Meeting Report

Consensus Recommendations for the Conduct, Training, Implementation, and Research of Perioperative Handoffs

Agarwala, Aalok V. MD, MBA; Lane-Fall, Meghan B. MD, MSHP; Greilich, Philip E. MD, MSc, FASE; Burden, Amanda R. MD; Ambardekar, Aditee P. MD, MSEd; Banerjee, Arna MD, FCCM; Barbeito, Atilio MD, MPH; Bryson, Trenton D. MD; Greenberg, Steven MD; Lorinc, Amanda N. MD; Lynch, Isaac P. MD; Pukenas, Erin MD; Cooper, Jeffrey B. PhD

Author Information

Department of Anesthesia, Critical Care, and Pain Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, [email protected]

Department of Anesthesiology and Critical Care, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania

Department of Anesthesiology & Pain Management, University of Texas Southwestern Medical Center, Dallas, Texas

Department of Anesthesiology, Cooper Medical School of Rowan University, Camden, New Jersey

Department of Anesthesiology and Pain Management, University of Texas Southwestern Medical School, Dallas, Texas

Department of Anesthesiology, Vanderbilt Medical Center, Nashville, Tennessee

Department of Anesthesiology, Duke University & Veterans Affairs Medical Center, Durham, North Carolina

Department of Anesthesiology and Pain Management, University of Texas Southwestern Medical School, Dallas, Texas

Department of Anesthesiology, NorthShore University HealthSystem, Dallas, Texas, Department of Anesthesiology and Critical Care, University of Chicago, Chicago, Illinois

Department of Anesthesiology, Monroe Carell Jr Children’s Hospital at Vanderbilt, Nashville, Tennessee

Department of Anesthesiology and Pain Management, University of Texas Southwestern Medical School, Dallas, Texas

Department of Anesthesiology, Cooper Medical School of Rowan University, Camden, New Jersey

Department of Anesthesia, Critical Care, and Pain Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.

Anesthesia & Analgesia 128(5):p e71-e78, May 2019. | DOI: 10.1213/ANE.0000000000004118

Free
SDC

Health care–related transitions of care (ie, handoffs or handovers) are often associated with patient harm.^¹ Care transitions have been identified as a high-priority safety concern by providers in several safety climate surveys^² and are an area of focus of health care regulatory agencies.^¹^,^³ Poor communication at the time of handoff has been implicated in adverse outcomes including diagnostic testing errors,^⁴ delays in diagnosis and treatment,^⁵^,^⁶ increase in patient harm,^{^7–10} higher hospital readmission rates and costs,^¹¹^,^¹² and an increase in malpractice claims.^¹³^,^¹⁴ One study estimated that up to 80% of serious medical errors involve communication failures between care providers during the transfer of patients.^¹⁵

Although most of the evidence on handoffs is not focused on perioperative care, handoffs during perioperative care are equally prone to failures.^¹⁶ Yet, there are few studies and little credible evidence identifying the key sources of risk and best practices for improved performance. An association between the occurrence of intraoperative handoffs and increased adverse events and morbidity and mortality has been demonstrated^{^17–21} as has improvement in intraoperative information transfer with standardization.^{^22–24} However, evidence that intraoperative handoff interventions have an impact on outcomes is lacking.

Given the paucity of evidence regarding handoffs in perioperative care, the Anesthesia Patient Safety Foundation convened a group of experts to establish consensus on several key aspects regarding perioperative handoffs and their safety implications, with the intent to suggest best practices and areas for further investigation. The development of consensus guidelines by experts within a field can be helpful when there is a lack of high-quality evidence.^²⁵ Our objective was to identify areas of consensus on 6 areas related to perioperative handoffs: (1) process elements and behaviors; (2) metrics and measurement; (3) important research questions; (4) best ways to manage training and education; (5) how best to implement; and (6) how to incorporate patient and family goals into handoffs. The meeting used a robust process used to achieve consensus. We report briefly on that process and focus on the results, which are intended to guide implementation and improvement of perioperative handoff processes and research.

METHODS

This consensus project consisted of a prospective, 4-round, modified Delphi process^²⁵ culminating in an in-person consensus meeting held on the first day of the 2017 Anesthesia Patient Safety Foundation Stoelting Conference.^²⁶ A more detailed description of the methods is available in Supplemental Digital Content 1, Appendix 1, https://links.lww.com/AA/C766.

