The following four drugs showed no quality problems at GMS and retained their potency at the facility level: amoxicillin capsules, acetylsalicylic acid tablets, phenoxymethylpenicillin tablets and tetracycline capsules. For three other drugs (ferrous sulfate tablets, retinol tablets and ampicillin injection) the outcome in facility samples reflected the problems of initial quality without additional loss of active ingredient. These seven drugs therefore did not show any problems with instability.

Incidental, clinically unimportant failures in facility samples occurred with three drugs: ampicillin capsules, doxycycline capsules and epinephrine injection. Measurement error is a possibility.

When matched batches at the central medical stores and facility level were compared, the mean loss of potency was statistically significant for two drugs: ergometrine injection (p<0.0005) and procaine benzylpenicillin injection (p<0.001). With ergometrine injection, the initial quality problem worsened at facility level: 72% of facility samples failed, compared with 65% failures at GMS.

The mean assay in the facility sample group was 73.5% of stated content compared with 82.4% at GMS, For procaine benzylpenicillin injection the assay was consistently lower in the facility sample group (mean 98.3%) compared with GMS samples (101.0%) with a mean sample interval of only 4.3 months.