Tag: <span>speak-up</span>

New paper – modelling “internal whistleblowing” in healthcare

A new study in BMJ Open (follow this link for full version of the paper) has been published by Paul Rauwolf, Bangor University and Aled Jones – who, below, reflects on the paper.


The paper makes a distinctive contribution to the debate about speaking-up (or ‘internal whistleblowing’ as referred to in the paper). It was an interesting challenge to work on a different approach to generating and analysing data. Agent based modelling is a computational modelling approach, which simulates the actions and interactions of both individual or collective entities (such as organizations or groups) and assessing their effects on the system as a whole.

The work focuses on how different speaking-up rates compare with each other in a variety of environments. These models also demonstrate how much institutional nuances can alter the effectiveness of a given speaking-up rate.

Internationally, healthcare staff are being encouraged and supported to raise and respond to concerns in the workplace (see here for an example from a previous post). Given this, the paper aims to answer questions such as:

  • Are resources within the system able to cope with potentially hundreds or thousands of complaints per month?
  • What are the effects, if any, of any delays in responding to concerns on patient care and on employees’ propensity to raise concerns in the future?
  • Will patient care improve as a result of speaking-up?

The paper calls into question the a priori belief that speaking-up always improves patient care. Given resource constraints, whistleblowing /speaking-up policies should seek to understand the complex relationship between accuracy of information, efficiency in responding to concerns, formal and informal speaking-up channels and the utility of speaking up.

The results show that an optimal model for speaking-up include:

  • Organisations where employees’ practices are fully transparent to others and employees within an organisation have a propensity to raise concerns;
  • When introducing roles such as Freedom to Speak Up Guardians, organisational resources and efficient processes are required to ensure concerns are responded to in a timely way. Inefficient processes and response delays suppress speaking-up;
  • It is almost always beneficial to encourage and permit informal channels for speaking up. A combination of informal and formal speaking-up processes perform better than formal speaking-up in isolation. Cheap, timely and informal but less transparent and possibly accurate channels for speaking-up, mixed with more accurate, formal but costly channels leads to excellent patient care.


10 lessons for speaking-up: learning from Virginia Mason’s PSA system.


Virginia Mason (VM) Medical Center based in and around Seattle consists of one 350 bed hospital facility and 9 satellite units. They employ 5000 staff, 500 of which are physicians.

The Patient Safety Alert (PSA) system was introduced in 2002 following a staff survey. This showed that staff were fearful of speaking-up about concerns. Also, staff doubted that information generated from concerns would improve the safety of care. At the time of the survey VM staff wishing to raise concerns had to complete a quality incident report (QIR). Typically, QIRs would sit on a shelf collecting dust for months, then filed away and forgotten. As we know from countless reports and commentaries into safety failures in healthcare and other industries, perceptions of fear and futility around speaking-up are inimical to creating a positive speak-up or open culture.

Patient Safety Alerts (PSAs).

The philosophy of PSAs is based on Toyota’s “Stop the line” policy, which  empowers any employee to stop the multi-million pound production line to prevent a defect occurring. Specifically, the philosophy being replicated at VM is that instead of criticizing or silencing employees who raise concerns, they offer support and resources to fix the problem. PSAs have a broad remit and are intended to capture all events involving the safety and well-being of a patient.

Results of implementing PSAs.

Result 1: From March 2002 to January 2014 staff raised concerns a total of 43,615 times. This has grown from around 10 or so reports a year in the period 2002-2004, to 850 reports a month in January 2014.

Result 2: Between 2004 and 2014 there has been a 74% reduction in insurance liability premiums paid by staff, which is an indicator of a major shift in “risk perception” by insurers. This shift in risk perception is probably based on insurers having paid out less in liability payments related to claims against VM staff – think “no claims bonus” with car insurance! Complaints and claims management go down as PSAs go up.

Result 3: PSA reporting has provided intelligence to drive improvements to quality. An improvement system where priorities are often identified and owned by sharp-end employees, who are supported by senior leaders and executives in the organization. Bottom up meets top-down.

Lessons from Virginia Mason…..

The rate change in PSAs reported in 2004-05 and
2009-10 seem to be linked to specific events which the
executive team subsequently used to publicise the
PSA’s philosophy to all employees. These events
seemed may have built employee trust in the system. Examples of events included what was termed “You call in a
PSA, we’ve got your back” which publicized how the PSA system supported a nurse speaking-up about a physician’s lack of compliance with safety procedures. Similarly, another event publicized how a senior physician used the PSA to report his own drug error. The “showcasing” of these series of events enabled the executive team to establish a new narrative supplanting traditional norms of skepticism and silence about reporting concerns with a “new” narrative of trust , openness and improvement stemming from raising concerns.

10 lessons for speaking-up in the NHS

1. VM didn’t get the PSA system right first time. It took almost a decade of learning.
2. They still haven’t got it right all the time, but reporting of concerns and responses to concerns are significantly better.
3. The PSA system has undergone continuous improvement.
4. Executive leadership is a prerequisite to building a culture of speaking-up.
5. Extensive staff education around listening and responding to employee concerns needed.
6. Publicizing positive case studies can dictate a safety narrative and create a new norm around
7. Mechanisms for speaking-up should be easily accessible, with multiple options available.
8. Rapid feedback to staff in the first instance. Keep staff updated with progress thereafter.
9. “Significance” is in the eye of the beholder: Open the floodgates to all concerns.
10. Be prepared to change as the organization learns from speaking-up.


Kenney C. A Leadership Journey in Health Care: Virginia Mason’s Story. New York: CRC Press; 2015.

Furman C, Caplan R. Applying the Toyota Production System: Using a patient safety alert system to reduce error. Jt Comm J Qual Patient Saf. 2007; 33(7): 376-86.

Embedding a System to Protect Patient Safety


Project update March 2021

We completed data collection in January 2020, just as Covid-19 was reaching the UK. The last 12 months have been dedicated …

New paper – modelling “internal whistleblowing” in healthcare

A new study in BMJ Open (follow this link for full version of the paper) has been published by Paul Rauwolf, Bangor University …

10 lessons for speaking-up: learning from Virginia Mason’s PSA system.

Background.  Virginia Mason (VM) Medical Center based in and around Seattle consists of one 350 bed hospital facility and …