Clinical trials of new drugs are conveniently classified into:
PHASE 1: in which the drug is given to a small number of normal volunteers in closely controlled and supervised conditions to study its kinetics and pharmacological effects.
PHASE 2: in which the drug is given to a relatively small number of patients with the disease for which its use is proposed. The therapeutic efficacy, correct dosage and pharmacokinetics of the drug are determined by comparing the data with those for normal subjects to obtain some evidence of its safety.
PHASE 3: in which the clinical evaluation of the drug is extended to large numbers of patients (perhaps hundreds). Trials include comparisons with placebos and with established treatments.
PH ASE 4: (postmarketing surveillance, PM S): in which long-term assessment of the safety and efficacy of the drug is made in thousands of patients, following the licensing of the drug for marketing. Experience has shown that careful observation of patients in phase 2 and phase 3 clinical trials is only likely to detect those adverse reactions that occur in 1% or more of patients exposed to a drug. Adverse reactions with an incidence of less than 1% require detection in phase 4 (PMS) studies.
Several countries have developed systems for collecting information about suspected adverse drug reactions. In the UK two systems are of particular interest-the prescription event monitoring and yellow card system.
Prescription event monitoring (PEM)
The PEM scheme involves identifying doctors and their patients as the prescriptions for a particular drug pass through the central Prescription Pricing Authority office. Relevant prescriptions are photocopied and the copies are sent in confidence to the Drug Surveillance Research Unit in Southampton. Each ‘test’ drug under investigation is matched with a ‘control’ drug that is chemically or pharmacologically similar and already marketed for the same indications. Similar numbers of patients receiving each drug are selected and a simple questionnaire is sent to their general medical practitioners, requesting information on age, new diagnoses or events that have come to the doctor’s attention, and reasons for any referral to a consultant or admission to hospital. PEM should be able to identify adverse drug reactions that have an incidence of 1 in 3000 or greater.
Yellow card system
The voluntary yellow card system has been the most productive to date in the UK for identifying important adverse drug reactions. Yellow reply-paid cards are supplied to doctors and dentists, who are encouraged to use them to report any suspected adverse drug reactions to the government’s advisory Committee on Safety of Medicines. Although the rate of reporting is low, this system has drawn attention to the association of oral contraceptives and thromboembolism, hepatitis and methyldopa, jaundice and halothane, and extrapyramidal effects and metoclopramide. At present, only this system is potentially capable of detecting risk at all levels of incidence. When suspicion has been aroused through the yellow card system, the existence or otherwise of a true association between a reported event and the implicated drug must be demonstrated epidemiologically by case control or cohort studies, and by clinical pharmacological and toxicological studies of the possible mechanisms involved. The problems associated with long-term surveillance of many thousands of patients must not be underestimated. Such studies are costly in both time and money, and it is difficult to maintain the integrity of the study cohort, the interest and commitment of the doctors, and the compliance of the patients.