Harmonising adverse drug reaction terminology: the role of the Council for International Organizations of Medical Sciences.

Health professionals from different countries are known to differ considerably in their use of medical terminology, including the terminology used for adverse drug reactions (ADRs) and in the exact meanings attributed to terms. To remedy this situation, the Council for International Organizations of Medical Sciences (CIOMS) has attempted to provide definitions and basic requirements for proper use of ADR terms. The work has concentrated on terms liable to be misinterpreted and those used for serious and frequently reported ADRs. For every selected term a short monograph has been prepared. It consists of a preamble, definition, basic requirements for use of the term and additional comments, if any. In cooperation with medical experts, drug regulators and the pharmaceutical industry, 13 papers have been published so far. Approximately 160 terms have been defined and work on another 50 terms continues. The full collection of monographs will eventually appear in the form of a book and CD-ROM intended to help doctors fill in case reports, and regulatory agencies and the pharmaceutical industry assess reports. Pharmaceutical companies receive numerous reports of suspected ADRs from medical practitioners and other prescribing professionals. Each company is required to transmit these reports to the drug regulatory agency of the country, or countries, in which the drug is used. Therefore, in addition to receiving the correct name of the ADR, collecting and evaluating centres, regardless of whether they are part of a regulatory agency or a pharmaceutical company, need to be provided with sufficient supporting data to be convinced that what is reported was what was actually observed, and that the ADR term used represents the observed event.

Establishing requirements for the use of terms for reporting adverse drug reactions.

The establishment of a project for the standardization of international reporting of adverse drug reactions (ADR), sponsored by the Council for International Organizations of Medical Sciences is described. The need to establish requirements for properly diagnosing a given suspected ADR and to describe it with the correct terms is demonstrated. However in view of the difficulties involved, it is suggested that the present deliberate approach needs to be reconsidered.

Recognition and assessment of adverse drug reactions in children.

Recognition of a drug safety problem is connected with the perception of severity, rare occurrence of the suspected event in non-drug related pathology, knowledge of such drug-ADR association occurring and clearness of the clinical picture. With regard to children knowledge of differences between adult and child is of further help in particular in prevailing pathologies. A lesser variety of pathologies means a lesser variety of drugs used, what should facilitate recognition of an ADR. In the process of assessment the question of causality arises. In epidemiological studies the higher incidence of the event in the treated group suggests causal relationship in general probability terms. With individual cases it is done on the grounds of the strength of evidence provided in each case. The assessment of causality is done either by experts according to their knowledge or with a method of standardized assessment. A true expert does not need a method of standardized assessment but a less experienced evaluator would benefit from such a method or at least using it in addition to his unstructured judgement would benefit from a sort of a double-check. (Tab. 5, Fig. 2, Ref. 6.)

Potential role of postmarketing research.

To consider beneficial and adverse effects during one and the same postmarketing study seems to be perfectly logical, but is difficult to accomplish from a methodological point of view. This is caused by the particular conditions and design of clinical trials in Phases I-III as opposed to those in the postmarketing phase. Postmarketing research becomes to a large extent an activity independent from the pharmaceutical company, yielding information about groups of patients, concomitant therapies and many other factors that were only partly considered in Phases I-III. The heterogeneity and volume of data produced after marketing, requires for their intelligent integration and interpretation, an effort to assure a minimum quantity of data in every case report or in every study and of acceptable quality. Another requirement is the wider use of generally accepted terminologies, criteria for categorization etc. Without these prerequisites a lot of information is being lost or risks being misinterpreted. More attention to these factors would considerably improve the cost-efficiency of postmarketing research.

[Drug surveillance throughout the world]. / La pharmacovigilance dans le monde.

There are several systems of international drug surveillance, such as the W.H.O. system, the "Boston University drug epidemiology unit" in the U.S.A. and a similar system used in Switzerland, or the "drug surveillance research unit" in Great Britain. These different centres are described, and the problems of international drug surveillance are discussed.

Updating of a method for causality assessment of adverse drug reactions.

The paper represents an update of a method of standardized assessment of causality that was published in 1980. Because of experience that was gained in the meantime, weighting of certain types of information was modified and several weighting tables were introduced to better reflect the influence of type of adverse reaction in the process of evaluation. Comparison of results with the described approach with opinion of medical evaluators is given and discussed.

