‘The Politics of Clinical Trials’ Professor Stuart Blume (University of Amsterdam and Innovia Foundation) on the pervasive influence of social, political and economic contexts on research participation.
Norma asked me to talk about the big picture and I thought I ought to say a couple of words about my qualifications for doing so. You [addressing the Chairman] said some of them, but basically a very, very long time ago I was secretary of a government committee in the UK called the Black Committee, which studied social inequality in health. And thereafter I got interested in what kind of questions did new healthcare technology solve and what kind of questions did it create or not solve. Why didn’t it, you know, why didn’t it lead to reduced inequalities? And I started out on a series of studies of how health, where new health technologies come from, how they get introduced into practice and what they do, which has been keeping me busy for the last 30 years. Some of them indeed [are], nowadays, vaccines and health technology, people with disabilities.
Anyway, the big picture: since I've been away from the UK for 27 years my big picture is not specifically focused on the big British picture because I'm less acquainted with all the things that have been happening. I started out by thinking well, we’re going to be talking about the role of volunteers in medical research which is a part of the issue of recruitment, so I thought, well, let’s see what... and I knew that recruiting people for medical research and clinical trials, especially big phase III trials, was quite an issue and often it was difficult and when it was difficult things got slowed down. So I thought I’d start at looking at the literature on recruitment and clinical trials to see what people have been writing about it, because I didn’t know that much about it.
And I found, wow, unsurprisingly my little search with clinical trial and recruitment produced 2,000 publications, 60% from the US, unsurprisingly, and just a quick sort of idea of the sorts of things that people write about. I mean, I obviously didn’t do a systematic review and I didn’t read all the 2,000 papers; I just picked out a few of the more highly cited ones to get an idea of what was going on, what were all these 2,000 odd authors writing about.
Now, a lot of it is about disparities in participation, differences in rates of enrolment between men and women, younger and older patients, different ethnic groups, etc. There was another whole huge set of papers about recruitment strategies, I mean how you ought to enrol people, looking at the advantages and disadvantages of going through doctors and health centres, through newspaper or other advertising, through fliers or direct mail and all sorts of things.
Clearly, recruitment has become not only important, but so much an issue in itself that I found that quite a lot of papers publishedwere actually meta analyses of writing on recruitment strategies or they were prospective trials of different recruitment strategies to see ones that worked better. Some people tried to work out cost-effectiveness in terms of yield per unit of currency, to the extent that a colleague of mine, a sociologist of science called Steve Epstein, wrote a paper on what he called recruitmentology as though it were turning into a field in itself, certainly becoming professionalised and commercialised. And, of course, as you all know, there are now companies whose principal business it is to recruit trial participants, generally in poor countries with little tradition of ethical regulation.
Now, going back to the literature, there's obviously a big difference between comparing the demographic characteristics of typical volunteers and non-volunteers and in trying to understand what motivates participation. It’s obviously a methodological difference, but it’s more than that. In the one you can look at external characteristics of people; in the other you're obliged to go and talk to them and try to find out, you know, to understand participation or non-participation from their point of view.
It’s a bit like the sort of discussion – no, not discussion. There's a big difference because of course there are sorts of researchers who are not accustomed to going to talk to people and other sorts of researchers who are. There are many similarities, I thought, between the discussion about the writing about compliance and noncompliance and many sociologists finding the notion inappropriate.
So there are a few studies where people have gone and talked to people and asked about their reasons. The results of those studies are not particularly surprising. One meta study reviewed 400 articles dealing with recruitment to surgical trials and they found – it won't surprise you – they found that the main reason for non-volunteering or non-participation was that people had a preference as to which form of treatment they wanted and they didn’t like the idea of randomisation. It’s just something that keeps coming back in many of the studies; people apparently are unhappy with the idea of being randomly allocated, the possibility of being randomly allocated to a placebo arm of a trial.
One fascinating study looked at... there have been a number of studies in the US regarding involvement of African-Americans and one of them found, rather interestingly, that when asked what people wanted to know, what counted as, you know, when would they feel sufficiently informed, what would they like to know about the study, it turned out that many men but not many women wanted to know who was paying for the trial and who was likely to profit financially from it. Those authors conclude that maybe though there ought to be a notion of informed consent perhaps having to made gender specific.
[Chairman: Pink and blue paper as well.]
Age specific too. One of the most interesting studies I found was done in Denmark. You know, I haven't focused on UK literature because I don't live here. In Denmark they carried out qualitative interviews among women cancer patients who’d either accepted or declined to participate in one of three randomised trials. They found that most people understood that there was a moral dynamic, there was a degree of altruism involved, that people were conscious of the fact that it was because predecessors had agreed to participate in trials that they had the treatment options that they had. And they accepted that and altruism played a role, but it was not the most for participants it was not the dominating factor; it was the possibility of personal benefit that weighed most heavily. And here too people were unhappy with randomisation.
The point was that when entrusting themselves or their bodies to the medical profession, people are accustomed to and in fact do hope for the best available treatment, or rather treatment based on an assessment of their therapeutic needs. And it’s not really surprising that they find the idea of the treatment they're being given deriving from some notion of randomisation as problematic.
