Medical device risk acceptability

As a general rule, clinical data are required as evidence to support conformity with the requirements of the Active Implantable Medical Devices (AIMD) and the Medical Device (MD) directives with regards to safety and effectiveness under the normal conditions of use, evaluation of undesirable side effects, and the acceptability of the benefit/risk ratio. Risk analysis should be used to establish key objectives that need to be addressed by clinical data, or alternatively to justify why clinical data are not required (mainly for Class I devices). The risk analysis process should help the manufacturer to identify known (or reasonably foreseeable) hazards associated with the use of the device, and decide how best to investigate and estimate the risks associated with each hazard. The clinical data should then be used to establish the safety and effectiveness of the device under the intended use conditions, and to demonstrate that any of the residual risks are acceptable, when weighed against the benefits derived from use of the device. 

Medical devices need to be assured in a manner which is acceptable to a regulator and the requirements are usually embodied in law. Regulatory conformance often involves being compliant with one or more standards linked to the local regulatory framework. Paradoxically providing assurance can in some ways be easier for medical devices as the risk and quality management expectations are clearly set out and well understood by practitioners. 

At this point it is worth mentioning how the ALARP approach to risk acceptability changes markedly when it comes to medical devices. For example. International Standard ISO 14971 [10] affords manufacturers the ability to justify risks on an ALARP basis. The European Commission Medical Device Directive (MDD) [11] cites ISO 14971 as a Harmonised Standard however the directive contains a number of Essential Requirements on risk acceptability which appear to conflict with ISO 14971. 

Note that the international version of the standard and that used to achieve compliance with the EC Medical Device Directive vary slightly and in particular in the approach to risk acceptability aiteria (see Sect 3.3). 

Abstract A recent standard for medical device software lifecycle processes, ISO/IEC 62304 (ISO 2006), assumes and specifies a software safety classification scheme, where docmnentation, verification and validation tasks to be carried out depend on the safety classification. This means that a risk-driven approach has become an accepted standard for medical devices. 

Risk Analysis takes all inputs and products to produce a list of risk items to be properly dealt with the appropriate measures. As a general rule, risks in medical devices are usually avoided or mitigated. On the contrary, direct acceptance of a risk is not an option except for cases of highly beneficial medical devices. Even in that case, it shall be possible only for the most superficial, harmless and improbable risks that can hardly be reduced or mitigated. Mitigation and avoidance can be achieved by means of additional requirements, safety checks, boundary control, labelling, etc. Furthermore, software can hardly reduce the severity of a risk, instead it can reduce its probability to happen and/or increase its visibility should it happen. In the end, all risk items should have been brought to an acceptable level of residual risk. 

The risk of failure of medical devices in general and for each device in particular should be considered acceptable. When good performance and absence of adverse events can be predicted with sufficient certainty, the outcome can be discussed with the patient in order to allow him to balance risks and benefits. 

Although a blood-compatible polymer should be non- or less-hemolytic, in practice several medical devices can cause hemolysis. Nevertheless, when hemolytic effects take place, it is important that the valnes of hemolysis are within acceptable limits and clinical benefits overcome the remaining risks. 

In developing a neuroprosthetic device, it is particularly important to understand the function that is to be restored and how this aspect of the disability is treated medically. The technology must be not only functional, but must also be deployable by clinical practitioners (physicians, therapists, and nurses) whose appreciation of the complexity of the technology may be limited. The design must also meet the requirements of the user, such as an acceptable level of risk, time commitment, and the effort required for implementation and training. The neuroprosthesis must not only function acceptably, but it must also be easy and natural to use and easy to put on. Acceptable function may be less than full, normal function. 

The Standard requires manufacturers of electro-medical equipment to have a formal risk management system in place. Manufacturers must estimate the risks relating to their device and take action dependent upon how that risk compares to predefined levels of acceptability. There are objective pass/fail criteria and one may choose simply to follow such requirements in the design of their device. 

---