5 Tools for Performing Medical Device Risk Analysis

Various tools can help you identify the hazards for your risk analysis, and the choice on which and how many of them will be used is yours. Of course, the use of more than one tool reduces the likelihood of missing some hazards and their causal chains. But, on the other hand, the performance of an additional 2 or 3 analyses consumes significant resources that might be currently scarce.

Two of the most common, and later discussed here in this post, tools for risk analysis are the Fault Tree Analysis (FTA) and the Failure Modes and Effects Analysis (FMEA). In addition to these two techniques, the popular Mind Map Analysis and the P-Diagram will also be presented.

Continue reading to learn about some of the tools you can use for identifying hazards!

1. Fault Tree Analysis (FTA)

The FTA technique, developed by Bell Labs in 1962, was first used on the Minuteman missile system. Afterwards, it was widely applied in civil aviation, space and military applications. The Fault Tree Analysis is a deductive top-down reasoning process, which starts from the undesired system outcomes. It attempts to identify all credible sequences of events that could potentially result in unwanted system outcomes. Visually, it is a graphic model that depicts the logical connections among the parallel and sequential combination of events which could lead to the event at the top of the tree.

The FTA risk analysis technique can:

• Model both normal and fault conditions under various operational and environmental scenarios;
• Identify and model fault dependencies and common cause failures.

The logical simplification and reduction are typical for the Fault Tree Analysis. The fault trees consist of parallel and sequential events interconnected by logic gates (e.g. ‘and’ and ‘or’). A single correct fault tree cannot be used to describe an entire system but multiple logically equivalent fault trees can. The FTA can provide you with:

• Minimal cut sets of the tree, where a cut set is the smallest set of basic events which if occur will lead to the occurrence of a top event.
• Qualitative component importance.
• Knowledge of cut sets susceptible to common cause failures.

The FTA is a qualitative model that can be assessed quantitatively. It can be applied to new products before their design details are available and to already existing products. When it comes to new products, the FTA tool can:

• reveal, at a high level, probable event sequences that might result in system hazards;
• alert the design team to safety-critical aspects of the system.

In regards to already existing products, the FTA can:

• help identify design weaknesses
• aid in the identification of design upgrades that could make the system safer.

2. Mind Map Analysis

The Mind Map analysis, popularised by Tony Buzan, is a very old technique for graphically organising thoughts and ideas. It uses diagrams to map information graphically, and it is considered as a very useful tool for brainstorming. Moreover, it’s a good alternative to the Fault Tree Analysis because it is simpler to learn, easier to use, and the software for mind mapping is either free, or very low cost.

The primary purpose of using this technique is to graphically tell the story of how an undesired event can occur. The information will later be captured in the Sequence of Events in the Preliminary Hazard analysis. Unlike the above described FTA tool, Mind Maps don’t use logic gates, and all connections are interpreted as ‘OR’ gates. However, a workaround could be to create an ‘AND’ node, which output would occur if all the incoming branches, fed into it, occur.

3. Preliminary Hazard Analysis (PHA)

The PHA is a risk analysis technique usable early in the medical device’s development process for identifying hazards, hazardous situations and events that could cause harm. It is highly recommendable to be used for new and novel product development. This technique makes possible:

• Identifying the safety-critical parts of the system concept;
• Estimating the potential risks associated with the system;
• Guiding the medical device’s design team to prioritise and focus resources on the highest risk parts of the system.

The Preliminary Hazard Analysis is an early version of the Hazard Analysis Report (HAR), and it uses the same RACT template. However, in comparison to HAR, the PHA has access to little actual information on the design and uses many estimations. It is seen more as project guidance and a reference for the HAR.

Main inputs to the PHA can be: system requirements; concept architecture; intended use, user and environment for use; risk acceptability criteria, list of clinical hazards and harms assessment list.

4. Failure Modes and Effects Analysis (FMEA)

FMEA, developed by the US military in 1949, was first used as a reliability analysis tool. Later on, it was applied in many space programs of NASA, and today it is widely used in many industries for improving the product quality. FMEA is a systematic, inductive/ bottom-up method of exploring how a product or process might potentially fail to achieve its objective. Additionally, it explores the effects of the possible failures on the performance of the system, or the process, or the environment and personnel.

The subject of the FMEA can be the entire medical device, a subsystem, a component, a process, or anything related to the medical device that the analyst chooses. Once selected, the subject is decomposed into elements. During the analysis, the failure modes of each element are identified, and the consequences on the whole subject are considered.

Some of the main benefits of FMEA are:

• It examines every element in the scope of analysis for their failure modes and effects;
• Helps to detect and to eliminate product failure modes, thus improving product reliability and quality;
• Reduces product development costs;
• Helps discover missing or wrong requirements.

As weaknesses of this medical device risk analysis tool can be mentioned the following:

• Inability to detect end effects that require multiple failure modes;
• Unable to catch hazardous end effects which don’t result from failures;
• Time-consuming to perform it and challenging to master it.

Different types of FMEA exist and serve different purposes. Some of the popular types are:

• Process Failure Modes and Effects Analysis (PFMEA) – it is a structured approach that helps identify weaknesses in process design and assign levels critically to each process step;
• Design Failure Modes and Effects Analysis (DFMEA) – it helps to assess the weaknesses that are present in the product and can be controlled by design. It is best if this tool is used early in the design and development phase;
• Use-Misuse Failure Modes and Effects Analysis (UMFMEA) – it is a tool that analyses the effects of failures in the use/misuse of a medical device.

All FMEA types are used to identify hazards and estimate the likelihood of their occurrence.

5. Parameter Diagram (P-Diagram)

The P-Diagram is another risk management tool which visually represents a model of a system and its behaviour under various conditions. The main elements of the model are:

• System – this is the entity that processes the inputs, under the control and noise factors, and delivers the output.
• Inputs signals – it describes the items that the system needs to fulfil its objective.
• Ideal function – this is the intended output of the system.
• Noise factors – anything that we cannot control and can influence the output of the system
• Control factors – any factors which we can control and change.
• Error states – unintended outputs of the system, resulting from the influence of the noise factors on the system’s operation.

This tool is beneficial in the development of FMEA, because the error states can help identify the failure modes in an FMEA and the noise factors can help with the identification of the causes of the failure modes.

Source:

1. Elahi, B. (2018). “Safety risk management for medical devices”. Academic Press