The Impact of a 22-Month Multistep Implementation Program on Speaking-Up Behavior in an Academic Anesthesia Department.

BACKGROUND: Speaking-up is a method of assertive communication that increases patient safety but often encounters barriers. Numerous studies describe programs introducing speaking-up with varying success; the common denominator seems to be the need for a multimodal and sustained approach to achieve the required change in behavior and culture for safer health care.
METHODS: Before implementing a 22-month multistep program for establishing and strengthening speaking-up at our institution, we assessed perceived safety culture using the "Safety Attitudes Questionnaire." After program completion, participants completed parts of the same Safety Attitudes Questionnaire relevant to speaking-up, and preresult and postresult were compared. In addition, levels of speaking-up and assertive communication were compared with a Swiss benchmark using results from the "Speaking-up About Patient Safety Questionnaire."
RESULTS: Safety Attitudes Questionnaire scores were significantly higher after program completion in 2 of 3 answered questions (median [first quartile, third quartile), 5.0 [4.0, 5.0] versus 4.0 [4.0, 5.0], P = 0.0002, and 5.0 [4.0, 5.0] versus 4.0 [4.0, 4.0] P = 0.002; n = 34). Our composite score on the Speaking-up About Patient Safety Questionnaire was significantly higher (mean ± SD, 5.9 ± 0.7 versus 5.2 ± 1.0; P < 0.001) than the benchmark (n = 65).
CONCLUSIONS: A long-term multimodal program for speaking-up was successfully implemented. Attitude and climate toward safety generally improved, and postprogram perceived levels of assertive communication and speaking-up were higher than the benchmark. These results support current opinion that multimodal programs and continued effort are required, but that speaking-up can indeed be strengthened.

Speaking-up is a method of assertive communication by which concerns, such as threats to patient safety or the presence of unsafe conditions, are stated with persistence until there is a clear resolution.[1-3] According to the Joint Commission’s sentinel event data from 2015, the failure to speak up was 1 of the top 3 root causes for adverse events in the perioperative period.[4] Withholding voice despite safety concerns is a common behavior among health care professionals. A Swiss multicenter study reported that 19% to 39% of health care workers had chosen to withhold voice within the past 4 weeks.[5] Several barriers for speaking-up have been identified in the perioperative setting, including perceived ineffectiveness, presence of patients, and authority gradients.[1,6]

Research on the implementation of speaking-up has mainly focused on single groups, including nursing students,[7] medical students,[8] and residents.[9,10] In general, implementation of speaking-up has demonstrated varying success,[11,12] but common themes include the following: necessity for an implementation program involving all members of staff, education to support a transformation in organizational culture,[13] and addressing norms and communication behaviors.[14] In short, strengthening a culture of speaking-up is not only an ongoing challenge[15] but also crucial to increasing patient safety.

To establish and strengthen speaking-up in our department, we developed and used a 22-month multistep implementation program. To measure the effect of the program, we compared perceptions of speaking-up before and after the intervention using elements from the “Safety Attitudes Questionnaire,” a validated questionnaire for perceptions of patient safety–related attitudes, as our primary outcome. As a further measurement and a secondary outcome, we compared postintervention levels of speaking-up and assertive communication with comparable Swiss institutions using the “Speaking-Up About Patient Safety Questionnaire.”

METHODS

Study Institution and Population

The study was performed in the Cantonal Hospital of Baden, a 382-bed teaching hospital of Zurich University, which annually treats more than 20,000 inpatients and more than 170,000 outpatients. All staff members of the department of anesthesia, that is, nurse and physician anesthetists (both residents and consultants) employed at any time during the 22 months, were exposed to the implementation program. The requirement for approval of our study, as well as for written consent, was waived by the ethical committee “Nordwestschweiz” and by our institutional legal board. Participants gave verbal consent. Materials were deidentified before any analysis and destroyed hereafter in conformance with legal requirements.

A total of 117 staff members participated in the implementation program at some time during the 22 months, but because of staff fluctuations, availability, and study requirements, the number of available participants varied over time. Details are presented on the timeline of the project in Figure 1.

FIGURE 1

Implementation program: of 177 members of staff present at some time during the intervention, 57 participated in the baseline survey, of which 34 completed the repeat survey, providing data for the primary objective. Independent of participation in the baseline survey, 65 members of staff completed the program and were available for the Speaking Up About Patient Safety survey, the secondary outcome.

