SHELF LIFE |
Manufacturer A: 2 years |
ASSAY METHOD |
Spectrophotometric method, BP 1988 p, 840/Addendum 1989. Contents of 10 vials mixed and measured three times; mean taken. |
ASSAY LIMITS |
90 - 110% (BP) |
SAMPLES OBTAINED |
Manufacturer A |
GMS samples |
no. |
% |
Total |
25 |
100 |
Harare |
16 |
64 |
Bulawayo |
9 |
36 |
Facility samples |
no. |
% |
Total |
72 |
100 |
Age > 50% SLife |
2 |
3 |
Facility type = PCH |
58 |
81 |
Climate = hot |
53 |
74 |
Transport = slow |
46 |
64 |
Longitudinal series |
|
|
Total |
42 |
100 |
SUMMARY ASSAY RESULTS
GMS samples |
All Manuf. |
|
|
|
Mean age (mths) |
|
1.5 |
|
|
|
Mean assay |
|
101.0% |
|
|
|
95% CL |
99.3 - 102.8% |
|
|
|
Mo. and % fail |
0 |
0% |
|
|
|
No. and % low fail |
2 |
3% |
|
|
|
No. and % high fail |
0 |
0% |
|
|
|
Facility samples |
Manufacturer A |
|
|
|
Mean age (mths) |
|
6.0 |
|
|
|
Mean assay |
|
98.3% |
|
|
|
95% CL |
96.9 - 99.7% |
|
|
|
No. and % fail |
3 |
4% |
|
|
|
No, and % low fail |
2 |
11% |
|
|
|
No. and % high fail |
0 |
0% |
|
|
|
Longitudinal series |
at manufacturer |
at GMS |
at facility |
Mean age (mths) |
|
0 |
|
1.4 |
57 |
Mean assay |
|
101.2% |
|
101.2% |
97.3% |
95% CL |
100.2 - 102.1% |
99.9 - 102.6% |
95.3 - 99.4% |
No, and % fail |
0 |
0% |
0 |
0% |
3 |
7% |
| |
Mean interval (mths) |
4.3 |
| |
Mean loss (-) or gain (+) |
-3.6% |
| |
95% CL for loss/gain |
-6.3 to -0.9% |

Figure 1 - Results of GMS samples

Figure 2 - Results of facility samples
Findings:
a) No expired samples were found.
b) All 25 GMS samples were within assay limits (mean 101.0%). There was one borderline sample (92.0%).
c) Three of 72 facility samples (4%) failed; all failures were below the lower limit (72.2, 83.0, 85.0%). All three failures were from the same batch, the same geographical area and were transported by rail.
d) 42 GMS/facility sample pairs showed a mean loss of potency (-3.6%); the difference was statistically significant (t-test one tail, p=0.0001).
e) No initial quality problem was found.
f) A modest instability problem was seen. Since the samples were in the early part of shelf-life, this is potentially more serious.