When I was a medical student approximately 197 years ago, a senior resident gave me great advice on responding to a hospitalized patient who needed CPR (aka a code blue). He told me that the first thing to do at a code is to check my own pulse. The following year, I heard an attending physician say something similar when dealing with an acutely sick patient: don’t just do something; stand there!
Both pieces of advice are counterintuitive upon initial reflection. Our instinct as first responders is to respond! Start barking medication orders or initiating lab studies. The patient needs your help, and the only way to help is to do something. Not so fast! We can often cause more problems by doing something—anything. As physicians, we take an oath to do no harm. To be true to that oath, we often need to do less, not more.
Low-value care has been defined as “the use of a health service for which the harms or costs outweigh the benefits.” Examples of low-value care might include ordering MRI scans for nonspecific low back pain or electrocardiograms for patients with no heart problems who are scheduled for routine surgery. These tests or procedures are not harmful per se; it’s just that there is no evidence that they’re actually helpful. Why would physicians order such tests? The most common reason is that this is how they were trained. The test or procedure might have been the standard of care back in the day. But evidence mounts, treatments are developed, and things change.
When the EHR aids in low-value care
Ironically, the electronic health record (EHR) can be another cause for low-value care. Order sets contain physician instructions for medications, labs, imaging studies, and many other interventions patients need while hospitalized. Ideally, they’re created following specific care guidelines and best practices, but they can become outdated from an evidence perspective. EHRs can make it easy for doctors to continue following outdated practices.
An interesting study was published last month in BMJ Open Quality. Researchers examined whether they could effectively help physicians identify and discontinue low-value orders in the EHR. They focused on recurring hospital orders, such as daily labs, continuous monitoring, and overnight vital checks. Often initiated without ongoing clinical necessity, these orders can disrupt patient recovery by interrupting sleep, drive up costs through redundant testing and unnecessary monitoring, and contribute to clinician burnout by overloading staff with non-essential tasks.
The intervention: a bundled de-implementation order set
To tackle this issue, researchers developed a bundled de-implementation order set integrated within the EHR. This tool targeted nine common care components prone to overuse, including telemetry, overnight vital sign checks, Foley catheters, and daily lab orders. Each component was paired with alternative, lower-intensity order options. The intervention included dynamic display features that presented relevant de-implementation options based on the patient’s active orders, reducing cognitive load for clinicians. Its user-friendly design grouped care components into subcategories, such as “sleep-friendly vital signs” or “reduce frequency labs,” and made the order set readily accessible in the “Suggested” section of the EHR interface.
The bundled approach allowed clinicians to review and de-implement multiple recurring orders simultaneously, streamlining workflows and minimizing unnecessary care. Pilot implementation took place over several months at a large academic hospital, during which hospitalists were encouraged to adopt the tool through meetings, emails, and group messaging.
Key results and findings
The pilot implementation of the order set demonstrated promising results. During the study period, the tool was used in 48 unique hospitalizations, leading to the placement of 80 de-implementation orders. The most frequently utilized options were discontinuing overnight vital sign checks and telemetry monitoring. Nearly half of all encounters involved two or more de-implementation orders, underscoring the efficiency of the bundled approach.
The intervention resulted in a statistically significant reduction in total high-frequency recurring orders per patient day, with overnight vital sign orders showing the most pronounced decline. While changes in other components, such as daily labs, were not statistically significant, the overall reduction in unnecessary orders highlights the potential impact of bundling interventions within EHRs. Although the study did not directly assess patient outcomes, existing literature suggests that reducing orders like overnight vital sign checks can improve sleep quality and patient satisfaction—key priorities for healthcare systems focused on value-based care.
Lessons for U.S. healthcare leaders
The findings from this study provide several important lessons for healthcare leaders. First, bundling interventions into a single EHR tool can simplify clinicians' decision-making, reducing effort and improving efficiency. This strategy also underscores the importance of targeting areas with the greatest potential for impact, such as recurring orders that disrupt patient care or drive unnecessary costs. Starting with a pilot program focused on low-risk patients, as this study did, offers a low-stakes way to test interventions before broader implementation.
Clinician engagement is also critical to the success of such initiatives. Adoption of the order set was driven by multi-channel communication, including meetings and direct messaging, which emphasized the benefits for both patients and providers. Additionally, leveraging technology to integrate user-friendly, dynamic tools can help nudge clinicians toward high-value care decisions without adding to their cognitive burden. By prioritizing design and accessibility, healthcare systems can enhance both adoption and effectiveness.
Future directions and broader implications
The study highlights the potential for bundled de-implementation strategies to reduce waste and improve care quality. Scaling this approach across healthcare systems will require regional and national collaborations, as well as adaptations to diverse organizational settings. These interventions also align closely with broader value-based care goals, offering opportunities to tie de-implementation initiatives to metrics linked to payer incentives. Future research could explore expanding bundled approaches to other areas prone to overutilization, such as imaging or specialty consults.
Emerging technologies like AI and predictive analytics also hold promise for enhancing de-implementation efforts. For example, AI-driven risk stratification tools could identify patients most likely to benefit from reduced monitoring or testing, enabling more targeted and efficient interventions. As healthcare systems increasingly adopt such digital tools, integrating these capabilities into bundled order sets could further amplify their impact.
Conclusion
The bundled de-implementation order set represents a simple yet impactful solution to a pervasive problem in hospital care. By reducing unnecessary recurring orders, healthcare systems can improve operational efficiency, enhance patient experience, and drive cost savings—all without compromising care quality. The takeaway for clinical and operational leaders is clear: small, well-designed interventions can yield significant benefits when implemented thoughtfully. It is time to evaluate recurring order practices and consider adopting bundled de-implementation strategies. Better care does not always mean doing more; sometimes, it means doing less but doing it smarter.