Guidance on Patient Reported Outcomes
-
Hello - Based on the Guidance on Patient Reported Outcomes, are non-significant data presentations acceptable, if they are in a claim-neutral manner? In areas such as Oncology, treatments aim to have no worsening/deterioration in HRQoL vs. comparators. While this equates to a non-statistically significant difference, it is actually a positive clinical outcome.
-
Hello - Based on the Guidance on Patient Reported Outcomes, are non-significant data presentations acceptable, if they are in a claim-neutral manner? In areas such as Oncology, treatments aim to have no worsening/deterioration in HRQoL vs. comparators. While this equates to a non-statistically significant difference, it is actually a positive clinical outcome.
Hi @alee
A non-inferiority test that results in a "statistically non-significant difference" is a form of statistical analysis and would be required to support such a conclusion, and it is not appropriate to conclude non-inferiority (or similarity) based on nonsignificant test result in a study designed only for superiority as noted in the Guidance Document for Claims Based on Non-Inferiority Trials. The appropriate interpretation of this observation is that the test product failed to achieve statistical significance for superiority (i.e. it would not be considered a “positive clinical outcome”), which does not support a claim of “similarity”. Please also note that all clinical claims must be pre-defined, directionally consistent with the primary endpoint (confirmed through statistical evidence), and similar to the description in the domains section, overall scores require statistical significance as per sections 3.4, 3.5 and 3.6 of the Guidance document for the Use of Patient Reported Outcomes in Advertising, respectively .