Background Open communication between healthcare professionals about care concerns, also known as ‘speaking up’, is essential to patient safety.
Objective Compare interns' and residents' experiences, attitudes and factors associated with speaking up about traditional versus professionalism-related safety threats.
Design Anonymous, cross-sectional survey.
Setting Six US academic medical centres, 2013–2014.
Participants 1800 medical and surgical interns and residents (47% responded).
Measurements Attitudes about, barriers and facilitators for, and self-reported experience with speaking up. Likelihood of speaking up and the potential for patient harm in two vignettes. Safety Attitude Questionnaire (SAQ) teamwork and safety scales; and Speaking Up Climate for Patient Safety (SUC-Safe) and Speaking Up Climate for Professionalism (SUC-Prof) scales.
Results Respondents more commonly observed unprofessional behaviour (75%, 628/837) than traditional safety threats (49%, 410/837); p<0.001, but reported speaking up about unprofessional behaviour less commonly (46%, 287/628 vs 71%, 291/410; p<0.001). Respondents more commonly reported fear of conflict as a barrier to speaking up about unprofessional behaviour compared with traditional safety threats (58%, 482/837 vs 42%, 348/837; p<0.001). Respondents were also less likely to speak up to an attending physician in the professionalism vignette than the traditional safety vignette, even when they perceived high potential patient harm (20%, 49/251 vs 71%, 179/251; p<0.001). Positive perceptions of SAQ teamwork climate and SUC-Safe were independently associated with speaking up in the traditional safety vignette (OR 1.90, 99% CI 1.36 to 2.66 and 1.46, 1.02 to 2.09, respectively), while only a positive perception of SUC-Prof was associated with speaking up in the professionalism vignette (1.76, 1.23 to 2.50).
Conclusions Interns and residents commonly observed unprofessional behaviour yet were less likely to speak up about it compared with traditional safety threats even when they perceived high potential patient harm. Measuring SUC-Safe, and particularly SUC-Prof, may fill an existing gap in safety culture assessment.
- Safety culture
- Graduate medical education
Statistics from Altmetric.com
Open communication regarding concerns, also known as ‘speaking up’, is vital to keeping patients safe and preventing errors.1–4 Healthcare professionals commonly encounter situations that call for speaking up about traditional patient safety threats such as inadequate hand hygiene, breach of a sterile field or insufficient hand off. They may also encounter safety threats perceived to be professionalism-related including falsified documentation, failure to disclose an error or disrespectful behaviour. Despite potential implications for patient safety, some clinicians may find speaking up in these situations challenging.3 ,5–7
For interns and residents, low on the medical hierarchy, substantial barriers exist to speaking up when they observe threats to patient safety. Interns and residents often lack adequate role models since faculty may be similarly ill-equipped to communicate effectively about these issues.3 ,7 In addition, interns and residents may perceive that assertive communication and transparency conflicts with securing a positive evaluation. A strong desire to gain the acceptance of team members can trump the moral courage needed to ‘speak up the hierarchy’ about concerns related to patient safety. While formal patient safety curricula are increasingly common, the hidden curriculum and behavioural norms may be more powerful influences on learners' attitudes and behaviour.8 ,9
In addition to hierarchy, research indicates that speaking up behaviour is influenced by subjective beliefs about when, how and to whom it is acceptable to speak up and is highly context and person specific.10 ,11 Several factors may influence speaking up about patient safety threats including the potential for harm to the patient, collegial relationships or relationship concerns, past experiences, perceived efficacy of speaking up and characteristics of the work environment, including psychological safety, leadership, culture and workload.2 ,12–18 However, whether and how these factors differ by nature of the safety threat (ie, professionalism-related compared with traditional safety threats) is not well understood.
There are several reasons to anticipate differences in speaking up by nature of the safety threat. In contrast to traditional safety threats, professionalism-related safety threats may be perceived as deliberate or disrespectful, and are more likely to be attributed to character traits (eg, arrogance).19 ,20 Thus, speaking up about professionalism-related safety threats may be viewed as less likely to result in change, more confrontational and less acceptable than speaking up about traditional patient safety threats, with greater risk of jeopardising professional relationships.15 ,21–23 While speaking up about traditional patient safety threats has been promoted and incorporated into patient safety curricula,24 ,25 speaking up about unprofessional behaviour remains relatively underemphasised.26 As a result, clinicians may feel less encouraged to speaking up about professionalism-related safety threats and some clinicians may have trouble recognising the associated patient safety implications.
