EP01 - e-Poster Session 1 (ID 52)
- Event: e-Health 2018 Virtual Meeting
- Type: e-Poster Session
- Track: Clinical Delivery
- Presentations: 7
- Coordinates: 5/28/2018, 10:00 AM - 11:30 AM, e-Poster Station 1, Parq Grand Ballroom, Conference Level
EP01.01 - Why Digital Health Needs Automated Testing? (ID 462)
Purpose/Objectives: Why Digital Health Needs Automated Testing Digital health is increasingly being used to capture and view clinical records, write orders electronically and track their status, view diagnostic tests, utilize clinical decision support, and many other functions. Health organizations are relying on them to improve transitions of care, chronic disease management, care coordination, and improvements to the patient experience. The digital health care ecosystem is becoming more complex and interconnected given the breadth and depth of the aforementioned elements. These systems are contributing to a better experience for patients and clinicians, better outcomes, increased productivity and lower costs associated with the delivery of high-quality health care. At the same time, like any technology and a business reliance on it, there are introduced risks. Each additional digital element added to the systems must be 100% reliable and interact with all of the other systems. Therefore, each application or device increases the risk to the delivery of high-quality and safe health care. Do health care organizations have the programs and tools in place to mitigate risks in the deployment and operation of this ecosystem? Automated testing is crucial in ensuring the safety of digital health applications and medical devices (also referred to as eSafety).
Methodology/Approach: The biggest inherent risks in this ecosystem are with interoperability and user interface testing. These areas are where there is the greatest configuration, customization, and systems integration. So, what is the most effective way to verify all of these system elements are functioning correctly, in other words with eSafety? One of the most effective tools we have at hand, but are underutilizing in health care, is automated testing. Test automation is the use of software, under a setting of test preconditions, to execute tests and then determine whether the actual outcomes and the predicted outcomes are the same. Automated testing can be used to validate that the messages sent and received between systems are as expected for a specific test scenario. Automated testing can be used for user interface validation. The user interface is one area that is often configured to a specific deployment. Therefore, the team deploying must validate that all of the fields are functioning as designed. Are the data captured and validated on the screen as expected? Are combinations and permutations of related data valid?
Finding/Results: Automated testing can reduce the testing and validation cycle of Health IT systems from months to hours. It offloads a high percentage of manual test scenarios freeing up staff (i.e. clinicians, business analysts, IT staff) from the time-consuming data entry validation and data presentation and help them focus on higher valued tasks and patient care instead. Presentation includes 2 case studies on results and ROI.
Conclusion/Implications/Recommendations: In summary, test automation is a practical and efficient way to ensure that the user interface and interoperability within a health care digital ecosystem are functioning as designed. It offloads mundane and repetitive tasks from staff, reduces errors, and results in better breadth and depth of test coverage at a lower cost and with reduced timelines.
140 Character Summary: Maintaining the quality and safety of digital health systems can be achieved in a cost-effective, automated and timely manner, by the use of test automation.
EP01.02 - Digital Quality Improvement Survey Improves Accessibility and Patient Engagement (ID 566)
Purpose/Objectives: Black Creek Community Health Centre (BCCHC) provides holistic care and program services to a diverse client population in Toronto, Ontario Canada. We compared traditional paper-based data collection with an interactive mHealth platform Tickit® for the HQO Client Experience Survey and specific BCCHC questions to evaluate and implement quality improvement initiatives efficiently and in real time. Objective Compare traditional paper-based data collection with an interactive mHealth platform Tickit® for the HQO Client Experience Survey and specific BCCHC questions to evaluate and implement quality improvement initiatives more efficiently and in real time. Aim: Increase the response rates by 400% to be more equitably representation of organization population.
Methodology/Approach: Implementing change management, LEAN Six Sigma processes and continuous improvement via the Plan Do Study Act model. Staff and patients were empowered to integrate workflow in the clinic by automating processes to collect the patients voice in a responsive and value-added process: Survey development Workflow and process mapping Review and revise survey drafts Promotional materials Identify staff champions Training Implementation Check in and revision As an added value measure, patients completed the survey pre- and post- appointment.
Finding/Results: Results Section 1: Overall Experience Section 2: HQO Quality Dimension -Access Use of an mHealth tool on a tablet resulted in Advance Access by reducing 3rd next appointment from 30 days to 2-3 days by engaging patient population with Tickit. Section 3: HQO Quality Dimension: Client-Centred As per HQO technical specifications, the number of responses for Always (96) and Often (33) were combined, and divided by the total of responses, excluding non-respondents Average time to complete: 3 minutes A Louder Patients Voice 429% Survey response rate increase with Tickit Tickit n=713 vs Paper n=167
Conclusion/Implications/Recommendations: Staff was instrumental in reminding and encouraging clients to participate in the survey, and this was demonstrated by the high rate of responses within the first 5 months. Regular reporting to staff at Sheridan and Yorkgate by QIP committee members regarding the number of surveys completed, along with a thermometer diagram at Sheridan to track progress and report to clients also helped in encouraging clients to complete the survey. Success Factors: Staff champions Harness analytics power for regular monitoring and reporting Check-ins and revisions throughout process Reminders to staff and patients Mobility of technology Lessons Learned: Integration of new Client Engagement Tool Opportunity for intersectoral collaboration and co-design process Identify and improve on current workflows Increase efficiencies in data monitoring, reporting and analysis
140 Character Summary: Using an mHealth tool validated quality improvement initiatives for staff satisfaction, patient-centric care and advocatng for the patient's voice.
EP01.03 - Youth SBIRT Implementation Support: Tools, Training, and Technical Assistance (ID 595)
Purpose/Objectives: To support health systems in developing processes and procedures to provide high-quality SBIRT in the context of comprehensive adolescent health screening To improve providers skill and comfort in addressing substance use with adolescents To provide comprehensive screening of adolescents and targeted intervention to reduce substance use through the use of technology
Methodology/Approach: Phase 1: Customize Tool and Approach to fit Community Needs.We met with leaders and stakeholders from the participating projects to conduct a needs assessment and benchmark usual care, including resources and characteristics that may affect the need for technical assistance and training. Phase 2: Implementation Planning & Training. We engaged clinic staff from each clinic to specify implementation processes for their site, including which team members will be sent Check Yourself results, protocols for follow up, and tools for documentation in the electronic medical record. Phase 3: Roll-Out of Implementation with Quality Improvement. As the implementation plan rolled out, we conducted regular meetings with healthcare staff and thought leaders for monitoring progress, evaluation and debriefing. Phase 4: Evaluation. We conducted interviews with all levels of participants from each site potentially including administrative staff, program coordinators, thought leaders, providers, and project advisory board members to understand the implementation outcomes (changes in practice thinking and practice behavior, changes in organizations/systems), implementation processes, and next steps.
