Ethical Issues in Intraoperative Neuroscience Research: Assessing Subjects’ Recall of Informed Consent and Motivations for Participation

Abstract

Background:

An increasing number of studies utilize intracranial electrophysiology in human subjects to advance basic neuroscience knowledge. However, the use of neurosurgical patients as human research subjects raises important ethical considerations, particularly regarding informed consent and undue influence, as well as subjects’ motivations for participation. Yet a thorough empirical examination of these issues in a participant population has been lacking. The present study therefore aimed to empirically investigate ethical concerns regarding informed consent and voluntariness in Parkinson’s disease patients undergoing deep brain stimulator (DBS) placement who participated in an intraoperative neuroscience study.

Methods:

Two semi-structured 30-minute interviews were conducted preoperatively and postoperatively via telephone. Interviews assessed participants’ motivations for participation in the parent intraoperative study, recall of information presented during the informed consent process, and participants’ postoperative reflections on the research study.

Results:

Twenty-two participants (mean age = 60.9) completed preoperative interviews at a mean of 7.8 days following informed consent and a mean of 5.2 days prior to DBS surgery. Twenty participants completed postoperative interviews at a mean of 5 weeks following surgery. All participants cited altruism or advancing medical science as “very important” or “important” in their decision to participate in the study. Only 22.7% (n = 5) correctly recalled one of the two risks of the study. Correct recall of other aspects of the informed consent was poor (36.4% for study purpose; 50.0% for study protocol; 36.4% for study benefits). All correctly understood that the study would not confer a direct therapeutic benefit to them.

Conclusion:

Even though research coordinators were properly trained and the informed consent was administered according to protocol, participants demonstrated poor retention of study information. While intraoperative studies that aim to advance neuroscience knowledge represent a unique opportunity to gain fundamental scientific knowledge, improved standards for the informed consent process can help facilitate their ethical implementation.

Introduction

In recent decades, there has been an increasing number of studies utilizing intracranial electrophysiology in human subjects (Chiong, Leonard, and Chang 2018; Tekriwal et al. 2019). Some of these studies take advantage of intraoperative situations, such as Parkinson’s disease patients being implanted with deep brain stimulation (DBS), to conduct a wide variety of research. Whereas some studies aim to further clinical understanding and improve patient outcomes, other “opportunity studies” (Mergenthaler et al. 2021) seek to advance fundamental neuroscience knowledge regarding aspects of cognition and other aspects of human brain function (Tekriwal et al. 2019; Ramayya et al. 2017, 2014). In recognition of the promise of such studies, two recent National Institutes of Health reports highlighted the extraordinary research opportunities provided by the neurosurgical context (BRAIN Initiative 2014, 2019).

However, the involvement of neurosurgical patients, particularly in studies that aim to generate basic neuroscience knowledge, raises important ethical considerations (Chiong, Leonard, and Chang 2018; Vaishnav and Chiong 2018; Hendriks et al. 2019). For example, patients undergoing neurosurgery for treatment of progressive neurodegenerative diseases, like Parkinson’s disease, may have impaired capacity to understand study procedures and provide informed consent (Karlawish et al. 2013; Moelter et al. 2016). Research participants may also have difficulty distinguishing standard therapeutic procedures from research protocols (Appelbaum et al. 1987; Kim et al. 2015). These concerns are especially salient given that these studies may involve additional risks beyond those related to the surgical treatment itself, such as potential increased risk of infection due to the extended surgical time necessary to obtain intracranial recordings (Mergenthaler et al. 2021).