While a growing literature describes factors influencing speaking up about patient safety threats,13–17 ,27 far less is known about how often residents encounter such threats and whether they speak up about them.28 Moreover, research to date has largely considered safety threats uniformly, without discriminating between traditional and professionalism-related safety threats, even though important differences may exist between the two. A better understanding of these differences could shape organisational priorities, mobilise evidence-based efforts and inform strategic interventions. Thus, our study aimed to (a) determine the self-reported prevalence of traditional and professionalism-related safety threats in the clinical learning environment and how often interns and residents speak up about them, (b) compare attitudes towards speaking up about traditional patient safety threats versus professionalism-related patient safety threats including perceived barriers and facilitators and (c) identify factors associated with speaking up about traditional and professionalism-related patient safety threats using two hypothetical safety vignettes.
We surveyed all interns (postgraduate year 1) and residents (postgraduate years 2 and above) in internal medicine, general surgery, plastic surgery, neurosurgery, orthopaedic surgery, urology, and obstetrics and gynaecology from six large US academic medical centres from different geographic regions. The survey methods were previously described in a related article.29
Interns and residents were anonymously surveyed via email using REDCap V.5.0.8 (Vanderbilt University, Nashville, Tennessee, USA). Up to three reminder emails were sent to non-responders every two weeks. Two of the medical centres were surveyed between March and June of 2013 and four additional medical centres were surveyed between March and June of 2014. Participation was voluntary and consent was implied by survey completion. All eligible participants received an upfront $5 gift card as an incentive to participate
Survey items (see online supplementary appendix) included (a) respondents' self-reported exposure to and speaking up about traditional patient safety breaches and unprofessional behaviour; (b) barriers, facilitators and attitudes towards speaking up about patient safety breaches and unprofessional behaviour; (c) new and validated measures of patient safety culture; (d) respondent characteristics and (e) two patient safety vignettes: traditional safety threat versus professionalism-related safety threat (table 1). The survey was developed by physicians and researchers (WM, LSL, EJT, JME, SKB) with expertise in patient safety, professionalism, ethics and psychometrics.
To assess speaking up behaviour, respondents were asked whether they had observed a patient safety breach or unprofessional behaviour at least once during their most recent inpatient month. Those who reported observing an event were then asked whether they discussed any of the unprofessional behaviours or patient safety breaches they observed with the person(s) involved. ‘Unprofessional behaviour’ was defined as conduct of a health professional that demonstrates disrespect or lack of compassion, commitment to ethical principles, integrity or accountability towards patients or coworkers. ‘Patient safety breach’ was defined as an act or omission that unnecessarily increases the risk of accidental or preventable injuries produced by medical care. Definitions included examples of patient safety breaches (eg, improper sterile technique, inadequate hand hygiene, poor handoff) and unprofessional behaviour (eg, covering up an error, falsifying documents, insulting others) derived from the literature.30–32
Barriers and facilitators to speaking up were drawn from a review of the literature.2 ,3 ,12 ,15 ,26 ,33 Respondents were asked to select the most significant barriers and facilitators to speaking up about patient safety breaches and unprofessional behaviour.
Validated measures of patient safety culture were used to assess respondents' perceptions of particular aspects of the patient safety climate within the clinical learning environment and included two standard measures: the safety and teamwork climate domains of the Safety Attitude Questionnaire (SAQ)—Short Form. This is consistent with other studies using SAQ to measure individual attitudes about patient safety climates.34 ,35 Because ‘speaking up’ is under-represented in current patient safety culture metrics,36 ,37 several of the authors (WM, LSL, EJT, JME, SKB) developed two new measures focusing on the extent to which respondents perceive that speaking up about patient safety and unprofessional behaviour is supported in the clinical learning environment: Speaking Up Climate for Patient Safety (SUC-Safe) and Speaking Up Climate for Professionalism (SUC-Prof). The development and psychometrics of these two new measures within this sample are described in a related article.29 Items from the two scales were used to assess interns' and residents' attitudes regarding the climate for speaking up about patient safety concerns and unprofessional behaviour within the clinical learning environment. Summary scores for each patient safety-related climate scale (ie, SAQ-Teamwork, SAQ-Safety, SUC-Safe and SUC-Prof) ranged from 0 (worst) to 100 (best).29 ,36 A summary score of ≥75 was interpreted as a positive perception of the patient safety-related climate being measured.29 ,38
Respondent characteristics included the Moral Courage Scale for Physicians (MCSP), a new measure of moral courage for physicians (ie, physicians' predisposition to voluntarily act upon their ethical convictions despite barriers) developed by several of the authors (WM, JME, SKB, LSL) and described elsewhere.39 MCSP scores range from 0 (worst) to 100 (best). Scores in the 75th percentile suggest a high level of moral courage.39 In addition to basic demographics, respondents were asked if they had received formal training in patient safety during residency.
