Evaluation (Validation and Verification) - Medical Device Design: Innovation from Concept to Market - page 235

And that is irrespective of precedents, evaluations, and design studies and it applies in a
similar way in all markets around the world. This ruling is also starting to be applicable to all
implants and to all new software.

Is it possible to test more than one hypothesis at a time? Clearly, if you are testing the safety
of a device you can also evaluate clinical benefits as often the measure is simply, do the
benefits outweigh the risk? However for clinical studies it is difficult, if not impossible, to
conduct a proper study if more than one hypothesis is being tested. So the simple answer is
no; try to avoid multi-hypothesis studies.

The study is a part of your design process so it needs documenting. Hence it is preferable to
have a pro forma that states the hypothesis you are intending to test, and gives any further
information/aims that may be required. Figure 9.20 illustrates a typical pro forma that bares a
remarkable resemblance to a statement of need.

9.6.2 Investigation Specification
You should not be surprised to see that the starting point is in fact none other than a PDS
for a clinical study. You should be designing your study using the same methodology as the
design of the device, and the starting point is the PDS. A word of advice: always include
a statistician (or at least someone with a working knowledge of statistics) at this phase.
Too many studies have collected results from a poorly designed study (from a statistics
perspective) only to result in no meaningful results being produced. Hence one should always
start with a PDS for the study. You can use the same pro forma as shown in Chapter 5, but
your sections will be different, as follows:

1. Target population : Who is the beneficiary? This may be an age group, a gender group,

maybe even an ethnic group. The beneficiary may not be a patient; it could be a member
of hospital staff, such as a nurse or surgeon.

2. Regulatory and statutory : If your study includes patients or human subjects it is

required to abide by the Helsinki Declaration 2 (now in its sixth revision); we will
discuss this further later. If your device is not cleared for market in the USA nor has
a CE mark then you will need to seek approval from the appropriate regulatory body.
Does your study fall under any specific national or international standards? Do you
need to meet any standards before the study can start? If you are going to capture
patient data will you be bound by data protection laws or freedom of information
requirements?

2 The Helsinki Declaration is concerned with the ethical issues related to experimentation on human beings.