• Communication
• Safety culture
• Healthcare quality improvement

Communication failure constitutes a key contributor to healthcare errors.1 ,2 In addition to poor communication and poor hand-offs, failure to speak up when one recognises a potential safety problem—unsafe acts or unprofessional behaviour—represents an important example of communication failure. Despite on-going calls for clinicians to speak up when they notice threats to patient safety, speaking up remains difficult due to fear of repercussions,3 ,4 power differences and authority gradients,1 ,5 among other factors. Some evidence suggests that clinicians may not speak up even when they perceive substantial potential for harm.6

In an effort to tackle the issue of speaking up, Martinez et al7 present preliminary psychometrics for a new measure of speaking up climate. This paper makes an important contribution to the literature, as the field can certainly benefit from a measure that focuses on perceptions and enablers of ‘speaking up’. Martinez et al study two scales. The first measures the climate for speaking up about traditional patient safety concerns (SUC-Safe), such as improper sterile technique or an inadequate hand off. The second scale focuses on perceptions of speaking up about professionalism-related safety concerns (SUC-Prof), such as covering up an error, false documentation or disruptive behaviour. Both of these areas—traditional patient safety concerns and unprofessional behaviours—clearly represent important targets for ‘speaking up’ by members of the care team.

Several of the findings from Martinez et al hold interest. Their results comparing per cent positive scores on their speaking up climate scales when compared with more general safety attitudes scores (measured using the Safety Attitudes Questionnaire8) highlight residents’ overall reluctance to speak up in general, but particularly regarding professionalism issues. These findings can drive change as they help pinpoint some of the more tangible aspects of safety climate that we need to address. Their findings are consistent with other work in which medical trainees report that they often do not feel they could approach someone who was engaging in unsafe care practice.9 Their multivariate results showing that patient safety training explained a significant amount of variance in perceptions of speaking up climate about safety, coupled with their findings showing a significant relationship between perceptions of speaking up climate and actual speaking up behaviours, are also encouraging for the field and suggest that patient safety training can be efficacious.

Other aspects of the approach used by Martinez et al provide an opportunity to reflect on further research needed in this area. In particular, the field would benefit from (i) additional conceptual work on the definition and dimensionality of speaking up climate and related implications for systematic measurement of this construct and (2) greater attention to the patient/family side of speaking up. I will comment briefly on each of these areas.

Martinez et al argue that professional norms, training exposure, personal stakes involved and other factors that influence speaking up around traditional safety concerns versus professionalism-related concerns may differ. Thus, speaking up climates for safety and for professionalism may differ within the same clinical setting. With this premise, they go on to use an approach to scale validation that treats ‘speaking up for safety’ and ‘speaking up for professionalism’ as two distinct concepts rather than treating them as two dimensions of the same concept of speaking up climate. This analytic approach, coupled with their results showing that speaking up about a patient safety breach was significantly related to perceptions of speaking up climate for both safety and professionalism, suggests that the concept of speaking up climate would benefit from a more detailed theoretical exploration that would help answer some important questions.

For instance, how is speaking up climate defined? Is the concept uni-dimensional or multi-dimensional? We also need to consider whether there are other salient dimensions of speaking up climate in healthcare. Martinez et al, themselves, mention speaking up about knowledge deficits—speaking up regarding concerns about clinical competence is certainly a critical area for ensuring patient safety. Perhaps speaking up about care decisions is also an important dimension of speaking up climate? Instances where a team member has questions about the course of clinical care but does not necessarily feel comfortable raising the question may pose an important threat to patient safety, and such instances likely arise more often than do the safety and professionalism concerns on which the authors focused.

Conceptually, one could also argue that the speaking up areas examined by Martinez et al represent dimensions of patient safety climate more broadly. Patient safety climate reflects staff perceptions of the importance and prioritisation of safety relative to other priorities on a patient care unit and in an organisation.10 Speaking up about safety demonstrates the extent to which patients safety counts as a greater priority to staff than characteristics such as camaraderie or professional hierarchy (in much the same way that senior leadership support for patient safety is a dimension of patient safety climate that shows the extent to which safety is a priority over goals such as productivity and efficiency).

Martinez et al note early in their paper that speaking up is under-represented in existing safety climate instruments and they have introduced much needed discussion about the importance of a speaking up climate. Future work might usefully examine conceptual overlaps among speaking up climate and the more established area of safety climate. Indeed, as the patient safety literature expands to include new measures in important areas like speaking up, it is crucial that there is a strong theoretical foundation for this work—this kind of nomological network with clear linkage between the conceptual and the observable (ie, a new measure) is central to establishing construct validity.11

Speaking up by patients also merits further research. Martinez et al7 studied residents from six large US academic medical centres and rightly point out that, because residents are low on the medical hierarchy, speaking up can be particularly challenging for them. Despite possessing a wealth of knowledge about their condition, patients and families are generally lowest on the knowledge hierarchy making speaking up particularly challenging for them too. The Safety Competencies Framework situates patients as members of the care team.12 Patients and their families are the most vested (and vulnerable) members of the care team. In addition, they often have the most comprehensive picture of what is going on with their care—particularly in an academic setting where a patient may be seen by a large number of medical trainees and staff physicians from more than one subspecialty area over the course of several days.

I tried to be part of day-to-day care decisions for my father during a recent hospitalisation for heart failure. The Joint Commission's Speak Up programme ‘urges patients to take an active role in preventing health care errors by becoming involved and informed participants on their health care team’. Despite the implementation of programmes such as these, barriers to hearing the patient's voice persist. Our request to include my husband, who is a physician, in my father's care discussions so that he could help the non-clinicians in the family understand what was happening, was met with a lukewarm reception: the staff physician offered to send him a discharge summary. Perhaps requests to be involved in care are perceived as questioning physician behaviour—something Pronovost13 notes that the healthcare culture still does not support. My profession is in the healthcare field and I have little trouble asserting myself in most situations. Yet, I was patently unsuccessful when I tried to speak up and ask questions regarding my father's care. This does not bode well for other patients and families, particularly those with low levels of health literacy, not to mention members of marginalised communities. In sum, consideration of including ‘speaking up by patients’ would constitute a welcome next step in developing measures of speaking up climate.

In this early stage of considering and measuring speaking up climate, hopefully we can avoid the pitfalls associated with the measurement of safety climate,14 where measures evolved and expanded in ways conceptually inconsistent with the definition of safety climate. Going forward, the field would benefit from deeper theoretical development and empirical inquiry regarding the definition of speaking up climate, its conceptual boundaries, its dimensionality, as well as its antecedents and outcomes. Finally, patients’ experiences with speaking up are central to a positive patient experience and good patient care. We would therefore be wise to include the patient perspective in any new speaking up frameworks that emerge.