Finding/Results: In total, 26 participants were recruited from the waiting area of the clinic, 5 through word of mouth, and 1 via flyer. In total, 32 interviews were conducted, though 1 interview was not analyzed as the participant provided predominantly single-word responses that were not felt to enhance understanding of the youths perspective. In general, participants reported that the Check Yourself tool was easy to use and that colorful images and interactive content increased their interest in the health information that was presented: All participants indicated that they would prefer the Check Yourself tool to pencil-and-paper screening. Some adolescents particularly appreciated that questions were presented one at a time such that responses to previous questions were not visible. They felt this feature would help conceal their responses from family members in a waiting area: Adolescents described distinct ways that they felt the Check Yourself tool could enhance their interactions with doctors. Many participants found it easier and less awkward to disclose health risk behaviors on the tool than face-to-face, and perceived the tool as helpful in reducing providers need to ask patients about sensitive topics during the appointment.
Conclusion/Implications/Recommendations: With the opportunity to create screening tools that go beyond collecting information from adolescent patients by providing education and feedback, it is useful to seek youth input in order to design content that is acceptable and effective for this population. Adolescents in this qualitative study were specific about their preferences for electronic, personalized feedback on their health behaviors. Additionally, participants valued feedback that enhances their ability for self-management by facilitating goal setting and offering ongoing technological supports. Future quantitative outcome research should test screening tools that incorporate these suggestions.
140 Character Summary: Check Yourself is a teen-friendly eHealth tool designed to promote motivational discussions between adolescents and their providers about substance use.
EP01.04 - Reducing Data Burden and Improving Data Timeliness (ID 337)
Purpose/Objectives: CIHI provides comparable and actionable data and information that are used to accelerate improvements in health care, health system performance and population health across Canada. The health information environment is one of proliferating data and data sources, and the capabilities and needs of CIHIs data submitters and those who access our data have grown and are evolving. To keep pace with the industry, CIHI has initiated a multi-year program that will oversee and influence the modernization of data supply and access models across all of its data holdings. The goal of this program is to accelerate the modernization of CIHIs data supply and access through business process re-engineering and effective use of new technologies, in order to reduce data burden and improve timeliness.
Methodology/Approach: CIHI has drafted a white paper to inform a multi-year roadmap and blueprint for its modernization efforts. This focuses on nine key themes: CIHIs secondary use role in the health care system; what data should be collected; where this data should be sourced; how this data an flow in an automated and efficient manner; timeliness of data; adopting a one person, one record concept for secondary use; internal & external governance; vendor partnerships and data access.
Finding/Results: The near-term and long-term recommendations from each of these themes will be described during this presentation. Guiding principles that CIHI has established to steer its modernization efforts will also be discussed. These recommendations will support the development of a roadmap that will guide the major changes CIHI will be making to its supply and access chain in the coming years. These changes will impact all healthcare organizations that currently submit data to and consume data from CIHI as such, change management and stakeholder engagement will be key to successful transformation.
Conclusion/Implications/Recommendations: Modernization of CIHIs data supply and access flows will have system-wide impacts. A phased approach is being applied to ensure smooth change management, starting with establishing strong governance structures that will oversee and align the numerous projects that will be undertaken across the organization on this work. Additionally, key projects will be launched in high-value areas such as integrating and automating CIHIs acute and ambulatory care systems.
140 Character Summary: Hear about CIHI's multi-year strategy to oversee and influence the modernization of data supply & access models across the country.
EP01.05 - Growing and Developing Digital Health Competencies Through Educational EHRs (ID 288)
Purpose/Objectives: To support the effective design, development, adoption, and optimization of electronic health records (EHR), there is a considerable need for Health Informatics (HI) students to have hands-on experience and training with using EHRs. However, many HI students in Canada have very limited exposure to EHRs. This lack of exposure results in student challenges with developing HI competencies that are pre-requisites for other courses, co-op work terms, and eventually the workplace. Consequently, HI students may feel inadequately prepared to design, develop, implement, and support EHRs as graduates. The integration of an educational EHR into HI curriculum is central to developing students' digital health (HI) competencies. To meet this need, McMaster University adopted an educational EHR for its Health Informatics and Health Information Management (HIM) diploma programs in August 2016.
Methodology/Approach: As a part of the course redesign efforts for HTH 105: Information Systems and Technology, an educational EHR was adopted by the course developer and instructor. A three-hour scavenger hunt activity using the education EHR was included as a pre-requisite for the course. During the first week of the course in September 2017, HTH 105 students were asked to complete a 54-item self-assessment questionnaire based on COACH's "Health Informatics Professional Core Competencies v3.0" using a Likert scale of 1-5 (No Demonstrated Achievement to Exemplary Achievement). The course is providing students 1-2 hours of weekly hands-on exposure to the educational EHR, as well as supplementary activities to gain additional practice. To measure any changes in HI competency development at the end of the course, students will be asked to complete the same HI Competencies self-assessment questionnaire in November 2017.
Finding/Results: The response rate for the baseline HI Competencies Self-Assessment questionnaire was 67% (n=16/24). Of the respondents, almost all s have less than one year of experience in the HI field (n=14/16, 88%). The overall average across all HI competencies was 2.66/5. Although students reported a higher self-assessment average for basic clinical terminology, they reported health information systems adoption and use, EHR architecture, and systems evaluation as their weakest competency areas. Nearly all HTH 105 students (n=22/24, 92%) have adopted the educational EHR and are completing the hands-on EHR activities. Evaluation and monitoring of the benefits and limitations of the educational EHR is ongoing until the end of the course (November 2017). It is anticipated that the educational EHR will help to significantly increase students' self-assessment of COACH competencies, especially related to the EHR.
Conclusion/Implications/Recommendations: With the introduction of hands-on exposure to the educational EHR, HTH 105 students at McMaster University are experiencing an educational EHR's features, learning how clinical and administrative data is inputted and retrieved, observing its workflows, evaluating its usability, and making recommendations for improvements in EHR design and adoption. These skills are foundational to the development of COACH competencies for HI students. Educational EHRs should be integrated strategically into other HI courses and programs across Canada to support the applied learning of HI students. Further, there is a need to conduct pre and post-course evaluations of COACH competency development for HI students.
140 Character Summary: McMaster University's integration of an educational EHR in Health Informatics curriculum is helping to significantly increase students' COACH competencies.
EP01.07 - Data Processing for Vital Signs Detection Using a Functional Bedsheet (ID 303)
Purpose/Objectives: Describe the development of a functional bedsheet using a non-invasive method for monitoring respiration patterns and changes in respiratory rate. The functional bedsheet captures millions of data points across a range of fabric pressure sensors, providing rich insights on individuals health, sleep patterns, apnea, respiratory rate, among others. In this study, we will validate the vital signs detection of the functional bed technology through a clinical study using user-centered evaluation (UCE).
Methodology/Approach: Our research team will conduct a clinical study evaluating the participants' interaction with the functional bedsheet, their movement in the bed, comfort, and the accuracy of data collection during such movement in bed. Another important aspect of this study is the validation of the functional bedsheet in parallel with other monitoring equipment, thus understanding possible interactions or interferences. The functional bedsheet consists of multiple sensors that cover a wide area that can determine pressure variation, the system is designed as an ecosystem where each sensor acts independently, yet is dependent of the neighbouring sensors. The functional bedsheet will be validated against several other clinical-grade sensors (SpO2, ECG, etc.) Our research team will use the data collected to develop a machine-learning algorithm for detection of respiration changes and apnea patterns, helping translate raw sensor data into useful and meaningful insights for patients and clinicians.
