Estimation of psychotropic drugs secondary effects on the vigilance
p. 101-110
Résumés
Cet article donnera une vue d’ensemble des développements dans le domaine des psychotropes et de la conduite automobile, en insistant sur l’importance des règles méthodologiques pour la recherche expérimentale portant sur les effets des médicaments en fonction des altérations psychomotrices qu ’ils induisent.
Il sera fait état du rapport international d’experts stipulant que les chercheurs, dans le domaine de la psychopharmacologie humaine, ont abouti à un consensus sur les aspects méthodologiques les plus pertinents et sur une catégorisation des médicaments, en fonction des doses et des produits en trois catégories.
L’aboutissement ultime des recherches expérimentales sur la toxicité comportementale des drogues devrait être une base de connaissance pour la classification de ces drogues dont l’usage est risqué pour la conduite.
Un nouveau système européen de catégorisation, basé sur un large consensus parmi les psycho-pharmacologues, serait à même de permettre de classer un grand nombre de produits utilisés fréquemment et potentiellement dangereux, comme les tranquillisants, les hypnotiques, les antidépresseurs et les antihistaminiques.
L’avantage d’une nouvelle classification, basée sur les niveaux de posologie, leur formulation et leur durée (pour les hypnotiques), au-delà des systèmes actuels de catégorisation concernant les seuls composés, sera discuté avantageusement.
La réduction des risques qui résultent des réactions toxiques au plan comportemental est essentielle pour transmettre les connaissances psychopharmacologiques en direction des médecins prescripteurs.
La réalisation de cet objectif devrait aboutir, à l’issue d’une série d’actions concertées, à obtenir un engagement total des responsables des services de contrôle du médicament, de la santé publique et de la sécurité routière.
Il est utile de rappeler que l’usage des psychotropes se trouve impliqué dans des accidents de la route provoquant 5000 décès et de nombreux blessés avec un coût annuel de 7 milliards d’écus.
Ce constat est amplement suffisant pour nous conduire à identifier les besoins et promouvoir des solutions basées sur la recherche.
This paper will give a brief overview ofthe importance of methodological guidelines for experimental research on medicinal drugs affecting driving performance and the catégorisation of medicinal drugs according to the psychomotor impairment they cause. Reference will be mode to international expert surveys stipulating that researchers in the field of human psycho-pharmacology hâve reached consensus on the relevant methodological aspects and a dose-related categorisation into three categories.
The ultimate outcome of experimental research on behavioural toxicity of drugs should be a knowledge base for the classification of drugs hazardous to driving. A new European categorisation System based upon consensus among psychopharmacologists who were able to classify a number offrequently used potentially hazardous drugs such as tranquillisers, hypnotics, antidepressants and antihistamines will be discussed. The advantage ofthe new classification based on the dimensions of dose, formulation and duration for hypnotics) over existing Systems categorising compounds only, will be stipulated as well.
The last part ofthe paper highlights the implémentation and updating ofthe présent knowledge. To minimise the risk that results from behaviourally toxic reactions it is essential to convey psychopharmacological knowledge to the practising health care professionals. However, this implementation should follow after a sériés of concerted actions aiming at the total commitment of those responsible for Drug Regulatory Affairs, Public Health and Transport Safety. It is emphasised that 5,000 deaths, over 150,000 casualties and 50,000 hospital admissions in which drug use by drivers was a contributory factor, with an annual cost ofmore than 7 billion ECU to the European Community is more than enough to identify the needs and to promote research-based solutions.
Texte intégral
I – Introduction
1Road safety deaths and injuries have grown to dimensions that the European Community can no longer accept. The social cost of road accidents, i.e. 50,000 road deaths and 1.5 million injuries each year is over 70 billion ECU (Source: White Paper on Transport Policy, COM 92/494, European Commission). It is widely recognised that for a substantial number (30-500) of road users involved in fatal road accidents, alcohol use was attributable to the cause of the accident. Many drugs (prescription, over-the-counter and illicit) are also known to impair driving ability. Available data allow one to conclude that users of the most frequently prescribed sedative and minor tranquillisers are more (2-5 times) likely to be involved in traffic accidents than non-users. It is a fact that these medicines are used regularly by a significant proportion (20-25 %) of the adult population in some EC Member States, e.g. France, Belgium and Spain (Baker, et al. 1983). Some experts estimate that 30-50 % of these patients use their medication on a chronic basis without proper indication and sometimes take more than therapeutically advised. In other words they misuse or abuse the medicines. The ultimate outcome of this description of transport safety in relation to public health is the estimation that at least 10 % of all people killed or injured in road accidents did use some kind of psychotropic medication that could be considered attributable to the cause of the accident. This represents a direct cost each year to the EC society of at least 7 billion ECU.
