We’re Rooted


In this series of posts under the heading ‘we’re killing our patients’ we explore why clinical error remains the third greatest killer in hospitals (greater than sepsis and trauma).

In many healthcare reporting systems adverse events which are sufficiently severe and likely to recur are subjected to a Root Cause Analysis (RCA).

Root Cause Analysis Definition:
‘A method used to investigate and analyse incidents to identify the root causes and factors that contributed to the incident. The process yields recommended actions directed at the prevention of a similar occurrence.’

Example – Central Line Related Air Embolus:
We have been advocating for central lines and attachments that open to air to be replaced with those that don’t. (See here) We don’t need our lines to open to air, and while they do they place patients at unnecessary risk of air embolus. When an air embolus event occurs it is concerning to hear the response ‘It’s okay, they’re going to do an RCA’. It’s like saying ‘we’re bringing the big guns out – if this ain’t gonna fix it nothing will’. Unfortunately numerous central line air embolus events have been subjected to an RCA but effective sustainable solutions are still yet to be put in place – and the air emboli continue to occur.

So why don’t the ‘big guns’ work?

Lack  of training
Most people performing RCAs have not received formal or adequate training. This may result in an inadequate investigation, failure to identify causal factors, and failure to develop a meaningful corrective action plan. It would be ideal if there were more human factors experts in the world – perhaps we would all benefit from greater understanding in this field.

Time pressure
In NSW – ‘The maximum time allowed for an RCA to be completed and the report to be submitted to the MoH is 70 calendar days from when the incident was notified in the incident management system.’

Hindsight bias
Many organizations only perform RCAs when a harmful event has occurred. As we advance in patient safety, RCAs should be used more proactively on near misses and no-harm events.

Lack of understanding & poor conclusions
‘No member of the RCA team should have been directly involved in the incident or in the care of the patient.’
Why not utilise the staff involved in an incident? They will often have ruminated for days over the event, will have likely become experts in this area, and would be best placed to reveal workable solutions. Those involved in the adverse event don’t have an avenue to propose their potential solutions to the problem. A process that involves these individuals would also be beneficial in reducing the ‘second victim’ effect.

Weak corrective action plans
Those involved in an RCA will often propose corrective action plans that typically focus on training and policy development. Instead there should be greater focus on the redesign of processes and products that will have more success in addressing human error.

The design and implementation of solutions is as important as the solution itself.

Were front-line users involved in designing the solution?
If not, and realistic workflow and environmental concerns were not considered, the solution may actually create another hazard. For example, a solution identified in a corrective action plan now causes a nurse to leave his area and go to another department to sign-out a drug or piece of equipment. Additional safety hazards may occur if there is no one monitoring his patients when he is away from the department. Workarounds are created when solutions are complicated and difficult to implement. As a workaround, this nurse may decide to take a piece of equipment he recently used on another patient from the dirty utility room and do a quick wipe-down rather than go to another department to sign-out a piece of equipment. The key to making any effective change is that it must support rather than hinder or create a barrier for the end-user.

Unwilling colleagues
Key people within the organization need to be involved in order to make the process effective; physician and leadership participation is vital to sustainable success. It is impossible to make true system-wide changes unless all disciplines in the organization take ownership of the safety issues involved.

Lack of spread
Often, teams that performed an RCA after an event occurred in a specific department do not feel empowered to make improvements beyond their own setting and therefore only focus on ways to address the immediate problem within that specific department.

Follow up
Once the corrective action plan is created, it should be monitored to ensure that 1) it is fully adopted into practice and 2) the problem is resolved. If neither of these objectives were accomplished, the RCA team should reconvene, ask “why”, and address those issues until a successful plan is implemented. Often limited health care resources prevent this from taking place.

Lack of confidence
Few organisations feel confident enough in their analysis to contact someone outside of their organisation, like a medical device manufacturer, about the identified need to redesign their product. Sustainable solutions which may require changing a piece of equipment or changing an established system may seem too far fetched and so may be off the radar altogether.
Through our own work we have become increasingly aware of this issue and would like to illustrate with the following examples. They demonstrate quite different responses from the companies involved.

1. Riotane Tinted Pink (Perrigo)
Almost colourless antiseptic solutions of chlorhexidine and alcohol have been mistaken for normal saline and inadvertently injected into the epidural space and central lines. (See here) There is no need for these solutions to be so indistinguishable when suitable alternatives exist.

We remember first seeing the lightly tinted solution and thinking ‘that’s dangerous, it could end up in someone’s back’. Several years later there was a near miss event at one hospital followed weeks later by the unfortunate and catastrophic injection of chlorhexidine in to a patient’s epidural space.

We presume that both events would have made it to an RCA. Perhaps these events are a good illustration of how incident reporting and RCAs fail. Years later we became aware that the solution involved is still supplied in some hospitals. Attempts to have the solution banned through state health procurement bodies were fruitless so we decided to contact the supplier directly.

Perrigo genuinely seemed unaware that their product posed a hazard. Subsequently they have been extremely helpful and are in the process of developing a safer, more obvious colouring of their product. We commend and thank them for their efforts and look forward to Riotane tinted pink chlorhexidine and alcohol being removed from our hospitals.

2. Draeger APL valves
There have been numerous case reports in the literature of sample lines and wiring becoming trapped under the Draeger APL valve. (See here) This is a hazard unique to Draeger anaesthetic machines, and most anaesthetists (even those who use the Draeger machine regularly) are unaware of the issue until it effects them or their immediate colleagues with the potential for disastrous consequences.


We would be interested in your thoughts.
(We lend very heavily on work published by the CHPSO in our discussion. See http://www.chpso.org/lessons-learned/how-effective-your-causal-analysis-process)


Please read other posts in the series:

We are killing our patients

Latent Errors – Equipment

Beneath The Radar

Prudence is seen as a virtue

‘No one else has complained about it’

‘I can’t work with you any more.’

Human Factors

Sexy Human Factors and the Clinical Error Taboo
We’re Rooted

Solution Feedback Loop

One thought on “We’re Rooted

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s