An additional worry is that the consent that patients provide to participate in intraoperative research may be unduly influenced by the role of the neurosurgeon as both clinician and researcher (Chiong, Leonard, and Chang 2018; Morain, Joffe, and Largent 2019; Council for International Organizations of Medical Sciences (CIOMS) 2016). Although the voluntariness of consent is a concern for all clinical trials, intracranial studies that aim to advance basic neuroscience knowledge are in many ways unique: they have features of clinical trials, in terms of both the clinical setting and patient population, yet they also have aspects of healthy volunteer studies, where the research knowledge gained may not hold the prospect of benefit for a disease or scientific endeavor that is relevant to the participant. With regard to motivations for participation, prior research has found that participants in clinical trials are variously motivated, with altruism or expected benefit to themselves being prominent (Truong et al. 2011; Locock and Smith 2011; Tromp, Michel Zwaan, and van de Vathorst 2016). In contrast, financial incentives are notable motivations for healthy volunteers (Stunkel and Grady 2011; Grady et al. 2017; Tishler and Bartholomae 2002; Almeida et al. 2007). However, it is unclear why patients participate in intraoperative studies. As a consequence, the extent to which questions of voluntariness arise in intraoperative neuroscience studies is unknown.

While to date, scholars have raised questions regarding ethical issues in intraoperative neuroscience research (Chiong, Leonard, and Chang 2018; Hendriks et al. 2019), there has not yet been an empirical examination of these issues in a participant population. The present study therefore aimed to examine ethical concerns regarding informed consent and voluntariness in Parkinson’s disease patients undergoing DBS placement who were recruited to participate in an intraoperative neuroscience study on flexible decision-making. By conducting preoperative and postoperative interviews with study participants, we sought to (a) examine concerns related to informed consent, by assessing participant understanding and recall of study purpose, protocol, risks and benefits, as well as participants’ postoperative reflections on, and recall of, the research task and (b) examine concerns related to voluntariness, by assessing participants’ motivations for participating in intraoperative DBS research and their postoperative reflections on participation, as well as why other potential subjects decline to participate.

Methods

Study procedures for both the intraoperative study and ethics study were approved by the University of Pennsylvania’s Institutional Review Board. Figure 1 depicts the timeline for both the parent intraoperative study and the supplemental ethics study.

Figure 1.

Study flow diagram.

Parent intraoperative study

Between May 2018 and July 2020, Parkinson’s disease patients who were scheduled to undergo unilateral or bilateral DBS electrode placement in the subthalamic nucleus were recruited to participate in an intraoperative study of flexible decision-making. The purpose of the study was to better understand how activity patterns of neurons in the substantia nigra pars reticulata (SNr) of the human brain, contribute to higher functions such as learning and decision-making. DBS electrode placement in the subthalamic nucleus requires identifying the ventral border of that structure by passing the electrode through to the adjacent SNr, which occurs while the patient is awake. During this phase, the therapeutic surgery was paused while the research team stepped in and had the participant perform a cognitive task (either ocular saccade tracking, or a probability game using handheld buttons). When the research task was completed, the DBS surgery resumed.

Study recruitment took place during the patient’s presurgical office visit approximately 1–2 weeks prior to surgery, where following an initial gauging of interest by a nurse practitioner, prospective participants met with a research coordinator to review the details of the study. As part of the informed consent process, research participants were informed that they could not expect to directly benefit from the study, and that risks included potential loss of confidentiality as well as increased risk of infection due to a maximum of 30 minutes of extended surgical time. No financial incentive was offered for participation in the intraoperative study.

Supplemental ethics study

All participants who were approached to join the intraoperative study between October 2018 and February 2020 were invited by the same research coordinator to take part in a supplemental study (“ethics study”). The ethics study consisted of two semi-structured, 30-minute interviews, one conducted preoperatively and one postoperatively via telephone (by AW). Preoperative interviews included questions assessing participants’ motivations for volunteering for the parent study and their recall of information presented during the informed consent process. Postoperative interviews included questions assessing participants’ surgical experiences and reflections on their experience in the research study. Participants also completed a brief questionnaire during the pre-surgical office visit that assessed demographic information. Participants were offered a $75 Amazon e-gift card for each interview as compensation for their time and effort. Individuals who were recruited to the parent intraoperative study but declined to participate were offered a $5 Amazon e-gift card in exchange for completing a brief questionnaire regarding reasons for declining participation in the intraoperative study.

All interviews were audio-recorded with participants’ consent and transcribed via Rev.com or SpeechPad. Codes were developed by two authors (RC and AW) iteratively for each interview question following an examination of the transcripts (DeCuir-Gunby, Marshall, McCulloch 2011). Final coding categories were refined after discussion with two additional authors (EK and SG). Two authors (RC and AW) coded the transcripts, and all disagreements in coding were resolved through discussion.

