Implementing a Human Factors Approach to RCA²: Tools, Processes and Strategies

The Joint Commission requires that all serious patient harm events undergo a thorough root cause analysis (RCA) to determine why the event happened and how similar events can be prevented from happening again.[1] Unfortunately, traditional RCA has not been effective in improving patient safety, as evidenced by the recurrence of similar harm events following the completion of an RCA.[2, 3] The problem lies in the fact that traditional RCA was not designed to analyze the cause of most patient harm events (i.e., human error). Rather, RCA is an engineering method grounded in the physical sciences, designed to uncover the causes of equipment failures and manufacturing defects.[4] As such, traditional RCA methods generally assume that failures are linearly linked and traceable to a single root cause.[5] This root cause, in turn, can be uncovered through a process of asking a series of 5 “why” questions.[6] The answer to each question serves as the basis for the next, with the answer to the fifth question revealing the root cause. However, such methods and assumptions are not valid when applied to human error. Errors are caused by a breakdown in non-linear interactions among multiple, tightly coupled system variables (e.g., human, environment, task, technology and organizational factors).[7] Thus, traditional RCA methods are inherently incapable of reliably determining why errors occur or how they can be prevented from causing further harm.

In light of these shortcomings, the National Patient Safety Foundation disseminated a set of guidelines called Root Cause Analysis and Action, or RCA² (pronounced “RCA squared”).[8] These guidelines help users translate traditional RCA into a process for investigating and preventing harm associated with medical errors. Numerous changes to traditional RCA are contained in these guidelines, including the use of RCA teams who are independent of the event, have a working knowledge of human factors, and are specially trained to conduct an RCA. Other significant changes include interviewing individuals involved in the event, developing causal/contributing factor statements, identifying and implementing corrective actions, and the engagement of organizational leadership throughout this process. Unlike traditional RCA, RCA² is more akin to a formal accident investigation process than an engineering failure analysis method. In fact, the authors of RCA² mused that, “While it might be better not to use the term RCA [at all], it is so embedded in the patient safety culture that completely renaming the process could cause confusion.” [8]

The RCA² guidelines offer fundamental improvements to RCA investigations of patient harm events. These improvements, however, center mainly on changes to the process and procedures for conducting an RCA investigation. The tools contained in the document still only support traditional RCA activities, such as performing risk assessments of events, developing timelines, and rating the strength of safety recommendations. Current RCA² guidelines stop short of providing a foundational set of robust human factors tools and methods that enable users to (a) reliably perform a thorough human factors analysis of patient harm events and (b) systematically generate effective human factors interventions to improve safety.[8]

To address this critical gap, our team systematically integrated a complement of well-established human factors methods and tools into the RCA² process. Our approach is grounded in the Human Factors Analysis and Classification System (HFACS), as well as the Human Factors Intervention Matrix (HFIX) and FACES methodologies.[9–12] These methods and associated tools were originally developed by members of our team to facilitate the analysis and mitigation of error-related aviation accidents. They have since been adapted and successfully applied by our team and others across numerous domains, including mining, petrochemical, construction, maritime, rail, firefighting, forensics, agriculture, food service and amusement park accidents.[13–20] They have also been shown to facilitate the process of identifying and correcting the underlying systems issues associated with patient safety events.[21–25] When combined with RCA², this complement of methods and tools creates a robust human factors process, called HFACS-RCA², that is specifically designed to identify and prevent human factors and systems issues associated with patient harm events.

To date, however, there is no single source that describes each of these human factors methods and tools (i.e., HFACS, HFIX and FACES), or the process by which they can be effectively integrated into the RCA² approach. Furthermore, little has been published regarding healthcare systems’ experiences associated with implementing HFACS-RCA². Consequently, there is scant data describing its utility relative to traditional RCA methods (i.e., does HFACS-RCA² improve the type of systems factors and corrective actions identified?). There is also very limited information on the types of organizational factors that either facilitate or hinder implementation success.

The purpose of the current paper, therefore, is to provide a general description of the major human factors methods and tools that are part of HFACS-RCA², as well as to explain the general HFAC-RCA² process. We will also present the results of an 18-month qualitative case study detailing (a) changes in RCA findings and recommendations following HFACS-RCA² implementation and (b) barriers and facilitators to HFACS-RCA² implementation success within a large, Midwest academic health center. We conclude with a brief discussion regarding the broader implications that HFACS-RCA² has for healthcare risk management in general.

Human Factors Frameworks, Methods and Tools

HFACS.

The HFACS framework is based on the Theory of Active and Latent Failures, better known as the Swiss Cheese Model.[26, 27] HFACS assumes that accidents and incidents are rarely linked to a single cause or contributing factor. Rather, they often are the result of several inter-related factors that interact across four functional levels of an organization (see Table 1). The first level refers to the unsafe acts of individuals, who are actively performing the work associated with the event. Examples could be an emergency room physician misdiagnosing appendicitis for gastroenteritis or an ICU nurse miss-programing an infusion pump. The second level, preconditions for unsafe acts, involves factors that can directly affect a person’s ability to perform safely. Issues such as mental or physical fatigue, teamwork and communication, technology design and environmental conditions represent causal and contributing factors at this level. Moving beyond the individual and working conditions, factors associated with supervision are at the third level. In many cases, a breakdown in supervisory oversight, excessive work pressures, or a failure to enforce rules can result in preconditions that lead to unsafe acts. Above the supervisory level, organizational factors such as safety culture, operational processes or decisions made by senior leadership about the allocation of resources, such as money, equipment, and people can all impact supervisors and the preconditions that foster unsafe acts.