II – Experimental research
2Several aspects of a drug’s driving impairing potential should be investigated before recommendations can be made for individual use. A drug’s hazard potential may vary depending on the dose, time after dosing, development of tolerance, the age and gender of the user, the user’s mental and physical State, interaction with other substances and the demand characteristics of the situation in which a person has to perform. Obviously, not all of these factors can be tested in one experiment. Therefore a research programme should be designed which provides ail the necessary information. It has already been suggested by the WHO in 1983 that new drugs should be tested first screened with a battery of relatively simple, but sensitive, behavioural tests, and where impairment is demonstrated, to proceed with progressively more complex tasks that approximate driving.
3The view of experts in the field of human psychopharmacology on methodological guidelines for experimental research on medicinal drugs affecting driving performance was investigated in a recent international survey (Vermeeren et al. 1993). For example, experts were asked to indicate which mental/behavioural functions and pharmacological effects of relevance for driving are absolutely necessary to détermine the impairing drug effects. Table 1 gives an overview of experts opinions according to the majority of experts responding in this survey.
Table 1. Mental/behavioural functions and pharmacological effects of relevance for driving that are absolutely necessary to measure drug effects on driving performance (Vermeeren et al. 1993)
Proportion of experts | Mental/behavioural function | Pharmacological effect |
>3/4 : | Divided attention Sustained attention/vigilance Continuons motor performance Discrète perceptual motor responses | Sedation, drowsiness |
> 2/3 : | Risk avoidance Perceptual disturbances | Visual disturbances |
> 1/2 : | Speed and accuracy of decision making Résistance to distraction Dynamic visual acuity Distance perception Movement perception Short-term spatial memory Multi-limb coordination | Fatigue caused by sleep disturbances Confusion Overactivation, euphoria Motor disturbances |
4The experts were unanimous in their opinion that it is absolutely essential to check for sédation or drowsiness. The next most important side effects relevant to driving safety were visual disturbances (79 %) and disturbances of perception (79 %), as well as sleep disturbances causing fatigue (67 %). Sixty-four percent indicated that confusion should be checked for and 56 % overactivation or euphoria. The latter being a problem for test design since 53 of the experts say that the same battery of tests could not measure both sédative and stimulating effects of drugs.
5A majority (90 %) of experts agreed that experimental studies aiming to establish a drug’s effects on driving should use subject samples representative of the driving population with respect to âge and gender. Young healthy male volunteers are not représentative of the whole driving population. Generalisation from this group to women and the elderly is problematic. It is necessary to study the effects of drugs in women and middle aged or elderly subjects. Assurance of reliable birth-control during the trial would be a prerequisite for participation of female subjects in a drug study according to nearly all experts (90 %). For elderly the majority of experts (78) recommended additional blood and urine chemistry and hæmatology examinations and 68 % felt that resting ECG should be included in the routine physical screening.
6Healthy volunteers are representative of actual users as they are unlikely to expérience different side effects than real patients. Furthermore, the drug’s therapeutic effect in patients is expected to attenuate the impairing side effect, thereby making the healthy volunteers a worst case example. The majority of experts (80 %) agreed that under most circumstances healthy volunteers studies are justified depending on the type of drug under study. Some drugs are used for treating disorders that probably have little or no influence on the driving ability of ambulant patients (e.g. H1- and H2-antagonists, non-opioid analgésies, NSAIDs, anticholinergics and anticholestero-lemics). Therapeutic relief would not be expected to improve patients’ performance, so the only effect the drug could have would be adverse. Patients treated with neuroleptics and antidepressants are often assumed to be impaired by the disorder itself. Predicting and distinguishing between therapeutic and impairing side-effects is difficult. Yet, the majority of experts believe healthy volunteer studies with these drugs are justifiable as preludes to patient studies. A spécial interest should be given to benzodiazepine users who have very different diagnoses and often no diagnosis at all. A large proportion of the users of these drugs are probably healthy in most respects, which make healthy volunteer studies justifiable.
7For new drugs administered in different doses across a therapeutic range, the majority (82 %) agreed that the study should involve at least two doses: the lowest and the highest dose that will be administered therapeutically, or, if there will be only one therapeutic dose, at least I x and 2 x this dose, given that the drug is well tolerated. A majority (53 %) went further to indicate that a dose 4 x the normal dose should be tested. According to 70 % of the experts, a frequent reason for single dose testing is budgetary constraints. Nearly half (48 %) also indicated that it might be due to the sponsor’s reluctance to discover a true drug effect.