The vignettes were based on actual cases and were designed by the authors based on review of the literature,1 ,12 ,33 ,40 personal experience and consultation with medical and surgical residents, nursing leadership and experts in patient safety. The two vignettes were designed to determine whether respondents' perceived likelihood of speaking up differed between a traditional patient safety threat and a professionalism-related safety threat while accounting for any differences in the perceived potential for harm. For each vignette, respondents rated their likelihood of speaking up (Likert scale from 1=not at all likely to 5=completely likely). Respondents made four judgements per vignette about their likelihood of speaking up to a nurse, intern, resident and attending. For our analyses, we categorised very or completely likely to speak up as ‘speaking up’ and not at all, slightly and moderately likely as ‘reticence’. In addition, respondents were asked, ‘What is the potential for harm to the patient in this situation?’ and given response options on a five-point Likert scale (1=very low to 5=very high). For the professionalism safety vignette, respondents were asked to rate the potential for harm separately for each of the four variations (ie, if the inattentive colleague was a nurse, intern, resident and attending physician).
Vignettes and survey items were pilot tested with 10 recent residency graduates from medicine and surgery. These physicians were asked whether the scenarios depicted in the vignettes were realistic and clear, whether any important information was missing which would assist respondents in interpreting and responding to the scenario and whether the scenario would be appropriate for both medical and surgical respondents. After completing the survey, they were asked to comment on the comprehensibility of the questions and ease of response, and to identify the presence of ambiguous wording. Their feedback was used to make minor modifications to the survey prior to its use.
Descriptive statistics were used to report survey responses. χ2 goodness-of-fit test was used to compare respondents' demographics to the total population surveyed. We used McNemar's test to analyse within-respondent differences in self-reported speaking up behaviour, barriers and facilitators to speaking up and attitudes regarding speaking up between traditional and professionalism-related patient safety threats.
Multivariate logistic regression was used to explore factors independently associated with speaking up in the traditional and professionalism-related safety vignettes. Factors potentially associated with speaking up were identified from the literature6 ,12 ,33 and assessed via scales and individual survey items. All hypothesised factors were included in each regression model. Covariates included level of hierarchy, perceived potential for harm to patients, perceived patient safety-related climates, gender, level of postgraduate training, specialty, moral courage, self-reported patient safety training and study site. Consistent with analysis of factorial survey data, the unit of analysis was the vignette, rather than individual respondents (sample size = number of respondents×number of speaking up judgements made in response to a vignette). We estimated a sample size of 2131 speaking up judgements for each vignette would provide 80% power to detect an effect size (OR 1.5) assuming moderate correlation (R=0.5) between covariates. This number translates to 533 respondents with four speaking up judgements per vignette (ie, likelihood of speaking up to a nurse, intern, resident and attending). To account for multiple comparisons, two-tailed statistical significance was set at an alpha level of 0.01.41 Analyses were performed using SAS V.9.4.
Of the 1800 interns and residents surveyed, 837 (47%) completed the questionnaire. Table 2 illustrates the characteristics of the respondents and the total population surveyed. The characteristics of these respondents were previously described in a related article.29
During their most recent inpatient month, 49% (410/837) of respondents reported observing a patient safety breach and 75% (628/837) of respondents reported observing unprofessional behaviour (p<0.001). However, respondents reported speaking up about the unprofessional behaviour they observed less commonly than speaking up about a patient safety breach (46%, 287/628 vs 71%, 291/410; p<0.001).
The majority of respondents (82%, 683/837) agreed that speaking up about unprofessional behaviour was important for patient safety. Greater than double the proportion of respondents agreed that it is difficult to speak up in their clinical area about unprofessional behaviour compared with patient safety concerns (38%, 322/837 vs 16%, 133/837; p<0.001, table 3) and substantially fewer forecasted meaningful change after speaking up in each setting (40%, 332/837 vs 60%, 504/837; p<0.001, respectively). While 65% (541/837) reported encouragement from colleagues to speak up about patient safety concerns, only 36% (305/837) reported the same for unprofessional behaviour; p<0.001. Respondents were least likely to report observing others speaking up about both patient safety concerns and unprofessional behaviour as a bystander (ie, when observing threats that did not directly involve themselves or their patients) (43%, 362/837 and 27%, 224/837, respectively).
