major challenge facing healthcare is the ability to process the genomic data in the patient's context and to return clinically interpretable dosing guidance to the physician in a realistic time frame. Delivery of these targeted therapeutics, made possible by clinical decision support systems connected to an electronic health record may help drive both the acceptance and adaptation of an electronic health record system, as well as provide personalized information at point-of-care, as part of the routine workflow. The realization of targeted therapeutics will depend on the concerted efforts of stakeholder groups as they address political, ethical, socioeconomical and technical challenges to achieve personalized medicine adoption through real-world implementation.

Damming the genomic data flood using a comprehensive analysis and storage data structure.

Data generation, driven by rapid advances in genomic technologies, is fast outpacing our analysis capabilities. Faced with this flood of data, more hardware and software resources are added to accommodate data sets whose structure has not specifically been designed for analysis. This leads to unnecessarily lengthy processing times and excessive data handling and storage costs. Current efforts to address this have centered on developing new indexing schemas and analysis algorithms, whereas the root of the problem lies in the format of the data itself. We have developed a new data structure for storing and analyzing genotype and phenotype data. By leveraging data normalization techniques, database management system capabilities and the use of a novel multi-table, multidimensional database structure we have eliminated the following: (i) unnecessarily large data set size due to high levels of redundancy, (ii) sequential access to these data sets and (iii) common bottlenecks in analysis times. The resulting novel data structure horizontally divides the data to circumvent traditional problems associated with the use of databases for very large genomic data sets. The resulting data set required 86% less disk space and performed analytical calculations 6248 times faster compared to a standard approach without any loss of information. Database URL: http://castor.pharmacogenomics.ca.

Integrating genomic-based clinical decision support into electronic health records.

There is a growing consensus that the first and most necessary step to improving the efficiency, cost-effectiveness and quality of healthcare systems can be achieved through the implementation of interoperable patient-centric electronic health record (EHR) systems across hospitals and clinics. Targeted therapeutics (including screening, prevention and disease management) through EHR-based clinical decision support delivery may drive both the acceptance and adoption of EHR systems by providing personalized information at the point-of-care. The realization of targeted therapeutics will depend on the resolution of current political, ethical, socioeconomical and technical challenges surrounding EHR implementation efforts. There is a growing need for broad-based consensus initiatives to foster an essential level of standardization for EHRs. The timeliness of these issues is underlined by the rapid emergence of private sector efforts in this potentially lucrative field, from direct-to-consumer testing to Google-, or Microsoft-owned personal health data. This review discusses the potential value for adopting healthcare technology, with a focus on personalized medicine, and highlights the challenges that remain to achieve this.

Génome Québec & Montreal Heart Institute Pharmacogenomics Centre: a translational pharmacogenomics platform--from R&D to the clinic.

The Génome Québec and Montreal Heart Institute Pharmacogenomics Centre (Montreal, Canada), created in 2006, is a translational pharmacogenomics platform whose main objectives are to conduct pharmacogenomics research, provide pharmacogenomics services to the academic, biotechnology and pharmaceutical sectors, and integrate pharmacogenomics solutions into the healthcare system. The Centre has brought together a multidisciplinary team of researchers with expertise in genomics, bioinformatics and clinical trial research. All the Centre's clinical research studies are supported by the Centre's unique Good Laboratory Practice facility framework, which has the ability to perform pharmaceutical clinical trials and deliver clinical diagnostics under the highest standards. The Centre has successfully leveraged its experience and expertise in technology development and pharmacogenomics clinical trial work to attract funding and collaborative partnerships in both the public and private sectors.