The WHO/INRUD drug use indicators are intended to measure aspects of health provider behaviour in primary health-care facilities in a reliable way, irrespective of who collects the data. The indicators provide information to health-care managers concerning medicine use, prescribing habits and important aspects of patient care. All the indicators have been extensively field-tested in many countries and found to be relevant, easily generated and measured, valid, consistent, reliable, representative, sensitive to change, understandable, and action oriented. DTCs can use indicator studies to:

• describe current treatment practices to determine whether there are problems in medicine use, and which facilities or prescribers have problems. When an indicator study shows unacceptable results, the DTC can investigate the problem in more depth and then take action to improve these results.

• show trends over time through the repeated measurement of the indicators so providing a monitoring mechanism. Prescribers and facilities whose performance falls below a specific standard of quality can be targeted for more intensive supervision.

• motivate health-care providers and DTC members to improve and follow established health-care standards.

• evaluate the impact of interventions designed to change prescribing behaviour by measuring indicators in control and intervention facilities before and after the intervention (see section 7.6).

In addition to showing the WHO/INRUD indicators (which are basically designed for primary health-care facilities), box 6.5 also shows a selected list of indicators for medicine use in hospitals, particularly for inpatients. These hospital indicators have not been field-tested and are not widely accepted as the WHO/INRUD ones are, and they cannot replace a drug use evaluation (section 6.5). Since most of these indicators do not relate diagnosis to disease, they cannot tell us exactly what proportion of people were treated correctly or the exact nature of the drug use problem; they can only indicate that there is a drug use problem. Furthermore, different disease patterns and prescriber type will greatly affect the indicators, so analysis should be done by diagnosis or prescriber type if these vary between the facilities to be compared.

The number of prescribing encounters per facility and the number of facilities which should be examined will depend on the objective of the study and are described in detail elsewhere (WHO 1993). If the objective of the study is to describe drug use problems in a sample of facilities that is representative of a majority, then at least 30 prescribing encounters in each of 20 facilities (a total of 600 prescribing encounters) should be examined. If fewer health facilities are examined, then more prescribing encounters should be examined. If the objective is to study prescribers in one facility, as may be the case for a hospital DTC, then at least 100 prescriptions should be obtained at the single facility or department; if there is more than one prescriber, 100 prescriptions for each individual prescriber should be obtained.