Table 1.

The Human Factors Analysis and Classification System (HFACS) for Healthcare

Organizational Influences

Organizational Culture: Shared values, beliefs, and priorities regarding safety that govern organizational decision-making, as well as the willingness of an organization to openly communicate and learn from adverse events.

Operational Process: How an organization plans to accomplish its mission, as reflected by its strategic planning, policies/procedures, and corporate oversight.

Resource Management: Support provided by senior leadership to accomplish the objectives of the organization, including the allocation of human, equipment/facility, and monetary resources.

Supervisory Factors

Inadequate Supervision: Oversight and management of personnel and resources, including training, professional guidance, and engagement.

Planned Inappropriate Operations: Management and assignment of work, including aspects of risk management, staff assignment, work tempo, scheduling, etc.

Failed to Correct Known Problem: Instances in which deficiencies among individuals or teams, problems with equipment, or hazards in the environment, are known to the supervisor yet are allowed to continue unabated.

Supervisory Violations: The willful disregard for existing rules, regulations, instructions, or standard operating procedures by managers or supervisors during the course of their duties.

Preconditions for Unsafe Acts

Environmental Factors

Tools and Technology: This category encompasses a variety of issues, including the design of equipment and controls, display/interface characteristics, checklist layouts, task factors, and automation.

Physical Environment: This category includes the setting in which individuals perform their work and consists of such things as lighting, layout, noise, clutter, and workplace design.  Task: Refers to the nature of the activities performed by individuals and teams, such as the complexity, criticality, and consistency of assigned work.

Individual Factors

Mental State: Cognitive/emotional conditions that negatively affect performance such as mental workload, confusion, distraction, memory lapses, pernicious attitudes, misplaced motivation, stress and frustration.

Physiological State: Medical and/or physiological conditions that preclude safe performance, such as circadian dysrhythmia, physical fatigue, illness, intoxication, dehydration, etc.

Fitness for Duty: Off-duty activities that negatively impact performance on the job such as the failure to adhere to sleep/rest requirements, alcohol restrictions, and other off-duty mandates.

Team Factors  Communication: The sharing of information among team members including providing/requesting information and the use of two-way (positive confirmation) communication.

Coordination: This category refers to the interrelationship among team members including such things as planning, monitoring and providing back-up where necessary.  Leadership: The team leader’s performance of his or her responsibilities such as the failure to adopt a leadership role or model/reinforce principles of teamwork.

Unsafe Acts

Errors

Decision Errors: Goal-directed behavior that proceed as intended, yet the plan proves inadequate or inappropriate for the situation. These errors typically result from a lack of information, knowledge or experience.

Skill-based Errors: These “doing” errors occur frequently during highly practiced activities and appear as attention failures, memory failures, or errors associated with the technique with which one performs a task.

Perceptual Errors: Errors that occur during tasks that rely heavily on sensory information which is obscured, ambiguous or degraded due to impoverished environmental conditions or diminished sensory system.

Violations

Routine Violations: Often referred to as “bending the rules,” this type of violation tends to be habitual by nature, engaged in by others, and tolerated by supervisor and management.

Exceptional Violations: Isolated departures from authority, neither typical of the individual nor condoned by management

According to the Swiss Cheese Model, adverse events occur when failures at each level of the system interact in a way that creates a window of opportunity for the event to occur. During an RCA investigation, HFACS and its associated tools (e.g., HFACS interviewing guide, causal-and-effect analysis tools, causal-factor statement templates) help facilitate the identification and analysis of the multiple causal factor pathways across the layers of an organization that led to the event. In doing so, RCA investigators are able to reliably and comprehensively identify a variety of opportunities for generating interventions to prevent the event from happening again.[9]

HFIX.

Although HFACS facilitates the identification of systems issues that need addressing, they do not support RCA investigators’ task of determining how such problems should be addressed. To support this activity, the HFIX methodology and associated tools were designed to facilitate the identification of corrective actions following an HFACS analysis. HFIX is modeled after the Haddon Matrix,[10] which was developed within the context of automobile accident/injury prevention. HFIX is a tool that helps users think broadly and strategically about how to reduce risk using established safety science principles. When utilizing HFIX, users juxtapose each HFACS causal-factor pathway with five specific types of intervention approaches. Each approach represents strategies commonly used by safety specialists within specific disciplines (e.g., systems engineering, behavioral science, and safety management).[10] Through a series of questions or prompts, the HFIX tool enables users to think systematically about how each approach to safety might be translated into possible interventions to reduce risk (see Table 2 for an abridged version of the HFIX tool with selected probes).

Table 2.

