A proactive learning approach toward building adaptive capacity during COVID-19: A radiology case study

Abstract

The COVID-19 pandemic has challenged organizations to adapt under uncertainty and time pressure, with no pre-existing protocols or guidelines available. For organizations to learn to adapt effectively, there is a need to understand the perspectives of the frontline workforce involved in everyday operations. This study implemented a survey-tool to elicit narratives of successful adaptation based on the lived experiences frontline radiology staff at a large multispecialty pediatric hospital. Fifty-eight members of the radiology frontline staff responded to the tool between July and October of 2020. Qualitative analysis of the free-text data revealed five categories of themes that underpinned adaptive capacity of the radiology department during the pandemic: information flow, attitudes and initiative, new and adjusted workflows, availability and utilization of resources, and collaboration and teamwork. Enablers of adaptive capacity included timely and clear communication about procedures and policies from the leadership to frontline staff, and revised workflows with flexible work arrangements, such as remote patient screening. Responses to multiple choice questions in the tool helped identify the main categories of challenges faced by staff, factors that enabled successful adaptation, and resources used. The study demonstrates the use of a survey-tool to proactively identify frontline adaptations. The paper also reports a system-wide intervention resulting directly from a discovery enabled by the findings based on the use of RETIPS in the radiology department. In general, the tool could be used in concert with existing learning mechanisms, such as safety event reporting systems, to inform leadership-level decisions to support adaptive capacity.

1. Introduction

The COVID-19 pandemic has been an unprecedented event that caught organizations by surprise. Hospitals, in particular, came under immense pressure to cope with the added challenge of caring for COVID patients while mitigating the risk of infection (Grigonis et al., 2021). There were few to no pre-existing protocols, norms, standards or guidelines to rely on to plan and respond to the crisis when it struck. Furthermore, there was varying and sometimes conflicting information about the virus, risk of infections, and appropriate mitigation strategies from public health institutions, including the Centers for Disease Prevention and Control (CDC) and the World Health Organization (WHO). Institutional leadership had to therefore make time-sensitive decisions under ambiguity and uncertainty. Hospitals’ management and operations have been disrupted and tested in this context. Some challenges include elevated patient volumes or increased risk profiles, risk of infection to caregivers and other patients, shortage of resources such as personal protective equipment (PPE), and lack of knowledge or coordination of resource availability (Auerbach et al., 2020; Becker et al., 2021; Grigonis et al., 2021; Liu et al., 2020; Woolley et al., 2020). Auerbach et al. (2020) report a potential risk of missed diagnoses of non-COVID-19 conditions in patients suspected with COVID-19 and missed COVID-19 diagnosis in patients admitted for non-respiratory reasons.

There is a growing body of literature on how hospitals have coped with the aforementioned challenges during the pandemic using adaptive strategies at various organizational levels. These include surge planning, patient monitoring dashboards, telehealth and remote medical services, and workforce protection (Romani et al., 2021). In all of the examples, adaptation at multiple institutional levels has been key. Despite the growing rate of vaccination, vaccine-hesitancy among the population and each new strain of the virus driving another wave mean that infection control measures will continue for the foreseeable future. The Delta variant of the SARS-CoV-2 virus, for instance, led to increased patient volumes at pediatric hospitals given its high transmissibility among the unvaccinated, including children (Weiland and Schaff, 2021). Therefore, there is a need and an opportunity to proactively learn about challenges as well as contributors to successful adaptation to allow hospitals to enhance their preparedness and adaptive capacity. Insights from such learning could be directly relevant to the current transitions from crisis-mode and strengthen crises preparedness measures.

Studies focusing on hospital adaptations to the pandemic have employed a variety of methods and sources of data. The majority of papers comprises case studies or descriptions of various adaptations by authors based on their own experience designing and implementing those changes in their units or hospital locations. For instance, adaptations have been reported by stakeholders in areas, including, intensive care, telemedicine, neurosurgery and pediatric emergency department (Arnaout et al., 2020; Becker et al., 2021; Sasangohar et al., 2020; Tan et al., 2020). Donelli et al. (2022) identified a range of resilient responses on a case study of an Italian healthcare organization's response to the COVID-19 pandemic. These included mobilizing physical (e.g. beds) and human (e.g. doctors) resources, enabling informational channels, and measures toward efficient service. Other studies have reported data collected through multi-institutional surveys with clinical and administrative leadership. These surveys largely relied on hospital medical leaders, physicians and researchers for data collection (Auerbach et al., 2020; Linker et al., 2021; Mitchell et al., 2021). There is a lack of studies reporting perspectives of frontline personnel on operationalizing policies and adaptations in their daily work. Perspectives of frontline staff are shown to reveal patterns of adaptation in everyday clinical work, which could be important in helping bridge the gap between work-as-imagined and work-as-done, and in turn, inform policy and design decisions that better reflect and support operational performance (Hegde et al., 2013, 2019; Hollnagel, 2015b, 2016). These adaptive patterns when synergized with policies and leadership-driven decisions contribute to the overall adaptive capacity of the organization (Rankin et al., 2014). In the context of this paper, we define adaptive capacity as the ability of a system to adapt to disruptions such that it is able to sustain its core functionality. The study explores the underpinnings of adaptive capacity of a radiology department within a large hospital, at the beginning of the COVID-19 pandemic.

Radiology is an expansive field with high demand. Patients receiving imaging studies come from all areas of the hospital, and those who are unable to travel will often have portable imaging equipment borough to their bedside. Additionally, radiology staff travel to the bedside throughout the entire institution, including areas with high patient volumes. The wide reaching demand for radiology studies means that radiology staff are exposed to patients of every risk classification for COVID-19 and other communicable disease exposure in a range of contexts from the emergency department where full patient conditions may not be determined, to highly-at-risk intensive care patients. Further, pediatric patients are often accompanied by caregivers and require more interaction, e.g. for sedation (Ertl-Wagner et al., 2020). The radiology department was therefore representative in terms of its scale and complexity as a sociotechnical system for this study (Brady et al., 2021).

