Effective use of medicines depends on the availability of reliable information. Where adverse reactions and interactions of drugs are concerned, existing systems of adverse drug reaction monitoring by health professionals are an important tool for collecting, collating and analysing data as a means of establishing new knowledge and generating early signals of possible drug complications not reported through clinical trials. Output from such adverse drug reporting systems complements the information appearing in the published literature and from other studies. However, for a number of reasons these systems have important deficiencies and limitations and they cannot stand alone. Adequate feedback is needed from users of medicines.

The following is an extract from a consensus document developed during the First International Conference on Consumer Reports on Medicines (CRM) held from 29 September to 1 October 2000 in Sigtuna, Sweden*. The meeting was attended by participants from the medical and pharmaceutical professions, drug regulatory authorities, consumer associations and the World Health Organization.

* The First International Conference on Consumer Reports on Medicines (CRM) was organized by the Consumer Institute for Medicines and Health (KILEN), Kammakargatan 7, SE-111 40 Stockholm Tel: +46 8-696 01 00. Fax:+46 8- 696 01 10. Website: www.kilen.org in collaboration with the Swedish International Development Cooperation Agency (SIDA), The National Board of Health and Welfare, Sweden, Dag Hammarskjold Foundation, Health Action International (HAI), People's Health Assembly, Uppsala Monitoring Centre, the Swedish Consumers Association and the Sigtuna Foundation.

There are both general and specific reasons to support the development of consumer reporting as an essential component of drug safety. It is also an important mechanism for consumer empowerment in health, which extends to processes such as regulation and clinical trials. Consumer reporting promotes:

1. Consumer rights and equity, including fairness since it is the patient who assumes the risks and the costs; the fundamental right to express oneself, to be listened to; and the right to informed consent.

2. Consumers have unique perspectives and experiences. They can provide information and insight which is crucial to advancing the effective and safe use of medicines. The sharing of information in a public reporting system also encourages consumers to help one another.

3. The medical establishment can profit from the discipline of consumer input since the standard of practice can only benefit from an ongoing and active encounter with consumer concerns. Consumers for their part need a willing ear and a channel through which to make their experiences and feelings known as they assume the risks and costs of using medicines. They are also exposed to the relationship between benefits, harm, and costs of drug use, and can themselves contribute to knowledge of that relationship.

Limitations of physician-based and pharmacist-based reporting

Adverse reaction reporting systems have evolved considerably since their creation and play a significant role in mobilizing data. However, they have certain serious limitations. Foremost among these is the fact that only a small proportion of practising physicians (often less than 5%) and a relatively small number of pharmacists (varying from country to country) contribute data. Consequently, information on suspected adverse reactions communicated by a patient to his physician stands a 95% chance of going no further and making no contribution to adverse reaction reporting.

There are various well-documented reasons why a health professional may fail to report to the system an adverse effect communicated to them by the patient. The effect may be sufficiently well-known already, or conversely it may be dismisses as being unlikely, vague or unrelated to the treatment. However it also seems clear that the prescriber may perceive risk and adverse events differently from the patient. While, for instance, the doctor is likely to accept the existence of any tangible or visible physical reaction (e.g. a rash), he or she is less likely to appreciate that a consumer's description of experiencing ill-defined malaise, unease or changes in mental state may equally constitute an early sign of drug complications. Some physicians appear particularly hesitant to report adverse reactions to a drug which they have prescribed for a non-approved indication ("off-label use"). In some countries, physicians have been more generally reluctant to submit adverse reaction reports to official systems because of a fear that, particularly where a reaction is serious, the prescriber may be blamed and accused of malpractice.

Above all, however, the physician who has no interest in communicating suspected adverse reactions to a central system will take no action. What may be significant information then remains buried in the patient records. It should be added that there is also a problem relating to underreporting by patients to physicians, especially where a patient is nervous about raising the issue. This situation is regrettable, but it exists, and some consumers in this situation would be less reluctant to report their experiences to a neutral consumerbased system. The prescriber could and should do more than is usually done to alert a patient to the need to report suspected adverse reactions, whatever reporting channel is used. This is particularly important where a new drug is prescribed, or an old drug is administered for a new or unusual indication.

A final problem with existing systems is that the input relates primarily to new drugs; reports of adverse experiences with products which have been marketed for many years are relatively few, despite the fact that new problems can emerge with such drugs even after decades of use and that old drugs may be redeveloped for new indications (e.g. thalidomide for treatment of acute lepromatous reaction). The establishment of direct consumerbased reporting of suspected unwanted events can compensate for these and other deficiencies in existing systems.

Benefits and challenges of consumer reporting

Some positive and direct benefits of direct consumer reporting can be identified.

1. There is evidence from the field that direct consumer reporting can result in an earlier accumulation of signals, some of which will point validly to new and hitherto overlooked reactions. In at least one comparative study, consumer reports have been found to cover a wider spectrum of organ systems and adverse reactions than reports entering a physician-based system.