Human Subjects

The investigation was reviewed and approved by the institutional review board of the University of Texas Southwestern Medical Center. The requirement for written informed consent was waived by the institutional review board; consent was assumed by participation in the preconference surveys and attendance at the conference.

Inclusion Criteria

Target participants included leaders and experts in perioperative quality and safety, as well as handoff experts from all fields who have published, received grant funding, and presented their findings to the scientific community. This included anesthesiologists, certified registered nurse anesthetists, anesthesia assistants, nurses, surgeons, educators, researchers in related fields (eg, human factors engineering, implementation science), and selected industry representatives with previous Anesthesia Patient Safety Foundation involvement. Remaining space was advertised through the “APSF Newsletter” and website, in the American Society of Anesthesiologists newsletter “ASA Monitor,” and on social media. Registration was complimentary, but travel and lodging expenses were not.

Exclusion Criteria

No interested people were refused registration or participation in the Delphi surveys, including conference planning committee members.

Consensus Rationale and Overall Process

We used a multistep Delphi process to develop statements to discuss in person at the conference to engage participants early and maximize the number of statements reaching the threshold for consensus. Based on Delphi principles,^²⁵ the consensus process unfolded in several iterative steps: round 1, open-ended question survey; round 2, closed-ended question survey; round 3, in-person small group discussions; and round 4, large group voting.

Delphi Round 1: Open-Ended Questioning

Based on the committee’s clinical and research experience and drawing from the published literature on handoffs, we created and distributed a 7-question open-ended survey (Supplemental Digital Content 2, Appendix 2, https://links.lww.com/AA/C767) about the conduct of handoffs, implementation strategies, and research priorities to the 107 registrants for the Anesthesia Patient Safety Foundation conference as of June 30, 2017. Sixty-two participants responded (57.9%). Responses were analyzed using a qualitative thematic analytic approach^²⁷ that allowed similar responses to be grouped into categories reflecting overarching themes.

Delphi Round 2: Closed-Ended Questioning

The second round of the Delphi process focused on reviewing and refining first-round content.^²⁷ To achieve this goal, the overarching themes from the first round were converted to statements. For example, the theme “process metrics: compliance/adherence” became “It is essential to measure process metrics such as handoff process compliance or adherence.”

Figure.:
Delphi process. Number of statements at each step and number of participants. ^aRound 1 of the Delphi process was conducted via electronic survey (Supplemental Digital Content 1, Appendix 1, https://links.lww.com/AA/C766) from June 1 to June 25, 2017. The number of distinct phrases were parsed from the open-ended responses (number of respondents to Delphi round 1). ^bAfter combining and compressing, themes were derived from the individual phrases identified in the previous step. These themes were further condensed into statements for presentation in the survey for round 2 of the Delphi process. ^cThe number of statements presented for consideration in round 2 of the Delphi. Round 2 of the Delphi process was conducted via electronic survey (Supplemental Digital Content 2, Appendix 2, https://links.lww.com/AA/C767) from August 24 to September 3, 2017 (number of participants for each group for Round 2). ^dThe results of voting from round 2 were used to prime in-person small-group discussion during round 3 (September 6, 2017). Small groups began with the same statements as presented for round 2. Statements reaching ≥75% consensus during round 3 were presented to the conference at large for voting by all participants during round 4 (number of participants in each small group). ^eThe final number of statements reaching ≥75% consensus during large-group voting (round 4) at the in-person conference (September 6, 2017).

Conference registrants (115 registered as of August 1, 2017) were randomly assigned to 1 of 6 groups corresponding to different aspects of handoffs with a second round of surveys: (1) handoff process elements; (2) handoff metrics; (3) handoff research topics; (4) handoff education and training; (5) handoff process implementation; and (6) patient and family involvement in handoffs (Supplemental Digital Content 3, Appendix 3, https://links.lww.com/AA/C768). These assigned groups would later be used to determine “breakout session” groups for round 3 of the Delphi process. In total, 87 participants responded (75.7%) (Figure).

In-Person Conference.