Previous work exists on the use of knowledge-elicitation techniques as an organizational learning approach. Lloyd-Smith (2020) have highlighted the need for organizations to enable key capabilities of improvisation and preparedness in order to respond resiliently to the pandemic. The Resilience Engineering Tool to Improve Patient Safety (RETIPS) has been previously used on a pilot-basis to learn about adaptive capabilities of anesthesia residents (Hegde et al., 2020). This study adapted RETIPS to the case of a pediatric radiology department at a major multi-specialty children's hospital in Mid-Atlantic United States to learn from its operational staff. The tool was implemented to learn about how frontline staff were coping with the COVID-19 pandemic, particularly during the first wave in 2020 when information and procedural guidelines were scarce. The longer-term goal of such learning is to identify the challenges and adaptive capacities in frontline patient care to inform future preparedness. To our knowledge, it is the first self-reporting tool used to collect narratives from frontline staff on their experiences adapting to pandemic times.

2. Methods

Brief description of RETIPS. The generic version of the tool is designed as a semi-structured questionnaire consisting of narrative-based questions and multiple-choice questions (see Hegde et al., 2019 for a full description). The tool starts by asking the respondent to describe a case or an example of an adaptation such as an adjustment to a workflow, a workaround, or a preventative measure. Themes that are key to understanding resilient capabilities are listed as cognitive probes for the respondent to consider in their narrative. This is followed by multiple choice questions to probe specific aspects of the example, such as types of challenges faced, factors that enabled success, and resources used. Free-text fields are also provided to allow the respondent to elaborate on their responses. Demographic information, such as years of experience and the area of specialization, is also collected.

Adaptation of RETIPS: The previous version of RETIPS was adapted for the purpose of the current study. To this end, the previous or ‘template’ version was reviewed by radiologists and experts in human factors and resilience engineering. The questions were reworded to refer specifically to the COVID-19 context within radiology. In keeping with the RETIPS template, the initial question-probe tailored to context, read: “Think of an ‘adjustment’ or coping strategy that you may have employed that helps ensure effective and safe care to patients during the COVID-19 pandemic response.” Subsequent question-probes were largely kept similar to the template version with minimal rewording for context-specificity. Response-choices were reworded as necessary, and in some cases new choices were included based on the expert stakeholders' consensus. Overall, the response-choices were listed with the goal of allowing the department to identify what enabled or constrained workers' adaptation during the pandemic. The revised tool, RETIPS_CovRad (can be viewed in the Appendix), included the following questions.

• −Role and years of experience
• −Detailed narrative: example of an adjustment or coping strategy or a preventive measure. Cognitive probes include types and sources of information shared, anticipation, innovation, infection control measures, monitoring behaviors, resource availability, policies and procedures, cooperation with colleagues, and engagement from leadership.
• −Surprises – What surprised you in working through the COVID-19 pandemic?
• −Discovered Opportunities - Could you describe any opportunities to improvise, that you discovered and leveraged to make the process safer?

Multiple choice questions (checkboxes) in the following categories.

• −What is Working Well – What factors contributed to the success of your response or practice?
• −Challenges and Concerns – What were some of the challenges/concerns that prompted the response described in your example?
• −Resources - What resources were necessary or helpful in this situation?
• −Area of Practice - Please indicate the primary clinical area/specialty that your example relates to.

The study protocol was reviewed and approved with an exempt status by the Institutional Review Board (IRB) of The Children's Hospital of Philadelphia (CHOP), which was the site of the research and implementation.

Setting: The radiology department of CHOP is positioned to serve the needs of the entire hospital which provides quaternary care for pediatric patients and is a level 1 trauma center. Through the principal pandemic timeframe of 2020 and 2021, the radiology department saw on average 735 patients per weekday, and 298 per weekend day. Patients originated from outpatients coming in for routine imaging, emergency department requests of varying patient acuity, and inpatient needs from all units of the hospital, including intensive care. The department recorded a cumulative 34.5% positive tests for COVID-19 among its patients and employees since the beginning of the pandemic until mid-June of 2020 (1 month preceding this study), the time period approximately coinciding with the first wave in the United States. As part of the largest pediatric hospital system in a major metropolitan area, this radiology department represents a sociotechnical system of substantial scale.

Distribution of the tool: The tool was made available via Google Forms to all radiology and related support department employees on July 29^th, 2020. The study and the tool were announced by direct emails to the department, posters, and face to face recruitment. In order to encourage responses, the tool was endorsed by the leadership of the department. The majority of responses were received over the course of 10 weeks, after which the analysis of data began.

Analysis: The responses, produced in a spreadsheet format, were organized into rows of text for qualitative data analysis. The responses to the detailed narrative question were coded by a team of 4 coders to identify themes relating to resilience. All of the coders have graduate engineering degrees (master's or doctoral) with a human factors engineering focus, and previous experience in qualitative data analysis. All coding was performed in either Google Spreadsheets or Microsoft Excel. The data were divided into two halves and the coding tasks were distributed such that 1 primary coder and 1 secondary coder were assigned to each half of the dataset. The coding process was done in multiple 'runs'. In the first run, coders looked for themes relating to resilience engineering concepts, such as anticipation, monitoring, preparedness, adaptation and coordination. Each coder developed their own codebook and updated it as new thematic codes were created. The primary and secondary coder coded their datasets independently and later compared codes. All coders then synchronously compared their codebooks as well as application of codes to instantiations in the data, consolidated codes that were similar or synonymous, and reconciled differences in labeling or application of codes. In the second run, coders revised their codes for consistency based on the consolidated codebook. Additionally, new codes were created during this run that were focused on more specific operational strategies and tactics indicated in the responses, such as workflow changes, resource mobilization or other adaptations. In the third run, one of the primary coders and a fifth coder identified and eliminated remaining minor inconsistencies and redundancies in codes (e.g. synonymous codes). Further, the codes were now translated as ‘themes’ and grouped into higher-level categories, based on underlying similarities. Narrative responses to the ‘surprises’ and ‘opportunities’ questions were far fewer in number when compared with those to the detailed narrative field, and were not included in the thematic coding runs. These responses were reviewed and found to confirm the themes that were already identified based on the analysis of the detailed narrative and did not generate new themes. The themes were then grouped into categories based on their semantic closeness or interrelatedness, based on a consensus among the coders.