2. Direct reporting can provide exceptionally informative, vivid and more complete accounts of unwanted experiences which do not readily lend themselves to classification in any rigid system. These may include mental reactions but also adverse changes in the quality of life which can be very important, real and distressing to the medicine user yet are unlikely to be clear to a prescriber.

3. Because consumer reporting can cover situations about which the physician is not informed and about which he or she therefore cannot report, it is bound to provide information which is not normally available to existing systems. Where, for example, a patient discontinues a drug because of an adverse effect, and does not return to the prescriber, the latter will have no knowledge of the event. Again, if patients experience a suspected reaction to a self-medicated drug, or to a traditional, herbal or alternative medicine which they have taken at their own initiative, no prescriber is likely to know about it. It is clear that in much of the world there is extensive self-treatment involving drugs which formally are restricted to prescription sale, and these include new and potent products with still poorly defined risks.

Similarly, if a patient is being treated by two physicians (e.g. a general practitioner and a specialist) the one may not be aware of drugs prescribed by the other, and adverse reactions or interactions will be missed. The onus of ensuring proper collaboration and communication in this matter clearly lies with physicians and the health system, but where it fails, a consumer-based system may well provide important input.

Present barriers to consumer participation

In addition to the above challenges, a series of reasons can be advanced as to why consumer experience so far has not been mobilized as a tool to develop new knowledge and monitor possible adverse reactions. The most obvious of these is the fact that in most countries, reports submitted directly by patients or their representatives are not accepted by the existing physician-based systems. They may be formally excluded, or those managing the systems may simply be reluctant to accept them.

In addition, however, consumers have not yet played an active role. Reasons for this include lack of knowledge regarding drugs and their reactions, a difficulty which is universal but especially severe in countries where lack of education and illiteracy are widespread. In some countries medical mysticism is still dominant, characterized by a desire to maintain secrecy around medical facts and thinking. Improved public education and information services, coupled with more modern attitudes on the part of professionals, can do much to remedy this.

The fact that drugs are often prescribed and used irrationally creates other difficulties, for example where many drugs are given for a single condition when only a single drug is indicated. In much of the world, the patient has only a brief interview with the physician before receiving a prescription or therapeutic advice, and there is neither an opportunity nor a positive tradition favouring information and advice about the drug that has been prescribed or recommended.

A particular problem in some developing countries is that a patient may not know with which drug he/she has been treated or which dose he/she has received. Not only are drugs sold under a multitude of names, but in some instances they are referred to anonymously ("take two of the blue tablets and three of the green ones every day") or dispensed without labelling (in sachets containing a few loose tablets, for example). In such situations, the active participation of users will be difficult to achieve for many years to come.

In industrialized countries, medicine users increasingly report their experiences on internet-discussion sites, or to patient and consumer organizations. These reports however tend to be left as anecdotal experiences. They are not taken seriously as consumer reporting. Even in countries where the above barriers are severe, a start can be made by mobilizing consumers in the more educated classes to take part in a limited venture which will subsequently extend into a wider movement.

A structure for consumer reporting

No single structure is likely to prove ideal for all countries. However, a number of firm impressions as to preferred structures can be recorded.

Responsible body

The scheme should be handled by an independent body at arm's length from government and industry as far as data gathering and interpretation of information are concerned. The body should be able to call upon independent experts, including pharmacists, pharmacologists and physicians with experience in assessing cause-effect relationships, but it should also have expertise in consumer affairs and patients' rights.

Reporting channels

Reports should be submitted through the most appropriate route, depending on the current situation in the country; the submission procedure should be as simple as possible. In some parts of the world the Internet provides a widely available means of reporting, whereas in others the mail, telephone or personal contact are more appropriate.

Form of reports

Existing physician-based systems have shown that it is helpful to receive the essence of a report in a standardized manner. Well-designed reporting forms should ideally be used. Reports received by telephone, letter or through personal contact should be transferred to such forms, but with retention of a complete description of the event which has been provided, e.g. by scanning a letter or recording the description given by telephone. The data as recorded on the standard form should be sufficiently exact to permit digital analysis, while a plain text description of the complication should also be provided since it can be valuable in recognizing unusual events not readily amenable to coding. As in professional reporting systems, great care must be taken to ensure anonymity so that personal medical data do not become common knowledge.

Terminology

As became clear at an early stage with professional reporting systems, there will be a need to standardize terminology, particularly bearing in mind the fact that information will arrive from different cultures and in different languages. It is possible that in this respect the system will be able to benefit from the work carried out by the Uppsala Monitoring Centre.

Examination

Every adverse reaction report received should be examined and considered. As with existing physician- based systems, it will be important to have adequate information about the user, the treatment details, timing and course of the reaction itself. A causality analysis should then be attempted, recognizing that there is no exact methodology to this end. An attempt should be made to determine whether the report is novel or whether it has previously been described in existing systems or in the literature. The methods of analysis will not differ greatly from those developed in the best existing professional-based systems, but a consumer perspective ("how was this event experienced by the user?") should be included.

