Discussion

Research participants’ motivation for direct personal benefit1 ,4 and the involvement of their own doctors as researchers are perhaps the two most discussed potential sources of the TM.1 ,7 In our study, whether one had primarily a personal benefit motivation versus an altruistic or dual motivation was not associated with TM-related phenomena. The two groups’ understanding of the purpose of research was similar. Although 5 (9.3%) in the personal benefit group incorrectly stated that the study was primarily intended to help those participating, those five subjects’ specific understanding of arm assignment, randomisation probability, rationale for sham arm, and difference in procedures between two arms was excellent, with only one instance of an incorrect answer and one ‘not sure’ answer (out of 20 opportunities, ie, 5 subjects×4 questions). The direct personal benefit group and the altruism/dual motivation group showed no differences in understanding of various research design elements, in perception of direct benefit, and in recollection of research risks. Both groups were also quite optimistic in their replies to likelihood of personal benefit, but the overwhelming majority in both groups said this was not based on what they were told by the researchers or the informed consent form; instead, the participants cited as bases for their views of potential benefits a variety of personal interpretations and reasons.

The altruism/dual motivation group was more likely to perceive no or very low chance of risk (82% vs 60% in direct personal benefit group). Thus, being motivated solely by desire for therapeutic benefit does not seem to lower or dampen perception of risk. In fact, it may be that therapeutic motivation as part of an overall regard for one's welfare may, rather than blinding one to the dangers of research (out of desperation), make someone more likely to pay attention to the risks of research as a matter of guarding or promoting one's own welfare.

What was the effect of having one's own doctor as the site investigator? One might expect persons whose own doctors were also the site investigators to have more of a therapeutic motivation, since they would have been more likely to see the research participation as receiving treatment from their own doctor. This was not the case. Instead, although the difference (51% of own doctor group had primarily therapeutic motivation vs 68% of those whose site PI was not their own doctor) was not statistically significant, the direction of the effect was opposite of expected. None of the five subjects who stated that the research was primarily intended to help the subjects of the study were in the own doctor group.

Further, the doctor status variable was not associated with differences in understanding of procedures or in perceptions and expectations of benefits and recollection of risks. The own doctor group had a higher proportion (81% vs 60%) of subjects who said there was no or very low chance of adverse events.

These results regarding the doctor status are generally not consistent with the traditional TM interpretation which says that having one's own doctor as the site investigator will blur the line between research and treatment, leading to TM. Instead, it appears that those who do not have their own doctor as the site investigator (ie, those who seek out research opportunities of their own initiative) are more likely to be concerned about potential direct benefit from research (and perhaps more concerned about risks as well, as a matter of an overall concern for their own welfare). By contrast, when patients are recruited by their own doctors—patients who may not have been actively ‘looking for research’—there may be a greater prevalence of altruistic motivation (although in this small sample study, the difference was not statistically significant). They may also be less concerned about risks, perhaps, we speculate, due to a greater trust in the researcher who is their own doctor.

We highlight three ethical implications of our findings. First, our results reinforce the need to distinguish between motivation and understanding in the ethical assessment of informed consent. Consistent with previous studies,4 ,16 ,17 we saw that research participants understand the purpose of research and its various elements regardless of their motivation for participation. As others have noted, it is important not to assume that being motivated by a desire for benefit implies a faulty understanding of research17 ,18—just as one can rationally buy a lottery ticket desiring to win, all the while understanding that a lottery’s purpose is not to enrich the buyer of tickets, but rather to raise funds.

Second, the traditional TM framework makes empirical assumptions that may not be warranted. For example, it focuses on the potential that therapeutic motivation will blind subjects’ understanding of research purpose, design, benefits and risks. But this ignores the self-protective function of self-interest which may, in fact, enhance understanding (eg, understanding the scientific rationale for the sham design) or heighten sensitivity to risks.

In fact, based on our results, some may even ask if it is those with altruistic motivation who need special protections, if their motivation leads to decreased sensitivity to risks. Our sense from the data is that there is not a need for special protections. The personal benefit group and the dual/altruism group had good understanding of the purpose and design of their clinical trials, and the two groups did not differ on their recollection of research risks. It is also arguable as to which group is ‘more accurate’ regarding their perception of likelihood of adverse events. That is, ‘very low’ chance and ‘modest’ chance could imply similar numerical probabilities for different people, and we do not have an objectively ‘correct’ answer to the perception of risk question. However, the potential divergence on perception of risks based on motivation for research participation does suggest that in enrolling subjects into clinical research, we should not assume altruistically motivated subjects are immune to potential misperceptions. Perhaps learning of a subject's altruism should alert the person obtaining informed consent to especially focus on the risks and burdens of participation.

Third, another unexpected finding of this study raises questions about how we assess subjects’ understanding of purpose of research—the jumping-off point of all discussions about the TM.9 Our study revealed unexpected complexity on this issue. Even when asked specifically: ‘Is the primary goal to benefit the subjects participating in the study, or future PD patients?’ some insisted on endorsing both purposes—the correctness of which is unsettled even among experts in research ethics.9 Curiously, the five people who clearly stated the wrong answer showed nearly perfect understanding of all the dimensions of research design having to do with lack of individualised treatment in research; this casts considerable doubt on whether we should interpret their ‘incorrect’ responses as an indication that they truly believe that research is being conducted primarily as a form of treatment for them (as would be interpreted by standard TM view).

This study has several limitations. First, the sample size was quite modest, and for the qualitative questions, we were able to conduct only posthoc statistical tests. However, a highly powered comparison of groups (which generally requires large samples) within sham surgery trials in PD is quite unlikely given the usual small sample sizes of such studies, and our results which combined several such studies may represent the limit of what is feasible. Second, qualitative research involves interpretive judgments in development of coding schemes and in the coding itself. Third, sham surgery trials in PD are a highly specialised type of research and any generalisations must be made with caution. Finally, in combining subject data from three different clinical trials, we interviewed subjects (due to sponsor requests) at various time points from the initial point of informed consent for their trials, with long lags in some cases.

In conclusion, we found that therapeutic motivation for entering sham surgery trials is very common, and having one's own doctor as the researcher is common. These factors did not specially increase the TM-related phenomena, and overall understanding of the purpose of research and of the research procedures was high. Although most subjects gave overly optimistic responses about potential benefits, their explanations present a complicated picture, as has been shown in other studies.5 ,17 We also saw an alternative role that therapeutic motivation may play: rather than leading people to false understandings and inaccurate assessment of risks and benefits, therapeutic motivation may indicate a ‘self-interested’ decision maker who is more rather than less likely to pay attention to factors pertaining to their welfare and rights. Finally, our results show that eliciting subjects’ understanding of purpose of research is more complicated than is commonly assumed, by methods that use one or two closed-ended questions to assess it. In future research, such assessments may need specific validation to ensure that intended phenomena are being measured.