Guidance For Industry - Assessment Of Abuse Potential Of Drugs - U.s. Department Of Health And Human Services - 2017 Page 32
ADVERTISEMENT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37Contains Nonbinding Recommendations
2. Does the test drug (T) produce mean responses that show less abuse potential compared to
positive control?
µ
µ
δ
µ
µ
δ
−
≤
−
>
δ
≥
H
:
H
:
versus
where
0
.
0
C
T
2
a
C
T
2
2
3. Does the test drug produce mean responses that show similar abuse potential compared to
placebo?
µ
µ
δ
µ
µ
δ
δ
−
≥
−
<
>
H
:
H
:
0
versus
where
.
0
T
P
3
a
T
P
3
3
The actual values of δ
, δ
, and δ
vary according to such factors as subjective measures, drug
1
2
3
class, and route of drug administration. All the margins should be pre-specified and justified in
the protocol. The statistical tests yield multiple comparisons (all doses of positive control drug
versus placebo; all doses of the test drug versus each dose of the positive control drug; and all
doses of the test drug versus placebo) for each of the subjective measures collected. For each
hypothesis, the statistical significance of the test should be achieved on all doses. Thus, no
multiplicity adjustment is recommended.
Statistically, HAP studies should be evaluated as safety studies. Thus, the null hypotheses for
the test drug should be constructed based on the presumption that the test drug produces abuse
potential similar to the positive control and therefore differentiates from placebo. In order to
demonstrate that the drug has no abuse potential, the null hypotheses should be rejected
statistically. This is in contrast to the statistical evaluation used in efficacy studies, in which the
null hypothesis is constructed on the presumption that the test drug is not efficacious and does
not differentiate from placebo.
The statistical analysis of the data from a HAP study should begin with descriptive statistics of
the mean, standard error, and other summary statistics such as minimum, first quartile (Q1),
median, third quartile (Q3) and maximum for each subjective measure, each treatment and each
paired difference among treatments. These data should then be used to create tables and graphs.
Primary Analyses
The statistical model that should be used in HAP studies is a linear mixed-effects model, which
includes period, sequence, and treatment as fixed effects, and subject as a random effect. The
primary analyses of abuse potential should be based on testing the differences between the means
from the primary measure(s) at the peak of drug response effects (E
) produced by the test
max
drug, the positive control, and placebo, using proper tests and appropriate statistical methods at a
significance level of 0.05 (1-sided).
29
ADVERTISEMENT
0 votes
Related Articles
Related forms
Guidance For Industry And Food And Drug Administration Staff - U.s. Department Of Health And Human Services - 2012
Legal
Department Of Health And Human Services Family Medical Leave/family Illness Leave Request Form
Business
Form Nih 2762-3 - Undergraduate Scholarship Program (ugsp) - Academic Enrollment Certification And Service Obligation Deferment Request - U.s. Department Of Health And Human Services
Legal
Form Acf-196t - Administration For Children And Families - U. S. Department Of Health And Human Services
Legal
Related Categories
Parent category: Legal