Results

Enrollment and demographics

A total of 22 preoperative interviews (conducted at a mean of 7.8 days following informed consent and a mean of 5.2 days prior to DBS surgery) and 20 postoperative interviews (conducted at a mean of 5 weeks following surgery) were analyzed. Figure 2 depicts the study enrollment flow diagram. Four patients declined to participate in the intraoperative study. All four completed the questionnaire and indicated that the primary reason for declining was health risks due to increased surgical time. Other factors rated as important included feeling overwhelmed by the surgery (n = 3), potential discomfort due to additional surgical time (n = 3), and friends/family advising against participation (n = 1). Table 1 depicts the demographic characteristics of the study population. No demographic differences were qualitatively apparent between the small number who declined to participate and those who enrolled in the study.

Figure 2.

Study Timeline.

Table 1.

Demographic characteristics of ethics supplement study population.

	Total N=22 (%)
Gender
Female, n (%)	5 (22.7)
Age (mean years ± SD)	60.9 ± 10
Time since Parkinson’s diagnosis (years)	8.9
Education
High school graduate	3 (13.6)
Some college, no degree	5 (22.7)
Associate’s degree	2 (9.1)
Bachelor’s degree	5 (22.7)
Master’s degree	3 (13.6)
Professional or doctoral degree	4 (18.2)
Household Income, n (%)*
Less than $15,000	3 (15.0)
$50,000–74,999	4 (20.0)
$75,000–$99,999	5 (25.0)
$100,000–149,999	2 (10.0)
$150,000–$199,999	3 (15.0)
$200,000–$250,000	1 (5.0)
More than $250,000	2 (10.0)
Prior Unilateral DBS Implant
Yes	3 (13.6)
No	19 (86.4)

^*Missing household income for 2 patients.

Recall of informed consent

Table 2 depicts participant recall percentages for study purpose, protocol, benefits, and risks. Interview questions were open-ended and phrased in a way to minimize anxiety (i.e., “Do you remember if the study team mentioned anything about study risks?” rather than “What are the study risks?”). With regard to study purpose, approximately one-third (36.4%) recalled that it was related to neuroscience and/or brain function. Recall of study protocol was coded generously, with any mention of elements of the task (e.g., “lights and two buzzers” or “the game with the cards”) coded as “correct.” Notably, only one participant volunteered a clear and accurate description of both the cognitive task to be performed and when it would occur. In relation to study benefits, “no direct benefits” as stated on the informed consent form, as well as “advancing science” and “helping others” were all coded as correct responses, for a total of 36.4% correct recall. With regard to study risks, even though loss of confidentiality and increased risk of infection were both stated as risks on the informed consent form and were reviewed verbally with participants by research coordinators, most (n = 17; 77.3%) reported that there was no risk or that they did not remember the risks. Overall, no demographic differences were apparent between those who performed well (as defined by >50% on recall) as compared to those who did not (<50% recall).

Table 2.

Subjects’ (N = 22) recollection for the main aspects of informed consent, assessed at a mean of 7.8 days following informed consent and 5.2 days prior to DBS surgery. Answers coded as correct responses are italicized.