Baseline Survey

Before implementing the program, the 57 current members of staff available completed the German language version of the Safety Attitudes Questionnaire. This questionnaire is a validated tool to assess[16,17] health care workers’ perceptions of patient safety–related attitudes in various clinical areas. Depending on the version, it comprises 30 to 60 items measured on a 5-point Likert scale covering 6 aspects of the safety climate: teamwork climate, job satisfaction, safety climate including perception of speaking-up, stress recognition, working condition, and perception of management. The German translation was recently validated[18] and successfully tested in 10 Swiss hospitals[19] and transcribed to the Survey Monkey online platform (Survey Monkey software: Momentive, AI; San Mateo, CA) for our survey of baseline values.

The Implementation Program

After the baseline survey, the multimodal implementation program was initiated in August 2019 and incorporated into the entire anesthesia department over a course of 22 months. It consisted of various elements including an awareness campaign, an online course, simulation based team trainings, and explicit invitation to speak-up incorporated into daily practice.

To begin the program, all current staff members were required to participate in the online course developed using the hospital’s native e-learning software, easylearn schweiz ag (Easylearn Schweiz AG, Hünenberg, Zug, Switzerland), comprising 3 components. First, background knowledge and the rationale for speaking-up were presented together with instructions including the 2-challenge rule[20] and providing coaching in advocacy inquiry with specific examples. The second element was a video featuring the department head as the recipient of speaking-up. Finally, there was a multiple choice exam testing participant’s knowledge on rationale and barriers for speaking-up, the effect of the authority gradient, and identification of the correct wording of speaking-up using crisp advocacy inquiry in various described situations. This exam was graded, and a pass was required. One year later, members of staff were again exposed to the same mandatory online course module as a refresher.

Complementing the teaching, we performed 3 high-fidelity in situ simulations with variations of opportunity for speaking-up throughout the implementation program, to which we assigned as many staff members as rostering allowed during the pandemic:

interdisciplinary team training for obstetric anesthesia staff with scripted opportunity for speaking-up during the scenarios (40 participants from our department) in December 2019

anesthesia induction sequence with scripted speaking-up situations with an acting instructor (75 participants) in October 2020

interdisciplinary team-training sessions for same-day surgery teams and obstetric anesthesia teams, with special focus on speaking-up in debriefings (29 participants from our department) in April 2021

Scenarios and teaching elements were developed and tested before study use by the author C.S., a trained instructor for medical simulation with experience developing standardized scenarios for measurement and research,[21] then refined by the authors C.S., F.W., and M.H. using a modified Delphi approach, and finally tested by fellow simulation instructors.

In addition, the program was accompanied by a continuous awareness campaign including various lectures and workshops reiterating the topics of the online course (background knowledge and the rationale for speaking-up, instructions and suggestions for providing speaking-up, and coaching in advocacy inquiry with specific examples), and an interview with the head of the department in the hospital newspaper, in which he discussed hierarchy and status issues, introduced the concept of, and called for, speaking-up.

Finally, as of January 2020, we incorporated speaking-up into our daily clinical practice by augmenting the preinduction checklist and team briefing with the request to perform speaking-up made by the highest-ranked team member. This action served a dual purpose—as an ongoing reminder of leadership commitment to speaking-up and a tool to reduce the barriers of hierarchy by the mechanism of leader inclusiveness—words and deeds by leaders that invite and appreciate others’ contributions which can take nature off its course, helping to overcome status’ inhibiting effects on psychological safety.[22]

Primary Outcome: Pre-Post Comparison Using the Safety Attitudes Questionnaire

For our primary outcome, we interviewed all current members of staff who completed the whole implementation program and had participated in the baseline survey (n = 34) using the following 3 questions from the Safety Attitudes Questionnaire used for the baseline survey, which specifically focus on assertive communication and speaking-up, after the implementation period of 22 months and compared scores:

In this clinical area, it is difficult to speak up if I perceive a problem with patient care.

In this clinical area, it is difficult to discuss errors.

I am encouraged by my colleagues to report any patient safety concerns I may have.

Both cohorts contained the same participants, and results were compared unpaired.

Secondary Outcome: Comparison of Results From Our Institution With the Benchmark of Comparable Swiss Institutions Using the Speaking-up About Patient Safety Questionnaire

Sixty-five members of staff participating in the implementation program from the beginning and available at the time of the survey completed the Speaking-Up about Patient Safety Questionnaire, a validated questionnaire developed by the Swiss Patient Safety Foundation focusing on speaking-up and assertive behavior among health care staff.[23] Specifically, the questionnaire assesses the 2 theoretical constructs of speaking-up and withholding voice, while covering 3 speaking-up climate-related subscales: psychological safety for speaking-up, encouraging environment, and resignation. The questionnaire has been used in 22 Swiss hospitals and in 5 comparable departments, which allows valuable cross-hospital comparisons of speaking-up behaviors and climate.