The Human Factors Intervention Matrix (HFIX) with example probes

Example questions for investigating fixes to Causal/Contributing Factors
Environment	Task	Technology	Individual/Team	Supervisory/Organizational
How could the workplace be redesigned to improve performance or team interactions?	How can the task be restructured so that it requires less reliance on human memory (i.e., use checklists or technology that signals the next step in the task)?	How could automation help in reducing the dependency on human performance of certain tasks?	How can the method of training delivery be improved or modified to enhance its impact on an individual’s knowledge and skills (e.g., use of simulation)?	How could methods be developed to improve a communication between supervisors and staff?
How could the noise level be modified or reduced to reduce fatigue, improve concentration, or enhance communication?	How could immediate feedback be integrated into the task to allow operators to know when they have done things correctly or incorrectly?	How could warnings or alarms be improved to increase operators’ awareness of hazards or the presence of abnormal conditions?	How could an individual’s stress and fatigue be reduced or monitored to improve safety and performance?	How could the awareness and appreciation of hazards and risk by supervisors be enhanced?
How could clutter be reduced or housekeeping be improved to make the working environment more conducive to safe and productive work?	How could errors in performing the task be reduced by having another team member check/verify important steps in the procedure?	How could controls be more easily identified and/or better designed in terms of shape, size, movement and other relevant considerations?	When individuals are working as a team, how can the responsibilities of each team member be more clearly defined?	How could the organization improve its process for recruiting and hiring people who are better qualified or more experienced?
How could the number of distractions in the environment be reduced to allow the operator to focus attention more fully on the task?	How could procedures be re-written so that they are less ambiguous or more germane to safety critical tasks operators perform?	How can tools or technologies be redesigned to enter into a “failsafe” mode when problems occur?	How could the content of training be developed or modified to improve an individual’s knowledge of procedures and/or tasks	How could the organization better promote, reinforce, or encourage safe practices?

FACES.

The HFIX tool was designed to foster divergent thinking about a problem (i.e., help generate as many different interventions as possible for addressing a causal-factor pathway). It was not designed to help RCA team members decide on which action or set of actions to recommend for implementation. Hence, to avoid this crucial set from being completed simply by the use of intuition or opinion, an adjunct decision tool for HFIX was developed. This adjunct tool is designed to help quantify the potential value of each intervention identified through HFIX using five well-established criteria: Feasibility, Acceptability, Cost/Benefit, Effectiveness, and Sustainability. Referred to as FACES, these criteria are integrated into a user-friendly rubric that supports a standardized and robust process for scoring, ranking and selecting a final set of safety recommendations (see Table 3).

Table 3.

FACES criteria and scoring rubric

	Can the change be implemented relatively easily or quickly?	The intervention does not exist today nor is it likely to become available in the near future; it is highly impractical and not suitable for your organization.	The intervention exists but is not readily available or will require modifications to better fit the context in which it is intended to be used.	The intervention is readily available and could be implemented in a relatively short period of time without much effort.
	Will those being impacted by the intervention readily accept the change?	The intervention will not be tolerated by those it impacts. People are likely to consistently resist the change and attempt to work around the change.	The intervention will be tolerated by those it impacts. There may be moderate resistance but attempts to undermine the change will not be wide spread.	The intervention will be readily accepted by those it impacts. People are likely to welcome the change and make every attempt to ensure it works.
	Does the benefit of the intervention outweigh the costs?	The cost of the intervention is exorbitant relative to its minimal expected impact on safety and performance.	The intervention is moderately expensive but cost could be justified by its expected benefit. Return on investment (benefits) is relatively equal to cost.	The cost of the intervention is nominal relative the expected impact on safety and performance.
	How effective will the invention be at eliminating the problem or reduces its consequences?	The intervention will not directly eliminate the problem or hazard and it relies heavily on willful compliance with the change and/or requires humans to remember to perform the task correctly.	The intervention reduces the likelihood of the problem or hazard occurring but relies in part on the human memory and/or willful compliance with the change.	The intervention will very likely eliminate the problem or hazard and it does not rely on willful compliance with the change or require humans to remember to perform the task correctly.
	How well will the intervention last over time?	The impact of the intervention will diminish rapidly after it is deployed and/or will require extraordinary effort to keep it working.	The benefits of the intervention may have a tendency to slowly dissipate over time and will require moderate efforts to maintain its benefits	The impact of the intervention will persist over time with minimal efforts being required to maintain its benefits.

HFACS-RCA² Process

HFACS-RCA² integrates its complement of human factors tools into the various steps of the RCA² process (see Table 4). As described in Textbox 1, these tools help support trained, interdisciplinary RCA teams in performing activities, such as (1) collecting factual information via stakeholder interviews to determine what happened, (2) performing cause-effect analyses to explain why the event happened, and (3) developing recommendations for preventing the same or similar events from happening again. The HFACS-RCA² process involves active support by senior leaders within the organization (i.e., executive sponsors), engagement of clinical stakeholders involved in the patient harm event, input from operational owners who are responsible for implementing changes, and interdisciplinary committees that provide oversight of the RCA process and implementation of recommendations.

Table 4:

Human Factors models, methods and tools

Model/Method Description	Associated Tools
The Human Factors Analysis and Classification System (HFACS): HFACS was designed to facilitate the identification and analysis of human factors issues during an accident investigation. The model represents system failures across four levels: unsafe acts, preconditions for unsafe acts, supervisory factors, and organizational influences. Specific types of failures are further differentiated and defined at each level within the framework using concepts derived from analyses of accident cases and human factors research.	1. HFACS reference guide including tables, definitions and exemplars for each of category within the framework 2. HFACS interviewing guide including questions and templates 3. HFACS cause-effect analysis and diagramming tools for identifying causal factors pathways 4. HFACS CCF tools and templates for developing CCF statements
The Human Factors Interventions Matrix (HFIX): HFIX is a structured problem-solving method that fosters divergent and creative thinking. When utilizing HFIX, users consider how each CCF statement identified during the HFACS-RCA2 process could be mitigated using five different approaches to safety. Each approach represents strategies commonly used by safety specialists within specific disciplines (e.g., systems engineering, industrial psychology, occupational safety and safety management). Through a series a questions and prompts, the method requires the user to think systematically about how each approach to safety might be translated into possible interventions to address a causal factor pathway.	1. HFIX reference guide including tables of intervention categories and descriptions of the HFIX process and tools 2. HFIX problem-solving checklist including questions and probes for thinking about solutions from 5 different perspectives (human, technology, environment, task, and supervisor/organizational approaches) 3. HFIX templates for performing the HFIX exercise on each CCF statement
FACES Framework: FACES is a weighted, multiple-criteria decision analysis methodology. The use of FACES enables RCA teams to evaluate their potential HFIX solutions for each CCF statement using 5 well-established criteria: Feasibility, Acceptability, Cost/Benefit, Effectiveness and Sustainability (FACES).	1. FACES guide, including dentitions and descriptions of each criterion 2. FACES scoring rubric that describes how ideas can be weighted or scored on each criterion on a scale from 1 (low) to 5 (high) 3. FACES template for listing and scoring each recommendation and computing total scores.

