Fault Tree Analysis for Medical Devices

By 22 August 2017 News No Comments

There may be many of you out there who are confused by that name, or you may have heard the words but just don’t know what it is? Perhaps you know what it is but have you ever used it in Medical Device Design? Until recently I knew very little of Fault Tree Analysis let alone how to interlace this into the intricacies of Medical Device Risk Management but this is exactly what we had to do for one client, which meant we ended up with a tool which could fault find and measure the risks. I will come on to Fault Tree Analysis (FTA) in a moment, but it is the background of risk management which underpins the reason for using this.

The FDA recommends using ISO14971 for risk management of medical devices and although it doesn’t stipulate which risk methods to use it does state you must use some form of tool to analyse the risks. Most, if not all medical device companies these days will be using some form of FMEA tool for their design processes and this is fine for most situations, however by only using this then they are potentially missing failures by making this their only approach. Meaning they will be susceptible to recalls, device failures and ultimately user harm.

Whilst FMEA is a bottom up approach, it looks at the failures of components in a specific system. FTA is quite different and the opposite of FMEA’s. It uses a top down approach, providing analysis of the system (including component failures) which lead to the occurrence of a defined effect or event.

Although not initially developed by them, FTA has been used extensively by NASA especially since the Challenger disaster in 1986. This probabilistic risk assessment method, once realised, became one of the most important reliability and safety analysis techniques available.

Medical device design would appear to be in its early days of realising this tool and as such has just started to understand its potential to prevent risks developing into harm and even death.

FTA works on the principle of a system failure being a top event, determining the causes by working backwards with a logical tree illustration to connect the events and their relationships to result in the undesired event. In its most simplistic form it connects relationships with AND and OR gates, where two events would happen together (AND gate) or happen either/or (OR gate). This allows the risk assessor to build a picture of where the root causes are by building a picture of the various interactions in the system. FMEA would only look at the component failures, not the interactions between them and the overall system.

It is important for medical device companies to understand where they may be missing an opportunity to prevent risk situations from occurring by using a tool such as FTA. At the moment, the FDA may not stipulate that both these tools should be used, however anyone who isn’t may well find themselves in a difficult situation when it comes to explaining why risks weren’t exposed. How can your company defend not using a tool such as FTA when the methods have been used and proven to work in other harmful risk industries?

In the image above. Shows a simple FTA for Fire being the top event and the events that occur that lead up to this risk.

Matt Wyre, Head of Systems – Mechanical Design Consultant