Biomedical Engineering Reference
In-Depth Information
evaluated but stopping short of delineating precise
test methods.
using a transparent development process (i.e., open to
public scrutiny) are eligible for recognition. Stan-
dards may be proposed for recognition from within
the agency based on regulatory needs or by outside
persons.
17.11.2 The Standards
Development Process
The standards development process differs among
organizations, but is generally voluntary and
consensus oriented. Three major standards organi-
zations with an interest in medical devices are the
ASTM International, the International Standards
Organization (ISO), and the Association for the
Advancement of Medical Instrumentation (AAMI).
Members of these groups include representatives
from health care organizations, academic laborato-
ries, government agencies, manufacturers, testing
companies, and trade associations. The development
and publishing of a standard, therefore, is
a consensus process with these persons of varied
backgrounds working together to agree on a standard
method or practice. A standard is typically initiated
by a technical committee or subcommittee of
members with a shared interest in a particular area,
who then write and vote on a series of drafts until
a final document can be agreed upon and published.
Standards are described as voluntary because the
organizations that issue them have voluntary
membership and they have no legal authority to
impose or enforce their implementation. In the
United States, the CDRH has a standards recognition
program by which consensus standards may be
evaluated by the FDA and recognized for use in
satisfying a regulatory requirement, but manufac-
turers are not legally required to use these standards.
The goal of standards recognition is to allow manu-
facturers the option to declare partial or total
conformity to a recognized standard. This conformity
is voluntary but is intended to reduce the time and
burden necessary for clearance or approval of
a device because both parties (the manufacturer and
the FDA) are already familiar with the techniques or
protocols described in the standard. The manufac-
turer should therefore not need to provide a detailed
explanation of those methods in their reporting as
they would have to do for an original test protocol.
The FDA updates the list of recognized standards in
the Federal Register approximately once per year
and more frequently, if necessary. National or inter-
national standards that do not conflict with any
statute, regulation, or policy under which the FDA
operates and that are developed by an organization
17.11.3 Benefits and Limitations of
Using Standards
The three primary benefits of standards in the
regulatory process are that they: (1) establish
a consensus in the technical community as to the best
currently available test procedure for a specific
application; (2) improve communication between the
manufacturer and regulatory agency; and (3) allow
comparisons of data among different device designs
and different testing facilities. First of all, a fully
developed standard should ideally be supported by
experience and data obtained from testing, so that the
users have some confidence that the techniques
described can produce repeatable, reliable results and
thus they do not need to constantly reinvent new test
methods. Second, when a manufacturer chooses to
use a standard, it should reduce the amount of time
and discussion needed for a regulatory official to
understand which tests were done and how they were
performed. Even if the user must make modifications
or adaptations to a standard to fit their specific device
design, beginning with a published standard method
or practice gives the manufacturer and the reviewer
a common point of reference. Finally, it is hoped that
standards allow comparisons of data not only among
different device designs tested at the same facility but
also devices tested at different facilities. The
certainty with which these data can be compared is
defined in terms of the precision (intra-lab vari-
ability) and bias (inter-lab variability) of the test.
Unfortunately, numerical estimates of precision and
bias are not included in many standards because of
the complexity involved in performing multi-
laboratory or “round-robin” testing. This short-
coming of many standards reflects the fact that
development of a standard test method is not finished
upon its initial publication, but rather requires
ongoing multilaboratory testing as well as periodic
review in light of technological advances.
Standards have several limitations as well. First of
all, standards are typically focused on methods of
measuring specific characteristics of a device or
material in vitro. They provide useful tools for
comparison of devices, but results cannot necessarily
