Defining compensable injury in biomedical research.

Author:Larkin, Megan E.
Position:III. Using Ethical Justifications to Define 'Compensable Injury' through Conclusion, with footnotes, p. 353-382

  1. Causation

    All justifications for compensation of research related injury incorporate a requirement of causation. A duty of risk-minimization based in nonmaleficence requires compensation only for those injuries that foreseeably arise as part of the research process. (267) The justification for this approach is that the duty bearers within the research enterprise cannot be held responsible for risks outside of those that they have an obligation to minimize. (268) This requires that compensation be limited to those injuries that are caused by the subject's participation in research. (269) Similarly, a duty based in distributive and compensatory justice requires compensation for those injuries occasioned by the research participant's decision to participate in biomedical research, thereby foregoing the standard of care. (270) It is this decision that produces the benefits of medical research and scientific progress that are received by both the research enterprise and society as a whole. (271) Injuries that can be traced back to that decision should be compensated as research related injuries. Finally, the duty of reparative justice requires compensation for those injuries that have been caused by the negligence or malfeasance of someone within the research enterprise. (272)

    The path that causation must tread in these cases is analogous to that required by courts in evaluating claims of negligence. (273) As illustrated above, different international systems of compensation require different proof of causation when it comes to research related injuries. In order to more clearly determine what it means for an injury to be caused by research, it is necessary to look at theories of causation more broadly and to examine causation in the research context. In particular, causation cannot be isolated from an analysis of whether participants received the appropriate standard of care. In so-called therapeutic research, it is often difficult to determine whether a patient-subject's worsening condition is the result of a research related injury or of the natural progression of the individual's underlying disease state. (274) These considerations must be incorporated into a comprehensive system for determining which injuries can justifiably be said to have been caused by the research.

    1. A Theory of Causation

      The concept of causation is deceptively difficult. Although the language that is used when discussing causation is both intuitive and intended to be interpreted by lay people, an analysis of causation as it is used in the law leads to complex questions about both the origin and the attribution of events. At this point, it is useful to draw upon the work of H.L.A. Hart and Tony Honore, who have undertaken a thorough examination of causation in the law. (275) Their analysis of causation aims to explicate what separates a cause from a mere condition of the occurrence of an event, and it also explores the conditions under which the determination that an event is the cause of another leads to an attribution of responsibility for the consequences of that event.

      1. Causes and Conditions

        Causes, as they are used in the legal or historical lexicon, must be distinguished from the causal concepts used by scientists and philosophers. (270) In general, lawyers are concerned with the attribution of responsibility for the consequences of an action. (277) Hart and Honore draw upon the work of Hume and Mill to conclude that for any given occurrence, a number of factors are necessary to produce the ultimate effect. (278) They therefore argue that "when we identify single events as causes it appears that we choose one element from such a set, although each of the members of the set is equally required for the production of the effect." (279) A classic example is the case of determining the cause of a fire when a lighted match was dropped into a wastepaper basket. It is intuitive to say that the lighted match caused the fire, even though the presence of oxygen was also a necessary background condition. (280)

        The natural question presented is, therefore, what separates an element that is considered to be a "cause" from other factors necessary to produce the effect? (281) One explanation is that a cause "is an interference in the natural course of events which makes a difference in the way these develop." (282) This definition is particularly useful for "simple cases" where an individual produces "some desired effect by manipulation of an object in [the] environment." (283)

        One important feature of this approach is that it takes into account the normal background conditions in which a cause operates. It rests on a "presumption, normally fulfilled but rebuttable, that when we deliberately intervene in nature to bring about effects which in fact supervene, no other explanation of their occurrence is to be found." (284) Normal conditions are those that are present both in the ordinary functioning of the item or process in question, and in the case of an accident or mishap. (285) They are therefore ruled out as a cause of any particular accident. (286) "[N]ormal conditions (and hence in causal inquiries mere conditions) are those conditions which are present as part of the usual state or mode of operation of the thing under inquiry." (287) For example, when considering what caused a train accident, a bent rail is likely to be considered a cause, because it, rather than the speed and weight of the train operating as usual, made the difference between the train's normal functioning and its derailment. (288) This is the case even though the crash would not have occurred as it did if the train were not traveling at that speed or if it did not have the same weight. (289)

        Causal definition in this manner is particularly useful in the research context. To begin with, it supports the proposition that participation in research must be a cause in fact, or sine qua non, of a research related injury. (290) It also tells us, however, that the inquiry must not stop there. To say that an injury would not have occurred but for the subject's participation in research only identifies research as one of the background conditions that was necessary for the injury to occur. It does not pluck it out from among those conditions and identify it as a cause of the injury. (291) This definition, without more, would not rule out compensation for injuries that, for example, occur due to traffic accidents on the way to the research clinic. (292) In order to identify research participation as the cause of the injury, it is therefore necessary to determine what research activity has caused a deviation from the normal conditions that otherwise would have prevailed.

        This definition excludes the natural progression of a patient/participant's illness as a cause of a research related injury if the illness would have progressed regardless of the individual's participation in research. For example, imagine the case of an individual with advanced cancer, for whom all standard treatments have failed. This individual's prognosis is poor, and death is likely to occur. Because death is the expected outcome for this participant, due to the natural progression of his or her disease, failure of a clinical trial intervention to avert that outcome is not a change in the normal operating conditions that will make a difference in the way those conditions develop. The search for a cause, as distinguished from a mere condition for the occurrence of an injury, should also guide the search for an explanation for any given injury.

        What counts as a normal condition for a particular occurrence may often be the product of human habit, custom, or convention. (293) This observation is of particular importance in the medical context because, with the knowledge that some conditions are harmful in the absence of human intervention and are commonly ameliorated with appropriate medical attention, it becomes the normal state of affairs to intervene, and the lack of intervention is therefore a deviation from the normal course of events. (294) An example is helpful to understand how this principle operates in practice. Imagine that Ms. Smith has been diagnosed with pneumonia. In the ordinary course of events, it is known that prompt treatment with antibiotics is generally curative, and that in the absence of treatment, the pneumonia may become significantly worse. (295) In this case, it could be said that the failure to treat Ms. Smith's pneumonia was the cause of the advancement of her disease and her ultimate death.

      2. Causes and Human Action

        In identifying a cause of an event, special importance is given to voluntary human action. It "is often regarded both as a limit [on how far back we trace the chain of causation] and also as still a cause even though other later abnormal occurrences have provisionally been recognized as causes." (296) An ordinary causal inquiry, in which we are seeking to explain an occurrence, will not ordinarily be traced "through a deliberate act." (297) Abnormal consequences that follow a deliberate act are also likely to be regarded as mere means by which the act reaches its natural end. (298) A deliberate human act is therefore a natural boundary for the causal inquiry. We do not trace the path of a later event back through it, and we pass through intermediate causes of other kinds on our way to it. (299)

        Our inquiry thus far should lead to the conclusion that a cause will differ from other necessary circumstances for the occurrence of a later event because it will be an abnormal condition, and it will often be brought about by a deliberate human action. A cause, once identified, should also be unproblematically connected to the consequences or effect it produced. (300) This means that it should fit within the broad generalizations about what types of action produce what types of effects, and it should...

To continue reading