Writing a Clinical Evaluation Plan for the EU MDR

What is a Clinical Evaluation Plan?

A Clinical Evaluation Plan (CEP) is an important technical document that sets out a clear plan for conducting a Clinical Evaluation of a medical device.

However, a CEP is much more than just a ‘check box’ document - it is a vital tool for ensuring that clinical evaluations are performed properly and according to a process laid out in advance.

An effective Clinical Evaluation Plan acts like a recipe, guiding and harmonising the clinical evaluation process and ensuring sufficient rigor and objectivity from start to finish.

Because production of a compliant Clinical Evaluation Plan is mandatory for all medical devices under the Medical Device Regulation (EU) 2017/745, it is essential that manufacturers are clear on requirements for CEPs and fully understand how to structure, write, and update a CEP for their medical devices.

What are the requirements for writing a Clinical Evaluation Plan under the MDR?

Article 61(3) of the MDR states that a clinical evaluation must “follow a defined and methodologically sound procedure”, meaning that a Clinical Evaluation Plan needs to be established in advance and should define how the evaluation shall be conducted.

MDR Annex XIV Part A provides further details on requirements for CEPs, stating that it should include at least the following:

• an identification of the GSPRs that require support from relevant clinical data;
• a specification of the intended purpose of the device;
• a clear specification of intended target groups with clear indications and contra-indications;
• a detailed description of intended clinical benefits with relevant and specified clinical outcome parameters;
• a specification of methods to be used for examination of qualitative and quantitative aspects of clinical safety with clear reference to the determination of residual risks and side-effects;
• an indicative list and specification of parameters to be used to determine, based on the state of the art in medicine, the acceptability of the benefit-risk ratio for the various indications and for the intended purpose or purposes of the device;
• an indication how benefit-risk issues… are to be addressed; and
• a clinical development plan

However, this list constitutes the bare minimum of what should be included in a CEP and, as written in the MDR, isn’t easy to follow. Let’s look in more detail at how to structure a Clinical Evaluation Plan.

How should a Clinical Evaluation Plan be structured?

When planning the structure of a CEP, it’s worth having a simple framework in mind that can be used as a reference point. One such framework of key requirements is as follows:

• identification, classification and general description of the device
• background and purpose of the CEP
• summary of CEP methodology (e.g. use of literature / equivalence route, etc)
• a hypothesis to be tested during the evaluation
• intended purpose, clinical benefits, indications and contraindications
• identification of similar devices
• analysis of relevant and non-relevant GSPRs with justification for any non-relevant
• a detailed search protocol for identifying, appraising and analysing clinical evidence.

Each section should be clearly written in sufficient detail to enable people other than the author to precisely follow the ‘recipe’.

Two aspects of a Clinical Evaluation Plan that commonly cause confusion are the need to construct a hypothesis and the requirement to set out a Search Protocol. For this reason, each of these deserves specific consideration.

Developing a Clinical Evaluation hypothesis

Done properly, a Clinical Evaluation is a process that uses clinical evidence to test an idea; properly formed, that idea should be expressed as a hypothesis. A hypothesis can be defined as a ‘testable statement’ - in other words, a statement that can be examined for validity in the face of evidence.

In a clinical evaluation, best practice is to develop a quantitative hypothesis developed using outcomes from previous literature reviews or from prior experience with similar devices. An example of a quantifiable hypothesis is:

When used as intended, the subject device will result in a mean reduction of pain VAS that is at least 5 points on a 0-10mm scale at 3 months post use.

In many cases, it is best to express a quantitative hypothesis in terms of non-inferiority - in other words, that the subject device will be non-inferior to an external threshold for a variable. An example might be:

When used as intended, the subject device will be associated with a frequency of adverse events (as a percentage of device sales) that is non-inferior to that seen following use of comparable alternative devices.

Although not specifically quantifiable, this example is essentially numerical as it calls into consideration data from comparable alternatives that will be derived from the literature reviews that need to be performed during the clinical evaluation.

