Presentation
Understanding Burnout in Healthcare: A Contextual Design Approach and Emergency Medicine Case Study
SessionApplied Methods (HE9)
DescriptionBACKGROUND: Burnout is characterized by emotional exhaustion, depersonalization, and a low sense of personal accomplishment in the workplace resulting from a variety of work-related stressors, psychosocial vulnerabilities, and coping styles. Although burnout can affect any employee, burnout in healthcare is particularly widespread, with 49% of all healthcare professionals (HPs) meeting burnout criteria. Emergency HPs are markedly affected, with reported burnout rates ranging from 35% to 65%. Unaddressed burnout negatively impacts HPs’ physical and mental health, patient care, healthcare costs, and health disparities. As the ED is critical to our healthcare system, it is imperative to address emergency HP burnout.
The 2019 National Academy of Medicine (NAM) report on HP burnout includes a Systems Model of Clinician Burnout and Professional Well-being that details factors contributing to burnout. The report emphasizes that burnout is not a personal failing of HPs but is a failing of the system in which HPs practice. The model’s systems approach provides a framework for understanding environments in which burnout is more likely. The report gives broad recommendations for reducing burnout, but the steps that are appropriate in one organization or unit may not be appropriate for another.
The US Surgeon General’s recent advisory on HP burnout further recommends seeking HP input to improve the healthcare system. Here we present a research design grounded in Contextual Design principles to incorporate HP perspectives and informed by the National Academy of Medicine’s systems framework to identify, contextualize, and prioritize workplace stressors associated with burnout. We also present the results of its application in a large ED. Our goal is that researchers in other healthcare systems will implement this approach to effectively assess burnout and identify and prioritize the unique stressors associated with burnout in their own organizations.
METHOD: The research design is grounded in Contextual Design principles, which place users and their interactions with systems at the center of the analysis. It also follows a funnel approach, beginning with broad data collection methods to elicit a general overview of the workplace from a larger number of people, and progressing to more specific and nuanced techniques to elicit details about individual experiences.
The process starts with a survey to provide a broad understanding of burnout-related stressors. Next, focus groups add contextual information that the survey cannot capture. Contextual inquiries then allow observation of HPs in their work setting and permit them to share individual insights not covered by the survey or focus groups. The qualitative data is synthesized into an affinity model that represents a shared understanding of the work. Last, all HPs have the opportunity to validate the model, prioritize the key stressors to address, and assess the impact and effort required to address each stressor.
PARTICIPANTS: Emergency HPs working in the ED of a large community hospital were recruited by the researchers and ED leadership in-person and through email and posted flyers.
RESULTS: Survey - The survey response rate was 35% (n=63/182). Eighty-one percent (n=51/63) of respondents met criteria for burnout. Participants ranked the NAM model’s workplace stressors by severity and priority for improvement (scale is 1 to 5) with the top two items being inadequate staffing and inefficient workflows. Participants could also provide comments if desired.
Focus Groups - We held six focus groups with 19 participants. Five groups were conducted in-person and one via video conference. Participants provided context about the key workplace stressors identified in the survey: inadequate staffing, inefficient workflows, patient-related stressors, and time pressure. Participants prioritized stressors after the focus group and inadequate staffing and inefficient workflows remained top priorities.
Contextual Inquiries - Contextual inquiries were conducted with twenty-two participants: seven physicians, five nurses, four medical technicians, three physician assistants, two patient relations specialists, and one health unit coordinator. Emergency HPs were observed for 3-5 hours per contextual inquiry while they went about their work. Total observation time was 77.5 hours (mean=3.5 hours).
Validation and Prioritization - The research team iteratively classified breakdowns identified in the qualitative survey, focus group, and contextual inquiry data according to the NAM model’s stressors to develop an affinity model. The model was shared with HPs for validation and prioritization. Validation was conducted with four nurses, three physicians, one certified nursing assistant, and one medical technician. The overall rate of agreement for the model was 83%. One hundred percent agreement was achieved for 53% (n=159/300) of stressors.
Impact and Effort Ratings - An impact and effort rating survey for the nine overarching priorities identified by participants in the validation phase was disseminated via email to all emergency HPs and completed by 44% (n=80/182: 29 physicians and advanced practice providers and 51 nursing and other staff). Opportunities rated as high impact and high effort were improvements to physical work environment, workflows, staffing, patient safety, extrinsic motivation, and psychiatric patient procedures. Opportunities classified as low impact and low effort were improvements to patient-related stressors, communication, and technology.