The practicing physician as generator and user of adverse reaction data.

Information about adverse drug reactions (ADR) helps the practicing physician to optimize his pharmacotherapy, however, he also generates most of the original observations and thus he is largely responsible for the quality of ADR data. As direct communication from physician to physician is limited other means had to be developed such as scientific papers, communication from health authorities and from the pharmaceutical industry. The paper discusses the characteristics of the elements involved in the process of generating assessing and disseminating ADR data and of their interplay.

Phase IV research by pharmaceutical companies.

The paper reviews the objectives and methods of phase IV studies performed by a pharmaceutical company with psychoactive drugs. After the introduction of a new drug additional information is needed especially regarding the safety of the drug. The term Post-Marketing Surveillance (PMS) covers all the methods used to collect this tolerability data. They include spontaneous monitoring by physicians, monitored release and related techniques, intensive monitoring typically in a hospital setting, case-control studies and cohort studies. Typical examples of new patient populations which are not usually studied extensively during the premarketing phases of drug development and which therefore need to be studied in more detail in phase IV are children and the elderly. In both of these groups certain special aspects have to be taken into consideration when planning and conducting clinical trials. Extension of the drug to new indications often starts spontaneously as observations of alert clinicians which then lead to systematic controlled trials. The methods used in the monitoring of physicians' prescription habits include the use of available commercial statistics, ad hoc physician surveys and the collecting of information through the medical representatives of the company.

Aspects of standardization as applied to the assessment of drug-event associations.

Individual case reports about suspected adverse drug reactions are an important source of drug safety data. Their evaluation involves, among other things, the assessment of causality. Usually it is done by expert medical evaluators according to their knowledge and experience. Subjectivity plays a big role as demonstrated by comparison of assessments of the same cases by several evaluators or by the same evaluator at a later date. In recent years, methods of causality assessment based on algorithms were introduced. In these formal approaches, items of information contribute to the assessment in a predetermined and standardized way with a logic that allows operational application. Results obtained with different methods may differ because items of information considered may not be the same and/or strength of evidence attached to them vary from one method to another. The main advantages of standardized assessment are: clear identification of items of information involved, improved communication, reproducibility of results, and the checklist function. Various implications of standardized assessment are discussed and ways to improve their performance outlined.

Adverse drug reactions--again.

The allowance made by the medical profession, the health authorities, or the public for the inevitable chance that the author of information may be wrong is much lower in the case of issues relative to drug safety than for other issues with comparable safety components. Recognition of the fact that true knowledge requires several studies so that various errors and biases can be eliminated would help to dedramatize the issue, and this would benefit everybody. Since no single method or system of monitoring will solve all the problems, proceeding through a sequence of objectively oriented approaches is necessary. Such an approach more optimally manifests the limitation of each method used and helps to interpret the results. An organizational structure that permits closer cooperation between all interested parties is necessary. It should not be left in the hands of any single body, and medical and methodologic audits should check its activities. Moreover, information should be more easily accessible to interested parties than is now the case with most monitoring systems.

How good are articles on adverse drug reactions?

A study was carried out of the quality and completeness of articles on adverse drug reactions: 5737 articles from 80 countries published between 1972 and 1979 were studied. Only 61% of the articles included information on the number of patients treated and the number with adverse drug reactions, yet these are essential for calculating the incidence of adverse reactions. In only 55% could the incidence of a particular adverse reaction be calculated. Great importance is placed on articles on adverse reactions, particularly those that report on many patients. Authors and editors should ensure that articles include the following information: drug regimens, numbers of patients treated, numbers of patients developing adverse reactions, and nature and incidence of adverse reactions.

Standardized assessment of drug-adverse reaction associations--rationale and experience.

The assessment of the causal relationship between the suspected drug and the adverse event represents one of the most important goals in the evaluation of individual case reports. Because of the variability of clinical judgement, there is a growing interest in systems of assessment based on clearly defined, operational criteria. In this paper we describe our approach toward this goal, the results achieved and the practical experience gained in the evaluation of over 600 case reports processed by the drug monitoring unit of our company.