Skip to main content

Nudge Unit Supports Physician, Patient Behavioral Changes Towards Medical Decisions that Improve Care Value and Quality of Care

December 23, 2020

Nudges are a change in the way choices are presented or information is framed that can have a large, but predictable, impact on medical decision-making, for both patients and providers without actually restricting individual choice. The Nudge Unit at Penn Medicine focuses on a range of different care improvement projects, including safety initiatives, with this framework in mind that are designed to improve workflow, support evidence-based decision-making, and create sustained changes in patient engagement and daily behaviors.1

Innovation Patient Safety Focus

The framework for nudge projects, like The Nudge Unit at Penn Medicine, is to guide patient or provider behavior in a way that enhances health-related actions or enhance care delivery, ultimately improving patient safety. This approach can be applied to a wide variety of quality improvement projects, including those with a specific focus on activities or actions that impact patient safety.

Resources Used and Skills Needed

Resources and skills needed for a nudge project is highly dependent on the project. However, all projects seek to include a multidisciplinary team wherever possible. Frontline staff should be included and consulted in the project planning process to ensure the intervention aligns with existing workflows and minimize unintended consequences.

Use By Other Organizations

The Penn Medicine Nudge Unit hosted the Inaugural Nudge Units in Health Care Symposium in 2018. Representatives from 22 health systems across the US and Canada attended.a To-date, approximately 30 organizations have engaged Penn Medicine with interest in establishing their own nudge/behavioral health units, including Geisinger, UCLA, and the Ministry of Health in Saudi Arabia.2

In addition to other organizations who have established their own nudge projects, Penn Medicine Nudge Unit team members serve as principal investigators for nudge projects. For example, Penn Medicine research are currently engaged in a nudge project across 42 sites designed to reduce opioid prescribing.

  1. Health system attending included: University of Washington; VA Puget Sound Health Care System; University of California at San Francisco; University of California at Los Angeles; University of Southern California; University of Colorado Anshutz; Sanford Health: BJC HealthCare/Washington University; Northwestern Medicine; McLaren Health Care; Michigan Medicine; St. Michael’s Hospital; Toronto General Hospital; University of Pittsburgh Medical Center Health Plan; Indiana Regional Medical Center; Geisinger Health System; Penn Medicine; Duke University Health System; New York City Health & Hospitals; Yale New Haven Health; Dartmouth-Hitchcock; Brigham and Women’s Hospital.

Date First Implemented
Problem Addressed

Regardless of how successful an intervention is demonstrated to be, it can only be effective if the intended target chooses to implement it. The ability to change human behavior is the final stumbling block for many advances in patient care and how to effectively guide an individual to choose a behavior or action is an ongoing question for patient safety and quality improvement professionals. Decisions and behaviors are heavily influenced by the environment in which they occur.  For example, small changes to the way choices are presented and information is framed within the electronic health record can lead to significant differences in the way clinicians order tests and treatments. Social networks and norms influence how individuals eat and exercise. In many cases, these influences go unnoticed or are not given much thought.  Nudge projects and Nudge Units can help develop systematic approaches to designing interventions that better align our decisions and behaviors with long-term goals. In implementing patient safety interventions, requiring a provider or patient to deviate from a previous course of action can be incredibly challenging to both initiate and then sustain if the intervention is difficult or time consuming to implement. Nudge projects can make it easier and more seamless for individuals to make evidence-based decisions and practices.1

Description of the Innovative Activity

A nudge is a change in the way choices are presented or information is framed that alters people’s behavior in a predictable way without restricting choice. Nudges should be transparent and never misleading, easy to opt out of if needed, and aligned with the welfare of the person being nudged. Nudges in clinical care can focus on both clinician and patient behaviors. Nudges to clinicians focus on the design of practice environments that aim to improve workflow and steer decision-making towards evidence-based care. Nudges to patients focus on daily health behaviors and interventions that lead to significant and sustained changes in patient engagement.1 Examples of nudges include changes to the EHR to make options more accessible or to change who receives alerts to better align with the workflow and reduce fatigue. Nudges may also include making the advantages of particular choices more apparent through how the information is framed, data transparency, or by allowing peer comparison.3

In many ways, nudges that target clinicians can be similar to clinical decision support that already exists within electronic health records. However, nudges are actually intended to improve these types of tools by applying a behavioral science approach that can provide a scientific framework to guide the design, prioritization, and implementation of interventions. Additionally, the principles of behavioral science behind nudge projects can help to ensure that interventions are well integrated into existing clinical workflows.4