Barriers and facilitators
Table 4 illustrates respondents' perceived barriers and facilitators to speaking up about patient safety breaches and unprofessional behaviour. Getting someone else in trouble, fears of conflict or eliciting anger and alienation from team members were the three most commonly endorsed barriers to speaking up about both patient safety breaches and unprofessional behaviour. Getting someone else in trouble (46%, 389/837) was the most commonly reported barrier to speaking up about patient safety breaches, while fear of conflict or eliciting anger (58%, 482/837) was the most commonly reported barrier to speaking up about unprofessional behaviour. Respondents more commonly endorsed nearly all the barriers provided for speaking up about unprofessional behaviour compared with traditional safety breaches (p<0.002 for seven of eight barriers, see table 4). Evidence that speaking up results in meaningful change and an anonymous reporting mechanism were the two most commonly reported facilitators to speaking up for both patient safety breaches and unprofessional behaviour (see table 4). About a third advocated greater commitment from leadership to facilitate speaking up about both types of safety threats.
Likelihood of speaking up in vignettes
Table 5 shows respondents' likelihood of speaking up in traditional versus professionalism-related patient safety vignettes across levels of hierarchy (ie, speaking up to a nurse, intern, resident or attending). Significantly fewer respondents reported that they would likely speak up in the professionalism-related patient safety vignette than the traditional patient safety vignette across all hierarchy positions, and these differences persisted even among respondents who perceived a high potential for harm to the patient in both vignettes (see table 5, p<0.001 for all comparisons). The fewest number of respondents reported that they would likely speak up to an attending physician for both the traditional patient safety and professionalism-related vignettes (64%, 537/836 and 9%, 78/836, respectively).
Factors associated with speaking up
Table 6 shows the results of two exploratory multivariate regression analyses of factors associated with speaking up in the traditional and professionalism-related safety vignettes. In the traditional safety vignette, perceived teamwork climate, SUC-Safe and SUC-Prof were all positively associated with speaking up. In the professionalism-related safety vignette, only perceived SUC-Prof was associated with speaking up. Surgical specialty and a high level of moral courage were also each positively associated with speaking up in both vignettes, while being an intern was negatively associated with speaking up in both. Male gender was positively associated with speaking up only in the professionalism-related safety vignette. Self-reported patient safety training was not associated with speaking up in either vignette.
Our study is the first to our knowledge to directly compare speaking up about traditional versus professionalism-related patient safety threats and our results highlight several key findings. First, about half of residents reported observing a safety threat during their most recent inpatient month, and about 70% spoke up about it, similar to a prior study of >1000 oncology doctors and nurses, where 54% recognised a colleague making a harmful error and 37% remained silent at least once.16 However, the prevalence of professionalism breaches was higher, with three-quarters of interns and residents reporting that they had observed such behaviour. And although the majority of respondents agreed that speaking up about unprofessional behaviour was important for patient safety, less than half reported doing so.
Second, while our findings demonstrate positive attitudes and behaviours related to communication about traditional patient safety breaches, they also reveal important safety deficits in the clinical learning environment. The majority of interns and residents perceived little support for speaking up about professionalism-related safety threats and did not perceive an environment that enabled them to be safety advocates in the setting of risky unprofessional behaviours. This was particularly notable when they considered being in a bystander role, where only a third of residents reported a supportive environment for speaking up. Given prior research demonstrating that perceived personal responsibility for patients, perceived efficacy of speaking up and encouragement from senior team members may facilitate speaking up,2 ,13 ,14 ,42 our results underscore the importance of fostering collective accountability, caring and capable leadership, and continued focus on cultivating a supportive clinical learning environment.43
Third, our findings add nuance to the barriers and facilitators of speaking up described in the literature (table 7).2 ,14 ,27 ,40 ,44–46 Speaking up about unprofessional behaviour was more commonly perceived as confrontational, more likely to cause self-alienation, less likely to be efficacious and less of priority than speaking up about traditional patient safety breaches. Fears of conflict or eliciting anger, and alienation from team members were the most common barriers to speaking up about unprofessional behaviour. This finding underscores the aspects of perceived personal safety that matter most to interns and residents, and the particularly high salience those aspects have to speaking up about unprofessional behaviour. Organisational leaders must commit to establishing, publicising and consistently adhering to a policy of non-retaliation and conflict-resolution curricula.