Recall Element	Language from Informed Consent	Coding Category	N=22(%)	Representative Quotation
Study Purpose	“…To better understand how the activity patterns of neurons in various structures in the human brain contribute to higher functions like learning and decision-making.”	Neuroscience-related	8 (36.4)	AD9: I think that they try to learn how people think, how people’s brain work.
		Remembers task, not purpose	5 (22.7)	AD4: Only that there was some kind of procedure where they could look into your eyes and read some of the responses that were going to happen.
		Don’t know/remember	5 (22.7)	AD18: I don’t remember, my memory these days. I don’t have a real good short term memory.
		Parkinson’s related	2 (9.1)	AD5: One of the goals is to help future patients that have Parkinson’s.
		Other	2 (9.1)	AD10: It was just to see if I lost any cognitive ability during the course of this process.
Study Protocol	“You will undergo cognitive testing at your preoperative, intraoperative and postoperative visit. Additionally, you will undergo microstimulation and neural recording during your intra-operative visit.”	Don’t know/remember	5 (22.7)	AD10: I don’t think so, at least I didn’t remember. I don’t remember them saying that.
		Recalled aspects of cognitive task	11 (50.0)	AD14: The light goes on, the light goes off and you push the two things on the pad they give you to push a button on.
		Unclear	4 (18.2)	AD20: I’ts easier to pinpoint the spot, I guess, in the brain that they need to use to fulfill the procedure.
		Misunderstood protocol	2 (9.1)	AD5: I think, if I remember correctly, during the awake part of the procedure, I will receive some questions.
Benefits	“You will not benefit directly from this study. It is hoped that the information learned from this study may benefit other people with Parkinson’s disease in the future.”	Don’t know/remember	12 (54.5)	AD7: No. Probably did, but, like I said, I don’t remember.
		No benefits	1 (4.5)	AD8: I should not expect to derive any personal benefit from the study.
		Helping others	2 (9.1)	AD2: Helping people is a benefit.
		Advancing science	3 (13.6)	AD22: Just to gain a better understanding of brain function.
		Both helping others and advancing science	2 (9.1)	AD9: I hope it will help patients in the future, and it will help doctors to understand better how our brain works, so they can help the future patient better.
		Other	2 (9.1)	AD3: If data is being collected, maybe it works to see how I did before, and then how I did during it.
Risks	“There is a potential loss of confidentiality associated with having your chart reviewed… This research proposal will extend the intra-operative collection time of microelectrode recordings for up to 30 minutes. The increased surgical time required to perform the task may increase the risk of infection.”	Don’t know/remember; was not mentioned	15 (68.2)	AD13: No, I don’t remember them saying anything about any type of risk.
		No risks	2 (9.1)	AD6: What I recall is that there was no risk.
		Recalled surgical risks	4 (18.2)	AD12: I do not think there are any risks, besides being up to a half an hour more than normal.
		Recalled loss of confidentiality	1 (4.5)	AD8: Only loss of confidentiality.

Undue influence and motivations for participation

Participants were asked both via open-ended and structured questions about their primary motivations for participating in the intraoperative study. All (N = 22) cited altruism (i.e., helping others, specifically future Parkinson’s patients) or advancing medical science as “very important” or “important” in their decision to participate in the study; most (n = 18) also agreed that trust in the clinical team played a significant role in their decision to participate. In response to the open-ended question regarding motivation, two of the three individuals who had previously been implanted with unilateral DBS reported participating out of gratitude to the neurosurgeon. One stated: “Since Dr. X improved my life for four years, anything for Dr. X because I understand the whole entire study is—he’s in charge of it.” Another said: “I’m going to do it because I want to help Dr. X in any way I can and try to give back to what he has done for me.”

Research participants were asked to characterize the nature and timing of their decision to volunteer for the intraoperative study. Six (27.3%) participants reported that they decided to volunteer as soon as they were introduced to the idea by the nurse practitioner, prior to learning any study details from the research coordinator. Six additional participants (27.3%) decided “right when asked” or “instantly,” suggesting that they agreed as soon as the informed consent process was initiated. All participants (n = 22) reported feeling like they had free choice to take part in the study.

Postoperative reflections

Postoperatively, three of 20 participants (15%) had no memory of the surgical experience, and therefore no memory of performing the research task. For the remainder (n = 17), recall of the surgical experience, and the awake portion in particular, was hazy, which was likely a result of anesthesia: research participants had fleeting recall of being in a “twilight,” “dream state,” or “a very deep fog.” Of these same participants (n = 17), 13 agreed that the boundaries between the research and therapeutic components of surgery were clear. For the remaining participants (n = 4), the entire experience was blurry; one stated: “I grabbed ahold of something… I’m not sure if that was part of it [the study]… I can’t really remember much.”

