Probabilistic information and decision making in the health context : The package leaflet as basis for informed consent
Thèse de doctorat : Università della Svizzera italiana, 2007 ; 2007COM009.
Medical decisions are paradigmatically uncertain. Awareness of this uncertainty at the policy level has produced a strict regulation of the information exchange related to medical issues, e.g. by means of the institute of the informed consent or through tight administrative norms regulating the pharmaceutical communication towards the public. As a special support of such sort of information,... PlusAjouter à la liste personnelle
- Medical decisions are paradigmatically uncertain. Awareness of this uncertainty at the policy level has produced a strict regulation of the information exchange related to medical issues, e.g. by means of the institute of the informed consent or through tight administrative norms regulating the pharmaceutical communication towards the public. As a special support of such sort of information, also the drug package leaflet has been object of thorough legal regulation, especially developed in the European and German legislation. However, precisely in Germany, recent court decisions concerning damage compensation for information faults have delivered contradictory judgments in relation to package leaflet information and the patient’s contributory negligence. Court argumentations in these cases range from the appeal to the patient’s duty of taking notice of the PL’s content to general observations concerning the irreplaceable role of doctor’s information. These observations are also backed by considerations regarding PLs’ information design and layout, which, in despite of great changes in the course of their history, are still object of improvement efforts (see also the recent conferences at the German Federal Institute for Drugs and Medical Devices – BfArM – devoted to the analysis and testing of PL readability: 15th February 2006 and 5th September 2007). Moreover PL information is blamed by health professionals of hindering compliance. Indeed, empirical findings emerging from psycho- sociological literature hint at contradictory attitudes in relation to health risk information in general: desire for this sort of information is not monotonically dependent on perceived uncertainty and present research falls short of providing coherent models of health information seeking behavior, which is expected to influence information processing and eventually lead to a decision change (e.g. therapy interruption). Apart from few exceptions (Koyuncu, 2005, 2006, Schlund, 1999) the legal literature on package leaflet fails to adequately address the disclaiming function of package leaflets and their role in the lay therapeutic choice. This may be partly explained by the fact that PLs have been developed within safety regulation, in which PL information is principally called to warrant for correct and safe use (Hart, 2003 a.o.), and also by the fact that liability for adequate information mainly concerns the doctor. By tacking into account the liability constraints determining the institutional functions of PL information, the dissertation investigates the role of PL information within the lay therapeutic choice and outlines the articulation of the responsibility spheres among doctor, pharmaceutical firm and patient as far as the therapeutic risk is concerned. The main conclusion resulting from my analysis is that legal function of PL information is not exhausted by safety purposes: as a privileged form of product instruction, it is also subject to liability regulation with a consequent reallocation of damage responsibility through risk disclosure. Therefore two main legal-communicative functions are identified for PL information: 1. Warning and risk prevention function, in observance of the consumer’s right to safety; 2. Disclosure of residual risk, in observance of the constitutionally protected right to self-determination. Safety information should prevent avoidable risk to occur; instead self-determination information should declare the unavoidable (residual) risk, so as to insure that the drug user is aware of it when he decides whether to undertake the therapy or not. The notion of residual risk is fundamental in this setting, because it is the risk which the beneficiary party needs to shoulder unless he has not previously been informed about it by the doctor and/or by the pharmaceutical firm through the PL. Once it has been established that PL information is not only a means of safety protection, but also part of the therapeutic informed consent, the point is then to evaluate whether PL information can be considered adequate for consent to be “informed”, i.e. valid. Because, from a normative perspective, information prior to consent should enable the patient to make an informed choice, the role of PL information has been evaluated on the basis of its contribution to epistemic accuracy within the therapeutic choice. By drawing on Andersson and Littkens’ two components model of individual health decision (2003), consent has been defined to be informed to the extent that it approaches a decision under risk, i.e. to the extent that the decision maker can assign a probability measure to each relevant health state and that he knows whether the act of taking the drug (a) shifts the probabilistic distribution towards stochastic dominance with respect to the act of not taking the drug. The model represents in fact the health decision as composed both by states over which the agent can assign probabilities and states shrouded by total ignorance. A complementary weight factor allows representing the proportion of ignorance and risk in the choice at hand. Against the framework of this model, PL information seems to provide the drug consumer with data whose personal relevance, and therefore prognostic value for the individual, is difficult to assess. This means that consent on the basis of PL information rather approaches a decision under ignorance than one under risk: as a prognostic device, PL information fundamentally asks more questions than it answers. Finally, a quantitative study (of limited sample size: n = 55) and a qualitative research have been carried out with the aim to provide a descriptive account of PL information as basis for the therapeutic decision.