Finding/Results: The expected results are the creation of a product that requires minimal interaction by the user (zero-effort technology), which is an important characteristic for our final user-population (older adults living independently). Additionally, our research program aims to use the data collected in the study to train the system to identify breathing patterns and diagnose apnea using machine learning. The technology is being designed for rapid diagnosis, prevention of associated diseases and reduction of visits to physicians.
Conclusion/Implications/Recommendations: Finding ways to improve the care and monitoring of patients in your own home is a necessary solution to reduce costs and increase the wellness of the user through uncomplicated equipment and easy access. The functional bedsheet aims to assist in this goal by making it possible to monitor vital signs and apnea continuously in the comfort of your bed.
140 Character Summary: Vital signs detection using a functional bedsheet and use of machine learning to identify respiratory patterns and apnea.
EP01.08 - The Clinical Impact of Workstation Single Sign-on (ID 16)
Purpose/Objectives: Following implementation of SSO, CHRISTUS set out to assess the impact in reducing clinician time logging in to various clinical software programs, and in financial savings from migrating to a thin client that enabled replacement of traditional hard drive computer workstations. A total of 65,202 logins across 6 CHRISTUS Health hospitals were sampled systematically during a 7-day period among 2256 active clinical end users for time saved in 6 facilities when compared to pre-implementation. Dollar values were assigned to the time saved by 3 groups of clinical end users: physicians, nurses and ancillary service providers.
Methodology/Approach: We share quantitative data on clinician time savings and recurrent computer hardware expenditure savings resulting from the implementation of SSO. We selected a 7 day observation period of SSO usage in May 2016 across 5 general community/general hospitals and 1 childrens hospital in Texas and Louisiana. There were 65,202 logins to the enterprise EHR. Potential SSO clinical users across the enterprise are 22,011. Of these, 5078 were based in the 6 hospitals evaluated, of which 2256 were active users (44.4%). Mean login durations were multiplied by the number of total first of shift and subsequent logins across all 6 hospitals for a 7 day period of evaluation and reporting in May 2016. We calculated the dollar cost savings or value in liberating time for each of 3 categories of clinicians utilizing SSO physicians, nurses and ancillary personnel. In our SSO implementations in these 6 facilities, 28% of clinical users were physicians,54% were nurses, and 18% were from ancillary departments. For nurses, we utilized the national average wage of $34.50. We collapsed physical therapists, dieticians and respiratory therapists into a single category of ancillary users and averaged their respective average hourly wages as $32.20. We collapsed all other medical specialties into a single category and averaged the physician hourly wage rate reported in the U.S. Department of Labor occupational employment statistical database rate ($95 per hour) with that reported by Beckers Hospital Review 2015 data ($165 per hour) for a rate of $130 per hour.
Finding/Results: *Results*: The reduction of total clinician login time over the 7-day period showed a net gain of 168.3 hours per week of clinician time 28.1 hours (2.3 shifts) per facility per week. Annualised, more than 1461 hours of mixed physician and nursing time is liberated per facility per year. The annual dollar cost savings of this reduction of time expended logging in is $92,146 per hospital per year and $1,658,745 per year in the first phase implementation of 18 hospitals. The computer hardware equipment savings due to desktop virtualization increased annual savings to $2,333,745. Qualitative value contributions to clinician satisfaction, reduction in staff turnover, facilitation of adoption of EHR applications, and other benefits of SSO have also been realised.
Conclusion/Implications/Recommendations: Based on this evaluation of the impact of SSO implementation,SSO is delivering substantial clinical value, recurrent annual ROI and net cost savings to the first 6 facilities implemented withinour hospital system.
140 Character Summary: Attend this session to learn about the research published in the International Journal of Medical Informatics on The Clinical Impact of Workstation SSO.
EP03 - e-Poster Session 3 (ID 54)
- Event: e-Health 2018 Virtual Meeting
- Type: e-Poster Session
- Track: Clinical Delivery
- Presentations: 8
- Coordinates: 5/29/2018, 09:30 AM - 11:00 AM, e-Poster Station 1, Parq Grand Ballroom, Conference Level
EP03.01 - Home Health Monitoring Patients Use Care 76% Less Than Average (ID 523)
Purpose/Objectives: In Canada, chronic diseases like heart and lung disease, cancer and diabetes cause 88% of deaths and consume 67% of direct healthcare costs. Much of these costs spring from emergency visits and hospital readmissions. A Canadian Institute for Health Information study found that 21% of discharged heart failure patients are re-admitted to hospital within 30 days, while 40% are re-admitted within 90 days. The same study discovered that 18% of COPD patients are admitted to hospital once a year; 14% are admitted twice. Clearly there is an opportunity to keep chronic disease patients healthier at lower cost, especially during that critical transition from acute to self-care. Home Health Monitoring (HHM) for chronic diseases such as COPD and Heart Failure are yielding exciting results. HHM solutions in the hands of community paramedics show great promise in helping seniors with chronic conditions in rural and remote communities. HHM programs let patients measure and manage their own health metrics at home while maintaining that critical daily connection with clinicians. In one provincial pilot, heart failure patients with HHM in place used healthcare services 76% less than heart failure patients not using home monitoring.  Noncommunicable diseases country profiles: Canada, World Health Organization, 2014.  Against the Growing Burden of Disease, Kimberly Elmslie, Centre for Chronic Disease Prevention, PHAC, 2012.  All-Cause Readmission to Acute Care and Return to the Emergency Department. Canadian Institute for Health Information, 2012.
Methodology/Approach: In Home Health Monitoring (HHM) programs, patients are set up with the devices they need to measure their pulse, blood pressure and other health metrics at home. Patients send data to an HHM nurse and answer questions through an online survey about how they feel. If clinicians have any concerns, they call patients, and vice versa. This dynamic lifeline lets patients heal more safely at home, where theyre healthier. It gives clinicians the tools and insights to address issues earlier, before they worsen.
Finding/Results: Patients in a recent provincial HHM pilot were very positive about the program: 97% would recommend it to others, with many wanting to hold onto the equipment beyond the pilot period. HHM enhances patients engagement in their own health. Many felt that both the equipment and clinician contact encouraged them to be more active in self-care, like following their action plan and doing daily recovery exercises. And as studies have shown, engaged patients become healthier patients. Ongoing communication also allowed earlier interventions, resulting in better outcomes at lower cost. The pilots 76% reduction in healthcare use bears this out. As compared with average heart failure patients, patients in the program incurred half the MSP billings of patients who were not in the program, 81% less acute inpatient days and 60% less emergency visits. And 86% of clinicians reported satisfaction with their ability to deliver care.
Conclusion/Implications/Recommendations: HHM is only in its infancy. The benefits of early pilots speak for themselves. Many more pilots are running today, each accompanied with rigorous evaluations. Their lessons are creating the rich store of insight that will help us know the very best way to offer HHM for the greatest benefit to patients in Canada and around the world.
140 Character Summary: Home health monitoring programs help patients with chronic conditions in urban, rural and remote settings stay healthier at home.