In addition to themes, a separate category of codes was created during the fourth and final run to indicate the organizational level at which each of the above themes occurred. Five levels of organization that have been previously applied (Hegde et al., 2020) were used as codes in this category – individual (level 1), team (level 2), department (level 3), hospital or multi-departmental (level 4), and beyond the hospital, e.g. governmental (level 5). Each theme was mapped to the corresponding level(s) involved in the described example. Fig. 1 depicts the analysis workflow comprising the four runs.

Fig. 1.

Analysis workflow: Stages or ‘runs’ of data coding resulting in thematic categories mapped to organizational levels.

Responses to each of the multiple-choice questions were aggregated numerically and represented through appropriate bar charts. The bar charts provide a visual comparison of participants responses in each question-category.

3. Results

The modified RETIPS tool was responded to by 58 out of the 489 (11.9%) members of the department between July 29^th and October 12^th of 2020 with over 90% received during the month of August. Among the respondents, 14 (24.1%) have leadership roles, 13 are technologists, 9 nurses, 7 radiologists, 7 research personnel, 6 administration, 2 child life specialists. Twenty-seven (46.6%) have been more than 9 years in their present primary role, 10 (17.2%) for 4–8 years, and 21 (36.2%) have been less than 4 years. In their response to the question “Please indicate the primary clinical area/specialty that your example relates to”, participants either selected their own areas or those of other roles with the following distribution: technologists at 17 responses (29.3%), followed by sedation/anesthesia (15.5%), radiologists (13.5%), research (12.1%), picture archiving and communication system (PACS) (10.3%), administration (6.9%) and nursing (5.2%). Other areas mentioned were the entire department (1.7%) and Quality and Safety (1.7%). Staff members in leadership roles comprised 14 responses (24.14%) and non-leadership roles, 44 responses (75.9%).

In their narrative responses, participants described examples of lessons-learned and indicated the success factors, challenges, and resources pertinent to their narrative. The thematic analysis of the descriptions of their examples of an ‘adjustment or coping strategy or preventive practice’ identified several prominent themes. Based on similarity and relatedness, the themes were grouped into five higher-level categories: ‘Information Flow’, ‘Attitudes and Initiative’, New and Adjusted Workflows', ‘Availability and Utilization of Resources’, and ‘Collaborations and Teamwork’. The 21 general themes are presented below with their overall occurrence count and distinctive definitions. The categories have been described in the decreasing order of total number of instantiations of the constituent themes. Excerpts from the actual response texts are provided to represent the category and several of the themes within.

3.1. Category A: Information flow

Information Flow encompasses themes concerning the movement of policy-related, procedural, and general information across different levels of organization. The role of leadership in keeping the staff informed was frequently mentioned, and summarized in the following quote: “providing critical needs or modifications to keep staff safe, transparency and sharing information, address(ing) requests and concerns promptly as well as to create a positive environment to ease staff's fears and anxiety”. Similarly, the hospital also had a process to communicate with patients and their families about COVID-testing protocols, scheduling changes and appropriate building-access protocols: “communication was sent out daily weekly by management keeping us in the loop for awareness and next steps … I cope with the pandemic by attempting to keep up with the local news, the Town Hall updates by (hospital name) and the email updates from our leadership team.”

• •Communication of Change (34): Related to information dissemination mechanisms for policy or procedural change. These included communication via email, town halls, huddles, and other forms of briefing. For instance, the picture archiving and communication system (PACS) team determined the number of employees who would need to be on site and those that could work remotely (particularly those that were considered ‘at risk’), and communicated this internally.
• •Timeliness and Clarity of Information from Leadership (34): Mention of the time frame information from leadership or higher is provided or the clarity of the message. Several responses acknowledged that the timely communication of policies, protocol changes, and other information from the leadership was helpful to their own workflows and decisions.
• •Availability of Policies Regarding PPE (17): Indication about having access to covid-specific instruction or documentation regarding personal protective equipment.
• •Availability of Information Regarding Procedures (15): Accessibility of information regarding new procedures or changes to existing procedures. This was the second most commonly mentioned factor that contributed to success 14 (31.8%) in non-leadership roles. E.g., new procedures relating to infection control while handling patients.
• •Availability of Information Regarding Social Distancing (6): Indication about having access to covid-specific instruction or documentation regarding social distancing.

The channels of communication and information flow established at the beginning of the pandemic were helpful to frontline workers in terms of knowing where to look for information. This was particularly important in enabling individual workers and teams to better align their own adaptations with the expectations of the organization, and synchronizing their activities with other individuals and groups. Such synchrony could be crucial to coordinating adaptive workflows organization-wide, especially under circumstances characterized by uncertainty and ambiguity.

3.2. Category B: Attitudes and initiative

This category comprises themes that reflect individual or collective attitudes, values and leadership, as factors that support successful adaptation and coordinated activity.

• •Leadership (23): Direct mention of supervisor or other management roles relative to the respondent. Leadership played an important role in terms of supporting local initiatives (e.g. team-level) by enabling resources and coordinating decisions across groups. Given the lack of procedures, formal policies and response mechanisms in the initial stages of the pandemic, such support was crucial to frontline staff adapting and synchronizing their efforts system-wide.
• •Culture and Attitudes (19): Indication of the nature and mindset of individuals or groups that the respondent encountered in their narrative. For instance, A culture of communication and sharing of information with staff and patients was exhibited through “… a very friendly, open and honest means of speaking to patients and staff to alleviate anxiety.” This theme reflects the heightened collaborative atmosphere in the hospital given that all stakeholders were faced with a common crisis, leading to greater cohesion and reciprocity.
• •Adherence to Protocol (9): Direct mention of protocol followed or not followed. E.g. continuously washing one's hands and keeping them sanitized.
• •Initiative (8): Clear display of individual going beyond protocol to fulfill need gaps. This theme included examples of individuals recognizing a need or opportunity in their environment for improvement that could benefit not just themselves but others as well, and acting accordingly on their own volition. This quote summarizes the theme: “had to step outside of scope to ensure operational issues or gaps that were not being filled with regard to patient scheduling, testing, canceling, rescheduling, communications, safety, communications and patient management in lieu of such by ordering providers, etc., at staggering lengths and volumes.”
• •Personal Concerns/Fears (4): Direct indication of personal concern or fear regarding procedure or virus spread as a driver of preventive or interventional approaches.