8Drug effects on performance are often tested after only a single dose. However, drugs are usually taken for longer periods. Tests of acute effects are relevant, because patients will experience the initial phase of treatment. But those effects are expected to change over time due to accumulation, tolérance and the therapeutic response. Consequently, the majority of experts (82 %) agreed that studies should generally involve multiple dosing over treatment conditions lasting until the drug’s plasma concentration has achieved stead-state, or, until any acute effect bas stabilised at a lower level owing to the occurrence of tolerance.
9Control conditions are needed in every experiment for correct interprétations of results. It was agreed by a majority of experts (71 %) that three control conditions should be applied in every study: (1) placebo; (2) an active control from the same therapeutic class; and (3) a verum, or a standard référencé, not necessarily from the same therapeutic class as the drug under study. Among the experts, 40-50 % indicated a preference for éthanol sufficient to raise blood alcohol concentrations to 0.5 or 0.8 mg/ml as a verum. Of course the major problem with alcohol is that it is very difficult to use in double-blind design. An equally preferred (40 %) verum was diazepam 10 mg.
10Sample size is one of the major factors influencing the probability of a Type II error. The power of a test dépends largely on the sample size: the smaller the sample size the less likely it becomes that equal drug effects will lead to significant results. Most experts (88 %) agreed that sample sizes have generally been too small in previous research. The reasons they gave for the small sample sizes were mainly budgetary (83 %) and time constraints (70 %). They believed, however, that more than 18 subjects would be required in the case of cross-over design and parallel group design; and are deemed necessary for achieving a sufficient degree of statistical power.
11Besides a series of guidelines on selection of statistical tests, content and construct validity, test-retest reliability and the use of common scale, the progression towards test realism could be proposed and agreed upon by a majority (68 %) of experts and was stated as follows; studies establishing the driving hazard potential of a particular drug should proceed from conventional laboratory testing to the most realistic test that can safely be applied. The final evidence that therapeutic doses of a drug would be safe or hazardous to a spécifie degree can best be obtained in the latter situation and should be based on the combined results of ail tests in the programme.
12Tests that are to be included in a research programme should be validated before their results are used for formal categorisation of drugs as présumable safe, minor/moderately or severely impairing. The objective of the survey by Vermeeren et al. (1993) was to provide a preliminary set of guidelines, based upon consensus of scientific opinion, regarding methodology of experimental research. Next, the drug regulatory authorities need to set requirements for drug screening procedures with respect to behavioural toxicity.
13It is clear that information on the drug’s hazard potential on driving is badly needed in order to develop effective warning Systems for prescribing physicians, dispensing pharmacists and patients who have to take the medicine.
III – Present and future warning systems
14Information concerning the increased potential for risk in reducing driving ability while using hazardous drugs should be communicated to patients. This is obvious in all EC Member States where, according to legislation, it is a punishable offence to drive a motor vehicle when driving ability has been impaired by a medicine. After the introduction of package warning Systems in The Netherlands (1973) and the Nordic countries (1983), both based upon a list of drugs by substance name, the European Community’s Committee for Proprietary and Medicinal Products (CPMP) approved a 3-tier warning System in 1991. The first two Systems being chemically oriented yielded very general statements as pharmacological class warnings in patient drug information with a lack of information regarding the differential drug and dose effects.
15In the Dutch System, according to a national agreement between the professional organisations of physicians and pharmacists, any pharmacist filling a « prescription for a drug on the list dispenses the drug with a label bearing the warning: This medicine can influence your driving ability ». Originally this label (in yellow with the warning in black) was affixed separately, but as a resuit of pharmacy automation the warning now appears as integral part of the dispensing label with dosing instructions and other warnings for the patient. The last evaluation of the Dutch System was carried out in 1987 following a general information campaign concerning the influence of drugs on driving. A questionnaire study revealed that 30 % of the patients receiving a drug with a yellow-and-black label said they changed their behaviour towards driving (by not driving, driving less, etc.). The majority, however, did not pay any attention al ail.
16In the Nordic System a red triangle is printed on all packages of drugs within categories entitled ‘especially dangerous’ or ‘potentially dangerous’. The Nordic System has not been examined extensively. The most recent évaluation of this System was carried out as an interview study in 1987 in Sweden with a représentative sample of the Swedish population between 15-74 years of age and revealed that 63 % of the respondents were aware of the meaning of the warning Symbol.
