|Home | About | Journals | Submit | Contact Us | Français|
Little is known about research participants' understanding of consent information over the course of a clinical study and the relationship of this information with participant behavior.
We conducted a cross sectional patient completed questionnaire of comprehension and satisfaction administered at the end of a Parkinson's disease clinical trial.
Scores on 9 comprehension items in a 30 item questionnaire covering the key elements of informed consent.
78% of eligible trial participants completed this sub-study. Greater than 90% of respondents showed good comprehension of the study purpose, method of treatment assignment, experimental nature of drugs, voluntary participation, and expected effect of the trial on their PD. However, 42.3% of subjects incorrectly endorsed that participating in the study was part of the “usual treatment” for their PD. We found no relationship between comprehension, compliance, and satisfaction with whether or not one's own neurologist was also the study doctor. Years of education and cognitive function at baseline were correlated with comprehension of study information.
Overall comprehension of key study information presented in the consent was high after 12 months of trial participation, although there were inconsistencies in responses that need further study.
Research participants' comprehension of study information has been widely studied near the time of initial informed consent . However, little is known about retention of this information over the course of a study and the effect of this information on participant behavior. If participants fail to appreciate key elements of informed consent, continuing validity of their consent could be compromised.[2, 3] This is of particular concern in progressive neurodegenerative disorders affecting cognitive function such as Parkinson's disease (PD), Huntington's disease (HD), and Alzheimer's disease (AD). 
In the clinical setting comprehension of treatment information is thought to enhance compliance and satisfaction, although recall of treatment information may deteriorate rapidly over time.[4-6] Participant retention of study information may affect clinical trials by affecting satisfaction, compliance with study procedures, and willingness to participate in future studies. We set out to determine if participants in a 12 month PD clinical trial appreciated key study information at the end of the study, how comprehension was related to cognitive function, and if comprehension was related to compliance and satisfaction with study procedures.
Eligible participants were 192 subjects enrolled in an NIH sponsored, Phase II PD clinical trial at 45 of 51 clinical trial sites in the U.S. and Canada. The trial was a 12 month, randomized, double blinded study, testing coenzyme Q10 and GPI-1485. Participants were randomized to one of three arms: active CoQ10 and placebo GPI-1485, active GPI-1485 and placebo CoQ10, or placebo for both agents. The clinical trial and this substudy were approved by the University of Rochester Research Subjects Review Board and the Institutional Review Boards of each participating site. Subjects provided written informed consent to participate in this sub-study on comprehension.
The primary outcome measure was a 30 item multiple-choice, patient completed questionnaire given at the final clinical trial visit. Here we report items assessing satisfaction with the study experience and comprehension of key study elements. The questionnaire was modeled on similar instruments found in the literature and on an instrument developed to assess the quality of informed consent in an oncology clinical trial (Personal communication, E. Emanuel). Pilot testing and face validity assessment were conducted using five individuals with PD not enrolled in the study. The instrument was approved by the clinical investigators conducting the clinical trial, the trial Data Safety Monitoring Board, and each site's Institutional Review Board (IRB). Six sites did not participate due to inability to get timely IRB approval. Compliance was the percentage of study drug taken at each study visit. Cognitive function, at baseline, was measured with the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) and PD severity by the Unified Parkinson's Disease Rating Scale (UPDRS).
The total sample size was fixed by the number of patients in the clinical trial and number of sites willing to participate in the substudy. With 80% subject participation in this substudy, we would be able to construct 95% confidence intervals of approximately 8% width or smaller for a range of observed proportions (0-100%) of participants giving a response to an individual question. The comprehension score was calculated by summing the total number of correctly answered comprehension items and could range from 0 to 9. Relationships between continuous variables were assessed with Spearman correlation coefficients. The chi-square test was used for dichotomous variables and the Kruskal Wallis-Test for ordinal variables.
Of 192 eligible participants, 149 (78%) completed the questionnaire. Responders were similar to non-responders on all demographic and baseline disease measures other than race (Table 1). The study investigator was the regular provider of PD care for 58% of respondents.
Table 2 summarizes the response distributions and frequency of correct responses for each of the 9 comprehension questions. The median comprehension score was 7 out of a possible 9 correct. On 6 of 9 questions, the correct answer was given by 90% or more of the respondents, on topics that included the purpose of study, random allocation of treatment, experimental nature of drugs, voluntary participation, and expected effect on their PD. On the placebo question (Q8), 67% correctly stated that they had had a 1/3 chance of having received a placebo, but another 28% stated a higher chance of having received a placebo. On question 1, 42.3% reported that participating in the clinical trial was part of the “usual treatment” for their PD and 47.6% (Q4) of subjects reported being informed that there were “…no risks to this study greater than my usual medical care”.