EP03.02 - Managing Senior’s Health Information in Assisted/Supportive Living Communities (ID 583)
Purpose/Objectives: Seniors are frequent users of the health care system due to their complex care needs. Managing these needs in long-term care or community care settings requires ongoing information sharing between the seniors, their family members, and their care team. Thus, digital technologies have the potential to help those involved with the collection, management, and sharing of health information. Our study objectives are to: (1) understand the health information technology needs and readiness of seniors living in assisted or supportive living communities, as well as the needs and readiness of their family members and the site staff involved in their care; and (2) to contextualize those needs in the operational context of assisted or supportive living communities.
Methodology/Approach: Our study is currently recruiting participants from three assisted/supportive living communities managed by The Brenda Strafford Foundation and Silvera for Seniors, located in Calgary, Alberta. We aim to have a sample of approximately 36 participants including: Residents (n=~12), resident family members (n=~12), and site staff (n=~12). Participants take part in semi-structured interviews where they answer questions about how they manage, interact, collect, and share resident health care information. Participants are also asked to demonstrate how they collect, share, manage and access resident health information through a contextual inquiry approach, which is filmed and included in the analysis. By combining these two research methods, we are able to hone in on both subjective and objective information, creating a rich basis for later project phases.
Finding/Results: Results from the semi-structured interviews and contextual inquiry will allow for a deeper understanding of the roles each participant group plays in the collection, management, and sharing of resident health information. These results will also help identify gaps in current practices that will inform later project phases.
Conclusion/Implications/Recommendations: Gaining a better understanding of current health information management processes, including identification of barriers that exist within assisted/supportive living communities, will help guide the development of digital health technologies that support the management and exchange of health information in these communities.
140 Character Summary: Understanding how seniors, families and care providers collect, manage, and share health information in long-term or community care contexts.
EP03.03 - Implementing a Provider Registry FHIR Interface (ID 490)
Purpose/Objectives: To design and implement an HL7 FHIR interface for the Provider Registry System (PRS) used in BC, AB, SK, and NL leveraging an existing application architecture.
Methodology/Approach: The Provider Registry System (PRS) is a foundational EHR component with a significant footprint in Canada. With over 16 years of service, PRS has seen considerable turnover of health integration standards. Predating HL7v3, the solution was designed with a proprietary messaging language. Over time, multiple iterations of the HL7v3 standard were also added as interface options, including support for current pCS versions, and leveraging Infoways Message Builder API. Lately interoperability talk is all about FHIR. While those existing HL7v3 integrations will still exist, how can legacy solutions like PRS, adapt to the next generation of digital health interoperability? Building on our experience with HL7v3 integration, we leverage our core, stable, messaging technology as an internal communication mechanism, and use the HAPI FHIR API to provide an external interface to our solution. To integrate with our existing data model and internal messaging formats, we follow a proven and adaptable design pattern for mapping and translation that isolates the API libraries from core PRS functionality.
Finding/Results: By leveraging existing application architecture, and design patterns, we were able to easily add FHIR interface options to our core solution. - By adopting the HAPI FHIR API, our solution is easier to maintain and extend. - We can implement revisions and new standards releases, or multiple versions with relatively low effort. - We offload the server interface components to those provided by the API - We focus our resources on semantics, data mapping, terminology, and our core business logic, rather than formatting, validation, parsing, and communication. While our solution includes a good alignment with the core FHIR model, there are some variances, and additional data in our model that is not part of the core standard. We chose to offer a pure, core standard interface as our primary offering and support an extended interface covering features unique to PRS through FHIRs extension capabilities.
Conclusion/Implications/Recommendations: By leveraging our existing application architecture and adding a quality FHIR API solution, we were able to add FHIR interface support to our exsing Provoder Registry solution. Our developers do not need to worry about the nuances of FHIR formats, or implementing a service layer to support it. We focused effort where it matters most; on mapping, data semantics, terminology, and our core business logic.
140 Character Summary: We extend the multi-jurisdictional Provider Registry System to include the HL7 FHIR standard by leveraging a proven application architecture and the HAPI FHIR API
EP03.04 - Real-Time Risk Scoring: Algorithms for Managing Expectations (ID 389)
Purpose/Objectives: Many efforts have been made to develop readmission prediction models using machine learning. Few, however, have actually been implemented. Significant barriers exist relating to the data access, technology, and change management required to meaningfully utilize real-time prediction scoring in a hospital environment.
Methodology/Approach: Data from an electronic health record system, outpatient pharmacies, outpatient clinics, census data, and more were combined to train a prediction model identifying heart failure patients at highest risk of 30-day readmission. These prediction scores were made available to clinicians and administrators involved in heart failure care at multiple acute care hospitals. They were used to help prioritize patients for interventions that included admission to a specialist heart failure unit, expedited outpatient follow-up, and targeted discharge teaching. Non-specialist physician services are often challenged to evaluate heart failure risk, and some inpatient units rarely see heart failure patients.
Finding/Results: The model development process highlighted that clinician opinions of what contributes most to readmission risk dont always align with the historical data. Most of the implementation challenges were not, however, related to model accuracy. Significant care variation exists between clinical services, units, and facilities because of differing resource availability and culture. This creates disparate demands and competing interests for how risk scoring should be used. It provided an opportunity for reflection on a number of key questions: Is predicting 30-day all-cause readmission always the right outcome measure? Is a risk scoring tool best suited to help bed placement clerks assign admission units or to support physicians making readiness for discharge decisions? Or to help prioritize rounding for clinical nurse educators? Can risk scoring successfully be implemented outside of EHR workflows? What level of readmission risk constitutes high? What level of model accuracy is sufficient? What are the most effective interventions that can be applied to heart failure patients to reduce readmission risk? Can some interventions simply be applied to all patients without the need for risk stratification?
Conclusion/Implications/Recommendations: Focusing on workflow and asking the right questions early in a project are critical elements to effectively exploiting machine learning for risk prediction. And equally important, knowing when to utilize other tools. Although incredible potential exists for improving workflow and clinical outcomes, successful implementation depends on a holistic approach to quality improvement and utilizing risk scoring alongside traditional efforts to standardize care pathways.
140 Character Summary: Effective use of risk stratification and machine learning requires careful management of competing interests. Lessons from a multi-site implementation.
EP03.05 - Age and Sex Patterns of Medical Care Utilization Through OTN (ID 456)
Purpose/Objectives: The Ontario Telemedicine Network (OTN) provides virtual health care services to the Ontario population, with an emphasis on those patients in underserviced regions. This research seeks to determine the associations between age and sex and medical care utilization rates faciliated by OTN and to determine if these associations are modified by geographical location.
Methodology/Approach: A historical, population-based, retrospective cohort study, utilizing record-level administrative billing data and census data was employed. Utilization was determined by Ontario Health Insurance Plan (OHIP) medical billing data that has OTN listed as the service location. Patient geography was determined by matching OHIP residence codes to Ontario census subdivision (CSD) codes. Patient geography was divided into four regions: rural north/south and urban north/south. North and south classifications were determined using the geographical boundaries defined by the Local Health Integration Networks and urban and rural classifications were defined using Statistics Canada's Statistical Area Classification System. Each OTN visit creates two or more distinct OHIP codes: one telemedicine encounter premium code and at least one specified medical service code. The telemedicine premium codes were used to determine general utilization and the specified medical service codes were used to determine specialty utilization. Medical service codes were organized into therapeutic areas of care. The top utilized specialties were distinguished and reported. The remaining specialties were grouped into an other category. Crude and adjusted utilization rates were calculated and associations between the exposures, outcome, and effect modifier were analyzed using multivariate Poisson regression.