There were another lot of studies, not many, but another lot of studies about motivations of healthy volunteers where clearly the expectation of personal therapeutic benefit is not the principle motivating factor. The general view in these studies, in most – well, I can't say most of which because they didn’t look at it like that – but probably they were largely studies from the US. But the general conclusion of the ones I looked at was that financial benefit, financial incentive, played a major role, and that was seen as problematic. It’s the same sort of argument as arose in regard to payment for blood supply, for example; it could lead to particular sorts of people participating which could have implications for the validity of the conclusions.
So one interesting thing, there was a review of studies; one person carried out a systematic review of studies of healthy volunteer participation in a number of countries. So we did a kind of... or they did, a kind of a meta analysis of studies of healthy volunteer participation. We found 13 carried out in different countries, and indeed financial incentives were the most important. Altruistic motives played a part. Only one study was way different, and that was a study in Malawi where – and I’ll come back to this very briefly at the end, which won't be in very long. In the Malawian study the principle reason for participating or volunteering for trials was access to healthcare, and hold on to that.
Now, I gave this talk the title, The Politics of Participation and you may wonder, well, what's political about the things I've been talking about, the knowledge that one age group’s more likely to volunteer than another or that people don't like randomisation; it doesn’t seem to have much to do with politics, does it, and whether or not it makes sense to appeal to altruism in trying to encourage people to participate doesn’t seem so political either, perhaps a little ethical but not so political.
Now, to understand the politics of it all, we need to stand back a bit and think about the rules and conventions governing how research is conducted and the sources of people’s expectations and anxieties, and trust is crucial, it seems to me. What the studies that I've referred to suggest is that in most cases people, when considering participating in a trial or a study, look beyond the details of that specific study. Looking at the context from which their options arise, they may have doubts that come closer to our notion of the political.
For example, the Danish study that I referred to found that many, many people had serious doubts about the ability of the physician, the clinicians or the researchers to resist influence from the pharmaceutical industry. This was true of both participants and non-participants. And there was a feeling that financial interest could outweigh the interests of patients in the way in which the trial was organised or conducted or reported, and many people felt that there was perhaps need for a higher degree of regulation of trials. And that's getting closer to the political, it seems to me.
A very different study, again with African-Americans, found widespread mistrust of the medical profession and of medical researchers in particular, with constant reference to – I’m sure all of you know the infamous Tuskegee experiment, and often people remembered it wrongly but that doesn’t matter. The point is, there was a kind of communal mistrust, a sense that black people had been exploited for reasons of medical research.
Now, you may say, OK, we’re not, it’s different here, but the point is about trust and the fragility of trust. So one reason for speaking of the politics of patient participation has to do with the sorts of reasons that people give for being willing to participate or not, which go beyond the details of the study and, as it were, refer to the trust in those who are doing the study.
The other reason for talking about the politics of participation has to do with the way in which particularly in the US, but I think with implications for all of us, the organisation of clinical trials became a matter of explicit political debate and legislation. So, again, I’m sure many of you know that in the 1980s health advocates in the US, particularly people from the gay community, HIV positive people, argued that the procedures for approving drugs were too much concerned with safety and too little concerned with benefit. And the argument was, why should people who’ve got nothing to lose, whose lives are at stake, be denied a drug because of some statistical scruples of a few methodologists.
That led, as I’m sure many of you know, to a kind of loosening up of regulatory process and notions of fast track approvals and what that brought with it – and this is important for all of us, I think – the notion of surrogate endpoints. If you couldn’t wait long enough to see whether people lived longer, you needed surrogate endpoints. And it seems to me that deciding what appropriate surrogate endpoints might be, especially in interventions that are concerned with sustaining people in chronic conditions rather than saving their lives, that deciding what should be appropriate surrogate endpoints in a whole variety of interventions, for people with chronic conditions, are things that ought to have been negotiated with patient groups themselves.
I’m coming to the end. The second set of issues that became politically controversial and debated politically in the US, but which again certainly had ramifications for all of us in Europe, was about who should participate in trials. The tendency had been to restrict trial populations to people who were kind of reliable and the argument that came up was – again, I’m sure all of you know – in the 1990s in the US was that in order to have a reasonable chance of benefit, or an equal chance of benefiting from the results of an intervention, an experimental intervention, then women and minorities and then later children and the elderly, should be represented.
So it was no longer... I mean, the question of what an appropriate trial population should be and who should be in it became politicised and you could argue and you could think about, if we think about the groups that gained, as it were, the right to participate, ethnic minorities and women in particular, in the first instance, and clearly group constituencies with a political voice, one might think about which groups have not been able to gain the right to be represented.
There's an interesting book by a legal scholar, ethicist, called Rebecca Dresser that some of you may know, who discusses the arguments that were used on behalf of participation. And she talks about the way in which the well-known bioethical principles of autonomy, beneficence, etc, applied and were used. People argued for a right to participate, a right to share the benefit, but she criticises it on the grounds of their sense of benefit is likely to be too influenced by the health professions and by the media which convey often a rather biassed sense of what's likely to come out of much medical research.