Nudge projects within the Penn Medicine Nudge Unit are co-designed in partnership with frontline clinicians, health system leadership, and patients and are implemented in a pragmatic and testable manner in pilot settings to inform larger scale deployment. Evaluations determine both positive and negative impacts and whether the project is suitable for spread across the system. Monthly meetings allow the team to discuss what is working and may be suitable for broader rollout, but also those interventions that are not working and should be suspended.1

Context of the Innovation

Penn Medicine began working on nudge projects in 2014. The first project focused on increasing the rate of generic prescribing. Generic medications are as effective as brand name prescriptions, and studies have found that patients are more likely to be adherent to medications when a generic is prescribed because they are better able to afford the cost. However, data indicated that generic prescribing at Penn Medicine was among the worst in the nation and the team sought to implement a project to specifically target these rates. A review of workflow practices found that while generics were listed among the EHR drop-down list of available prescriptions for providers to choose from, they were listed below brand name options if a provider searched for a brand name medication, requiring providers to scroll to reach them. The resulting nudge project focused on decreasing resistance to prescribing generics by implementing a single simple, cost effective change. If a provider searched for a brand name medication, only the generic options would initially be displayed unless the provider specifically selected they would like to see brand name options. This project was able to steer provider decision-making towards choosing a generic option.5 Following a pilot study, the intervention was expanded to the entire system, resulting in an improvement in the rate of generic prescribing from 75.4 to 98.4% (P <.001) and approximately $32 million in savings.6 The rate of generic prescribing has since been sustained over time.7 

The success of this first project spurred the creation of the Nudge Unit in 2016. The objective of this Unit is to have a team focused on systematically identifying and implementing high priority ideas efficiently. Since its creation, the Unit has worked on approximately 50 projects, many of which have focused on practices that improve patient safety. A number of the projects facilitated by the Nudge Unit have been identified through crowd sourcing competitions that ask frontline providers directly what challenges matter most to them.

A list of Nudge Unit projects can be found here:


In addition to the success of the nudge project promoting generic prescribing, the Penn Medicine Nudge Unit has implemented numerous successful nudge initiatives targeting both providers and patients. Several examples are captured below with more available at

Opioid Prescribing in Emergency Medicine

In an effort to decrease opioid prescription quantities, the Nudge Unit facilitated an intervention with the emergency medicine team to use a new electronic health record in the emergency department. In the electronic health record, they set a default prescribing option of 10 pills listed first, 20 pills listed second, 28 pills listed third, and then provided an option for manual entry of the pill count. Following implementation, the team observed a 22.8% increase in the proportion of prescriptions written for 10 pills, while seeing a 13.3% decrease in the proportion of prescriptions from 11-19 pills and a 6.7% decrease in the proportion of prescriptions for 20 pills. They did see a 5% decrease in the proportion of prescriptions for less than 10 pills, which indicated to prescribers the default prescription could be set even lower in the electronic health record. The success of this nudge project pilot formed the basis for a much larger intervention that is currently ongoing.8 

Unnecessary Imaging in Palliative Radiotherapy

 Clinical practice guidelines recommend that patients undergoing curative radiotherapy undergo daily imaging to augment positioning during treatment. However, for those undergoing palliative radiotherapy, daily imaging is considered unnecessary and can increase treatment time and expense for patients. This nudge project implemented a default imaging order in the electronic health record that specified no daily imaging during palliative radiotherapy. Physicians could opt out and select another imaging frequency, but the intention was to reduce unnecessary care for these patients. This project results in a significant reduction in daily imaging (P=.004) and demonstrated that simple nudges have the potential to reduce unnecessary care.9

Planning and Development Process
  • Solicit ideas from clinical staff. The Penn Medicine experience with the Nudge Unit has demonstrated that the frontline clinical staff are the best source of suitable project ideas. Regular crowd sourcing has helped to identify low-cost workflow or behavioral interventions as frontline staff are the most knowledgeable of their workflow and processes and are best positioned to recognize challenges faced and potential simple solutions.  
  • Take a systematic approach to understanding the problem and identifying solutions.3 
    • Identify ideas that address challenges. Aim for projects that are of high impact but feasible with available capacity and resources. Broad and targeted approaches may both be needed to address the scope of the challenge. Best practice guidelines, published initiatives, and insights from patients and family members can be good sources for solutions. 
    • Understand available resources to inform prioritization. Resources such as funding, team and stakeholder expertise, and infrastructure should all be considered when prioritizing projects. 
    • Identify available stakeholders. Outside of the team within your healthcare organization, stakeholders such as policymakers and community organizations can provide additional support or guidance for your project. 
    • Consider what data is needed versus available. Assess different types of existing data from various sources, including patient behaviors and characteristics, and ensure a complete understanding of what measurement would be needed and how that would impact provider burden. Use existing data to assess current practice variance and to identify gaps that can help guide the intervention. 
    • Create a clear hypothesis. Ensure the nudge project has a clear hypothesis. This hypothesis will then inform the design of the pilot, implementation, and evaluation strategy and may necessitate the collection of both qualitative and quantitative data.
Resources Used and Skills Needed

Resources and skills needed for a nudge project is highly dependent on the project. However, all projects seek to include a multidisciplinary team wherever possible. Frontline staff should be included and consulted in the project planning process to ensure the intervention aligns with existing workflows and minimize unintended consequences.

