10.4.1 Reporting and peer review

A system of annual reporting exists for all the QC laboratories, except the one in Uganda. Another way of assessing the performance of a QC laboratory is to have it reviewed by its peers. Several countries participate in review schemes. The Australian TGAL, the SAM Laboratory in Estonia, and both the General Laboratory and the Pharmaceutical Laboratory in Cyprus participate in the Network of Official Medicines Control Laboratories scheme of the European Directorate for the Quality of Medicines (European Pharmacopoeia). The Drug Analysis Division laboratory in Malaysia is a QC centre for ASEAN and a WHO Collaborating Centre for Regulatory Control of Pharmaceuticals, while the Venezuelan INHRR laboratories are Centres for International Reference on Biological Products.

10.4.2 Ability to meet demand

Table 10.3 presents for each country, for the period 1994-97, the number of drug samples submitted for testing, samples tested and the failure rate in those tests. Data are available from all 10 countries on the total number of samples submitted and the number of samples tested, and these show that seven countries- Australia, Cyprus, Estonia, Malaysia, the Netherlands, Venezuela and Zimbabwe -have been able to meet the demand for testing (92-100%). Uganda has a relatively low test rate of 56% for the submitted samples, followed by Cuba and Tunisia at 72% and 88%, respectively. The lack of certain equipment and materials, such as reference standards and reagents, constrains analysis in Cuba, Cyprus, Uganda and Zimbabwe.

Figure 10.2 Samples submitted for testing as a percentage of drugs registered*

* Data for registered drugs for Australia only approximate

The ratio of samples submitted to samples tested indicates the ability of the QC laboratory to meet demand. As stated earlier, however, these ratios do not indicate whether the current activities of the drug regulatory authorities keep pace with the overall need for pre-marketing and post-marketing drug QC. If the number of drug samples submitted for testing is viewed in terms of the number of drugs registered, the picture looks quite different. Figure 10.2 shows the number of samples submitted for testing as a percentage of drugs registered. Without taking into account the number of drugs submitted for registration, a significantly larger number of samples are submitted for testing in Estonia and Venezuela, as a percentage of drugs in the market, than in other countries.

The adequacy of testing in terms of regulatory needs can also be examined with respect to the kinds of test that the DRA laboratory has the capacity to perform. Laboratories that can perform only physicochemical analyses (see Table 10.2) probably lack the capacity to undertake drug QC.

10.4.3 Quality of drugs tested

The data on drugs tested by the DRA laboratories show that the percentage of drugs failing QC tests ranges from 1.4% in Cyprus to 23.8% in Uganda (Figure 10.3). Because drug regulatory authorities in some countries - Australia and Venezuela, for example - employed a targeted approach to QC testing by collecting samples of drugs which were suspected to have quality problems, test results from these countries are likely to show a higher failure rate than in countries where samples are collected randomly. The interpretation of QC failure rates is probably also affected by the way the terms “pass” and “fail” are defined. With a stricter definition, a larger proportion of samples will fail the tests. In Tunisia, for instance, the National Medicines Monitoring Laboratory considers defective packaging to be a failure to meet standards, and the batch concerned will be recalled.

Figure 10.3 Percentage of drug samples tested and failed during 1994-97*

* Data for failed samples not available for the Netherlands.

According to information gathered by the principal investigators, the average QC failure rate of drugs ranges from less than 5% to 20%, depending on the country and the type of product. Drug regulatory authorities themselves in Australia, Estonia, Malaysia, Venezuela and Zimbabwe estimate the average QC failure rate of drugs to be: higher than 5% for prescription drugs and 15-20% for complementary drugs in Australia; 5% in Estonia; 5% in Malaysia; 1% in Venezuela and 12% in Zimbabwe (Table 10.3).

The question whether sanctions are imposed-and which sanctions-when a product is found to fail drug QC testing is as important as detection of substandard products itself. The present data sets are, however, inadequate for establishing direct links between the findings of QC analysis and the imposition of regulatory sanctions (area E in Figure 4.1). Certain cases have been documented in Venezuela over the past 10 years. For example, 100 products were prohibited, and a laboratory manufacturing almond oil was closed down.