Presentation
DH3 - Challenges and Solutions of Developing Usable EHR Alerts
DescriptionThe development and implementation of patient care plans is a process encompassing significant levels of information flow, whose processing and synthesis can generate up to 50 million gigabytes of data each year (World Economic Forum, 2019). The clinicians involved in this process are subsequently tasked with decomposing this multi-faceted and multi-sourced information into clear and actionable care decisions. Subsequently, significant efforts have focused on identifying gaps in care and developing tools and technologies to assist clinicians in this work. These efforts have led to findings that show substantial safety and organizational health risks due to incomplete or missed information during care. For example, prior literature indicates that clinicians must navigate around the fact that upwards of 60%-70% of all prescribed medications in hospital settings have moderate to severe interaction risks (Rodrigues et al., 2017). Similarly, information-gathering processes to form clinical decisions suffer from disproportionate rates of unnecessary blood draws and lab tests (Vrijsen et al., 2020; Lewandrowski, 2011). Fortunately, many interventions have been implemented to assist clinicians in organizing, interpreting, and acting on information. One intervention, and the focus of this panel submission, is the Electronic Health Record (EHR) and Clinical Decision Support (CDS) alert systems.
EHR alerts represent informational and interactive popups within a medical database interface to inform the user about a patient's health or warn about a possible future health state. These alerts have a variety of functionality and intractability that range from informational (e.g., no required interaction), forced interaction (e.g., the user must select some option), or forced stop states (e.g., preventing an action from taking place within the system). They are applicable in a variety of settings, including during patient orders (e.g., medications and labs), information review (i.e., alerting that a patient has not received a medication), and also during non-direct patient care (i.e., alerts during supply management). Unfortunately, significant barriers to successfully implementing these alerts diminish their effectiveness (Jung et al., 2020). These include but are not limited to, 1) the intricate knowledge needed to interact with and implement alert system parameters, 2) knowledge and skill in developing in HTML, CSS, and Javascript, 3) lack of usability and design guidance for alert displays, and 4) the challenge of building alerts that attune to both the trigger scenario and the mental model of the clinician at the time of activating.
These barriers have led to an overwhelmingly lackluster response from the providers in their interaction and adoption of alerts into their workflow. Many reports show significant rates of user dissatisfaction in their interactions with EHR alerts, which may be reflected by the suggestion that up to 90% of pharmacists and physicians choose to overwrite these alerts (Brodowy & Nguyen, 2016). Alternatively, the clinicians may acknowledge alerts when triggering, but the required action may not be appropriate or available reflecting the challenges of implementing alerts in an appropriate and usable manner (Iii et al., 2020). In response, research initiatives have called for Human Factors and Human-Centered Design (HCD) approaches to combat these challenges and inform the development of practical and applicable alerts. Unfortunately, little focus has been applied to developing practical and implementable guidance for the analysts who create and apply alerts within hospitals.
This poster will present and discuss the challenges and applicable solutions for developing, designing, and implementing EHR alerts into practice within a hospital setting. Multiple perspectives will be represented via a frontline view from our Hospital’s Clinical Decision Support and alert development committee member and Director of Provider Informatics, an operational and implementation approach from embedded human factors and user experience practitioners, and an academic and research approach via user experience researchers.
