Whether treating sepsis in patients in an intensive care unit, or providing long-term maintenance for a chronic condition such as diabetes, healthcare providers make important decisions every day that affect patients’ health and quality of life. That’s why it’s so important that the act of putting patients’ safety and welfare first remain at the heart of care delivery.
While clinicians rely on their intuition and experience to decide on the best course of action for each patient they treat, they also use reference materials like websites and textbooks to help support their disease management strategies. Nowadays, clinical decision support (CDS) alert systems, which trigger recommendations that aid in decision-making, serve as another one of those reference materials.
These alerts are designed by clinical informaticists and teams of quality and patient safety staff. Triggered when appropriate based on information in each patient’s electronic health record, the goal is that the alerts improve health system processes, increase efficiency, reduce waste, and ultimately, improve patient outcomes. These teams of specialists creating these alerts, based on historical data about what worked best in similar cases, understand how to use electronic health record (EHR) alerts to make data-driven decisions and improve patient outcomes.
So what exactly goes into building an EHR alert, and how does an alert inform quality and process improvement? First, let’s take a look at what EHR alerts are, how they originate, and how they’re used.
EHR alerts are real-time, visual cues that provide clinicians recommended next steps and/or future actions for treating patients. While alerts themselves may be general, e.g. the same alert fires for all patients who come into the emergency room (ER) presenting with chest pain, their presence is patient-specific - meaning they will be triggered for an individual patient based on information in his or her health record.
Alerts typically fall into two main categories:
Medication EHR alerts warn clinicians about possible risks like drug interactions and known allergies to medications, whereas operational decision support alerts help providers make decisions about lab orders and suggested screenings, as well as other best practices based on the contents of the patient’s health record.
Alerts may either be interruptive or non-interruptive; an interruptive EHR alert requires some form of action from the provider. These actions may include:
Non-interruptive alerts won’t interfere with clinical workflow and serve more as general reminders about less urgent recommendations.
One can think of this interruptivity as akin to the experience of browsing a website. Oftentimes, advertisements will be embedded into the webpage and don’t interrupt your browsing experience. However, there may be instances when a pop-up window appears that requires an action before proceeding. The pop-up will grab your attention, but is more likely to annoy you if it isn’t relevant or interesting.
Health systems have alerts firing literally all day long, often hundreds of thousands of times a day, depending on the number of patients they’re treating. As a result, it’s extremely important that health systems use available data to better curate their electronic workflows and design better alerts that lead to better outcomes.
Understanding how to use EHR alerts to make data-driven decisions is imperative if health organizations want to improve quality, safety, and efficiency. Trends in alert data are often the first sign that a process needs to be reviewed or changed. For example, special metrics around alert fatigue can act as a clear indication an inefficient workflow is live in the system.
EHR alert data shows how often an alert is firing, how often end users are overriding the alert, whether clinicians are placing orders based on information contained within the alert, and so forth. Taking analysis a step further, clinical informatics platforms can be employed in assessing that data to provide additional insights, enabling the quality, safety, and clinical informatics teams to understand how an alert affects patient outcomes. For example, when that alert was deployed and responded to, did patient wait times decrease? Did fewer patients need to be admitted following an ER visit? Did readmission rates drop?
The data and insights gathered from EHR alerts and subsequent patient outcomes can then be used to initiate quality improvement (QI) projects, which health systems undertake to support process enhancements in both technical systems and clinical practice. QI projects aim to solve big problems -- with the end goal of creating a more efficient and effective healthcare system that improves patient outcomes. Without using EHR alerts to make data-driven decisions, QI as a methodology would be like shooting darts in the dark.
These EHR-based insights and resulting improvements need not be one-off events, either; data-driven decisions should become iterative and informed with even more data over time. Because EHR alerts are stored and tracked on an ongoing basis, this alert data can be used to measure year-over-year improvements or declines in outcomes. Stats to watch may include mortality rates, readmission rates, and average lengths of stay for various conditions and patient populations.
While healthcare organizations use EHR alert data and insights to improve processes and to kick off QI projects, patients and the providers who care for them benefit from data-driven decisions as well. These benefits include:
Analyzed consistently and correctly, healthcare data has a compelling story to tell - one that can help both clinicians and entire health systems make decisions that are in the best interests of the patients they serve. By using EHR alerts to make data-driven decisions, healthcare organizations can improve efficiency, lower the cost of care, and enhance the quality of care.