If resources were available, then every single report would constitute a signal, but in practice, some have used simply the number of reports for a particular reaction/drug combination as a cut-off.
WHAT CONSTITUTES A SIGNAL?
If
resources were available, then every single report would constitute a signal,
but in practice, some have used simply the number of reports for a particular
reaction/drug combination as a cut-off. This cut-off has been, for example, two
or more, or three or more reports. This is a reasonably sensitive but a very
non-specific test. The number of reports, whether the signal is a causative
effect or not, will depend on the number of patients exposed to the drug. The
first step in the process is to attempt to estimate incidence. The number of
reports is taken as the numerator but a question exists as to what is the
correct denominator. Possible alternatives are as follows:
• Sales,
• Prescriptions written,
• Prescriptions dispensed.
Even
if the data on prescriptions dispensed are available, they do not necessarily
relate to the impor-tant factors related to a causal effect. If it is simply
patient-years of exposure, then the total number of prescriptions dispensed is
a reasonable measure. However, this assumes that the risk of having the ADR is
constant over time. If this is not so, then we need patient-years grouped by
duration of treatment. This requires individual patient-based data or at least
the distribution of the number of prescriptions per patient.
A cut-off for a signal could then be an incidence rate that is greater than background. This is the basis of the Poisson method of examining signals. This has utility in some specific areas where the background rate is well known and rare, and where the reporting rate is known to be reasonably high or at least well known. It can be used to compare reporting rates of two drugs using sales data as a denominator.
Related Topics
TH 2019 - 2024 pharmacy180.com; Developed by Therithal info.