3.3. Category C: New and adjusted workflows

New and Adjusted Workflows examine the impact of the new way of life, especially in respect to the patient and the provider. New workflows had to be created for specific patient categories, such as persons under investigation (PUI), i.e. patients showing symptoms who have not yet tested positive for COVID-19: “Nursing developed a routine for patient arrival and waiting. We identified a suitable room for intubation and a work flow to keep staff and patient as safe as possible during an MRI with general anesthesia (GA) for PUI.” Participants also described adjusting their individual workflows or work-formats leveraging flexibility of work arrangements and connectivity, such as working remotely from home (to avoid risk of exposure to the coronavirus): “Working remotely helped increase work efficiency and decrease distractions.”; “working remotely required an adjustment in how I communicated about everything.” Respondents also mentioned several adaptations that reflected a patient centered process in terms of communicating with families, and being responsive to patient needs by revising scheduling and ordering procedures: “The sedation triage center decided to own all of the covid processes for the sedation/general anesthesia patients, where other services would defer that process to the ordering provider(s). We wanted to decrease the multiple communications to the families and cut out the ‘middle man’ to hopefully decrease confusion.”

• •Flexibility of Work Arrangements (19): Direct mention of alternate work arrangements as it relates to covid response. Respondents appreciated the flexibility to be able to work remotely as and when needed, such as attending daily huddles virtually, or the expansion of telemedicine capabilities allowing them to interact with patients remotely.
• •Patient Centered Process: Consulting Patient/Family (5): Indication of communication with patients/families as part of decision making. Increased communication and openness was important because patients/families were also under the duress of uncertainty and needed clarity and information given the new hospital procedures.
• •Patient Centered Process: Scheduling and Ordering (4): Indication of need identification in direct connection with patient scheduling, ordering or care, and revising those processes so as to serve patients while minimizing risk of infection.
• •Innovation (3): Direct indication of new procedure/process development and reasoning/methodology for actions (or need for new procedure/policy). E.g. “We experimented with methods to maximize participation and engagement including video and audio presence, touchdown at the beginning of the conference, technical monitor, chat box moderator, and fellows volunteering to take cases.”
• •Connectivity: Accessibility of Team (3): Mention of accessibility of coworkers and team in respect to alternative work arrangements, such as the ability to work remotely and participate in virtual huddles.

New and adjusted workflows reflect changes at system-levels that allowed individuals to adapt and function in new ways that were not previously possible. These were often departmental or organizational responses to the needs of both providers and patients.

3.4. Category D: Availability and utilization of resources

Availability and Utilization of Resources contains themes which examine the ease of access and use of resources. One participant displayed initiative and availability of resources where they would “Clean reading room/workstation myself, which required having adequate cleaning supplies and arriving earlier than scheduled.” Another participant mentioned interdepartmental coordination to mobilize resources for alternate work arrangements (AWA) to allow staff to work remotely from home: “working with other CHOP departments to get everyone everything they needed to be able to be fully functional for radiology AWA roles. This includes but is not limited to our Telemedicine ramp up.” At times, staff had to innovate and improvise to create new resources to perform care procedures safely: “During COVID pandemic response, the anesthesia, sedation, and radiology departments needed to invent a way to provide anesthesia services in an area outside the OR which does not have negative pressure capabilities or resources of OR.” As such, resources are inherent to adaptation and adaptive capacity.

• •Knowing Where to Find Resources (16): Indication that information or resources were sought out and retrieved rather than having to wait for them to be pushed to the participant. This theme is closely related to the earlier category of Information Flow in that staff were generally aware of where to look for resources based on the communications organization-wide regarding resources.
• •Availability of Resources (13): Indication as to the availability or lack thereof of proper resources to complete job duties properly and safely. A commonly cited category of resources was PPE, which critical to sustaining operations.
• •Experience and Knowledge of Coworkers (3): Reliance on the information provided by another coworker, potentially secondhand information from the ground truth source (CDC/WHO). This included BioResponse, a group responsible for centralizing and disseminating information relative to the pandemic.

3.5. Category E: Collaborations and Teamwork

Collaborations and Teamwork includes themes which focus on coordinated and combined efforts distributed across individuals and groups. The responses reflected multiple instances of collaboration and teamwork that supported adaptation at individual and group levels. One participant commented on the atmosphere of the team and department which enabled them to perform their work well: “There was a level of cooperation with colleagues to meet patient & family needs throughout the hospital even when I feel most comfortable on our radiology units. I was floated to the ED (emergency department) for half a shift in June and one of our CRNP's (Certified Registered Nurse Practitioner) helped me connect with our sedation RN (Registered Nurse) who was trained to do the fitting for N95s and allowed me to be fitted and have a mask before being floated to the ED. This made me feel more prepared just in case there was a trauma or other situation where I would be unsure if the patient had COVID but child life was needed in the room.”

• •Cooperation/Collaboration Between Coworkers (16): Indication that shared workload or collaboration is occurring. E.g. “My doctors, I supported (each other), and kept in constant contact through phone call, blue jeans meetings, and email.”
• •Shared Understanding Between Coworkers (6): Focus on the sharing of knowledge and attitudes between coworkers to ensure shared operating perspective and cohesiveness of team.
• •Interdepartmental Cooperation (4): Mention of multiple departments working together to achieve a common goal.

Collaboration, teamwork, and cooperation were key to adapting as an organization as resources, information needed to be shared, and specific issues had to be discussed and resolved in coordination between individuals and groups.

Following categorization, each individual instance of a theme was mapped to the applicable level of organization at which it occurred. The predominance of themes varied by system level. The relative dominance of each theme at each level is depicted in Fig. 2 .

Fig. 2.

Theme counts at each level of organization.

At level 1 (individual), the theme of ‘Culture and Attitudes’ was most frequently mentioned with eight occurrences. Second most mentioned was ‘Initiative’ with seven instances, followed by ‘Adherence to Protocol’ with six instances. The adaptability of staff at the individual level was apparent through the ‘Initiative’ theme appearing 7 out of 8 times at level 1.