17The new European System is based upon the détermination of the degree of psychomotor impairing effects of a drug/dose. This System was a resuit obtained in a world wide consultation of 45 experts who were able to rate the drugs they studied according to one of the following three categorisations (Wolschrijn, et al. 1991):
- presumed to be safe or unlikely to produce an effect;
- likely to produce minor or moderate effects;
- likely to produce severe adverse effects or presumed to be potentially dangerous.
18The enormous undertaking by the experts provided us with a catégorisation System, that received acceptance by the EC Drug Regulatory Authorities, but was based upon data concerning the acute effects of drugs. It lacked drug ratings about chronic drug treatment. The model was accepted, however, but the data need annual updating and completion pertaining to different drugs, dosages and conditions of treatment. Suggestions for label warning Systems and descriptions of the categories in patient package inserts were presented to the CAMP, but no significant response from DG III has been observed ever since.
19It is obvious that in the future one European warning System will be based upon the new drug categorisation and could be considered as an important contribution to traffic safety and public health. It is also clear that no single Member State will step forward in changing present national législation without an European Directive and that Concerted Action is needed to accomplish this.
IV – European Concerted Action Programme: Requirements and expected results
20The proposed Concerted Action Program for drugs and driving could be divided into seven topic areas with the respective following objectives:
- Harmonisation of research methodologies;
- Standardisation of drug registration procedures;
- Categorisation of drugs;
- Cost/benefit analysis of implementing drug catégorisation;
- Development of prescribing guidelines and warning Systems;
- Development of guidelines to define driving licence restrictions;
- Drug use evaluation and assessment of accident risk.
1. Harmonisation of research méthodologies
21Categorisation of drugs with respect to behavioural toxicity requires the application of valid performance tests within the context of well-designed experimental studies. Yet the relevant scientific discipline (i.e. experimental human psychopharmacology) has failed so far to develop and standardise the necessary methodology. Neither the desired test procedures nor circumstances covering their application have been defined. It is necessary to do so before the goal of categorising the impairing properties of drugs can begin. Methodological guidelines for experimental research on médicinal drugs affecting driving performance were recently published on the basis of an international expert survey, commissioned by the National Swedish Road Safety Office and the Netherlands’ Ministry of Transport and Public Works (De Gier & Laurell, 1992; Vermeeren et al. 1993).
2. Standardisation of drug registration procedures
22Drug manufacturers need criteria for standardised procedures in order to provide drug regulatory authorities with sufficient proof of absence or degree of drug impairment. Two further consensus meetings are proposed in order to achieve broader consensus among experimental investigators, drug regulatory authorities and pharmaceutical industry: firstly, to specifically focus upon the validity and practicality of the proposed methodological guidelines; and, secondly, to develop a System to apply research that follows these guidelines for the purpose of drug catégorisation.
3. Categorisation of drugs
23A sériés of annual consensus meetings, organised into several expert committees, each identified with the évaluation of therapeutic classes of drugs known or suspected to cause driving impairment is now needed. Experts in the field of human psychopharmacology, in particular drugs and driving research, should be invited to attend these meetings for review and rating of new drugs applying for Pan-European registration. The identification of a suitable data base for maintaining data on drug catégorisation will be part of preparatory activities. A workshop is needed to arrive at an intérim solution for meeting the EC’s stated requirement for research to categorise the behavioural toxicity for « old » drugs as well as those registered after 1 January 1992. It should be attended by experts in drug registration, human psychopharmacology and représentatives of pharmaceutical manufacturers.
4. Cost/benefit analysis of implementing drug catégorisation
24Economie implications of the implementation of drug categorisation are largely unknown. There is a need to estimate cost/benefit ratios of proposed warning Systems, based upon an algorithm specifically designed for this purpose and assuming the use of the least impairing drugs by drivers after their implémentation. Experts in économies, pharmacoepidemiology and pharmaceutical market analysis have to participate in workshops to provide data on the usage prevalence of different drugs, their respective costs and to estimate the contribution to économie loss through traffic accidents. The workshops will also focus on the cost saving by pharmaceutical manufacturers in conducting the minimum number of efficient, standardised studies on the behavioural toxicity of their products as compared with the present situation in which manufacturers are forced to support superfluous research to acquire a convincing body of evidence on the sheer basis of volume.