References
Brodowy, B., & Nguyen, D. (2016). Optimization of clinical decision support through minimization of excessive drug allergy alerts. American Journal of Health-System Pharmacy, 73(8), 526–528. https://doi.org/10.2146/ajhp150252
Iii, J. D. M., Mallozzi, C. P., Perkins, R. M., Shelov, E., & Schreiber, R. (2020). Reducing Alert Burden in Electronic Health Records: State of the Art Recommendations from Four Health Systems. Applied Clinical Informatics, 11(1), 1–12. https://doi.org/10.1055/s-0039-3402715
Jung, S. Y., Hwang, H., Lee, K., Lee, H.-Y., Kim, E., Kim, M., & Cho, I. Y. (2020). Barriers and Facilitators to Implementation of Medication Decision Support Systems in Electronic Medical Records: Mixed Methods Approach Based on Structural Equation Modeling and Qualitative Analysis. JMIR Medical Informatics, 8(7), e18758. https://doi.org/10.2196/18758
Lewandrowski, K. (2011). POC testing in the emergency department: Strategies to improve clinical and operational outcomes. Acutecaretesting. https://acutecaretesting.org/en/articles/poc-testing-in-the-emergency-department-strategies-to-improve-clinical-and-operational-outcomes
Rodrigues, A. T., Stahlschmidt, R., Granja, S., Pilger, D., Falcão, A. L. E., & Mazzola, P. G. (2017). Prevalence of potential drug-drug interactions in the intensive care unit of a Brazilian teaching hospital. Brazilian Journal of Pharmaceutical Sciences, 53, e16109. https://doi.org/10.1590/s2175-97902017000116109
World Economic Forum. (2019, December 5). Four ways data is improving healthcare. World Economic Forum. https://www.weforum.org/agenda/2019/12/four-ways-data-is-improving-healthcare/
EHR alerts represent informational and interactive popups within a medical database interface to inform the user about a patient's health or warn about a possible future health state. These alerts have a variety of functionality and intractability that range from informational (e.g., no required interaction), forced interaction (e.g., the user must select some option), or forced stop states (e.g., preventing an action from taking place within the system). They are applicable in a variety of settings, including during patient orders (e.g., medications and labs), information review (i.e., alerting that a patient has not received a medication), and also during non-direct patient care (i.e., alerts during supply management). Unfortunately, significant barriers to successfully implementing these alerts diminish their effectiveness (Jung et al., 2020). These include but are not limited to, 1) the intricate knowledge needed to interact with and implement alert system parameters, 2) knowledge and skill in developing in HTML, CSS, and Javascript, 3) lack of usability and design guidance for alert displays, and 4) the challenge of building alerts that attune to both the trigger scenario and the mental model of the clinician at the time of activating.
These barriers have led to an overwhelmingly lackluster response from the providers in their interaction and adoption of alerts into their workflow. Many reports show significant rates of user dissatisfaction in their interactions with EHR alerts, which may be reflected by the suggestion that up to 90% of pharmacists and physicians choose to overwrite these alerts (Brodowy & Nguyen, 2016). Alternatively, the clinicians may acknowledge alerts when triggering, but the required action may not be appropriate or available reflecting the challenges of implementing alerts in an appropriate and usable manner (Iii et al., 2020). In response, research initiatives have called for Human Factors and Human-Centered Design (HCD) approaches to combat these challenges and inform the development of practical and applicable alerts. Unfortunately, little focus has been applied to developing practical and implementable guidance for the analysts who create and apply alerts within hospitals.
This poster will present and discuss the challenges and applicable solutions for developing, designing, and implementing EHR alerts into practice within a hospital setting. Multiple perspectives will be represented via a frontline view from our Hospital’s Clinical Decision Support and alert development committee member and Director of Provider Informatics, an operational and implementation approach from embedded human factors and user experience practitioners, and an academic and research approach via user experience researchers.
References
Brodowy, B., & Nguyen, D. (2016). Optimization of clinical decision support through minimization of excessive drug allergy alerts. American Journal of Health-System Pharmacy, 73(8), 526–528. https://doi.org/10.2146/ajhp150252
Iii, J. D. M., Mallozzi, C. P., Perkins, R. M., Shelov, E., & Schreiber, R. (2020). Reducing Alert Burden in Electronic Health Records: State of the Art Recommendations from Four Health Systems. Applied Clinical Informatics, 11(1), 1–12. https://doi.org/10.1055/s-0039-3402715
Jung, S. Y., Hwang, H., Lee, K., Lee, H.-Y., Kim, E., Kim, M., & Cho, I. Y. (2020). Barriers and Facilitators to Implementation of Medication Decision Support Systems in Electronic Medical Records: Mixed Methods Approach Based on Structural Equation Modeling and Qualitative Analysis. JMIR Medical Informatics, 8(7), e18758. https://doi.org/10.2196/18758
Lewandrowski, K. (2011). POC testing in the emergency department: Strategies to improve clinical and operational outcomes. Acutecaretesting. https://acutecaretesting.org/en/articles/poc-testing-in-the-emergency-department-strategies-to-improve-clinical-and-operational-outcomes
Rodrigues, A. T., Stahlschmidt, R., Granja, S., Pilger, D., Falcão, A. L. E., & Mazzola, P. G. (2017). Prevalence of potential drug-drug interactions in the intensive care unit of a Brazilian teaching hospital. Brazilian Journal of Pharmaceutical Sciences, 53, e16109. https://doi.org/10.1590/s2175-97902017000116109
World Economic Forum. (2019, December 5). Four ways data is improving healthcare. World Economic Forum. https://www.weforum.org/agenda/2019/12/four-ways-data-is-improving-healthcare/
Event Type
Poster Presentation
TimeTuesday, March 264:45pm - 6:15pm CDT
LocationSalon C
Digital Health
Simulation and Education
Hospital Environments
Medical and Drug Delivery Devices
Patient Safety Research and Initiatives