Finding/Results: There were 885,761 completed patient sessions facilitated through the OTN from 2008/2009 to 2014/2015, providing care to 185,061 unique patients throughout the province. Provincial utilization rates were highest in family and general practice, followed by addiction medicine. Addiction medicine was approximately 7-fold higher in young adults (20-44 years of age) residing in Northern Ontario compared with young adults in Southern Ontario. Utilization rates among children (0-19 year of age) residing in rural Ontario were highest in psychiatry and dermatology services, whereas children residing in urban Ontario utilized family and general practice services most frequently. Oncology, internal medicine and surgery services had the highest rates of use in male, older adults (65 years of age and older) residing in Northern Ontario.
Conclusion/Implications/Recommendations: This project provides a detailed description of telemedicine use in Ontario by sex, age, rurality, and region. While the majority of completed patient sessions through OTN are related to addiction medicine, the findings of this study highlight the variability of telemedicine usage by sex and age across the province. Results provide insight for e-health networks, such as OTN, and local health networks on what services are being utilized and by whom. Future research will investigate OTN service utilization trends, particularly in older adults, and compare OTN service rates with provincial, in-person care services.
140 Character Summary: Telemedicine services facilitated by OTN have increased access to specialized medical care for all patients, young and old, especially in Northern, rural Ontario.
EP03.06 - Breathing Easier from Home: Telehomecare Results (ID 448)
Purpose/Objectives: Chronic Obstructive Pulmonary Disease (COPD) has the highest rate of admissions among major chronic illnesses in Canada, as well as the highest readmission rate. 18% of people living with COPD are admitted to hospital once a year, while 14% are admitted twice . Canadian hospitals are admitting more clients with COPD flare-ups than heart attacks. COPD is the only chronic disease with an increasing mortality rate. Home health monitoring (HHM) is gaining recognition as a clinically transformative health service that supports and accelerates transitions to community-based, integrated primary care. A pilot project introduced HHM to clients to help them manage their condition from home. HHM helps clients with chronic diseases, their caregivers and families access services, enhance understanding of their condition, and reduce hospital visits. Ultimately, HHM results in high levels of client satisfaction, with 100% of clients surveyed recommending HHM to others. HHM for COPD has been operational since November 2016, and is currently in the evaluation phase. The evaluation framework has been developed and is being executed; as client data is evaluated and analyzed, results will be presented.  Canadian Institute for Health Information, All-Cause Readmission to Acute Care and Return to the Emergency Department (Ottawa, Ont.: CIHI, 2012)
Methodology/Approach: Best practice for solution design and delivery was followed for the implementation of HHM for COPD. Team members are now analyzing extensive client feedback, evaluating impact on acute care utilization by these clients, evaluating overall program utilization, and identifying potential areas of improvement. Below are the key components of the evaluation phase: Client Surveys: Telephone interviews to capture qualitative data on HHM program experience. HHM Clinician Surveys: Online survey capturing data on implementation, training and overall program deployment. Acute Care Utilization Evaluation: Data regarding emergency department visits, inpatient care admissions, and length of stay for clients before, during and after HHM will be analyzed. Only encounters specific to COPD will be included. Program Utilization (Referral) Evaluation: Look at referral data and areas for improvement.
Finding/Results: Preliminary analysis from client and clinician survey data shows that the majority of HHM clients and clinician users are highly satisfied with the HHM service. Detailed results will be shared. Analysis of the acute care utilization data by HHM COPD clients is expected to show a decrease in emergency department encounters, decreased inpatient admissions and shortened inpatient length of stay for COPD-related visits during and after HHM participation. Detailed results will be shared. Referral analysis is expected to show areas for improvement in recruiting and maintaining clients that are relevant to all telehomecare programs in Canada. The referral evaluation is also expected to show the barriers to achieving increased benefits in acute care utilization and client satisfaction. Detailed findings and recommendations for effective HHM programs will be shared.
Conclusion/Implications/Recommendations: HHM projects have shown a positive impact on clients living with COPD and their care teams. Providing better access to patient health information, particularly for patients that are located in rural and remote areas, can reduce healthcare system costs, increase patient satisfaction and improve overall population health. With the number of COPD cases on the rise in Canada, solutions such as HHM will become increasingly important.
140 Character Summary: HHM for COPD aims to increase service effectiveness, efficiency and capacity to manage high-risk clients in the community.
EP03.07 - The Importance of a Dedicated Education Team in Health Informatics. (ID 503)
Purpose/Objectives: The Fraser Health Authority department of Health Informatics (HI) manages the Health Care Information System (HCIS) for Acute Care. Clinical Informaticists (CIs) work to manage the Acute Care HCIS; in addition to working with stakeholders to plan, design, implement, review and evaluate HCIS processes, CIs create, conduct, evaluate, and revise HCIS education and training. In the HCIS, are eight modules; CIs are expected to be module experts for two of them. Leadership consisted of four coordinators who led two modules each; and report to a department manager. This meant CIs were reporting to two coordinators and at times this lead to uneven workload. Furthermore, this resulted in inconsistent development and delivery of education; for example, only one of the four modules education materials had identified learning objectives. In February 2017, the HI department reorganized into four teams. Three teams had two to three modules and eight CIs; and one team was a team with five CIs dedicated to creating, evaluating, and revising HCIS education. The purpose of this presentation is to highlight the benefits of a dedicated education team for HI.
Methodology/Approach: The education team was developed from employees within Health Informatics who expressed interest in education. Team members were encouraged to complete the Educator Pathway, a 4 level education model to facilitate advancement of education within BC health authorities. This allowed the incorporation of principles of adult education into the training materials. Existing training scripts were reviewed and learning objectives created in consultation with module teams responsible for content. Education team members are integrated with specific module teams, attending module meetings, anticipating and facilitating modifications to training materials ensuring currency and comprehensiveness. Formation of the education team has led to collaboration with stakeholders including Online Learning Services, Student Placement and Professional Practice.
Finding/Results: Scripts including learning objectives enable new hires within HI to grasp key content when learning to instruct courses. Computer application instruction to health authority staff is identified as the most important function of the department and staff report greater confidence in providing education with learning objectives contained within scripts. The separate education team has reduced module team burden of maintaining and updating training materials, allowing them to focus on module-specific concerns such as planning, building and testing new application functionality. A dedicated education team has also enabled creation of two computer based courses, which are so successful in providing flexible education that HI is now expanding its complement of computer based training.
Conclusion/Implications/Recommendations: Limited department resources have been effectively managed through the creation of an HI department education team. Training resources are updated concurrently with system changes to ensure accuracy of content introduced to learners. A designated education team has removed the requirement of the module teams to update training materials, allowing them to focus more attention on modules. Module teams have shown acceptance and appreciation for this support. Importantly, having a designated education team allowed the department achieve its goal of creating online learning modules, which will be expanded upon in the months and years ahead.