Comprehension was not correlated with age, gender, PD severity, or satisfaction, but did correlate with years of education (r=0.30; p<0.01) and cognitive function on the RBANS (r=0.18; p=0.03). The response to Q1 on “usual treatment” was not associated with whether the study investigator was the subject's regular provider.
Mean compliance as assessed by tablet count was 93.7%. There was no correlation between comprehension and compliance (r=-0.06). Ninety-seven percent of subjects were moderately or very satisfied with their overall experience. Ninety-five percent of subjects reported that they received all the information they wanted and were most likely to ask the study coordinator or investigator if they had questions rather than outside sources. Less than 5% of participants reported feeling pressure from any source to join or stay in the clinical trial.
Overall, the majority of these early Parkinson's participants understood the key aspects of the study, including random allocation and the chance of receiving placebo. Compliance with study drug and satisfaction with the study experience were very high. This may have contributed to the lack of an observed relationship between comprehension and compliance with satisfaction. Years of education and cognitive function were correlated with comprehension, consistent with studies conducted near the time of consent to participate.
The high level of appreciation of randomization and the chance of receiving placebo is unusual in informed consent studies and may reflect the level of education of participants. While high education levels are common among subjects in PD studies, this finding also suggests that subjects may have learned more about the study during participation.
The most striking finding was the discrepancy between the excellent performance on 7 of 9 comprehension items versus performance on the remaining 2 items that asked about the differences between research and routine care and the understanding of risks (Q1 and Q4, Table 2). One explanation is that the poor performance on the two questions shows that the respondents were exhibiting therapeutic misconception (TM), conflating research and treatment, or grossly underestimating risks of harm.[10,11] This interpretation raises the interesting question of why the rate of apparent TM is so low in the other 7 questions, some of which were also probes for TM. For example, the phrasing of the placebo question (Q8) specifically allowed testing of the respondents' ability to appreciate how the design of the study might have affected them, not merely to factually understand, the chance of having received a placebo. But only 5% denied the possibility of having received a placebo and 95% of subjects responded that they had a 1/3 or greater chance of receiving placebo.
Another interpretation of poorer performance on Q1 and Q4 is that it reflects the difficulty of operationalizing probes for TM. The phrase “usual treatment” may have been interpreted as describing common medical practices involved in PD treatment, such as taking oral medications. Subjects' perceptions of risk may have been shaped by actual experience in a trial that involved generally safe, well-tolerated agents, and by the respondents' reading of the response option “no risks to this study greater than my usual medical care.” This interpretation is supported by the fact that the responses to questions 1 and 4 were not correlated with whether or not the study investigator was the subjects' usual treating neurologist. This factor should have strongly correlated with subjects' responses if the problem were TM, since one of the main triggers for TM is thought to be one's own doctor acting as a researcher,  which is often the case in many PD clinical trials.
The overall findings of this study are encouraging, showing generally good understanding of study information after 12 months of participation among early PD subjects. This may suggest that rather than deterioration of comprehension over time as has been observed in some clinical settings, research participants may retain or even improve comprehension through participation. However, less educated and mildly cognitively impaired PD participants may be at risk for poorer comprehension of study information. This is of particular concern in PD since cognitive function declines over time. The cross sectional designs of the study limits our ability to assess whether comprehension changed over time or was shaped by experience in the study. Future studies in PD should follow subjects longitudinally to assess changes in comprehension and further explore the effects of progression and cognitive decline. This will help to inform procedures for maintaining informed consent in long-term observational studies and clinical trials in PD.
The apparent incongruity in the responses may in fact also be an important finding, raising key questions for future research on informed consent in PD. Although the explanation could be that TM occurred, there are countervailing reasons to doubt this as the definitive interpretation. An alternative interpretation is that what the respondents mean and what researchers mean by phrases such as “usual treatment” and “usual medical care” may be a source of unexpected ambiguity for less educated or mildly impaired laypersons. The closed-end format of the questions limits our ability to assess how the respondents understood the questions, especially given the somewhat inconsistent nature of the responses. In future research on informed consent in PD, it will be important to keep this possibility in mind, by adding a qualitative interview component to probe the answers of the respondents.
We thank the NET-PD investigators and coordinators who conducted this study and the patients who participated in this study.
Funding/Support: This study was sponsored by the NIH (National Institute of Neurological Disorders and Stroke), U01NS043127 and U01NS043128. The authors have no conflicts to declare.
Contributors: BR conceived and designed the study and oversaw the implementation, analysis, and interpretation. He was primarily responsible for drafting and revising the article. CS and JE conducted the analysis and assisted in the interpretation and writing of the manuscript. CK, KK, and SK were involved in the design, implementation and interpretation of the study and contributed to editing and revising of the manuscript. KK was the principal investigator of the parent study.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.