Medicine-use researchers, health care providers, policy makers, and managers collect data to describe patterns of medicine use, to address medicine-use problems, and to monitor medicine use over time. They use two basic methods: quantitative methods, to measure what is being done, and qualitative methods, to provide information on why it is being done. Both methods are used to better understand the causes of problems before intervening to correct them. The intervention design must include outcome indicators that are meaningful, reliable, and measurable. Appropriate study sites and a relevant comparison group must be randomly selected; outcomes need to be measured before and after the intervention in both groups. If an appropriate comparison group cannot be identified, study designs that measure change over time (interrupted time series) may be used. Medicine-use data can also help evaluate the effect of interventions. Reliably measuring medicine use requires standardized indicators to provide consistency. In 1993, the World Health Organization (WHO) and the International Network for the Rational Use of Drugs (INRUD) produced a manual that defines core medicine-use indicators and provides a methodology for measuring these indicators for general outpatients in health care facilities. Similar indicators are needed to measure pharmaceutical use in hospital inpatients, in private pharmacies, and for pharmacy benefits for insurance schemes, as well as to measure adherence to antiretroviral therapy (ART).
Sources of data differ according to the intended use and setting. Common sources of quantitative data include pharmaceutical supply orders, stock cards, patient registers, medical and prescription records, medicine-use databases, and patient exit surveys. For qualitative studies, data routinely come from patient interviews, questionnaire surveys, patient observation, and focus group discussions. Each method has strengths and weaknesses and is appropriate for different circumstances.
Medicine-use investigations can occur in public health care facilities, in private-sector facilities, and in the community. Methods for investigating use in different settings may differ; for example, mystery shoppers or simulated patients who pose as customers with specific health problems are useful for studying practices in private-sector pharmacies, whereas public health facilities may rely on readily available, routinely collected data. After a medicine-use study, meetings to discuss the results help health care providers, managers, and policy makers identify specific problems related to the medicine use process and design appropriate interventions to address them. The effectiveness of different interventions depends on a number of factors, including the intervention itself, the setting, and the implementation process. Evaluating interventions is necessary to assess the overall impact of the program or to compare the relative effectiveness of different interventions.