Notably, evidence of meaningful change after speaking up was the single facilitating factor most commonly endorsed by interns and residents, above communication skills training, personal protection from retaliation and other interventions. While many medical centres have processes in place to improve systems of care in response to traditional patient safety concerns, processes focusing on addressing unprofessional behaviour are more recently emerging.47–50 Organisational leaders should develop a reliable and transparent system for tracking concerns raised and holding individuals accountable who persist in their unprofessional behaviour.47 ,48 Such a system could also allow for anonymous reporting which nearly half of respondents felt would facilitate speaking up and which also serves as a protective mechanism.47 As organisations debate the merits of confidential versus anonymous reporting, these data may help shape additional inquiry or policy decisions. Although there were some notable differences, overall similarities in barriers and facilitating factors for speaking up about traditional and professionalism-related safety threats suggest that organisational leaders can help shape a more supportive clinical learning environment with streamlined interventions emphasising a conscious shift in organisational safety culture policies and perspectives that are inclusive of unprofessional behaviour.
Fourth, this study reveals differences between traditional versus professionalism-related safety threats, highlighting an important distinction for researchers and organisational leaders to further explore. We observed consistent differences in self-reported behaviour, barriers, attitudes and vignette results with regard to speaking up about traditional patient safety breaches versus unprofessional behaviour. While perceived risk of patient harm has been demonstrated as a predictor of speaking up,2 ,12 ,15 interns and residents were significantly less likely to speak up in the professionalism-related safety vignette than the traditional patient safety vignette even when they thought the potential for harm to the patient was high. Overall, interns and residents reported a higher likelihood of speaking up about traditional safety threats than previously reported,12 ,15 although our data show considerable variability when parsed by hierarchy (ie, speaking up to attending vs intern). One explanation is that prior studies examined speaking up more globally, diluting the distinct effect of the nature of the safety threat (ie, traditional vs professionalism-related) which may present greater obstacles to speaking up. For example, some incidents reported in prior studies by oncology doctors and nurses include ‘rule violations’ such as not using a mask or gloves when required, rude communication of prognosis to patients and not responding to persistent requests to evaluate a worsening patient,15 ,45 each of which could be perceived as professionalism-related safety threats.
Fifth, our exploratory regression analysis of factors associated with speaking up in two vignettes suggests that standard measures of patient safety culture (ie, SAQ teamwork and safety climate) may be limited in their ability to assess elements of patient safety culture related to speaking up. These scales were not associated with the likelihood of speaking up about a professionalism-related safety threat, while the SUC-Prof scale was. We previously reported that the SUC-Safe and SUC-Prof perceived by interns and residents in this sample was poor even when SAQ safety and teamwork climates are relatively good.29 The findings reported here further suggest that measures of safety and teamwork climate may not be adequately capture perceptions of the speaking up climate, especially around professionalism-related safety threats. Organisational leaders should consider adding the SUC-Safe and SUC-Prof to existing assessments of patient safety culture to detect potentially unmeasured obstacles to patient safety.29 Explicitly measuring SUCs may also signal to others that enhancing speaking up climates is an organisational priority and provide a way to track progress in response to interventions to improve speaking up.29
We found self-reported patient safety training was not associated with the likelihood of speaking up in either vignette. Simply offering training on speaking up without addressing the organisational and cultural elements that influence speaking up is unlikely to succeed and may increase the moral distress of participants by raising behavioural expectations without providing a supportive clinical environment. Indeed, few interns and residents felt communication skills training would help and these responses resonate with a prior study in which simulation training alone did not improve speaking up among faculty anaesthesiologists.51 However, as part of more comprehensive efforts including strong leadership and meaningful changes after speaking up, advanced communication skills training that emphasises a specific speaking up rubric,25 provides strategies to address conflict or anger with opportunities to engage in practice9 and highlights institution-specific policy and escalation pathways,14 ,48–50 may prove effective. Incorporating medical ethics education may also bolster the moral courage associated with speaking up.52 ,53
We found that surgical specialty was positively associated with speaking up, especially in the professionalism-related vignette. This may represent inherent personality differences between surgeons and internists,54 or elements of surgical versus medical culture that were not captured by the safety culture metrics in our study.55 ,56 In addition, male gender was associated with speaking up in the professionalism vignette but not in the traditional vignette, perhaps due to the more confrontational nature of addressing unprofessional behaviour21 ,22 and previously reported gender-based differences in confidence and empowerment.8 ,57 Finally, site differences persisted even after controlling for various patient safety-related climates and respondent characteristics and may reflect elements of organisational culture not captured in our study or the influence of regional differences in assertive communication.58 These preliminary findings merit further research.