Respondents indicated a high level of cooperation and sharing between colleagues and coworkers. These were related to sharing of information and other resources, as well as supporting each other with workload and flexibility of schedules. At level 2 (team), the narrative theme counts 12 mentions of ‘Cooperation/Collaboration Between Coworkers’. The second most counted theme, ‘Culture and Attitudes’ had just half the mentions. ‘Shared Understanding Between Coworkers’ was the third most popular theme at level 2 with just four occurrences. Across system levels 3 (department), 4 (hospital), and 5 (beyond), ‘Communication of Change’, ‘Timeliness and Clarity of Information from Leadership’, then ‘Leadership’ were the most frequently identified themes in order of their popularity. The high frequency of the first two most popular themes at levels 3, 4, and 5 are indicative of the value of efficient information flow. Two themes were mentioned across all five levels of organization: leadership and communication of change. This result is indicative of the influence that leadership has in communicating change both across and within organizational layers. Information flow themes (category A) were frequently identified in levels 3, 4, and 5. This further emphasizes that employees valued communication and availability of information, enabled through leadership at various organizational levels, as factors facilitating adaptive capacity. Collaborations and teamwork (category B) themes were most identified at level 2, which indicates coordination among groups within and across departments. New and adjusted workflow (category C) themes were most identified at level 1, indicating adaptations by individual frontline staff. In general, themes at levels 1 and 2 represent adaptive patterns at the ‘sharp end’ that are often informal in nature but enable resilient performance. Themes at levels 3, 4 and 5 represent coordination of distributed work across the organization, either in a centralized (e.g. via policies) or decentralized (e.g. inter-departmental coordination) manner.

Overall, the most theme counts were attributed to category A, ‘Information Flow’, with 106 theme instances in total. It was apparent throughout participant responses that the timely access and clarity of information was what influenced their overall adaptability.

Additionally, participants responded to the multiple-choice sections of the tool on challenges and concerns, and resources that were used in their examples of adaptation. Fig. 3, Fig. 4 show the counts for each of the responses in these sections. Among challenges and concerns (Fig. 3), respondents most frequently identified “uncertainty or ambiguity of situation” and “perceived risk of infection” as challenges or concerns which prompted response. These concerns were equally reflected through the responses to earlier covered open-ended questions.

Fig. 3.

Challenges and concerns that prompted the responses. PUI: person under investigation; PPE: personal protective equipment.

Fig. 4.

Resources that enabled adaptive performance in the participants' responses. PE: Personal Protective Equipment.

Seventeen responses were received in the free-text field under Challenges/Concerns. A majority of these related to ambiguity or varied information from multiple sources. In addition, in the beginning of the pandemic, there was some misalignment between the hospital and department-level policies and protocols which led to confusion among the workforce. This was reflected in a couple of free-text responses to this section of the tool: "Clearer communication between management and employees could have been a bit better", and "once or twice there were conflict between official policy, and how things were carried out between departments", indicating that the synergy between organizational layers wasn't quite perfect. However, these issues seem to have been addressed as time progressed and an alignment of policies with consistent and timely communication was achieved. Other challenges indicated in the free-text field included, the additional time and effort needed to clean rooms when a patient was perceived to be positive for COVID, and inconsistencies noticed in adherence to appropriate protocol among coworkers.

Among resources (Fig. 4), non-PPE Infection controls, PPE, and hospital-provided information were identified most frequently, all with 36 instances. Non-PPE infection controls included measures such as social distancing protocol, hand-sanitizers, and scheduling practices.

4. Responses to “What surprised you in working through the COVID-19 pandemic?”

Many responses to this open-ended question concerned infection rates, teamwork, and work-from-home measures. Respondents generally appreciated the hospital as an organization doing well to sustain its functioning through the pandemic. Specific attributes identified were.

• •The speed at which the hospital adapted its systems at scale: “The speed of adaptability; within 2 weeks we had complete alternate work arrangement[s].”
• •How communication and information played an important role in this adaptability: “Ability to communicate well and share information in real time.”
• •Flexibility of processes: “The need to be flexible with practice for the service changed according to systems and supplies changes. Impressed with the due diligence of the hospital safety team.”

5. Responses to “Could you describe any opportunities to improvise, that you discovered and leveraged to make the process safer?”

Although most of the responses to this question reiterated the aforementioned themes, respondents reported several specific strategies, workflows and interventions that weren't described in the initial open-ended question. These responses specifically reflect a typical feature of adaptation, which is to improvise upon, augment, or repurpose existing resources and workflows to meet evolving needs. Some of the specific strategies included.

• •The expanded use of virtual and remote interaction technologies for patient care: “Virtual conferencing improved our conference participation and virtual review of imaging with referring clinicians using BlueJeans, business Skype or FaceTime improved patient Care and physician collaboration.”
• •Forums for frequent and regular communication: “Standard mandatory team communication process every day less than 24 h apart.”
• •Streamlined patient-flow procedures for infection control: “We kept a team together that included clean and dirty technologists. We traveled to the floor to retrieve the patient so we kept them on one bed. When arriving in MRI we went directly into the warm zone space … When completed we took the patient back upstairs. This minimized exposure to areas in our department and to staff not involved in the care of the patient. A PPE monitor from the department traveled with the team as well opening doors clearing hallways and providing guidance for donning and doffing in the environment.”

Additionally, several respondents used this question to raise issues they've faced or suggested ways to improve existing policies and processes.

• •Need for enterprise-wide licensing and enabling of applications for remote/virtual format: “Unfortunately, the hospital doesn't have business Skype installed on hospital computers just on laptops. There needs to be a uniform hospital wide platform.”
• •Expanding current ‘local’ best practices in other areas system-wide: “a patient who has a trach or is vented when visiting any modality within radiology, is treated and handled using proper precaution … as a positive coronavirus patient would be with staff wearing all precaution … these policies should be implemented and communicated to all modalities.”