140 Character Summary: Dedicated education resources have reduced burdens while enhancing education materials and teaching perspectives within the Health Informatics department.
EP03.09 - Health Services Integration: The Role of Virtual Community (ID 136)
Purpose/Objectives: Health Services integration is a challenge in healthcare; the continuum of care extends beyond the point of care to the social services and social determinants of health (e.g. housing, income). However, the current eHealth tools target the healthcare delivery within a strict biomedical model of health that excludes the social determinants of health; the result is translated in inefficiencies and higher cost (e.g., patients that finished their episode of care but cant be discharged without organizing services that address their social needs once discharged; people whose health needs - such as mental health - are linked to the fulfillment existence of social needs). A truly patient centered healthcare need to take into account the social determinants of health. While social services exist in Canada; eHealth solutions that integrate biomedical as well as social care around the patient are still lacking. This paper argues that Health Virtual Communities enable professionals to coordinate biomedical and social care around a patient; such eHealth tools can contribute to a better healthcare delivery while overcoming some inefficiencies in the health system.
Methodology/Approach: Health Virtual Communities research published recently in a book by the author (Novel Applications of Virtual Communities in Healthcare Settings) show effectiveness and efficiency of health virtual communities in diverse areas including mental health, cancer care, pain management, personalized medicine, elderly care, and community engagement.
Finding/Results: Health Virtual Communities enable individuals with common health-related conditions to provide mutual social support, and enables healthcare providers to provide care at a distance especially in chronic disease management. Research has demonstrated effectiveness and efficiency effects linked to Health Virtual Communities for multiple health conditions (chronic kidney disease, pulmonary hypertension, cancer, pain management, elderly care), especially the virtual community members engagement. As an eHealth tool, Health Virtual Communities enable coordination of care among healthcare providers by sharing the patient's care needs and coordinating their actions. They enable providers to discover services available and book services for the individuals under their care.
Conclusion/Implications/Recommendations: Health Virtual Communities provide an opportunity to coordinate biomedical and social care in order to support a truly patient-centred care. They can support truly coordinated care across a holistic continuum of care that includes the social dimension.
140 Character Summary: Health Virtual Communities provide an opportunity to coordinate biomedical and social care in order to support a truly patient-centred care.
EP05 - e-Poster Session 5 (ID 56)
- Event: e-Health 2018 Virtual Meeting
- Type: e-Poster Session
- Track: Clinical Delivery
- Presentations: 8
- Coordinates: 5/29/2018, 02:00 PM - 03:30 PM, e-Poster Station 1, Parq Grand Ballroom, Conference Level
EP05.02 - Devops is Coming (ID 496)
Purpose/Objectives: This discussion will provide you with a common understanding of DevOps, its approach, methodology, and tools, as well as the benefits and implications for the management of your IT systems. But implementing DevOps will also extend into how you hire and motivate your staff, how you structure your organizations roles, how you approach governance and decision making, how you deliver your services, how you layout your facilities and choose to co-locate your staff.
Methodology/Approach: We will speak briefly to this image to explain DevOps:
Finding/Results: The practical benefits of DevOps and what it means for eHealth 1. Shorter development cycles, faster delivery of features and fixes a. Means that new or enhanced capabilities are available more frequently for your clients. 2. Reduced deployment failures, rollbacks, and time to recover a. Means less interruption to dependent internal processes and shorter downstream system outages. 3. Increased efficiency through automation of tasks from code development through to deployment a. Means you achieve consistency, reliability, and auditability of IT processes and more time for your teams to spend on higher level tasks. 4. Increased collaboration and reduction of team and business silos a. Means a reinvigoration of your software engineering teams as they find new ways to innovate to meet business needs.
Conclusion/Implications/Recommendations: DevOps is coming. If you want to implement DevOps successfully and reap the benefits, then you must think about changing the culture of your organization to allow it to flourish. 1. Encourage experimentation and innovation so teams can Fail Fast, Fail Small, Recover Fast. 2. Shift the traditional concepts of accountability and responsibility so people feel empowered to embrace the idea of expanding the boundaries of their work and the definition of done. 3. Redefine recruiting, professional development and traditional career paths to account for a wider range of skills and experience. 4. Break down the boundaries between teams, even to shifting how you design the workspace to promote collaboration. 5. Communicate with your clients to include them in the changes and get them excited about the benefits they will get from the faster release of features.
140 Character Summary: DevOps is inevitable in the way IT systems are developed and operated in the future; the implications extend beyond IT to an organization's culture and operations.
EP05.03 - Streamlining a Multi-Facility Electronic Health Record (EHR) Change Request Process (ID 365)
Purpose/Objectives: Establishing a joint EHR change request process in a HIS cluster was integral to ensuring clinical standardization and adherence to system change control processes. In the past, change requests were managed using Microsoft Word documents that were stored on a joint SharePoint site. Challenges arose: understanding the process; version control; determining approval status, auditing changes; reporting common characteristics. These challenges presented an opportunity to develop an electronic solution to create, track, and store EHR change requests.
Methodology/Approach: Software was utilized to create an EHR Application Change Request form and portal as shown below: This software enables EHR change requests to be created and tracked electronically. It allows for business intelligence rules to be built in, facilitating adherence to the change request workflow, such as clinical committee approvals at the applicable time in the review process. The solution allows for each change request to flow through the approval process in the correct manner before being flagged as ready for adoption into the live EHR system. The solution also enables discrete data collection and reporting, and includes an audit trail that is useful for reviewing the changes to the change request form. The system also allows for keyword searching and filtering which is valuable to view historical changes. The solution use eliminates the need to have copies of Word documents and hyperlinks on agendas as the one electronic copy that everyone can view and contribute to is used. The use of business intelligence rules facilitates the automation of the change request process, automatically adding the change request to the required committee agendas.
Finding/Results: Using an electronic solution for change requests has resulted in increased compliance of change requests. Efficiencies in time savings were found as Clinical Informatics staff no longer need to submit change requests to be added to agendas and administrative support was no longer require to compile committee agendas or for the storing of completed change requests. Staff also reported gains in being able to understand where each of their submitted change requests were in the change request process and were better able to report on workload and timelines. Staff reported increased satisfaction in using the new software.
Conclusion/Implications/Recommendations: Introducing an electronic solution for EHR change requests in a multi-hospital HIS cluster environment is a robust solution compared to the manual creation, tracking, agenda creation, and storing of change requests using Microsoft Word. The ability to use business intelligence rules allowed for the change request process to be built into the software producing increased compliance to the process and significant efficiencies. The adoption of an electronic solution for EHR change requests is recommended.
140 Character Summary: A multi-facility EHR collaboration required a new EHR change request process. A new electronic system was successfully implemented and is receommended.
EP05.04 - Alberta EHealth Support Services: Evolutive Adoption and Support (ID 421)
Purpose/Objectives: Alberta is on an exciting journey, with Alberta Healths mission to set policy and direction to lead, achieve and sustain a responsive, integrated and accountable health system. Providing the right information in order to for the right caregivers to provide the right services to citizens in the right setting and the right time is a hallmark of an advanced and proficient eHealth program. This is achieved through Alberta eHealth. What is the eHealth Delivery Model and why are eHealth Support Services important?
