Making soft intelligence hard: a multi-site qualitative study of challenges relating to voice about safety concerns
Martin and colleagues ask why organisations fail to uncover concerns even when they are, in principle at least, eager to, and propose a further possible influence: that certain properties of formal channels for raising concerns may, perversely, deter voice. They interviewed 165 people in three academic hospitals in two countries, 57 leaders and senior managers at the blunt end of care and 108 at the sharp end. Leaders said they valued voice and saw formal channels as a key route for it, and many described an instinct towards formalisation: to be knowable at the blunt end, soft intelligence had to be made hard, properly documented, formalised and amenable to verification, through procedures attentive to due process, fairness and defensibility against litigation, complaints of bullying or discrimination, or union action.
The way these systems operated meant that some concerns were never voiced. Participants were much happier to report system or equipment failures formally than concerns about individuals; central to that reluctance was a widely shared understanding that raising one formally passed it from the sharp end into the blunt end's world of processes and procedures. When unsure a concern could survive harsh scrutiny, they were likely to stay silent, not wanting to catalogue evidence and, some feared, to share in consequences that might escalate or prove disproportionate; anxiety about evidence and process was particularly relevant where the intelligence was especially soft, feelings or intuitions hard to resolve into a coherent account. Informal sense-checking among peers handled some concerns in situ but, once local handling became the norm, could lead to implicit tolerance of unacceptable behaviour. Making soft intelligence hard thus risked creating forbidden knowledge (a borrowed term), dangerous to know or share. Formal systems are vital, they say, but their findings suggest no effort to improve them will fully overcome this, and they see the managerial emphasis on due process as consistent with a wider trend towards legalism. They offer two possible routes: supporting informal, peer-oriented voice with an escalation plan behind it, and leaders who are visible and trustworthy, listen informally and abandon the quest for clarity and certainty, for which they found hints in their data.
The bridge to psychological safety, a term the paper does not use, is ours: we read it as suggesting that whether people speak can depend on what a channel will do with a half-formed concern, not only on how they expect to be treated for raising it. Its stated limits: interviews only, with no means of verifying behaviours, practices or impacts; no way to say whether confidentiality secured a breadth of informants or complete accounts; and three large academic medical centres, each focused on improving its safety culture, in two high-income countries. The first hospital commissioned the initial study, with freedom to publish agreed from the start.
Save this paper to a bibliography, keep notes on it, and export citations. Create a free account →
Neighbours (7)
- Helping and Voice Extra-Role Behaviors: Evidence of Construct and Predictive Validity
- Culture and behaviour in the English National Health Service: overview of lessons from a large multimethod study
- Implicit Voice Theories: Taken-for-Granted Rules of Self-Censorship at Work
- Amplifying Weak Signals
- Leadership in Healthcare (CRM)
- Whistleblowing & PS
- You Can't Fix a Secret