The part of the Anesthesia Patient Safety Foundation conference dedicated to handoffs was held on September 6, 2017 (Supplemental Digital Content 4, Appendix 4, https://links.lww.com/AA/C769). Ten short lectures were presented about topics relating to handoffs, including evidence supporting standardization, use of simulation in handoff training, and implementation science, and an overview of Delphi round 2 results.^²⁶

Delphi Round 3: Small Group Discussions

After the lectures, the 6 preassigned breakout groups met separately to discuss their group’s statements, led by trained facilitators and scribes, with the goal of achieving consensus on as many statements as could be addressed in the allotted time. After each breakout group’s session, there was a vote on the final edited list of statements. The number of participants in each group is presented in the Figure. Only those statements reaching the 75% threshold for consensus during this round were advanced to round 4 for voting.

Delphi Round 4: Large Group Voting

All conference participants reconvened to review breakout group statements meeting the threshold. Using an audience response voting system, thresholds for “consensus” (75%) and “strong consensus” (90%) were voted on first with 94% of participants in agreement with these levels. The meeting facilitator read each statement aloud and participants voted on each statement separately. A short discussion period was allowed during each group’s allotted presentation time. During this discussion, participants could propose edits to the statements to achieve consensus or could offer commentary about why they disagreed with a given statement.

RESULTS

Table 1.:
Demographics of Conference Participants

Table 2.:
Topics of Discussion for Each Small Group

Table 3.:
Final Consensus Statements From Delphi Round 4 for Process Elements and Behaviors and Measurement and Metrics

Table 4.:
Final Consensus Statements From Delphi Round 4 for Most Important Research Questions and Education and Training

Table 5.:
Final Consensus Statements From Delphi Round 4 for Implementation and Patients and Families

The demographics of all participants involved in voting (n = 99) are presented in Table 1. The flow of the Delphi process is illustrated in the Figure, which shows the number of statements or research questions that resulted from each stage of the Delphi process, the numbers of participants that contributed to each step and includes the final number of statements from each theme that arose from the discussions in each breakout group. Ninety-six attendees participated in the breakout groups. The 7 questions that formed the focus for the themes of the 6 breakout groups are listed in Table 2. Fifty-nine statements and questions related to the conduct, measurement, research, training, implementation, and patient involvement in handoffs reached the voted-upon threshold of 75% consensus, with 21 statements reaching 90% consensus or greater (Tables 3–5). A summary of the key aspects of discussion for each breakout group during round 3 is available in Supplemental Digital Content 5, Appendix 5, https://links.lww.com/AA/C770.

DISCUSSION DURING DELPHI ROUND 4

In the round 4 large group discussion, many issues were raised about statements/questions for which there was not at least 75% consensus. In some cases, these were modified and subsequently achieved at least 75% by the large group. At the end of the conference, several comments were made by participants with suggestions about spreading the results of the conference and for suggested future work, for example, that Anesthesia Patient Safety Foundation create a training video for perioperative handoffs like others it has produced for fire safety, medication safety, and postoperative visual loss.

The final statements that achieved ≥75% consensus in the large group voting are listed in Tables 3–5.

DISCUSSION

With a lack of clear evidence for best practices related to handoffs, and only limited evidence for the association of handoffs with outcomes, the opinions of this multidisciplinary group of perioperative professionals can provide guidance to those working to improve handoffs in their own institutions and to those interested in generating evidence to support handoff improvement. Consensus was achieved about a great number of statements and questions related to the conduct, measurement, research, training, implementation, and patient involvement in handoffs.

Many process elements and behaviors were deemed necessary for high-quality handoffs, too many to be adopted universally. Despite the face validity of these statements, more rigorous inquiry is needed to determine priorities. In the meantime, expert consensus may aid the selection of standardized handoff processes.

Six clear, broad themes for research were identified to guide implementation of evidence-based and consensus-based guidelines. These high-level research questions generally fall into 2 categories: (1) best practices for performing and implementing handoffs; and (2) how handoffs are related to outcomes, including by implication, the mechanisms by which handoffs impact outcomes. These questions may help focus the efforts of those interested in handoff research and provide guidance to funding bodies and organizations considering supporting handoff research. The full list of research questions from round 1 (Supplemental Digital Content 6, Appendix 6, https://links.lww.com/AA/C771) may stimulate further ideas about what questions remain unanswered.