Methodology/Approach: The eHealth Delivery Model was introduced in 2015. It is a new comprehensive and flexible model, methodology and approach, designed to meet the client goals and objectives by focusing on continuous improvement, innovation, automation, mobility, self-support and sustainability. The CGI eHealth Support Services (eHSS) Team supports the Alberta Health Digital Health initiatives and eHealth technologies to enable improvements in the delivery of health services in Alberta. Additionally the team support the enhanced functionality of Alberta Netcare to healthcare providers, clinical students, and the Personal Health Portal to Albertans.
Finding/Results: The graphics depict the evolutive nature of the model and a high level overview of where the benefits were realized.
Conclusion/Implications/Recommendations: Benefits were quickly realized. Five months into the contract CGI implemented the eHealth Avaya Contact Centre, receiving over 6,000 inquiries from healthcare providers. Over the first 4 months, it did not take long for physicians, nurses and other healthcare providers to gain a high confidence in the eHSS professional team of eHealth Consultants. The next benefit was at the 18 month mark when cost savings of $675K, in travel and training expenses were realized as a result of the new eHSS Delivery Model. Alberta Health quickly turned those savings back into eHealth, as new revenue to provide additional effort to the support new eHealth initiatives.
140 Character Summary: The new eHealth Delivery Model supports entire value stream of eHealth initiatives to deliver quality support to healthcare providers.
EP05.05 - Efficiency and Clinical Effectiveness of Telehomecare in Insulino-Treated Diabetic Patients (ID 156)
Purpose/Objectives: Prevalence of diabetes has drastically increased in the last decade. Clear correlation between elevated A1c and diabetes associated morbidity and mortality has been established. Unfortunately, over 50% of North American patients with diabetes have difficulty to control their sugar levels and consequently have an A1c above target. In this context, use of hospital resources for education and management of diabetic patients increased markedly. Given its chronic nature, the patient's involvement in the management of its diabetes on a daily basis is essential. The Telehomecare (THC) project was initiated at the ambulatory clinic of the Centre hospitalier de l'Université de Montreal (CHUM) in order to improve the follow-up of diabetic patients treated with insulin while maintaining the same quality of care and patient's satisfaction. The objectives are to evaluate the impact of THC on organizational efficiency, clinical efficacy and patients' satisfaction compared to the standard care in diabetic patients treated with insulin. Main outcome is to analyse impact of THC on diabetes management during the first 3 months of use of the teaching platform and 3 months after THC was stopped.
Methodology/Approach: A prospective non inferiority and controlled clinical trial was designed. A total of 105 patients was assigned to either an intervention group provided with a THC system during 3 months for transmission and online analysis of capillary glucose (n=55) or to a control group receiving standard care (n=50). Patients in the THC group either use a dedicated tablet provided by healthcare provider, or access services through a web browser via their personal computer/tablet/phone. Patients use THC to register their daily health activities personalized by their care provider and receive real-time feedback. Clinical data are securely transmitted, stored and accessible to caregivers allowing a prompt therapeutic adjustment if needed.
Finding/Results: Preliminary results: 49 patients in THC group and 22 patients in the control group have, to date, completed the study. THC group had an average of 0.6 medical visit compared to 1.0 in control group. However, an increase in nursing interventions (mainly emails or phone calls) was noted in THC group (n=14.8) compared to control group (n=1). A significant decrease in A1c levels was observed at 3 months with a reduction of -0.66% in THC group compared to -0.12% in control group. The program targeted diabetes type 1 and 2 patients with no discrimination on duration since diagnosis. One surprising result from the program is that even expert patients, with long-term experience managing their condition, seemed to have obtain improved results and knowledge from undergoing the program. Satisfaction in care was similar in both groups.
Conclusion/Implications/Recommendations: These preliminary results suggest an improvement in glycemic control in patients followed by THC, compared to usual care, which can be explained by an intensive glycemic monitoring and a rapid therapeutic adjustment. Also, it suggests a decrease in the number of medical visits needed for patients followed by THC. The final analysis of the data is currently pending to determine the impact on glycemic control at 6 months and the cost-effectiveness of THC compared to usual care.
140 Character Summary: E-Health Medicine to Improve Diabetes Management (Care and Cost Effectiveness): Evaluation of Telehomecare Management in Diabetic Patients on Insulin Therapy
EP05.06 - Evaluation of a Telehomecare Initiative Used in Gestational Diabetes Women (ID 157)
Purpose/Objectives: The prevalence of gestational diabetes mellitus (GDM) increases steadily in recent years. Pregnant women with GDM are at risk for obstetrical and neonatal complications. These patients therefore require close multidisciplinary follow-up to ensure proper management of GDM, which implies many appointments for patients and a significant use of hospital resources. The telehomecare (THC) initiative is a clinical Remote Patient Management project led by the Endocrinology division at the Centre Hospitalier de lUniversité de Montréal (CHUM). Primary objective is to evaluate the ability of THC to decrease medical visits in clinics or at the emergency room. Secondary outcomes are: 1) to improve diabetes control and possibly limit GDM related complications, 2) to increase access to care and dedicated team, 3) to evaluate patient's satisfaction and empowerment, and 4) to evaluate cost effectiveness of THC by assessment of direct and indirect costs.
Methodology/Approach: A prospective noninferiority and controlled clinical trial was designed. A total of 161 women was assigned to either an intervention group provided with a THC system for transmission and online analysis of capillary glucose data (n=80) or to a control group receiving usual care in clinic (n=81). Patients in the THC group were either given a tablet to use or an access to a web patient portal via their personal computer/tablet/phone. They register their daily health activities as outlined by their care plan. Automated and adjusted feedback is provided directly to the patient by the platform to promote better health behaviours. Clinicians will receive alerts based on predetermined algorithms and can adjust therapy quicker upon the symptoms and results registered by the patient. An innovative aspect of this project is that THC is integrated within traditional care as the responsibility for the remote care of patients is assigned to nurses working on the unit. Another innovative aspect is the delivery of personalize and interactive coaching through the technology. This automated and individualized process upon patients results generates motivation by promoting better health condition thus, facilitating patients empowerment.
Finding/Results: The results are the outcome of a large collaboration, led by clinicians (doctors, nurse, nutritionist, etc.) and patient partners, supported by a group of technologists and implemented in a platform that supports clinical processes. Preliminary result (n=80 THC and n= 81 Controls) indicate that patients in the THC group had an average of 1.45 compared to 3.32 medical visits, a decrease of 56%. However a 80% increase in nursing interventions (mainly by emails or phone calls) was noted. Maternal and fetal outcomes were similar between groups demonstrating that by using THC, there is no loss in the quality of care. Satisfaction in care was similar in both groups (9/10) however women in the THC group felt that access to their clinical team was easier.
Conclusion/Implications/Recommendations: THC monitoring appears to significantly reduce medical visits as well as improving access to care without compromising pregnancy outcomes and patients satisfaction. However, the impact on maternal and fetal complications rates and cost-effectiveness remains to be analyzed. The final analysis to establish statistical and clinical significance is currently pending.