Textbox 1: Description of the HFACS-RCA² Process.

The process begins when a potential or actual patient safety event is identified and reported. The event report is then routed to an Event Review Committee. This committee consists of an interdisciplinary team of individuals with various expertise (e.g., clinical practice, organizational operations, quality management) who received training in the event review process. Together, they objectively review the report, the patient’s electronic health record and other available information regarding the event. They then use a standardized methodology for scoring the actual or potential severity of the event. An event with a severity score that reaches or exceeds an established threshold is required to undergo further RCA investigation. When this occurs, the committee chairperson activates the next steps in the HFACS-RCA2 process.	The second meeting typically occurs two weeks after Meeting 1. In this meeting, RCA team members share insights gleaned from their review of the interview summaries and other information. Based on this review, they finalize the original timeline and solidify “what” happened. Next, they use the HFACS framework and tools to structure their conversation around human factors and systems issues and organize their thoughts about “why” the event occurred. They use the HFACS analysis tools to perform a formal cause-effect analysis, identify critical causal factor pathways and develop causal factor statements. Next, they consider how the event and other similar events could be prevented from happening again. Using the HFIX and FACES tools, the team generates a list of potential solutions for each causal factor statement and selects their final set of recommendations.
When an RCA of an event is initiated, a senior leader within the clinical area in which the event occurred is identified and notified of this decision. This senior leader, who has received training in the HFACS-RCA2 process and is knowledgeable of their roles and responsibilities, serves as an executive sponsor for the RCA. The executive sponsor is responsible for providing oversight of the RCA investigation and subsequent recommendations. The executive sponsor attends and participates in meetings throughout the RCA process, provides guidance and support to the RCA team, and assures his/her administrative staff are available to help the RCA team with issues associated with logistics, communication and scheduling.	During the third meeting, which occurs approximately 1 week after Meeting 2, key stakeholders involved in the event and operational owners responsible for implementing changes meet with the RCA team and executive sponsor. Together, they discuss the team’s findings and recommendations. These stakeholders then provide feedback to the RCA team, particularly as it relates to the types of changes being recommended by the team (e.g., appropriateness, feasibility, potential barriers, etc.). The RCA team and executive sponsor use this feedback to create their report using a standardized HFACS-RCA2 report template.
Upon initiation of an RCA, the “on call” members of the RCA team are notified and activated. RCA teams consist of interdisciplinary, trained “core” members including physicians, nurses, risk managers (RM) and quality improvement specialists (QIS). Teams may also include trained, ad hoc members from other specialties as needed. The designated RCA team facilitator is a RM or QIS who has experience in team facilitation. Upon activation, the RCA facilitator uses their HFACS-RCA2 planning guide to gather initial information about the event and prepare for the team’s first meeting.	Following Meeting 3, the RCA facilitator finalizes and submits the team’s report to an RCA Oversight Committee for inclusion on the committee’s next meeting agenda. Members on this oversight committee include leadership stakeholders, trained in the HFACS-RCA2 process, from across various clinical services within the organization. During the committee’s meeting, a summary of the RCA report is presented by an RCA team member, the executive sponsor, and the operational owners of the proposed changes. The committee provides feedback regarding the appropriateness of the recommendations based on the RCA findings and their knowledge of prior efforts to remedy similar problems. The committee also helps identify resources needed to make the change(s) and develop strategies for facilitating implementation.
The first meeting occurs 3–5 days after the team is activated at a pre-established time on RCA team members’ calendars for the period in which they are “on call.” During this meeting, the RCA team discusses what is known about the event. They develop an initial event timeline and determine additional factual information they will need for their analysis of why the event occurred. Then, the team identifies individuals they want to interview and the goals of these interviews. Using the HFACS-RCA2 Interviewing Guide and Templates, they develop specific interview questions that focus on human factors and systems issues. Interview assignments are then agreed upon by the team. When possible, assignments are based on RCA team members’ clinical backgrounds and the backgrounds of the interviewees (e.g., nurses interview nurses, physicians interview physicians, pharmacists interview pharmacists, etc.). The team then adjourns to perform their designated activities.	Once recommendations are approved, the operational owners, with executive sponsor support, are responsible for implementing the changes. The RM or QIS who facilitated the RCA associated with the change is also available to provide advice and support as needed; however, they do not own the change or perform the work. Operational owners are required to submit regular progress reports to both the executive sponsor and an interdisciplinary Patient Safety Committee that oversees quality and safety improvement efforts across the organization. Milestones are tracked by the patent safety manager and reported to this committee. The Patient Safety Committee works with the operational owner and executive sponsor to overcome any barriers until changes have been fully implemented.
Following Meeting 1, the RCA facilitator works with the executive sponsor’s administrative assistant to coordinate times and locations of interviews. If there are barriers to scheduling interviews, the executive sponsor helps navigate and resolve issues (e.g., release clinical duties for those being interviewed). During each interview, RCA team members use the interview questions generated by their team to structure the conversation around the HFACS framework. After the interview, team members upload their notes into a shared folder located on a secured server. This folder serves as a share point for information regarding the case and affords team members the ability to access and review materials before the next meeting.	Upon completion of the implementation process, the Patient Safety Committee and Patient Safety Officer monitor change effectiveness by tracking the frequency in which similar events reoccur. This information is then used to provide feedback to RCA teams, executive sponsors, and the RCA Oversight Committee to inform their work on subsequent RCA investigations.