DISCUSSION: The survey indicated an 81% burnout rate among emergency HPs, higher than the 35-65% reported elsewhere, but many of the identified stressors align with the literature. The presence of inefficient workflows, insufficient staffing, ineffective communication systems, dated technology, and a desire for higher pay are reported by other articles. However, our research design yielded data on the specific workflows that are problematic, the exact shifts and roles that could use additional staff, the precise communication channels to improve, the most effective technology upgrades, and alternative compensation ideas.
We also identified stressors not reported in the literature. The in-depth systems approach was required to identify these stressors because they were often work practices that HPs were not able to articulate. These included an ineffective ED layout, inefficiencies in hospital admissions procedures, and bottlenecks apparent only with increased patient volume. In addition to identifying these stressors, we also collected specific contextual information to guide changes.
Recommendations from the literature, the NAM report, and the Surgeon General’s advisory are general guidance, making it difficult to translate those recommendations into concrete action items for a particular organization. Although other studies have identified stressors in healthcare, they usually use surveys and do not gather validated contextual data on specific issues. The results of this research design provide the information necessary to plan improvements tailored to the specific stressors in their organization. For example, the data establish not only that workflows in the ED are stressors but pinpoint exactly which workflows are problematic and why. We are unaware of this research design being applied to identify, contextualize, and prioritize stressors contributing to burnout in healthcare outside of our efforts, although the individual methods are widely applied in the human-computer interaction and design domains.
Finally, this research design affords a nuanced understanding of stressors’ impact on HPs. Informal feedback from HPs suggests that the systems approach effectively integrated the HP perspective into identifying burnout-related stressors, yielded valuable insights into their improvement priorities, and gave HPs a welcome forum to express their views.
CONCLUSION: Employing a unique research design based on Contextual Design to identify, contextualize, and prioritize specific stressors in healthcare, we established a burnout rate of 81% among emergency HPs, gathered contextual data concerning stressors contributing to burnout, and collaborated with HPs to identify and prioritize specific stressors for improvement. Our study results will guide tailored improvement efforts in the ED. Employing a similar approach, other healthcare organizations can identify, contextualize, and prioritize stressors contributing to burnout in their unique environment, make informed decisions about addressing stressors, and, in turn, promote better HP well-being and patient care.
The 2019 National Academy of Medicine (NAM) report on HP burnout includes a Systems Model of Clinician Burnout and Professional Well-being that details factors contributing to burnout. The report emphasizes that burnout is not a personal failing of HPs but is a failing of the system in which HPs practice. The model’s systems approach provides a framework for understanding environments in which burnout is more likely. The report gives broad recommendations for reducing burnout, but the steps that are appropriate in one organization or unit may not be appropriate for another.
The US Surgeon General’s recent advisory on HP burnout further recommends seeking HP input to improve the healthcare system. Here we present a research design grounded in Contextual Design principles to incorporate HP perspectives and informed by the National Academy of Medicine’s systems framework to identify, contextualize, and prioritize workplace stressors associated with burnout. We also present the results of its application in a large ED. Our goal is that researchers in other healthcare systems will implement this approach to effectively assess burnout and identify and prioritize the unique stressors associated with burnout in their own organizations.
METHOD: The research design is grounded in Contextual Design principles, which place users and their interactions with systems at the center of the analysis. It also follows a funnel approach, beginning with broad data collection methods to elicit a general overview of the workplace from a larger number of people, and progressing to more specific and nuanced techniques to elicit details about individual experiences.
The process starts with a survey to provide a broad understanding of burnout-related stressors. Next, focus groups add contextual information that the survey cannot capture. Contextual inquiries then allow observation of HPs in their work setting and permit them to share individual insights not covered by the survey or focus groups. The qualitative data is synthesized into an affinity model that represents a shared understanding of the work. Last, all HPs have the opportunity to validate the model, prioritize the key stressors to address, and assess the impact and effort required to address each stressor.
PARTICIPANTS: Emergency HPs working in the ED of a large community hospital were recruited by the researchers and ED leadership in-person and through email and posted flyers.