6. Impact on organizational learning and leadership decisions

Report to Leadership and Future Directions: Initial results from the RETIPS reports were presented to the leadership of the radiology department on November 19^th, 2020. The primary focus of the reporting was to introduce the leaders to the concepts of resilience, provide initial feedback on the COVID response to date, and provide suggestions for continued success based on findings. The leaders found the information captured through RETIPS highly valuable in terms of helping them understand how their frontline staff were coping with the uncertainty and challenges related to the pandemic. In particular, given the rapid pace of adaptations with limited resources available, the leadership found the tool to be a useful way to enable bottom-up communication and learning about evolving strategies at the sharp-end, as illustrated in this quote by the Director of Clinical Operations:

“the early days and first wave of the pandemic were a lot of rapid decisions and actions taken with incomplete information, having the RETIPS tool to help us collect and identify the actions taken and decisions made closer to the time they happened helps us respond better in the future”.

One such action taken was based on the finding that timely and clear communication from the hospital or department leadership played a crucial role in effective adaptation among caregivers. The emphasis on communication reflected in the RETIPS reports triggered an assessment of the completeness of Radiology's email distribution list. This effort led to the discovery that the sedation nursing team list was incomplete and therefore that group was not receiving the emails about COVID-related policies. As a result of this insight, the radiology and sedation nursing leadership came together to update the email lists to ensure the current assigned sedation nursing personnel were always included in the email distribution list. The importance of this intervention in department's (and the hospital's) overall effectiveness is summarized in this quote by the chief of Radiology Quality & Safety:

“The nursing team plays a central role in our operational effectiveness, and in helping achieve our overall quality and safety goals. The inconsistent coverage of nurses in our mailing list at the beginning of the pandemic was an important deficiency in our efforts to keep the entire workforce fully informed, that became conspicuous in light of the RETIPS reports. We were able to quickly fill this gap thereafter by having them included in our mailing list ensuring that all radiology involved staff are aware of the changing conditions and policies.”

The Quality & Safety chief also summed up, broadly, the importance of RETIPS as a learning tool during dynamic situations:

“I think the value of the RETIPS tool is it not only identified the problems we faced responding to an unprecedented pandemic, it also revealed the many ways our staff rapidly adapted to a completely new, and continuously shifting, clinical environment. Clinical operations usually place (understandably) emphasis on standardization and consistency; this is fine when the situation is stable, but when the ground of your clinical practice norms falls through, when information is constantly changing and sometimes conflicting, and when staff are fearful of contracting a deadly pathogen, rapid adjustment and learning is the key to keeping both patients and staff safe.”

Given the demonstrated usefulness of the tool, the hospital leadership encouraged continuous availability of RETIPS and revolving analysis.

7. Discussion

The study represents a Resilience Engineering (RE) approach toward understanding how things go well in everyday work based on frontline perspectives. The RE approach was also applied recently at long-term acute care hospitals, learning from ‘normal’ work processes and designing for response-readiness during the pandemic (e.g. Grigonis et al., 2021). In this study, the RETIPS tool was applied in Radiology-specific settings to allow the hospital to identify what enabled successful adaptation at operational levels during the COVID-19 pandemic. As a self-report method of identifying hospital adaptations during this pandemic, the study is unique in that it was conducted within an organization focusing on frontline workers, whereas previous surveys were multi-institutional and targeting leadership-level respondents (Auerbach et al., 2020; Linker et al., 2021; Mitchell et al., 2021). We received a high response rate in a short period of time coinciding with rapid transitions at multiple organizational levels. This could be an indication that caregivers and staff perceived the tool as an outlet to share their lived experiences, in addition to other forums such as townhalls and departmental meetings.

One recent study by De La Garza and Lot (2022) shared a similar focus on organizational resilience in a hospital during the first wave of the COVID-19 pandemic. The study was primarily based on the perspectives of the hospital's crisis management team gathered through interviews and recorded staff meetings, and found a strong relationship between anticipation and adaptation at various organizational levels. While our study throws some additional light on how a hospital department adapted during the pandemic, it also represents the use of a tool that could be used for continuous data gathering and organizational learning.

Exemplars of Adaptive Capacity from the Frontlines: The narratives reveal a number of processes and attributes of everyday operations that underpin the adaptative capacity of the organization as a whole. These are reflected in themes such as cooperation between coworkers, shared understanding and initiative taken by individuals. Several tactical decisions and innovations were made locally by individuals and teams in accordance with their operational needs and workflows. For instance, one of the interventional radiology rooms was converted into a negative pressure area temporarily as a location for handling known positive patients. The use of the 3D printing lab to generate face shields and goggles is another example. Several of the exemplars of adaptation identified through the current implementation of RETIPS at this hospital are similar to adaptations reported by other studies, such as altered entry pathways for patients, availability of procedures for use of PPE, and communication and information flow (Ertl-Wagner et al., 2020; Stennett et al., 2022; Tan et al., 2020). However, certain other adaptations reported in the literature were not mentioned, such as the creation of back-up teams, provision of mental health support to staff, or reconfiguring organizational crisis management structures (De La Garza and Lot, 2022; Stennett et al., 2022). Interestingly, the responses did not reflect the issue of elevated patient volumes, but did reflect concerns about handling infected or high-risk patients. This is not surprising given a previous study's findings that patient volumes in pediatric radiology dropped at the start of the pandemic. This was due to non-urgent and non-emergent cases being deferred to reduce potential exposure (Ertl-Wagner et al., 2020).

While some of these adaptations became normalized and known to the leadership, there were other local and informal adaptations that may not have been reported or widely known, such as a team leader purchasing safety googles for their team due to unavailability from hospital resources. Identifying such bottom-up adaptive processes can be helpful in better aligning top-down policies and interventions with emerging patterns that underpin adaptive capacity (Hollnagel, 2015a; Rankin et al., 2014). Enabling such alignment could help reduce latent errors being introduced or going unnoticed in the system (Reason, 1995). It appears therefore that the frontline workforce at the hospital was empowered to express sufficient initiative, at least to the extent indicated in the responses, to meet operational challenges during the pandemic. Expression of initiative is an important aspect of adaptive capacity that should be enabled by governance in organizational networks (Woods, 2019). While many studies report top-down driven strategies and adaptations, our study complements those studies with bottom-up perspectives on adaptation. Organizations can better align their policies and strategic thinking with frontline adaptive patterns by gathering such bottom-up perspectives (Hegde et al., 2019; Hollnagel et al., 2007).