Qualitative Case Study

As stated previously, prior research on HFACS-RCA² has focused primarily on the development and utility of the individual tools. However, there is scant information regarding whether or not the entire HFACS-RCA² process, as just described, can be successfully deployed. Therefore, we conducted an 18-month implementation study of HFACS-RCA² within a large, Midwest academic health center to evaluate its feasibility and identify any organizational variables that might serve to either facilitate or hinder implementation success. The results of our qualitative analyses of structured interviews conducted with key stakeholders involved in this process, including risk managers, quality improvement specialists, patient safety managers, and senior leadership are presented here. Our discussion will highlight findings associated with HFACS-RCA² outcomes (i.e., types of causal factors identified, and corrective actions generated), as well as the various barriers and facilitators/strategies to implementation that were identified. In addition, we will discuss some unintended, albeit positive, stakeholder and organizational consequences that also emerged as a result of HFACS-RCA² adoption (e.g., changes in culture).

METHODS

Data Collection

Case Description

This study took place within an integrated, nonprofit academic health system that serves over 600,000 patients each year in the Upper Midwest, with 1,400 physicians and 16,500 staff at 6 hospitals and 80 outpatient sites. The health system’s operations included a centralized quality office responsible for conducting all RCAs of patient harm events across the enterprise. Prior to this study, the healthcare system’s RCA process involved the use of ad hoc teams of untrained healthcare providers, facilitated by experienced quality improvement specialists and risk managers. The roles and responsibilities of RCA team members and the tools utilized during investigations were variable across teams.

During this study, however, RCAs of all serious patient harm events were performed using the HFACS-RCA² process (see Textbox 1). Implementation included an initial 8-hrs of HFCS-RCA² training for RCA facilitators (quality improvement specialists and risk managers), predesignated clinical RCA team members (e.g., physicians, pharmacists, and nurses), as well as middle managers and senior leaders involved in the RCA process. Following a run-in phase, in which stakeholders practiced using the HFACS-RCA² on a small sample of events (n = 3), a full rollout occurred, with stakeholders applying HFACS-RCA² to all serious safety events identified over the 18-month study period (n = 48), which involved either actual or potential harm.

Participants

Key stakeholders involved in the implementation of the new HFACS-RCA² process were interviewed upon study completion. These included RCA team members (quality improvements specialists [QIS; n = 3] and risk managers [RM; n = 2]), mid-level managers (MM; n = 3), and senior leaders (SL; n = 4). These participants were recruited and selected based on their familiarity with both the organization’s legacy RCA processes and experience with the new HFACS-RCA² process (i.e., they had been involved in several HFACS-RCA² investigations during the course of the study). Clinical RCA team members, as well as providers and patients involved in harm events, were not included in our sample. These individuals did not meet inclusion criteria and/or privacy and confidentiality issues precluded their recruitment. Prior to recruitment, the local Institutional Review Board approved this study.

Procedure

A semi-structured interview guide was developed based on a structure, process, outcomes (SPO) framework.[28] This paper focuses on responses to these interview questions regarding outcomes associated with the implementation of the HFACS-RCA² system, including causal factors identified and recommendations developed. Responses to questions regarding stakeholders’ experiences with different facilitators and barriers during the implementation process will also be presented. Each interview was conducted by one of the authors in a private room and audio recorded. Interviews lasted approximately 1 hour. Audio recordings were transcribed by a HIPAA-certified transcriptionist and then uploaded into qualitative analysis software NVivo 12.6.0 (QSR International) for coding.

Data Analysis

Data analysis was conducted according to directed content analysis.[29–30] Two researchers read the transcripts and coded excerpts related to HFACS-RCA² outcomes, which were defined as any statement that referenced an impact, effect, or consequence of implementing the new process. These outcome excerpts were placed into categories related to RCA findings or recommendations and analyzed for themes. Statements regarding outcomes that did not fit within either category were further reviewed to identify additional categories and themes. A similar process was followed for coding experts related to facilitators and barriers, which were defined as factors that supported or hindered the implementation process. Throughout the process, researchers independently coded each transcript and reviewed their findings together with the research team for consensus. Member checking and skeptical peer review were periodically completed.[31, 32] Discrepancies were resolved through discussions among coders and the research team. This process was repeated when new themes emerged.

RESULTS

Outcomes

Our analysis initially focused on RCA outcomes, particularly outcomes related to RCA findings and recommendations. However, several other outcome-related categories and subcategories emerged. These included Stakeholder Outcomes (RCA Team Members, Leadership, and Providers Involved in the Patient Harm Event) and Organizational Outcomes (Cultural Changes Surrounding Transparency). Each of these outcome categories, subcategories and related themes are described below.

RCA Outcomes: RCA Findings

Types of Causal Factors Identified.