Statistical Analysis

Results for the primary and the secondary outcome were examined by inspection of the histograms. Negatively worded items were reversed before statistics were performed. Two-sided P values <0.05 were considered statistically significant. All statistical analyses were conducted using R version 4.0.2.[24]

To compare the preimplementation and postimplementation results of the 3 relevant questions on the Safety Attitudes Questionnaire (primary outcome), a Mann-Whitney U test for nonpaired samples was performed. Because of the small sample size and lack of normal distribution, we present the median, and first and third quartiles.

Concerning the secondary outcome, we compared the results of the Speaking-up About Patient Safety Questionnaire to the benchmark values using Welch t test for unequal variances; here, we report the mean and SD according to previous analyses.[23]

RESULTS

Primary Outcome

Of the 57 members of staff initially completing the preimplementation Safety Attitudes Questionnaire, 34 (59.6%) completed the whole implementation program and were also available for the postimplementation survey with the 3 relevant questions from the questionnaire.

Scores after implementation were significantly higher in 2 of 3 questions surveyed and did not change significantly in the third question (Table 1).

TABLE 1

Comparison of Median (First Quartile, Third Quartile) Responses to Safety Attitude Questionnaire Items Before and After Implementation

Safety Attitudes Questionnaire (Measures on a 6-Point Scale; n = 34)	Median (1st Quartile, 3rd Quartile)		P*
	Preimplementation	Postimplementation
In this clinical area, it is difficult to speak up if I perceive a problem with patient care.†	4.0 (4.0, 4.75)	5.0 (4.0, 5.0)	0.0002
In this clinical area, it is difficult to discuss errors.†	4.0 (4.0, 4.0)	5.0 (4.0, 5.0)	0.0022
I am encouraged by my colleagues to report any patient safety concerns I may have.	4.0 (3.0, 4.0)	4.0 (3.0, 5.0)	0.7220

*P values: Mann-Whitney U test for nonpaired samples.

†Negatively worded items are reverse coded for the total score.

Secondary Outcome

A total of 65 members of staff who had completed the implementation program also completed the Speaking-Up About Patient Safety Questionnaire. Safety concerns were common among survey participants. The majority reported at least 1 patient safety concern during the past 4 weeks (92%). At least 1 episode of speaking-up during the past 4 weeks was reported by 94%. At least 1 episode of “withholding voice” was reported by 58%. The barriers reported by respondents as hindering them to voice their concerns were reaction of the actor not predictable (35%), presence of patients or relatives (34%), ineffectiveness of speaking-up (31%), unclear risk for the patient (29%), difficulty finding the right tone (12%), and fear of negative reactions (8%).

Overall responses to the climate survey items are reported in Table 2. Results obtained in this study were higher when compared with the Swiss perioperative care sample.[6] Respondents in our hospital reported higher levels of psychological safety and a more positive encouraging environment, and described less resignation toward speaking-up.

TABLE 2

Comparison of Mean (SD) Responses to Climate Survey Items for Our Department and the Swiss Comparison

Items and Scales (Measure on a 7-Point Likert Scale)	Mean (SD)		P*
	This Sample (n = 65)	Swiss Perioperative Care Sample (n = 360)
Psychological safety for speaking up, mean scale score	6.2 (0.6)	5.5 (1.1)	<0.001
I can rely on my colleagues (doctors and/or nurses), whenever I encounter difficulties in my work.	6.4 (0.6)	5.6 (1.4)	<0.001
I can rely on the shift supervisor (person in charge of a shift) whenever I encounter difficulties in my work.	6.4 (0.9)	5.6 (1.6)	<0.001
The culture in my unit/clinical area makes it easy to speak up about patient safety concerns.	6.2 (0.9)	5.4 (1.6)	<0.001
My colleagues (doctors and/or nurses) react appropriately, when I speak up about my concerns about patient safety.	5.9 (0.9)	5.4 (1.2)	<0.001
My shift supervisors (person in charge of a shift) react appropriately, when I speak up about my patient safety concerns.	5.9 (1.0)	5.5 (1.4)	0.009
Encouraging environment for speaking up, mean scale score	5.9 (0.9)	4.9 (1.4)	<0.001
In my unit/clinical area, I observe others speaking up about their patient safety concerns.	5.6 (1.2)	5.2 (1.5)	0.028
I am encouraged by my colleagues (doctors and/or nurses) to speak up about patient safety concerns.	6.0 (1.1)	4.6 (1.7)	<0.001
I am encouraged by my shift supervisor (person in charge during a shift) to speak up about patient safety concerns.	6.1 (1.1)	4.9 (1.8)	<0.001
Resignation toward speaking up, mean scale score	2.5 (1.1)	3.2 (1.4)	<0.001
When I have patient safety concerns it is difficult to bring them up.†	2.0 (1.1)	2.4 (1.6)	0.002
Having to remind staff of the same safety rules again and again is frustrating.†	3.1 (1.7)	3.9 (2.1)	<0.001
Sometimes I become discouraged because nothing changes after expressing my patient safety concerns.†	2.5 (1.5)	3.1 (1.9)	0.004
Total speak up climate score, mean across items	5.9 (0.7)	5.2 (1.0)	<0.001

*P values: Welch t test for unequal variances.