Cross-scale Information Flow to Harness Initiative: Several of the examples reported may be seen as ‘expressions of initiative’, which is fundamental to adaptive capacity. Woods (2019) defines initiative as when “the unit begins to adapts on its own, using information and knowledge available at that point, without asking for and then waiting for explicit authorization or tasking from other units”. Given the time-critical nature of patient care, frontline providers had to express initiative based on their recognition of a need and knowledge. However, as Woods cautions, unchecked, initiative can lead to fragmentation and mis-synchronization across the system. It is therefore important that an organization's governance structure enables expression of initiative, but proactively seeks to identify and synchronize initiative across its adaptive units. The radiology leadership at the institution of study acknowledged the role of RETIPS in helping them build awareness of how its workforce is coping at various levels of scale. In that sense, the implementation of RETIPS during the pandemic itself was an exemplar of enabling information flow across levels of scale. In terms of impact, a system-wide follow-up action, i.e. the decision to include sedation nurses in the regular mailing lists, is non-trivial given the scale and immediate nature of the intervention. Further, nurses are vital to a hospital's resilient performance as they are typically closest to the patient's point of care and central to clinical workflows. Keeping them in the loop of communications from leadership was therefore critical.

The role of communication and coordination in enabling adaptive capacity across organizational layers, specifically during the COVID-19 pandemic, is consistent with findings from previous studies (e.g. Auerbach et al., 2020; Grigonis et al., 2021). Given the uniqueness of the situation, a sudden departure from normative workflows, and constantly varying information about the virus, frontline workers depended on their leadership for guidelines. The hospital leadership's proactiveness with communicating changes frequently and providing clear information about resources and procedures allowed radiology caregivers to bring local adaptations in alignment with organizational policies and instructions. Previously, Linker et al. (2021) used similar stratification to categorize adaptations reported in their multi-institutional survey on workforce planning during COVID-19. Although few in number, the free-text responses to the challenges/concerns question seemed to offer a balancing perspective on some of the themes reflecting synergy. For instance, one respondent mentioned "Clearer communication between management and employees could have been a bit better", and "once or twice there were conflict between official policy, and how things were carried out between departments", indicating that the synergy between organizational layers wasn't quite perfect. Such feedback could be followed up on by system administrators and leadership to proactively address concerns, tensions and risks.

7.1. Implications for Proactive Organizational Learning

The current and previous studies implementing RETIPS, have led to findings that are qualitatively distinct from what typically results from adverse event reports and their analyses. These include patterns of everyday adaptation that underpin the ‘normal’ functioning regardless of outcome. Therefore, RETIPS could complement traditional incident reporting systems with its focus on everyday adaptations. While event reporting leads to analysis of incidents to identify contributing causes and risks, analysis of RETIPS reports could help identify ways to support the adaptations that characterize everyday performance. Some of these adaptations, such as workarounds, are often seen as violations of norm in the light of the occasional negative outcome. Such learning could also serve to provide additional context around complexity and risk inherent to everyday work that does not result in negative outcomes because of the ‘silent’ adaptive patterns in frontline patient care. Thus, RETIPS could be used to expand organizational learning into a more proactive approach that helps mitigate the hindsight bias associated with investigating adverse events (Fischhoff, 2003). Safety interventions and policies could be made less reactive to incidents, and more proactively supportive of evolving patterns of adaptation around constraints faced in everyday work. Some practical ways of implementing RETIPS include embedding it within the workflows of the hospital's existing incident reporting system as an additional form within the reporting portal. In addition to making RETIPS available as a tool for continuous learning, it could be tailored to specific situations (e.g. a period of surge in patient volumes), areas (e.g. emergency medicine, surgery etc.), and roles (e.g. nurses, physicians, administrators), for more focused learning in such specific contexts. The tool should be made available to stakeholders at all levels of the organization in order to capture perspectives from frontlines to leadership. This could help identify cross-scale interactions and opportunities to enhance synergy, as with the aforementioned example of the sedation nursing group being included in the communication network.

As such, RETIPS is one tool that represents a proactive approach to organizational learning. However, in order to operationalize such a learning paradigm at scale, it would be necessary to create a larger framework that integrates data from multiple sources, such as clinician notes, weekly staff huddles, and morbidity & mortality (M&M) conferences to be analyzed with the lens of adaptation and resilience (Hegde and Jackson, 2022). Such a framework would no doubt involve a substantial and ongoing effort in terms of analysis and would have to be supported through investment of institutional resources dedicated to quality improvement and safety. Additionally, recent advances in the use of data analytic techniques such as natural language processing could be used to support analysis at scale (Fong et al., 2021; Young et al., 2019). Fig. 5 , adapted from Hegde and Jackson (2022), represents such a learning framework.

Fig. 5.

An organizational learning framework to support adaptive capacity (adapted from Hegde and Jackson, 2022).

Limitations and Future Work: A caveat to the findings is that the themes relating to what went well do not imply that those aspects always went well or that those couldn't have been improved. In fact, in several cases, participants described the same aspects as lacking or needing significant improvement. For instance, while timely communication was largely acknowledged as an important factor in successful adaptation, lack of timely communication of change was also mentioned in a few responses, particularly to the ‘challenges and concerns’ question. This may be because efficiency of communication would have varied across specific instances, contexts, events, or issues. The variation in responses only further emphasizes the role of prompt and timely communication in the hospital's adaptation as an institution. Similarly, while many narratives reflected ownership and responsibility by individuals, there were a few that complained about coworkers not complying with precautions, such as wearing appropriate PPE. Additionally, several responses comprised broader descriptions, such as “support of staff”, or “teamwork”, without specifying how these factors were instantiated. The lack of more detail or specific information limits the potential to translate such responses into actionable insights. This could indicate potential to improve the current RETIPS tool itself in its ability to prompt information not just about what was done, but how. Additionally, implementing RETIPS as part of the aforementioned learning framework, combining with information from other forums (e.g. M&M conferences) can help mitigate biases emanating from singular learning approaches.