Stakeholders across all roles consistently indicated that HFACS-RCA² improved the identification and analysis of higher-level systems issues associated with patient harm events. Compared to their former RCA process that identified mostly preconditions for unsafe acts (i.e., local working conditions), HFACS-RCA² fostered the identification of a greater number and variety of supervisory and organizational factors, in addition to these preconditions. As noted by one MM, RCA teams are “…finding more cultural organizational-level issues as opposed to…staying down at that unit or local-level issue.” Some stakeholders stated that their former RCA process did not support them digging deeply enough into supervisory and organizational factors. However, the HFACS-RCA² provided them the impetus to thoroughly examine these issues and complete a “deeper and [more] meaningful analysis than [they] did before” (RM).

Interpretation of Findings.

Given the HFACS-RCA² methodology fostered the identification of more causal factors, some stakeholders indicated that organizing and interpreting RCA findings (e.g., writing causal factors statements) was potentially more difficult. Nevertheless, they also indicated that HFACS-RCA² processes and tools supported RCA teams’ abilities to analyze and draw connections between the various levels or types of causal factors identified. One QIS stated that RCA teams were “able to make those connections” between the “low-hanging fruit” (preconditions), and the higher-level systems issues. Stakeholders indicated that this process fostered a deeper understanding of the event and the ability to effectively communicate their findings when presenting their report to leadership. One SL described how the “information [they are] seeing is so much better organized…it’s distilled, because it’s going through the different levels of the event.”

RCA Outcomes: Recommendations

Quality & Characteristics of Recommendations.

Compared to the former RCA process, stakeholders in this study consistently indicated that HFACS-RCA² increased the number of corrective actions that targeted broader system-level issues, in addition to local changes within the system. One MM noted, “In our past, we would modify policies, or reeducate people. And that just wasn’t getting to the root causes. You can do that until you’re blue in the face, but until you block that hole up top, you’re gonna continue to have those issues…So I think we’re finding more organizational vulnerabilities and how to fix those.” However, some stakeholders also indicated that broader systems changes take considerably longer to implement and produce results. Therefore, actions are needed strategically across the different layers in the system, some of which can be implemented quickly while others will take more time. Stakeholders indicated that the HFACS-RCA² process supported the identification of, “…a much more comprehensive solution…that has a chance of working” (SL).

Stakeholder Outcomes: RCA Team Members

Collective Contribution and Commitment.

RCA team members (QIS and RM) consistently stated that the HFACS-RCA² process fostered a team-based approach to improving patient safety, as well as increasing collaboration and cooperation among RCA team members and across departments (e.g., Quality Improvement and Risk Management). Given HFACS-RCA² enabled RCA investigations to explore higher-level organizational issues, RCA team members stated they gained a deeper understanding of the healthcare system and the dynamics (formal and informal) within and across departments. RCA team members also expressed a sense of professional satisfaction in their work. Their ability to conduct investigations which identified substantive root causes that led to their recommendations being adopted by leadership allowed them to see tangible results from their labor. This feeling of making a difference tended to further increase the motivation of RCA teams. As stated by one QIS,“…people [on the team] really have a different hunger now…they’re coming in ready to make things better.”

Stakeholder Outcomes: Leadership

Trust in the Process.

The robustness of the HFACS-RCA² methodology fostered greater levels of trust in the RCA process among SLs and MMs, who served on the RCA review committee. As a result, members on this committee spent less time critiquing the work performed by the RCA team and instead spent more time discussing the RCA team’s recommendations and how to implement them. As one QIS noted,“…they know the quality of the analysis that’s been done…[leadership] knows there’s a rigor in the approach, [they] don’t have to worry about which team did it, [they] know that every team’s gonna do a rigorous job…So, they can feel much more comfortable now talking about, ‘Here’s the recommendation, how do we help that get done?’.”

Stakeholder Outcomes: Providers Involved in a Patient Harm Event

Sense of Closure.

Although providers involved in patient harm events were not interviewed as part of this study, the RCA stakeholders believed that the HFACS-RCA² process afforded these providers an opportunity to emotionally heal and gain closure from the event. Through the interview process, providers involved in the event were able to share their perspective without being accused of wrongdoing. Additionally, these providers were invited to attend the RCA team’s third meeting, during which findings and potential recommendations were discussed. Stakeholders believed that those providers who attended this meeting were able to see how their willingness to talk to the RCA team helped contribute to system-wide improvements, not to blame. The focus on inclusion and improvement facilitated feelings of relief and support among providers involved the patient harm event. One QIS stated,“… I think it gave some of the frontline staff some closure. That, you know, it’s not a finger-pointing thing, it’s not a ‘You messed up,’ it’s a, ‘This happened, we’re gonna fix this system…and hopefully prevent this stuff from happening in the future’.”

Organizational Outcome: Culture

Changes Surrounding Transparency.

In general, stakeholders believed that the organization became more transparent with regard to the RCA process and causes of patient harm. Leadership gained greater awareness of how systems issues can impact provider performance and began to talk about patient harm differently and openly. One SL commented, “I don’t know if we were that transparent with RCA reports before, and now we are. Even outside of…the RCA process, we’re going around to different groups and sharing awful cases…before people would have been really nervous about that. Like we’re not supposed to air that stuff out…[so it is a] big change, yeah.” One MM provided their perspective on cultural changes within the organization, stating, “…there are some qualitative things that we are already seeing change about the organizational culture that we didn’t measure but are palpable.”. Stakeholders also commented on the sustainability of these changes. As one SL described,“…the value of this [HFACS-RCA² process] is up and down the organization. I think if [someone] said we’re going to sunset the [new] RCA program, it’s not going to happen.”

Facilitators, Barriers, and Strategies

Several facilitators, barriers, and strategies related to both the HFACS-RCA² and implementation processes were identified. These are detailed in Table 5 and summarized below.