All participants who had memories of the surgical experience (n = 17) were asked to reflect on their experience of the study. All reported that they were happy to have participated. Finally, as it became increasingly clear from the preoperative interviews that only a small percentage of participants recalled the purpose of the study as being neuroscience-related, participants were asked postoperatively if they would have participated in the study even if its aim was solely to advance basic understanding of brain function and not to help Parkinson’s disease. All participants who were asked (n = 19) agreed that they would, with some stating that any advancement in brain science could help further knowledge of Parkinson’s disease and ultimately help others.

Discussion

Informed consent is a keystone for human-based research ethics. The process of obtaining informed consent is intended to ensure that research participants are aware of the aims of the project they are considering, informed about relative risks and benefits attendant on participation, capable of decision making, and voluntary in their decisions to participate or not (Emanuel et al. 2000). The choice to participate in a research study should ideally arise from a clear understanding of the relative risks and benefits. However, there is no consensus definition of what constitutes adequate understanding, how it should be measured, or when it is best measured (Verheggen and van Wijmen, 1996). The findings of the present study are relevant to all three of these, but particularly the last. When research participant understanding is assessed at a temporal distance from the informed consent process, as it is in the present study, it is highly dependent upon recall. The findings of the current study highlight the importance and limitations of using recall to assess participant understanding within the informed consent process.

Even though research coordinators were properly trained and the informed consent was administered according to the IRB-approved protocol, in the present study participants demonstrated poor retention of study information. Prior studies assessing recall—in both healthy and clinical populations, and for both treatment and research—have found that it is often poor (for review, see Falagas et al. 2009; Verheggen and van Wijmen, 1996). Still, our findings about accuracy of recall, particularly with regard to risk (22.7%) and benefits (36.4%), were on the low end as compared to previous studies assessing recall specifically in Parkinson’s patients participating in research studies (Ravina et al. 2010; Kim et al. 2015; Valadas et al. 2011; Finder et al. 2012). However, numerical comparisons should be interpreted with caution because of wide variations in study methodology, particularly with regard to the kinds of trials assessed, when the assessment was performed (e.g., ranging from initial clinical trial visit to a four-year follow-up visit), how recall was examined (open-ended questions, multiple choice questions, or asking patients whether they “felt informed about risks”), and how correct responses were coded and reported (Verheggen and van Wijmen, 1996).

As noted above, while the ability to recall relevant information might be a sign of greater understanding, it is possible that a participant may understand study information in the moment but fail to recall relevant details at a later time. In the present study, because of the length of participants’ presurgical visits, as well as logistical considerations, the pre-operative interview did not occur on the same day as recruitment, but rather via a telephone interview at a more convenient, later time for participants. It is therefore possible that participants had a clear understanding of the study during the office visit, but simply did not recall details at the time of the interview.

Poor recall of study information found in the present study (i.e., with regard to protocol, risks, purpose, benefits) could be a result of a number of factors. First, during the presurgical visit, the impending DBS surgery and its concomitant risks were almost certainly foremost on participants’ minds. Details of the intraoperative research study were likely less salient. In contrast, prior work assessing recall of research informed consent in Parkinson’s patients involved hypothetical trial participation (Finder et al. 2012), pharmaceutical clinical trials (Ravina et al. 2010; Valadas et al. 2011) or sham surgery controlled neurosurgical trials (Kim et al. 2012); none of these studies involved participants who were concurrently undergoing a significant surgical operation. Second, participants were recruited during their pre-surgical visit, which was a long and cognitively taxing day that involved up to several hours of travel time to and from the surgical clinic; cardiology screening and blood work; and meetings with the neurosurgeon, nurse practitioner, and research coordinator. Third, all participants had been living with Parkinson’s disease for nearly a decade (mean time since diagnosis was 8.9 years), raising the possibility of some mild cognitive impairments (Mamikonyan et al. 2009; Goldman and Litvan 2011; Foo et al. 2017), although all participants passed standard neurocognitive testing.

Notably, not a single participant had a therapeutic misconception (i.e., expected that the primary purpose of the study was to directly benefit them rather than for research purposes; CIOMS 2016; Henderson et al. 2007). In addition, although the ability to distinguish research from therapy is often considered an important component of an ethical research study, the drugs used for sedation during neurosurgery can cause anterograde amnesia. Thus, that three participants did not recall anything from their surgical experience, and four were postoperatively not able to distinguish the intraoperative research component of the surgical experience from the therapeutic one, are not unexpected findings.