Direct feedback

Experience with existing adverse drug reaction reporting systems shows the great importance of acknowledging promptly the receipt of every report, and following this up in due course with a full reply and, where possible, an explanation and reassurance.

Information to others

The individual report is essentially confidential and under no circumstances should the name of the reporter, or any data on the basis of which the reporter might be identified, be transmitted to other parties. However, general information on the nature of reports received on a particular drug should be made available, together with the relevant assessment, to all interested parties. These are likely to include the national professional- based adverse reaction reporting scheme, the official drug regulatory authorities, the health professions, agencies responsible for rational drug use policies and programmes, the manufacturer, consumer and patient organizations and others likely to need the data. The media should have full access to the analysed data.

When providing information, disclaimers similar to those employed by existing spontaneous reporting systems should be added, so as to reduce any risk of over-interpretation or unnecessary alarm. In many instances periodic reporting on data input (e.g. quarterly) will be adequate, but it will be appropriate to disseminate groups of related reports that suggest the emergence of a hitherto unrecognized problem more urgently.

Collaboration

At the very least, there should be an exchange of data between public reporting schemes in different countries to strengthen systems and a further acceleration in the identification of new data. There is strong interest in establishing a global centre for consumer reports, analogous to that existing for professional reporting to The Uppsala Monitoring Centre but separate from the latter. The Uppsala Centre is, however, open for an input of data from the consumer system, which could form a separate data base. The Centre currently holds the data from 59 national physician-based schemes and provides reports and analyses of this worldwide material, as well as conducting methodological research. Its experience and advice could be of great value in developing consumer systems.

Action required

A careful distinction must be made between a neutral system for collecting, analysing and disseminating data and an institution taking positive action in the light of emergent findings. While, as noted above, the system can and should bring its findings to the attention of other parties and agencies, it is likely to gain greatest respect if it concentrates single-mindedly on the most balanced possible analysis of data and the generation of signals. It would be more appropriate for bodies primarily concerned with the active promotion of patient and consumer interests to take whatever action seems justified in the light of such findings.

Hospital reports

Appropriate arrangements may be made to enable hospitalized patients to report directly to the proposed scheme when they experience the need to do so, either during their hospital stay or later.

Reports from trials

It would be very desirable for patients and volunteers taking part in clinical drug trials to report adverse reactions when they so wish to a central system. It is well known that many adverse reactions occurring in trials provide early pointers to drug injury, and that the data in question sometimes fail to become generally known either in a timely manner or at all. In particular, adverse reactions experienced by subjects dropping out of a trial may be lost to the records despite the fact that these events may provide the first indications of a serious problem. The principle of the subject's freedom to report should be laid down in guidelines such as those for Good Clinical Practice.

The same principles should be incorporated in clinical trial protocols subject to approval by ethics committees and analogous bodies. Confidentiality agreements concluded between an industrial sponsor and a clinical investigator should never be allowed to impede reporting of adverse reactions by participants.

Active reporting

Over the years, there have been several successful ventures into "intensive monitoring" or "active reporting", seeking to obtain integral registration of all adverse events from a particular patient group or at a particular location. Such efforts are justified, where there is a special reason to anticipate certain problems and to quantify them within a short period.

System monitoring

The functioning of all aspects of a system of this type should be monitored internally and its usefulness should be evaluated periodically so that modifications can be introduced as necessary. Field trials of modified procedures may be called for.

Secondary services

It seems clear from experience that once consumers have established contact with the proposed system they will seek supplementary services from it, in particular requesting advice and information. While the system should refer requests for medical advice to the most appropriate body (generally the patient's own physician) sympathetic assistance should be given where factual information or reassurance is sought. Such information may also need to be communicated to other bodies, especially where it points to a problem which, though it may not concern adverse reactions, demands attention and correction. Finally, much information regarding adverse drug reactions is now to be found on the Internet, although it is scattered and of variable reliability. Consumer reporting systems could in due course perform a useful role by periodically sampling the relevant content of the Internet and disseminating selected data that appear significant.

Pitfalls and problems

Past reasons for rejecting the principle of direct consumer reporting need to be borne in mind since they can provide pointers to necessary action. The opinion that information emanating directly from consumers is less reliable than that communicated by way of physicians (or pharmacists) is only valid insofar as the professional reporter may provide preliminary screening and useful background information when submitting the report. Provided there is proper feedback and careful expert analysis of reports, it will not necessarily be more difficult to eliminate mischievous, misleading or ambiguous reports than has been the case in professional-based systems, where entirely similar problems arise. A problem deserving particular attention, especially in international collaboration between systems, is that of translation and interpretation. Concepts regarding drug safety and subjective experiences (e.g. concerning the quality of life) may prove difficult to translate from one language to another, and plain text accounts of experiences may simply be difficult to comprehend when read in a different environment. These are intercultural as well as linguistic challenges.