140 Character Summary: Telehomecare Initiative to Ease Gestational Diabetes Mellitus Management in Pregnant Women Significantly Reduce Medical Visits and Improve Access to Care.
EP05.07 - Network Effects in Telemedicine (ID 609)
Purpose/Objectives: Over the past decade, a regional telemedicine provider in Canada has developed one of the largest telemedicine networks in the world. Last year, over 750,000 clinical events took place across the network. Over the past few years, the organization has gradually shifted its core business model from managing a telemedicine utility to developing a virtual healthcare platform. The platform business model leverages network effects to drive value and growth.
Methodology/Approach: With the shift to a platform business model, the organization has broadened its focus beyond simply providing telemedicine services over its network to catalyzing clinical program development with partner organizations, driving the spread of successful telemedicine business models across the province, disseminating information about new innovations and best practices, enabling and supporting new communities of practice, and creating business intelligence for its users.
Finding/Results: The repositioning of the organization from a service provider to a platform manager and redefinition of the network from technical to social in nature, has allowed it to leverage the innovation of its partners; drive the spread of best-in-class solutions and align the evolution of telemedicine services to health system priorities and transformational initiatives.
Conclusion/Implications/Recommendations: By removing itself from the centre of business processes such as product development, adoption and change management, and customer care, the value proposition of telemedicine in the province is rapidly increasing.
140 Character Summary: The adoption of a platform business model and leveraging of network effects has dramatically increased the value proposition of a provincial telemedicine system
EP05.08 - User Receptivity of BoosterBuddy: An App for Mental Health Disorder (ID 598)
Purpose/Objectives: BoosterBuddy is an app designed for youth between the ages of 15 and 24 years of age, and is intended to support and increase their self-management skills for mental health. BoosterBuddy uses gamification (e.g., quests, coins, levels) to improve motivation and to engage in behavioural activation focused activities. The study had the following two objectives: 1) To describe users feedback about the app (e.g., do users like it) and frequency of use, and 2) To examine whether users recommend the app to others via social media platforms. Each objective included different methodology, as outlined below.
Methodology/Approach: Objective 1: 1,047 unsolicited user feedbacks were coded using two complementary coding methods: first, using the Mobile App Rating Scale (MARS; Stoyanov et al., 2015), which is a framework used to evaluate the quality of apps; second, using predetermined categories that were created using a small subset of the data, where similar feedback was grouped to create coding categories that are specific to the BoosterBuddy app. The app was also evaluated using online questionnaires, where 118 app users responded to an online questionnaire that assessed demographics, app usage (e.g., frequency), and users satisfaction with BoosterBuddy. Objective 2: Examined social media posts from Twitter, Instagram, Tumblr, and a variety of vlogs and blogs between September 2014 and April 2017 for mention of recommending the app.
Finding/Results: Objective 1: Coding the unsolicited feedback indicated that 67% of app users reported that BoosterBuddy was helpful, and 49% found features of the app to be well-targeted to their demographic. The 188 users who completed the online questionnaire were 28 years old on average (SD = 10: min-max: 19-62) and mostly women (70%), gender fluid (7.7%), transgender man (6.8%), and dont know (5.1%), other (4.3%), and man (3.4%). A total of 33% were of British descent or selected other as their ethnicity (31.4%). Majority of users (75%) reported using the app for 1-10 minutes per day and 64% reported that the app was extremely/very useful. Table 1 outlines the features of the app and the percentage of users who rated each feature as useful. Critically for behavioral activation, ~60% reported agree/somewhat agree that BoosterBuddy helped them start their daily tasks. Objective 2: Trends emerged when observing the unsolicited feedback provided through social media platforms, with an overwhelming number of users recommending the app. Table 1. BoosterBuddy features and the percentage of users who reported the feature to be useful 'extremely useful' or 'very useful' (N=118) BoosterBuddy Feature Percentage of users who reported the feature to be useful Coping Library 62.7% Crisis Plan 39.8% The Quests 52.6% The inspirational quotes and encouragements 61.9% The check-ins about how I am feeling 64.4%
Conclusion/Implications/Recommendations: Users of BoosterBuddy report that the app is helpful and indicated some recommendations for improvement. A limitation is that the data is self-report and conclusions cannot be made about the impact of the app on changes in users mood and personal well-being.
140 Character Summary: Users of BoosterBuddy report high frequency of using the app and that it is useful for managing their mental health. The users had some suggestions for improvements.
EP05.09 - Exploring CIHI's Long-Term Care Quality Indicators Using Big Data Analytics (ID 324)
Purpose/Objectives: CIHIs Information Quality Framework includes quality monitoring and control activities that focus on quality at every stage of the information life cycle. Traditional activities employ simple data validation (e.g. checking for missing or invalid values) during data capture, submission and processing. Data surveillance is a newer quality approach that monitors information outputs (e.g. performance indicators) through a cyclical process of analysis, reporting, review, investigation and response and aims to mitigate the impact of data use on data quality. CIHI is using big data analytics and surveillance techniques to examine changes in health system performance indicators reported in its Your Health System (YHS) web tool. In collaboration with Health Quality Ontario (HQO), we are beginning with the Potentially Inappropriate Use of Antipsychotics in Long-Term Care indicator, which is based on information in CIHIs Continuing Care Reporting System.
Methodology/Approach: Regression modeling, trending analysis and facility-level comparisons determine whether indicator changes reflect changes in clinical practice or changes in data quality. A report summarizing the analysis highlights provincial-level findings and LTC homes with pronounced changes in data related to the indicator. Homes are engaged in a collaborative report review process that enables discussion and identification of the reasons for change, which will ultimately influence quality improvement across the sector.
Finding/Results: Indicator results show that, in Ontario and across Canada, there has been a decrease in potentially inappropriate use of antipsychotics over the last 5 years. Regression analysis confirmed there has been real system-level change in clinical practice and reduction of antipsychotic use among residents, particularly those with dementia. However, there also appear to be some changes in coding and data quality. There was a decrease in denominator cases due to an increase in the coding of exclusion criteria (mostly delusions, hallucinations, end-stage disease and hospice care). Some of this change may be due to improved data quality (e.g., more comprehensive assessment of risk factors that exclude residents from the indicator), but some variation may also be due to differences in understanding of the data (e.g., lack of clarity in assessing delusions and hallucinations). This is evident in the variation seen across homes in the prevalence of exclusions, with some homes having larger increases over time than others. Some of these homes also exhibited unusual changes in data related to the indicator.
Conclusion/Implications/Recommendations: Following up with LTC homes is essential to understanding the underlying factors associated with the observed data changes, and both CIHI and HQO have a role to play in this. The information gathered will influence and shape next steps, which could include the development of clear coding guidelines and education related to the exclusion criteria, and ongoing data monitoring. The information will also provide insights to surveillance analytics by strengthening statistical models and enabling identification of areas requiring further data quality improvement. It is important that all homes benefit from the data quality efforts undertaken by individual homes. Collaboration, openness and transparency are key to the success of data surveillance and ensuring continued trust in using the data for performance improvement.
140 Character Summary: CIHI is exploring using big data analytics to monitor changes in the quality of data behind it's Your Health System performance indicators.