This is the first study to assess participants’ motivations for participation in an intracranial opportunistic neuroscience study. All participants volunteered out of a desire to help Parkinson’s patients and/or advance medical science, even though the purpose of the study was to advance knowledge about basic neuroscience. Participants likely perceived that the study was helping Parkinson’s disease for two primary reasons. First, there was a sentence in the informed consent stating that it is hoped that information gleaned from the study would help benefit other Parkinson’s disease patients in the future (Table 2). Second, social context may have played a role. Namely, participants were recruited for the study within the neurosurgeon’s office during a presurgical visit for a therapeutic intervention for Parkinson’s disease. However, when asked if they would have participated if they were told the study purpose was related exclusively to brain science, and not Parkinson’s disease, all of the participants indicated a willingness to contribute to basic neuroscience research.

Our findings that 54.6% of participants decided to participate prior to learning study details, or as soon as informed consent was initiated, are in line with previous work suggesting that patient decision-making—for both research and therapy—is largely shaped not by the information communicated during the informed consent process, but rather by complex social processes such as setting, framing, and trust (Sachs et al. 2003; Sankar 2004; Sankar et al. 2006; Clapp et al. 2019).

Of the patients with prior experience being implanted with unilateral DBS by the neurosurgeon, two of three reported participating because they wanted to “give back” to the neurosurgeon. Because DBS can have a dramatic impact on patients’ quality of life, voluntariness may be a greater issue in this subset of participants as compared to other neurosurgical subjects, although gratitude need not compromise voluntariness. Furthermore, all patients reported that they felt like they had free choice to participate in the study: while this is certainly not a definitive measure of voluntariness, it is notable in that not a single participant expressed reservation throughout the course of open-ended interviews.

No demographic differences were observed qualitatively between enrollees and those who declined to participate, although the small number of those who declined preclude more substantive conclusions. A previous study assessing differences between decliners and enrollees in two phase II neurosurgical intervention trials for Parkinson’s disease found that decliners perceived higher risk of harm and had a lower tolerance for harm (Kim et al. 2012).

The present study has a number of limitations. First, our sample size was relatively small and limited by the number of enrollees in the parent study. Second, cognitive abilities of Parkinson’s disease patients can fluctuate due to the disease or the timing and dose of medications used to treat it (Cooper et al. 1992; Kulisevsky et al. 2000), thus recall ability may have varied amongst our participants in response to questions. Third, our study assessed participants approximately one week after the informed consent, therefore our results indicate recall and not necessarily comprehension of study information. Fourth, given the lack of control group or comparative measures, it is difficult to know how this population, or a similar one, would have performed on other measures of recall (e.g., recall for the standard DBS informed consent).

Future studies could examine comparative recall figures, as well as the extent to which the companions who accompany patients during the presurgical visit recall information for both the neurosurgical process and the intraoperative study. Additionally, to disentangle what participants understand at the time of their initial consent to participate from what they retain later, future research could assess participants at a time period closer to the informed consent.

Given that intracranial, intraoperative studies on neurosurgical patients are increasingly being conducted, future work should aim to develop interventions to increase participant understanding and retention of study information. Notably, previous research has found that some of the most optimal methods for improving the understanding of study information presented during the informed consent involve extended discussions and one-on-one interactions (Flory and Emanuel 2004; Nishimura et al. 2013). Given that these discussions did occur in the present study, we suggest modifying the time in which they are presented: for example, an in-office discussion of the informed consent could be followed with a telephone consent, after the individual has had time to review study materials. In addition, “teach-back” methods, which involve participants relaying information about the informed consent back to study personnel, have been shown to increase comprehension (Glaser et al. 2020) and could be employed in future studies. In sum, while intraoperative studies that aim to advance neuroscience knowledge represent a unique opportunity to gain fundamental scientific knowledge, an improved informed consent process can help facilitate their ethical implementation.