5. Development of prescribing guidelines and warning Systems
25All existing warning Systems fail to provide information regarding the differential drug and dose effects resulting in sceptical attitudes of prescribing physicians. Their prescribing practices hâve not been affected by guidelines for selecting a less impairing drug. Expert meetings will hâve to focus on the possible applications of guidelines for prescribing physicians and label warning Systems directed for patients. Workshops will be needed to understand the different attitudes in EC Member States with respect to législation and acceptance/resistance by national medical and pharmaceutical communities. Experts representing these areas, as well as représentatives of patient/consumer organisations will be invited to discuss, develop and propose final guidelines for health care providers and users of médicinal psychoactive products.
6. Development of guidelines to define driving licence restrictions
26Within the framework of EC-Directives driver licensing authorities hâve to meet their obligation for assuring applicants’ mental and physical fitness to drive in issuing or renewing driver licences. By identifying the behaviourally toxic drugs according to the catégorisation System, the patient and/or his physician need further information on how to react responsibly. The procedures to be followed might have an impact on the patient’s attitude to report his medicine use at moments he has to complete an application form. They will also affect the physicians attitude in notifying the responsible licensing authority on the patient’s inability to operate a vehicle safely. Experts in defming medical fitness to drive, medical ethics, in human psychopharmacology and driver licensing authorities need to formulate practical guidelines and recommendations for this purpose.
7. Drug use evaluation and assessment of accident risk
27Policy makers and responsible authorities in public health and traffic safety continuously point out that conclusive data to prove causative relationship between médicinal drug use and increased involvement in road accident are lacking. Traditional epidemiological studies including accident involved drivers, non-injured control groups and body fluid analysis to ascertain médication exposure will be more difficult to justify ; most costly and impractical. Therefore research strategies will have to be modified in finding other solutions to investigate the relationship between medication exposure and accident involvement. The use of drug prescribing and dispensing data from computerised Systems in medical and pharmaceutical practice, offer the opportunity to design pharmacoepidemiological studies, linking drug files of patients to data on hospital admissions and reported accident involvement. Concerted actions to initiale these studies are urgently needed. The results from these studies will influence policy makers, health care providers and patient/consumer advocacy groups in deciding on measures needed to minimise risks patients expérience as a resuit of behaviourally toxic drug reactions.
V – Conclusion
28A new drug categorisation System for médicinal drugs affecting psychomotor performance based upon expert consensus is feasible. However, the implementation of the System should follow after a sériés of expert meetings and workshops to cover different aspects of the new System and aiming at the total commitment of those responsible for EC Drug Regulatory Affairs, EC Public Health and EC Transport Safety. They must construct the framework needed for those who provide health care by prescribing and dispensing of drugs that are tested and evaluated according to well designed procedures resulting in practical guidelines for patients who receive the médications. Initiatives to provide the basis for an European Concerted Action Program for Drugs and Driving are recognised but not supported by those who guide policy decision making. Probably we need to show more than 5,000 road deaths, over 150,000 casualties and 50,000 hospital admissions each year with a social cost of more than 7 billion ECU in order to identify and promote research-based solutions throughout the European Community.
Bibliographie
Références
M. Balter, D. I. Mannheimer, G. Mellinger (1983), Cross-sectional comparisons of anti-anxiety/sedative drug use. Current Med. Res. Opin. 8 (Suppl. 4P) 5-20.
J. J. De Gier, H. Laurell (1992) Studies on man-machine interactions. Methodology in Man-Machine Interaction and Epidemiology on Drugs and Traffic Safety, Eds. Ferrara SD and Giogetti R. ARFI Research Monograph : 6 ; 33-50.
A. Vermeeren, J. J. De Gier, J. F. O’HAnlon (1993). Methodological guidelines for experimental research on médicinal drugs affecting driving performance : an international expert survey. Report IHP 93-27, Institute for Human Psychopharmacology, University of Limburg, Maastricht, The Netherlands.
Who (1983) Drugs, driving and traffic safety, WHO Offset Publication N° 78, World Health Organisation, Geneva.
H. Wolschrijn, J. J. De Gier, Smet Pagm de (1991). Drugs and driving : a new categorisation System for drugs affecting psychomotor performance. Institute for Drugs, Safety and Behaviour, University of Limburg, Maastricht, The Netherlands.
Auteur
Institute for Human Psychopharmacology, University of Limburg, Maastricht, The Netherlands
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Trajectoires familiales et espaces de vie en milieu urbain
Francine Dansereau et Yves Grafmeyer (dir.)
1998
Vigilance et transports
Aspects fondamentaux, dégradation et prévention
Michel Vallet et Salah Khardi (dir.)
1995