While education and training alone are insufficient, they are essential elements of effecting behavioral change. The IPASS group has shown that patient handoffs can be improved and adverse events reduced; in their case, success required a comprehensive, multidimensional approach.^{^28–30} The statements supported by the consensus process highlight the importance of leadership buy-in, inclusion of all members of the team, and suggest that experiential and team training, with coaching and feedback, and didactic education are essential for perioperative handoff improvement.

Implementation of new handoff processes necessitates changes in tasks, attitudes, and behaviors. To be successful, the value of changing behaviors or the process must be demonstrated.^³¹ Early engagement of key stakeholders should guide development of a project proposal or charter. Project leads with training in patient safety and quality improvement methodology should oversee the redesign process and recruit champions within each of the involved disciplines. Subject matter experts in team training, education, project management, human factors, and implementation science should ideally be included in the team, and timely feedback to and from all involved clinicians is important.

Patient and family engagement are not typically considered in perioperative handoffs; little consensus was achieved on how to accomplish this goal. Despite obvious challenges to such practices, we deliberately chose to include this topic given the importance of patient-centered care and the influence handoffs may have on satisfaction with care. The large group considered only 4 statements that reflect how patient and family involvement might be introduced (eg, via a patient hotline or quality improvement reporting mechanism). While challenging, we believe it worthwhile to consider how patients and families might be included in perioperative handoffs.

This study included participants from varying practice types across the United States. The prework associated with rounds 1 and 2 of the Delphi process engaged participants months before the in-person conference, priming attendees for important discussion, and allowing consensus to be achieved for a greater breadth of topics than would otherwise be possible in a 1-day conference. Overall, the consensus building process appeared to work well. With strong response rates to each round of the Delphi, and participation of nearly 100 perioperative quality and safety leaders and advocates, we believe the strength of our recommendations is robust.

However, given the complexity and magnitude of perioperative handoffs, it may have been too ambitious to try to address every topic within a 1-day conference. For example, we did not specifically discuss strategies to reduce the number of handoffs during perioperative care, such as optimizing long-term staffing and short-term staff scheduling, a technique that might well result in improved outcomes.^¹⁷ Nonetheless, based on the evaluations, conference participants felt that much was accomplished in a short amount of time. Because the Anesthesia Patient Safety Foundation conference was open attendance, participants were not necessarily representative of all anesthesia providers across the country; the majority were from academic backgrounds, and there were relatively few certified registered nurse anesthetists, surgeons, and nurses.

There are other possible biases (eg, unequal distribution of expertise in the small groups for round 3). Also, despite using anonymous electronic polling, it is possible that the open group voting in the fourth round of the Delphi could encouraged “group think” or peer pressure, limiting individuals’ ability to openly express disagreement. Finally, we did not have participation from individuals specifically representing the patient and family perspective.

While further research is certainly needed, we believe that these findings suggest activities that health care organizations and individuals can do immediately to improve their handoff practices and reduce harm. We urge that all organizations not already doing so review these findings and consider what the actions might be best in their settings and with their resources.

ACKNOWLEDGMENTS

The authors acknowledge David Gaba, MD, Stacey Maxwell, and Mark Warner, MD, for their support and assistance. Special thanks to Jay Vogt (Peoplesworth, Concord, MA) for essential contributions in designing the consensus conference process. The authors would also like to recognize the Multicenter Handover Collaborative for its efforts in continuing this work.

DISCLOSURES

Name: Aalok V. Agarwala, MD, MBA.

Contribution: This author helped conceive, design, and analyze the study, and draft and revise the manuscript.

Name: Meghan B. Lane-Fall, MD, MSHP.

Contribution: This author helped conceive, design, and analyze the study, and draft and revise the manuscript.

Name: Philip E. Greilich, MD, MSc, FASE.

Contribution: This author helped conceive, design, and analyze the study, and draft and revise the manuscript.

Name: Amanda R. Burden, MD.

Contribution: This author helped conceive, design, and analyze the study, and draft and revise the manuscript.

Name: Aditee P. Ambardekar, MD, MSEd.