Table 5:

Key facilitators, barriers and strategies

Category	Description of Facilitators, Barriers/Challenges and Strategies
HFACS-RCA2 System
Usability	RCA team members indicated that the HFACS-RCA2 materials, including the handbook, tools, and templates were well-designed and easy to use. However, they also indicated that some of the tools, such as the interviewing guide, were too generic and needed to be more specific for each individual case. To address this issue, stakeholders tailored the questions within the interviewing guide to make them relevant for the type of event being investigated. Stakeholders also indicated that the hardcopy of the tools were not flexible enough for use during team meetings. As a result, they developed digital versions of the templates in Microsoft Word and Excel to make them more flexible, easier to share, and simpler to archive.
Utility	Stakeholders indicated that the human factors frameworks (e.g., HFACS and HFIX) and their associated tools were appropriate and useful for investigating the types of patient harm events that were occurring within their healthcare system. They also felt that the methods and tools helped them become more efficient in structuring their investigation activities, including interviewing healthcare providers, analyzing data, developing causal factors pathways, creating causal factors statements, and developing recommendations. One of the challenges encountered was determining the best way to synthesize of the information and integrate it into the existing RCA report format. To address this challenge, stakeholders redesigned their RCA report template to make it compatible with the new HFACS-RCA2 system.
Training Curriculum	Stakeholders stated that the training they received was effective, and that the mock interview activities and team-based case study prepared them to perform these activities upon completion of the course. However, they stated that some RCA team members, particularly ad hoc members, could forget some of the material, because several weeks could go by before they participated in an RCA investigation. They also indicated that additional practice and use of some of the tools and techniques were needed to fully understand them. For example, they reported particular difficulty in developing casual and contributing factors (CCF) statements, even after training and practice. To address these challenges, a refresher video was created and an additional 1-hour follow-up training on CCF statements was provided.
Implementation Process
Planning and Preparation	Planning and preparation activities were highlighted by stakeholders as being very important for success. These activities included developing a communication strategy to inform both leaders and staff about the rationale for the change; strategically recruiting safety-focused clinicians to be part of the new RCA program, ensuring everyone was trained, anticipating changes in workload among RCA teams, negotiating changes in roles and responsibilities across the quality and risk management departments, establishing team schedules and team member rotations, and establishing milestones and metrics for success. Challenges included competition for time and resources with other organizational initiatives that were occurring. Aligning the HFACS-RCA2 program with these other initiatives and priorities, to make it a complementary rather than competing effort, was one strategy used to overcome this potential barrier.
Leadership Engagement	Leadership engagement was repeatedly mentioned by stakeholders as being vital to successful implementation. Leadership engagement was instrumental in recruiting clinical RCA team members, helping ensure that interviewees were allotted time away from their clinical duties to participant, and supporting RCA recommendations that required broad systems-level changes. However, a potential challenge was that not all leaders felt equally familiar with how to perform these tasks; consequently, leadership engagement was somewhat variable early in the process. Soliciting regular feedback from leaders, particularly executive sponsors of RCA teams, was a strategy used to identify problems and implement changes to support leaders in their new roles.
Roles and Responsibilities	Having a formal document that clearly described roles and responsibilities of stakeholders involved in the process helped facilitate implementation. In particular, clarity of roles and responsibilities helped RCA facilitators to feel comfortable distributing workload among the team and it helped team members hold each other accountable for doing the work. However, a frequently mentioned challenge was that it took time for RCA team members, managers, and senior leaders to become comfortable in their roles (see Leadership engagement). Occasional updates to the roles and responsibilities document were needed as stakeholders gained more experience with the process, to ensure that expectations for each role were clearly defined.
Management of Change	Mid-level managers played a key role in managing and coordinating change. Through the use of daily huddles, managers were able to guide RCA team facilitators as they shared experiences, challenges, and strategies for utilizing the new RCA system. Managers were able to monitor progress and ensure that actives and strategies for dealing with challenges did not undermine the fidelity of the HFACS-RCA2 approach. One particular challenge encountered by RCA teams was managing workload and completing tasks in the timeframe allotted. Strategies for addressing this challenge included developing detailed agendas each meeting, distributing workload among RCA team members and identifying tasks that could be completed between meetings, rather than during each meeting. As staff rotated on and off RCA teams over the 18 -month study period, another challenge was to ensure that team members were trained and that RCA facilitators possessed the necessary competencies to be effective team leaders. To address these challenges regular training sessions were conducted and new RCA facilitators shadowed experienced facilitators prior to leading their own team.

HFACS-RCA² System

Characteristics of HFACS-RCA² Method and Tools.

RCA team members generally preferred HFACS-RCA² compared to the former process. One MM stated,“…the HFACS and the HFIX [tools]…were something that [we] didn’t have, so I think [the method] was stronger.” RCA team members also indicated that the HFACS tools were well-designed, appropriate for types of patient harm events that were occurring within their healthcare system and were useful in supporting activities associated with each step of the RCA process. One QIS stated, “I like the way they’re designed…it’s helpful…from a usability standpoint, I haven’t had any issues using it. It’s been a good resource.” Stakeholders stated that the training they received was helpful, but sometimes RCA team members, particularly ad hoc members, would forget some of the material, because several weeks could go by before they participated in an RCA investigation. They also indicated additional practice and use of some of the tools and techniques were needed to fully understand them.

Implementation Process

Characteristics of the Implementation Process.

