Event Title
Improving the social value of clinical research
Start Date
26-6-2010 1:00 PM
End Date
26-6-2010 2:30 PM
Description
This presentation is part of the Social Values in Medical Research track.
“Most therapeutic trials are inadequately formulated, and this from the earliest stages of their conception”: in a landmark article published in 1967, statisticians Daniel Schwartz and Joseph Lellouch made this remarkable claim. They based their argument on a now-familiar distinction between explanatory and pragmatic clinical trials. Explanatory trials investigate hypotheses under ideal conditions while pragmatic trials investigate hypotheses under usual, or “real-world”, conditions. They suggested that medical researchers at the time were making two common errors. The first was that they believed they were conducting pragmatic trials when they were really conducting explanatory trials, and the second was that they failed to provide adequate ethical justification for the explanatory trials they ended up conducting.
One might be forgiven for thinking that while the inadequate formulation of therapeutic trials was a problem when Schwartz and Lellouch were writing in 1967—still the early years of clinical trials, relatively speaking—a more informed approach to the planning and design of trials prevails today. Unfortunately, this does not appear to be the case. The May, 2009 issue of the Journal of Clinical Epidemiology reprinted the famous article by Schwartz and Lellouch and contained a series of six back-and-forth response papers by two groups of highly respected clinical researchers, each of whom assessed the applicability of the historical argument to contemporary clinical research. Both groups agreed that researchers continue to make the same mistakes in the design of clinical trials that they made four decades ago.
Consider, for instance, eligibility criteria for trial participants, which is one of the factors determining the classification of a particular trial on the explanatory/pragmatic continuum. In 2007, Harriette Van Spall and colleagues conducted a systematic sampling review to determine the nature and extent of exclusion criteria in RCTs published in the top medical journals. On the basis of their analysis of articles published between 1994-2006 in the top nine medical journals, Van Spall and colleagues concluded that “RCTs published in major medical journals may exclude both large segments of the general population and also specific patient populations from the benefits of participation in clinical investigations.” For example, they found exclusions on the basis of age in 72.1% of the RCTs, sex or sex-related conditions (including pregnancy, lack of contraception and menopausal status) in 39.2% and co-morbidities in 81.3%. A significant majority of trials (84.1%) contained at least one exclusion criterion judged by the reviewers to be poorly justified.
This trend toward more exclusionary explanatory trials is more than simply odd, however; it is potentially harmful. In a widely cited article, Emanuel, Wendler and Grady identify seven requirements that must be met in order for clinical research on human subjects to be ethical. According to this account, the first question that should be asked during the ethical review of proposed research is whether the trial is socially valuable. The way this criterion is typically understood, a study investigating a “trivial” or “uninteresting” question, or that overlaps with already established research, is unethical because it does not promise any benefit, either to the subjects of the trial or to the general population. Principles of distributive justice, which raise concerns about appropriate use of social resources, support this criterion since wasteful spending has opportunity costs. Principles of social justice, which raise concerns about the inappropriate privileging of research that advances the interests of powerful social groups while neglecting others, also provide support for this criterion. This is because the requirement aims to protect the public good against, for instance, the good defined by powerful interests.
With this framework in mind, it is worth considering whether a general assessment might be made about the relative value of explanatory and pragmatic trials. Explanatory trials appear to investigate questions that are of indirect social value, since they don’t tend to yield results that are immediately applicable in clinical practice. Pragmatic trials have direct social value, by contrast, and so are more likely to be immediately applicable to practice. In this paper I argue that even if it isn’t the case that explanatory trials are uniformly unjustified (since they may have some, albeit indirect, social value), it may frequently be the case that pragmatic trials would be better justified than explanatory trials. In many cases, pragmatic trials would better fulfill the ethical requirement for socially valuable clinical research. Further, we may have reason to pursue the most valuable research rather than minimally valuable research because of necessary limitations in research resources, including a global shortage of research subjects.
One of the implications of this argument is that a burden of justification exists for those who would add restrictions to trials. Researchers should begin from the assumption that their trial will be designed to answer a practical question and enable decision-making in a particular context and then justify any idealizing elements added to the trial design. Greater reflection on the purpose of medical research is not just a matter for the ethics police in the form of various research ethics boards, funders, and other regulators, though some responsibilities certainly fall to these groups. Reflection on the purpose of research is part of the responsibility of the researchers themselves. Researchers need to be educated about the different purposes trials can serve so they can be in a position to provide justifications for their choice of research methods. Discussions about social value should be the starting point for clinical research.