Given the uneven distribution of roles and demographics among our participants, it was not possible to meaningfully parse themes by role and background. This challenge is further compounded by the fact that several participants may have had multiple roles, for e.g., attending radiologists may also have leadership or administrative roles in the department. However, it is definitely a consideration for future research and will require a much larger set of responses that would allow us to delineate thematic trends between diverse roles and backgrounds.

The questions relating to what surprised the respondent through the pandemic, and what opportunities to improvise were discovered, were entirely new additions to this version of RETIPS. While responses to these questions elaborated on the initial narrative responses, several also reflected challenges, issues, things that could have been done better, and opportunities to improve. The former question (surprises), however, did not yield substantive responses that could potentially inform policy. In future versions, this question could therefore be eliminated or revised. The latter question (discovered opportunities), however, does seem to be effective in eliciting concrete examples of specific adaptations in addition to the opening question, and should be retained for future versions of the tool.

While the tool remains available, the response rate drastically decreased after the initial wave of COVID as the hospital operations stabilized. To continue the process of learning proactively, it may be necessary to adapt RETIPS to the current situation and the hospital's transition beyond the pandemic. Future work involves adapting the tool to various clinical and procedural areas in radiology and other domains to support the overall care quality and safety framework at hospitals.

8. Conclusion

The study implemented a survey-tool, RETIPS, to help a radiology department at a large pediatric hospital to learn from its workforce, how they were coping at the onset of the COVID-19 pandemic, which as a large-scale disruptive event. The approach marks a shift from the traditional paradigm of learning from adverse events toward learning about patterns of adaptation in everyday work. In other words, learning about work-as-done. Analysis of responses from radiology staff and leadership revealed a number of themes underpinning the department's (and hospital's) resilience during the pandemic. These included the timeliness and clarity of communication from the leadership levels to the frontlines, adapted workflows enabled by flexible work arrangements, the availability of key resources, and a collaborative work environment. As such, the tool implementation represents a way for an organization to discover coping mechanisms and other adaptive patterns evolved across various levels of scale. Such an approach can help bridge gaps in information flow between the ‘blunt-end’ and ‘sharp-ends’ of the system and enhance synergy.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

APPENDIX. RETIPS_CovRad

RETIPS - Radiology

Health care is a complex and challenging field where patient safety is a major concern, particularly in the recent COVID-19 pandemic. Resilience in health care means overcoming the challenges and risks you encounter in everyday situations to ensure that both you and the patient are safe. We'd like you to help us understand, through examples from your own experience working during the COVID-19 pandemic response, how you ensure that the patient is protected from potential harm and maintain your own safety. The lessons you share are highly valuable in improving the hospital's ability to continuously improve the quality of care and patient safety.

To share your example of resilience, please complete the survey below.

ROLE

What is your Role (check all that apply).

• ●Administration
• ●Leadership
• ●Nursing
• ●PACS
• ●Radiologist
• ●Research
• ●Sedation/Anesthesia
• ●Technologist
• ●Other (specify)

Years in your present role.

• 0–4

• 4–8

• 9+

Think of an ‘adjustment’ or coping strategy that you may have employed that helps ensure effective and safe care to patients during the COVID-19 pandemic response.

(e.g. The COVID status (PUI/positive) is communicated well, and you were able to identify and obtain the necessary support materials, personnel and execute the procedures to enact patient care safely while confident you maintained your own infection risk safety.)

(e.g. being able to come into your work area and feeling confident that you could perform all of your job responsibilities safely in all conditions.)

(e.g. knowing what PPE or general infection protocols were in place and where to obtain appropriate materials)

DETAILED NARRATIVE

Describe in detail your example of an 'adjustment’ or coping strategy or preventive practice.

Themes to consider are: Sources of information, Types of information shared, when and how information was shared, availability of information, Anticipation; Innovation; Infection control Measures; Monitoring Behaviors (checks, reviews etc.); Resource Availability (PPE, hand sanitizer); Policies and Procedures; Cooperation with colleagues patients and families, engagement and support from leadership.

WHAT IS WORKING WELL

What factors contributed to the success of your response or practice (check all that apply)?

• ●Experience and Knowledge of co-workers
• ●Availability of information regarding procedures
• ●
  Availability of current standard practices

  • ○
    Policies regarding PPE
  • ○
    Policies regarding Social Distancing
  • ○
    Communication of change
  • ○
    Knowing where to find resources
  • ○
    Information regarding Aerosol Generating Procedures (proper PPE etc)
• ●Cooperation between Coworkers
• ●Shared understanding between Co-workers
• ●Culture and attitudes
• ●Leadership
• ●Timeliness and clarity of information from leadership and administration
• ●Other (specify)

If needed, please use this space to elaborate on your choices above, or to briefly describe any other factors that enabled the success of your response or practice.

CHALLENGES AND CONCERNS

What were some of the challenges/concerns that prompted the response described in your example (check all that apply)?

• ●Patient condition (PUI/Positive status) or behavior
• ●Communication issues (with staff or patient/family)
• ●Uncertainty or ambiguity in the situation
• ●Ambiguity or contradictions in information received from [hospital name]
• ●Limited resources (time, money, staff, equipment, technology, PPE, etc.)
• ●Policy issues
• ●Perceived risk of infection
• ●Other (specify)

If you would like to explain the challenges and/or concerns further, please use this space.

RESOURCES

What resources were necessary or helpful in this situation (check all that apply)?

• ●Adequate time
• ●Technology/Equipment
• ●Co-workers/Consults
• ●[Hospital name] provided information
• ●Radial
• ●Self retrieved information (Outside of [hospital name], i.e. WHO or CDC)
• ●Scientific Literature
• ●[Hospital name] Procedural Guidelines
• ●PPE
• ●Sanitizers, disinfectants, social distancing, and other non-PPE infection controls

AREA OF PRACTICE

Please indicate the primary clinical area/specialty that your example relates to.

• ●Radiologist
• ●Technologist
• ●Nursing
• ●Sedation/Anesthesia
• ●Research
• ●Administration
• ●PACS
• ●Leadership
• ●Other (specify)

SURPRISES

What surprised you in working through the COVID-19 pandemic.

DISCOVERED OPPORTUNITIES

Could you describe any opportunities to improvise, that you discovered and leveraged to make the process safer?