Stakeholders described several facilitators and strategies used to support implementation. For example, they emphasized the importance of crafting a change message and communicating, early and often, about the reasons for change. One SL noted that they went through a “… pretty in-depth process of evaluating…who needs to know, how will we communicate it, who will do [the communicating]…so, we had a pretty robust communication plan.” They also indicated that leadership engagement was vital to successful implementation, given their role as opinion leaders and their ability to remove barriers that might hinder adoption. The need to clarify roles and responsibilities and ensure that each RCA team member had the appropriate skill set to perform their tasks (e.g., facilitate RCA teams) was also mentioned by MMs. Finally, managing workload was a potential barrier, as the amount of work involved in the RCA process had increased. As one RM stated, “there’s definitely more work…preparing for each session.” One SL noted that during meeting 3, RCA teams may have difficulty getting through all their recommendations in the time allotted and may only spend “…a little bit of time on the FACES [tool].” RCA team members commented that they learned to distribute work among the team, rather than the facilitator completing most of the work, as with the former process. They also learned to improve efficiency, such as requiring team members to perform activities online in-between meetings, in order to create more time for discussing recommendations during meeting 3.

DISCUSSION

Stakeholders in this study included risk managers, quality improvement specialists, mid-level managers, and senior leaders within a large, Midwest academic health center. Members from each of these stakeholder groups consistently indicated that HFACS-RCA² improved the identification and analysis of higher-level system issues associated with patient harm events. Compared to their former RCA process that identified mostly preconditions for unsafe acts (i.e., local working conditions), HFACS-RCA² fostered the identification of a greater number and variety of supervisory and organizational factors, in addition to preconditions. These findings are consistent with prior HFACS research across numerous domains, including healthcare.[9–25] Additionally, our findings support prior research showing that integrating human factors into RCA activities fosters the development of more systems-focused corrective actions.[22, 33] Specifically, stakeholders in this study consistently indicated that HFACS-RCA² facilitated the identification of more comprehensive and feasible solutions compared to their former RCA process, which previously produced corrective actions that focused mostly on education and policy updates.

Several emergent, unintended outcomes were also identified. Specifically, RCA team members reported that using HFACS-RCA² improved their motivation and satisfaction with their work, because they witnessed their efforts making a difference. Leadership also developed a greater trust in the process, knowing that the methodology was robust and thorough. As a result, they were able to spend more time focusing on recommendations and solutions, rather than scrutinizing and debating the RCA findings. Stakeholders also indicated that by participating in interviews and contributing to solutions, providers involved in the patient event found the process cathartic. Such outcomes have important implications for aiding the recovery of second victims.[34] Finally, the focus on identifying underlying systems causes of events fostered more transparency in the RCA process and discussions of patient harm events throughout the organization. An SL stated that this change in culture, along with the other emergent outcomes, increased the likelihood of HFACS-RCA² efforts being sustained after completion of this study.

Numerous facilitators and barriers related to the HFACS-RCA² methods and implementation process were identified. Members of RCA teams indicated the HFACS-RCA² tools were user-friendly and helped structure their investigation activities, such as conducting interviews, identifying causal factors, and developing recommendations. Nonetheless, they reported that time and experience was needed to fully learn how to use these tools and that refresher training could be useful. Developing a strategy for clearly communicating the need for change and engaging leadership early on in the process was also highlighted by stakeholders as important for implementation success. Indeed, numerous implementation studies support this finding.[35–37] Managing workload was one of the biggest challenges associated with the implementation process. However, proactive scheduling of teams and the ability of RCA facilitators to distribute workload among team members and between meetings were strategies used by stakeholders to overcome this potential barrier.

Limitations

Results of this case study indicate that implementing HFACS-RCA² is feasible and may be more effective than traditional RCA in identifying and correcting systems causes of patient harm. These findings are consistent with, and expand upon, previous research.[20–23] However, caution should be taken in interpreting these findings too liberally. The current study was conducted at a single academic health center and involved the use of qualitative outcomes. Research is needed to determine the feasibility of implementing HFACS-RCA² within other healthcare systems, including smaller systems with fewer resources devoted to RCA activities. Analyses are also needed to explore the impact of HFACS-RCA² on quantitative outcomes, including data associated with causal factors, recommendations, and patient harm rates. Finally, a follow-up study is required to determine whether the HFACS-RCA² approach is sustainable.

Implications for Healthcare Risk Management

HFACS-RCA² provides a robust human factors framework for classifying, coding, and archiving causal factors across patient harm events. Hence, for Healthcare Risk Managers, this affords an opportunity to analyze causal factor data globally across the enterprise and identify systemic trends in the types of human factors that repeatedly cause harm. Such global data allows for the prevalence of system hazards to be more clearly identified and the need for major systems changes to mitigate risk more easily justified.[38] Furthermore, the ability to trend the occurrence of various types of causal factors across events, provides Healthcare Risk Managers with the empirical data they need to evaluate the efficacy of their efforts in reducing risk (i.e., return on investment). Such data is invaluable when deciding whether to reinforce or reallocate resources to various risk reduction efforts.

Conclusions

The RCA² guidelines offer fundamental improvements to RCA investigations of patient harm. However, these guidelines stop short of providing a foundational set of tools for conducting a thorough human factors analysis of these events. To address this gap, our team systematically integrated a complement of well-established human factors methods and associated tools into the RCA² process. The objectives of this paper were to provide a general description of these methods and explain how they can be used to support various steps in the RCA² process. Our goal was also to present results from a qualitative case study detailing improvements in the nature of RCA findings and recommendations, as well as facilitators and barriers to successful HFACS-RCA² implementation. In doing so, our expectation is that other healthcare systems will be more willing and able to adopt this promising approach to investigating and preventing patient harm in the future.