Improving the social value of clinical research
This presentation is part of the Social Values in Medical Research track.
“Most therapeutic trials are inadequately formulated, and this from the earliest stages of their conception”: in a landmark article published in 1967, statisticians Daniel Schwartz and Joseph Lellouch made this remarkable claim. They based their argument on a now-familiar distinction between explanatory and pragmatic clinical trials. Explanatory trials investigate hypotheses under ideal conditions while pragmatic trials investigate hypotheses under usual, or “real-world”, conditions. They suggested that medical researchers at the time were making two common errors. The first was that they believed they were conducting pragmatic trials when they were really conducting explanatory trials, and the second was that they failed to provide adequate ethical justification for the explanatory trials they ended up conducting.
One might be forgiven for thinking that while the inadequate formulation of therapeutic trials was a problem when Schwartz and Lellouch were writing in 1967—still the early years of clinical trials, relatively speaking—a more informed approach to the planning and design of trials prevails today. Unfortunately, this does not appear to be the case. The May, 2009 issue of the Journal of Clinical Epidemiology reprinted the famous article by Schwartz and Lellouch and contained a series of six back-and-forth response papers by two groups of highly respected clinical researchers, each of whom assessed the applicability of the historical argument to contemporary clinical research. Both groups agreed that researchers continue to make the same mistakes in the design of clinical trials that they made four decades ago.
Consider, for instance, eligibility criteria for trial participants, which is one of the factors determining the classification of a particular trial on the explanatory/pragmatic continuum. In 2007, Harriette Van Spall and colleagues conducted a systematic sampling review to determine the nature and extent of exclusion criteria in RCTs published in the top medical journals. On the basis of their analysis of articles published between 1994-2006 in the top nine medical journals, Van Spall and colleagues concluded that “RCTs published in major medical journals may exclude both large segments of the general population and also specific patient populations from the benefits of participation in clinical investigations.” For example, they found exclusions on the basis of age in 72.1% of the RCTs, sex or sex-related conditions (including pregnancy, lack of contraception and menopausal status) in 39.2% and co-morbidities in 81.3%. A significant majority of trials (84.1%) contained at least one exclusion criterion judged by the reviewers to be poorly justified.
This trend toward more exclusionary explanatory trials is more than simply odd, however; it is potentially harmful. In a widely cited article, Emanuel, Wendler and Grady identify seven requirements that must be met in order for clinical research on human subjects to be ethical. According to this account, the first question that should be asked during the ethical review of proposed research is whether the trial is socially valuable. The way this criterion is typically understood, a study investigating a “trivial” or “uninteresting” question, or that overlaps with already established research, is unethical because it does not promise any benefit, either to the subjects of the trial or to the general population. Principles of distributive justice, which raise concerns about appropriate use of social resources, support this criterion since wasteful spending has opportunity costs. Principles of social justice, which raise concerns about the inappropriate privileging of research that advances the interests of powerful social groups while neglecting others, also provide support for this criterion. This is because the requirement aims to protect the public good against, for instance, the good defined by powerful interests.
With this framework in mind, it is worth considering whether a general assessment might be made about the relative value of explanatory and pragmatic trials. Explanatory trials appear to investigate questions that are of indirect social value, since they don’t tend to yield results that are immediately applicable in clinical practice. Pragmatic trials have direct social value, by contrast, and so are more likely to be immediately applicable to practice. In this paper I argue that even if it isn’t the case that explanatory trials are uniformly unjustified (since they may have some, albeit indirect, social value), it may frequently be the case that pragmatic trials would be better justified than explanatory trials. In many cases, pragmatic trials would better fulfill the ethical requirement for socially valuable clinical research. Further, we may have reason to pursue the most valuable research rather than minimally valuable research because of necessary limitations in research resources, including a global shortage of research subjects.
One of the implications of this argument is that a burden of justification exists for those who would add restrictions to trials. Researchers should begin from the assumption that their trial will be designed to answer a practical question and enable decision-making in a particular context and then justify any idealizing elements added to the trial design. Greater reflection on the purpose of medical research is not just a matter for the ethics police in the form of various research ethics boards, funders, and other regulators, though some responsibilities certainly fall to these groups. Reflection on the purpose of research is part of the responsibility of the researchers themselves. Researchers need to be educated about the different purposes trials can serve so they can be in a position to provide justifications for their choice of research methods. Discussions about social value should be the starting point for clinical research.