†Negatively worded items are reverse coded for the total score.

DISCUSSION

Results

We found that the 22-month implementation program was associated with higher levels of self-reported speaking-up behavior, as evidenced by a significant improvement in 2 of 3 elements on the postimplementation Safety Attitudes Questionnaire items addressing assertive communication and speaking-up, and higher overall scores in the climate survey as compared with the benchmark of similar health care institutions in Switzerland.

Although our study did not investigate the effects of each separate intervention within the program, evidence does suggest that leader inclusiveness and leadership support are critical—as such, we feel that our head of department providing interviews, lectures, and a scripted video inviting to speaking-up was essential for the program’s success and patient safety climate in our department.

Although there was an improvement in 2 of 3 responses on the Safety Attitudes Questionnaire, the survey question “I am encouraged by my colleagues to report any patient safety concerns I may have” did not show any improvement after implementation. We believe that this might be because of the relatively high baseline value (4.0 on a 5-point scale) and the fact that our implementation program did not explicitly focus on peer support as much as the more prominent issues of hierarchy, leadership, and empowerment. Also, the request to perform speaking-up expressed by the highest ranked team member at every induction might have made encouragement by other team members seem less important. However, this evidence seems to show that strengthening of peer support to do the right thing might indeed need more focus in consecutive programs.

Although the higher overall scores in the Speaking-up About Patient Safety Questionnaire as compared with the benchmark of similar health care institutions in Switzerland suggest a positive effect of our implementation program, some results are sobering, albeit not unexpected. Although most respondents reported at least 1 patient safety concern during the past 4 weeks, more than half reported withholding voice within the same period—this is a stark reminder of the fact that even an intervention of our dimension is only one step on the road to patient safety. Reported barriers (unpredictable reaction of recipient of speaking-up, presence of patients or relatives, assumed or experienced ineffectiveness of speaking-up, an unclear risk for the patient, difficulty finding the right tone, and fear of negative reactions) persist and provide a road map for further interventions. Because we only implemented our program in the department of anesthesia, we must consider 1 barrier, the assumed or experienced ineffectiveness, in the context of interdisciplinary communication in particular: if the culture of patient safety and leadership support for speaking-up is less well established in a department closely interconnected such as surgery, there is a limit to the benefit for patient safety, which can be achieved by improvements in one department only.

Strengths of Our Study

To our knowledge, our study is one of the first to detail a longitudinal and multifaceted implementation program involving all levels of staff and leadership, addressing speaking-up and voice behavior, and providing objective measures of its success. A further advantage is our comparison of scores to a national benchmark.

Limitations of Our Study

Our study is limited by its small size and relatively small response rate. Because of the requirement that study participants completed the whole implementation program and staff fluctuation over the 22 months, overall numbers were smaller than expected. In addition, the prominence of leadership support in “safe behavior” makes a Hawthorne effect highly likely.

Furthermore, at the time of the study, we did not have a structured reporting instrument for near misses and adverse events in place apart from the critical incident reporting system, which, because of legal restrictions in Switzerland, cannot be considered a representative database. Improvements in reporting are a logical next step for the implementation program.

Another possible limitation is that this study was a single-center study in 1 department and cultural region; it is unclear in how far results are reproducible in another department, institution, or even country with different norms and cultures. Indeed, a department of anesthesia with a traditionally shallow hierarchy in Switzerland (being a country with low power distance index but relatively high scores on indices for individualism, masculinity, and uncertainty avoidance according to Hofstedes cultural dimensions) probably requires emphasis on different elements of a multimodal approach, as would a different department or population in another cultural setting. Because of this limitation, we feel that a rigorous investigation into perceived barriers before implementing such a program—as we performed using the Safety Attitudes Questionnaire—can provide valuable guidance to address these differences.

CONCLUSIONS

A long-term, inclusive and multistep program for establishing speaking-up was successfully implemented at our institution. Attitude and climate toward safety in our department improved after implementation according to “Safety Attitudes Questionnaire” scores; the Speaking-Up About Patient Safety Questionnaire respondents at our institution reported higher levels of psychological safety and a more positive encouraging environment, and described less resignation toward speaking-up, as in comparable Swiss institutions. These results seem to support a current opinion that, although a multimodal program and continued effort are required to assist the change in culture and behavior toward safer health care, increases in levels of speaking-up can indeed be achieved.