RESULTS: Survey - The survey response rate was 35% (n=63/182). Eighty-one percent (n=51/63) of respondents met criteria for burnout. Participants ranked the NAM model’s workplace stressors by severity and priority for improvement (scale is 1 to 5) with the top two items being inadequate staffing and inefficient workflows. Participants could also provide comments if desired.
Focus Groups - We held six focus groups with 19 participants. Five groups were conducted in-person and one via video conference. Participants provided context about the key workplace stressors identified in the survey: inadequate staffing, inefficient workflows, patient-related stressors, and time pressure. Participants prioritized stressors after the focus group and inadequate staffing and inefficient workflows remained top priorities.
Contextual Inquiries - Contextual inquiries were conducted with twenty-two participants: seven physicians, five nurses, four medical technicians, three physician assistants, two patient relations specialists, and one health unit coordinator. Emergency HPs were observed for 3-5 hours per contextual inquiry while they went about their work. Total observation time was 77.5 hours (mean=3.5 hours).
Validation and Prioritization - The research team iteratively classified breakdowns identified in the qualitative survey, focus group, and contextual inquiry data according to the NAM model’s stressors to develop an affinity model. The model was shared with HPs for validation and prioritization. Validation was conducted with four nurses, three physicians, one certified nursing assistant, and one medical technician. The overall rate of agreement for the model was 83%. One hundred percent agreement was achieved for 53% (n=159/300) of stressors.
Impact and Effort Ratings - An impact and effort rating survey for the nine overarching priorities identified by participants in the validation phase was disseminated via email to all emergency HPs and completed by 44% (n=80/182: 29 physicians and advanced practice providers and 51 nursing and other staff). Opportunities rated as high impact and high effort were improvements to physical work environment, workflows, staffing, patient safety, extrinsic motivation, and psychiatric patient procedures. Opportunities classified as low impact and low effort were improvements to patient-related stressors, communication, and technology.
DISCUSSION: The survey indicated an 81% burnout rate among emergency HPs, higher than the 35-65% reported elsewhere, but many of the identified stressors align with the literature. The presence of inefficient workflows, insufficient staffing, ineffective communication systems, dated technology, and a desire for higher pay are reported by other articles. However, our research design yielded data on the specific workflows that are problematic, the exact shifts and roles that could use additional staff, the precise communication channels to improve, the most effective technology upgrades, and alternative compensation ideas.
We also identified stressors not reported in the literature. The in-depth systems approach was required to identify these stressors because they were often work practices that HPs were not able to articulate. These included an ineffective ED layout, inefficiencies in hospital admissions procedures, and bottlenecks apparent only with increased patient volume. In addition to identifying these stressors, we also collected specific contextual information to guide changes.
Recommendations from the literature, the NAM report, and the Surgeon General’s advisory are general guidance, making it difficult to translate those recommendations into concrete action items for a particular organization. Although other studies have identified stressors in healthcare, they usually use surveys and do not gather validated contextual data on specific issues. The results of this research design provide the information necessary to plan improvements tailored to the specific stressors in their organization. For example, the data establish not only that workflows in the ED are stressors but pinpoint exactly which workflows are problematic and why. We are unaware of this research design being applied to identify, contextualize, and prioritize stressors contributing to burnout in healthcare outside of our efforts, although the individual methods are widely applied in the human-computer interaction and design domains.
Finally, this research design affords a nuanced understanding of stressors’ impact on HPs. Informal feedback from HPs suggests that the systems approach effectively integrated the HP perspective into identifying burnout-related stressors, yielded valuable insights into their improvement priorities, and gave HPs a welcome forum to express their views.
CONCLUSION: Employing a unique research design based on Contextual Design to identify, contextualize, and prioritize specific stressors in healthcare, we established a burnout rate of 81% among emergency HPs, gathered contextual data concerning stressors contributing to burnout, and collaborated with HPs to identify and prioritize specific stressors for improvement. Our study results will guide tailored improvement efforts in the ED. Employing a similar approach, other healthcare organizations can identify, contextualize, and prioritize stressors contributing to burnout in their unique environment, make informed decisions about addressing stressors, and, in turn, promote better HP well-being and patient care.
Event Type
Oral Presentations
TimeWednesday, April 211:15am - 11:37am EDT
LocationHarbour C
Hospital Environments (HE)