Funding Sources

Funding for nudge projects under the Nudge Unit comes from Penn Medicine.

Getting Started with This Innovation
  • Get the health system or hospital leadership’s support. Quick, low-cost wins with demonstrable return on investment can highlight the impact even a simple nudge project can have. This can build support for future initiatives or a dedicated nudge team, as in the case of Penn Medicine. 
  • Engage the whole team. Depending on the project, different team members will be required for implementation. However, involvement from a behavioral economist can provide critical insight into the design of the intervention. Additionally, prior to implementation the project team should engage representatives from all clinical areas related to the project workflow to ensure the intended interventions align well with the existing workflow and that the clinical team agrees with the root cause the intervention addresses. Clinical team buy-in is critical for successful implementation but you must also engage the team from your informatics or information systems department to ensure the project can be evaluated once it is underway.   
  • Establish a monitoring process. Ensure that there is a clear process of monitoring the progress of the project so success and any unintended negative consequences can be easily identified and projects suspended, if necessary.
Sustaining This Innovation

Some nudge projects will require a one-time change to the EHR. Those types of projects are very easy to maintain as they do not require active sustainment efforts. However, there are approaches to support sustainment in behavioral interventions. 

  • Include a social incentive. Social incentives include collaboration, competition, and support and help to reinforce behavioral change. This is particularly effective in patient behavioral interventions where family members are involved because the sustained familial relationships help to maintain that behavioral change. 
  • Provide individual feedback. Both individual feedback on behavioral change as well as peer to peer comparisons can provide motivation for continued and sustained change by allowing tracking of tangible progress and promotion competition.
References/Related Articles

A complete list of Nudge Unit publications can be found here:

Examples of Nudge Unit projects with a patient safety component include:


1.    Penn Medicine. The Nudge Unit. Accessed September 9, 2020. 
2.    Levins, Hoag. National Impact of Penn Medicine Nudge Unit Concept Spreads: Report From the 2019 Nudges in Health Care Symposium. Published October 2019. Accessed September 9, 2020. 
3.    Changolkar S, Asch DA, Balachandran M, el al. Key insights on launching a Nudge Unit within a health care system. NEJM Catalyst. Published May 1, 2019. Accessed September 9, 2020. 
4.    Patel MS, Navathe AS, Liao JM. Using nudges to improve value by increasing imaging-based cancer screening. J Am Coll Radiol. 2020;17(1 Pt A):38-41
5.    Patel MS, Dat S, Small DS. et al. Using default options within the electronic health record to increase the prescribing of generic – equivalent medications: a quasi-experimental study. Ann Intern Med. 2014;161(10 Suppl):S44-52. 
6.    Dubner SJ, How to Save $32 Million in One Hour (Ep. 397). Freakonomics. Published November 13, 2019. Accessed September 9, 2020. 
7.    Olshan D, Rareshide C AL, Patel MS. Longer-term durability of using default options in the electronic health record to increase generic prescribing rates. J Gen Intern Med 2019;34(3):349-350.  
8.    Delgado MK, Shofer FS, Patel MS, et al. Association between electronic medical record implementation of detail opioid prescription quantities and prescribing behavior in two emergency departments. J Gen Intern Med. 2018;33(4):409-411. 
9.    Sharma S, Guttmann D, Small DS, et al. Effect of introducing a default order in electronic medical record on unnecessary daily imaging during palliative radiotherapy for adults with cancer: A stepped-wedge cluster randomized clinical trial. JAMA Oncol. 2019;5(8):1220-1222. 

Original Publication
Original Publication indicates the date the innovation profile was first posted to the AHRQ Health Care Innovations Exchange website.
December 23, 2020
Date Verified by Innovator
Date Verified by Innovator indicates the most recent date the innovator provided feedback during the review process.
September 9, 2020
The inclusion of an innovation in PSNet does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or of the submitter or developer of the innovation.
Contact the Innovator

Contact the Penn Medicine Nudge Unit at

Related Resources