Abstract
A study was carried out to determine whether it would be feasible to use record linkage for monitoring adverse effects of drugs. For a total population of 43 117 people, all prescriptions dispensed during a two-year period were linked with records of hospital admissions, obstetric deliveries, and deaths (in or out of hospital). For another population of 10 453 people, prescriptions were linked with information about illnesses recorded by general practitioners. In both cases, the data were analysed to reveal associations between drugs and diagnoses. It was concluded that record linkage would be valuable for detecting unsuspected hazards of drugs, and for testing hypotheses from other sources. The opportunities for this type of research are greatest in countries with highly organized health services, such as the Scandinavian countries and the United Kingdom.

This publication has 15 references indexed in Scilit: