February 2024 SOCRA Source Journal - Journal - Page 67
•
•
What is the net impact
(risks/benefits) of the
product?
Are the patients
representative of the target
population?
The answer to each question
may be different for each study
that is part of the literature
review.
If the results are about the
device under evaluation but
for a different indication, their
applicability to the indication
under evaluation must be
addressed. If there is bias, a
financial disclosure statement by
the researchers and the name of
the sponsor must be included.
POST-MARKET DATA
Post-market data is another
key part of the CER (Table 5).
Include any direct and indirect
data for the subject device and
similar devices relating to the
following:
• Adverse events
• In-house complaint data
• Post-market clinical trials
• Recall databases
• Device registries
When using US data, it is
necessary to include a total
lifecycle report for the product
code. The report discloses
and analyzes all of the adverse
events. In-house complaint
data must be disclosed. The
MDR and the IVDR require
that device manufacturers be
proactive as well as reactive;
they must search for information
to be included about postmarket clinical trials.
Device manufacturers often
have questions about how
to summarize post-market
TABLE 4:
LITERATURE AND THE CER
Sources:
• Scientific literature databases
• Post-market databases
• Standards
• Registries
• Clinical trial registries
Literature review process:
• Databases to be searched
• Search terms
• Inclusion criteria
• Exclusion criteria
Critical appraisal:
• Are the results valid?
• Are there limitations?
• Are the results applicable?
• Is there bias?
surveillance events. The
summaries must identify the
subject device and equivalent/
similar devices, using both inhouse and externally available
data. Details should include
descriptions of the nature of
the event, geography, and the
impact on the patient.
The analysis should include
trends, rates, and risk
assessment. In-house postmarket data should be
compared to external postmarket data. Statistics should
be updated based upon new
information. It is important to
be careful with assumptions.
If the risk management plan
states, for example, that an
adverse event will occur in 1 in
1 million subjects and it occurs
in 5 or 10 in 1 million subjects,
the risk management plan
needs to be revised.
The CER should include
results from all relevant recall
databases, including those
of Health Canada, Swiss
Medic, the US Food and
Drug Administration (FDA),
and the Therapeutic Goods
Administration (TGA). Health
Canada has an outstanding
publicly available recall
database. Regulators pay more
attention to recall databases
pertaining to their geography
or the geographies they work
closely with.
Considerations related to postmarket data include paying
attention to regulatory actions.
Often, device manufacturers
think that a regulatory action in
a small county is not important;
however, regulators will
probably be communicating
with each other. It is better
SOCRA SOURCE © May 2023
67