While quantitative hypotheses are preferred, it is also permissible to include qualitative hypotheses where logical and reasonable. For example:

When used as intended, the subject device will be associated with types of adverse events (as a percentage of device sales) that are no more serious than those seen following use of comparable alternative devices.

Whether quantitative or qualitative, care must be taken to ensure that the hypothesis is suitable for testing in the clinical evaluation. Outcomes from a previous literature review can be really useful here - there’s no point stating that you will test reduction in pain VAS if this isn’t a variable examined by publications in the clinical field. If you need help in conducting a systematic literature review, be sure to discuss your requirements with our team.

Developing a Search Protocol

A vital element of all Clinical Evaluation Plans is a search protocol. This protocol is a plan for the identification, appraisal and analysis of clinical data. In many ways, this forms the core of the clinical evaluation process and a CEP will be held to be non-compliant without a clearly constructed search protocol.

When developing the search protocol, be sure to pay attention to all three elements. Let’s look at each in turn.

1. Identification of clinical evidence

This section should set out how you will identify relevant sources for inclusion in the evaluation. Evidence should be included from both the manufacturer and from independent sources such as journal publications, so a process for identifying evidence from each of these should be described.

When it comes to independent literature, a sensible starting point is to draft one or more research questions to be addressed by the literature review. Following this, it should be easier to define precise search terms that can be entered into clinical databases such as PubMed and Google Scholar.

Other things to consider are inclusion and exclusion criteria - for excluded sources, what justifications will be provided and how will these be documented in the CER?

The test for sufficiency of an identification protocol is whether someone independent could follow the instructions and obtain an output that is essentially identical to the next person. Use this test to ensure your evidence identification plan is sufficiently detailed.

2. Appraisal of clinical evidence

The next step is to describe how the quality of a source will be determined and recorded. Simply speaking, appraisal is used to assess how well conducted a study is and how relevant it is to the evaluation; it is a necessary precursor to weighting, which describes how much emphasis should be placed on the results of a source in the evaluation.

Appraisal criteria should account for multiple features of a study; for example, simply having a variable for ‘method’ is unlikely to be sufficiently granular. A better approach would be to establish a scoring or appraisal scheme for sample, size, follow-up period, statistical methods, clinical relevance and other more individualised aspects of study quality.

Establishing an appraisal scheme in the CEP means that it cannot be ‘reverse engineered’ to suit the quality of favourable studies subsequently identified. In turn, this helps demonstrate that the Clinical Evaluation meets the test of objectivity imposed by the EU MDR.

3. Analysis of clinical evidence

The final component of the CEP search protocol is a system for analysing clinical evidence. In short, analysis considers what ‘message’ can be derived from identified and appraised evidence. Specifically, what does it say about:

• the device’s suitability for intended purpose?
• the device’s benefit-risk profile?
• conformity with relevant GSPRs?

In writing a plan for analysis of evidence, once again ensure that the plan is sufficiently detailed to be followed independently and consistently, leaving no room for subjective adjustment of outcomes following conclusion of the evaluation.

General Safety & Performance Requirements (GSPRs)

Another core component of an EU MDR Clinical Evaluation Plan is to set out an analysis of the MDR Annex I GSPRs for relevance.

This should comprise a listing, in full, of those that are relevant along with a justification for each held to be non-relevant to the device.

Where can I find help writing a Clinical Evaluation Plan?

Mantra Systems offers a range of support intended to ensure that your CEPs are fully compliant with the EU MDR. Our comprehensive Clinical Evaluation Plan CEP template provides line-by-line guidance for each section of the CEP, along with illustrative examples to show writing style and phrasing. It’s suitable for anyone with some existing experience of technical / regulatory writing for the EU MDR.

We also offer a full CEP writing service where our experts will develop your CEP on your behalf from start-to-finish. This is the surest way to ensure your CEPs meet with all requirements of the MDR and can be engaged with or without our market-leading CER writing service.

For any enquiries relating to Clinical Evaluation Plans or other aspects of working with the EU MDR, contact our team today.