Whereas prior studies of speaking up have primarily focused on a single specialty, practising physicians and nurses, and traditional patient safety threats (such as adherence to hand hygiene or a simulated clinical error),2 ,12 ,15 ,51 ,59 ,60 our study adds to the existing literature by measuring speaking up across multiple specialties and US medical centres and focusing on interns and residents who experience some of the greatest barriers to speaking up. In addition, as professionalism-related safety threats are increasingly recognised, we provide quantitative data on both prevalence and resident responses to such behaviours in the clinical learning environment, as called for by thought leaders to help better frame the problem and motivate solutions.28 Similar to prior studies,12 ,13 ,15 ,17 ,25 ,46 we found the likelihood of speaking up to be highly context-dependent, offering additional insights for organisational leaders to best support such advocacy. Although a study of labour and delivery nurses and obstetricians showed that bravery and assertiveness were not independently associated with speaking up, we found that moral courage was associated with the likelihood of speaking up, highlighting the importance of this specific virtue among physicians, and the potential unique role for this metric.39
Our study has important limitations. Despite a response rate consistent with other surveys of physicians, non-response bias could have affected the results. Respondents who were males, surgeons, interns or from study site F were somewhat under-represented relative to the total population surveyed. However, the majority of our analysis focused on within-respondent differences in attitudes and self-reported behaviour between traditional and professionalism-related safety threats; and those differences were highly significant. The results reported here do not advance the generalisability of findings regarding the psychometrics of the SUC-Safe and SUC-Prof scales that we reported previously within this sample.29 Future studies should aim to confirm these findings with higher response rates in different samples and among other health professionals (eg, nurses and attending physicians). Respondents were asked to recall past events, and thus, the findings are subject to recall bias. However, we limited the recall period to the most recent inpatient month and asked respondents to recall speaking up about unprofessional behaviour and patient safety breaches—relatively salient events that may have been easier to remember. Nonetheless, because the study relied on self-reported behaviours and self-assessment, our findings were subject to social desirability bias, and respondents may have overestimated or underestimated their likelihood of speaking up. While research shows that judgements made in response to vignettes are often similar to those made in practice,61 decisions to actually speak up are likely influenced by contextual factors that could not be simulated in our vignettes. Furthermore, only two vignettes were used and speaking up judgements may not be generalisable to other speaking up scenarios involving other traditional or professionalism-related safety threats, and further research with multiple scenarios and actual or simulated events is needed. Because the dependent and independent variables used in our regression analysis as well as the validation measures of the SUC-Prof and SUC-Safe scales29 are all from the same data source (ie, respondents from a single survey), our models are subject to common-method variance. In addition, clustering (eg, vignette judgements within participants and participants within institutions) may have impacted our regression analysis by underestimating variances and increasing effect sizes. Thus, our vignette findings and regression models should be viewed as preliminary. Finally, additional research is needed to correlate improved speaking up climates with clinical safety outcomes.
In conclusion, interns and residents frequently observe unprofessional behaviour; however, they are less likely to speak up about it compared with traditional patient safety threats even when they perceive high potential for patient harm. Our findings offer important insights into the factors that may influence the likelihood of speaking up and may be used to design interventions to address current deficits and improve patient safety. In addition, we highlight new metrics (ie, SUC-Safe and SUC-Prof) that may help organisational leaders more comprehensively assess safety culture and track progress over time.
Twitter Follow William Martinez @DrWillMartinez
Contributors Study conception and design: WM, LSL, EJT, JME, GBH and SKB. Acquisition of data: WM, JTS, DWB, AMS, JAB and NBM. Analysis of data: WM and JME. Interpretation of data and critical revision: all authors. Drafting of manuscript: WM.
Funding American Philosophical Society; National Center for Advancing Translational Sciences (UL1 TR000445); The Arnold P. Gold Foundation; Brigham and Women's Hospital; National Institute of Child Health and Human Development (1K24HD053771); Health Resources and Services Administration (T32HP10251).
Competing interests None declared.
Ethics approval IRB at each study site.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.