Abstract
Scientific advancements in spinal cord injury (SCI) have focused on early interventions. However, research in the acute SCI setting presents ethical challenges, particularly the durability of informed consent. Given trauma’s effects on cognition, we assessed participants’ recollection of the consent process and willingness to participate in future research. A 15-item questionnaire was administered to 119 participants previously enrolled in one of three acute SCI studies at our institution between 2010 and 2022. Of 32 responses, 47% recalled the consent process, 30% were unsure, and 23% had no recollection. Recall of specific details was also inconsistent. Of 11 participants who recalled receiving a study treatment, only 4 (36%) were in an interventional arm. While injury severity and level did not affect recollection, time since injury and age at the time of the survey were associated with recall differences. Notably, all 18 subjects who recalled their enrollment also recalled follow-up visits from study personnel, suggesting ongoing contact may reinforce awareness of their involvement. In the acute SCI setting, physical and psychological distress challenge the informed consent process. Our survey revealed significant inaccuracies in participant recall, undermining the intent of informed consent. We propose that practices like continuous consent, coupled with regular study-related interactions, may improve participants’ understanding and retention of information, thereby strengthening the ethical foundation of acute SCI research.
Introduction
Paralysis from spinal cord injury (SCI) is an incurable, life-altering condition. The severity of this condition has spurred research efforts across the spectrum of scientific research and discovery that have the shared goal of elucidating disease-modifying early interventions that can offer hope to those afflicted by SCI. 1 The relative scarcity of successfully completed interventional studies in the acute phase is due to a confluence of variables that include the prioritization of life-saving medical interventions, resource demands in the critical care neurosurgical setting, and the logistical and regulatory burden of trauma research studies. Beyond these challenges lie the bioethical complexities of obtaining informed consent from patients in severe physical and psychological distress. Given these limitations, the informed consent process ought to be given special consideration. 2
Our institution has extensive experience with SCI clinical trials3–8 across the spectrum from acute and subacute 9 to chronic injury.10,11 This experience has highlighted the unique challenges of the informed consent process in the early period after an acute SCI. To better understand the effectiveness of our consent procedures, we surveyed former participants from our acute SCI clinical trials. The goal was to use this knowledge to identify potential areas for improvement and optimize future consent processes.
Prior studies on the efficacy of consent during periods of severe illness have shown that while obtaining robust informed consent is difficult, it is achievable through staff education and standardized processes.12–16 However, despite the use of standard operating procedures and other tools, the severe distress associated with major injuries can significantly affect a patient’s ability to recall events from that time. Indeed, poor recall of informed consent is a documented issue across various medical specialties, including procedural settings like cardiology. 17 We aimed to assess participants’ recollection of both the informed consent process itself and the details of their participation in an acute SCI research study.
Methods
This was a retrospective cohort study. We designed a 15-item questionnaire (Table 1) to be administered by telephone interview, intended to capture prior SCI clinical study participants’ recollection of the consent process and their subsequent participation. The questionnaire was designed to capture responses addressing key metrics used to assess the quality of the informed consent process in accordance with Good Clinical Practice guidelines. The survey collected data on participant age (ranging 20–66) and year of injury (ranging 2010–2022) to allow for analysis of how these factors might influence recollection. The questionnaire then assessed participants’ recall of their injury classification, for example, ISNCSCI grade 18 and level of injury (LOI), their enrollment in a clinical trial, and their interactions with the research team. A section was also included to gauge interest in future research participation.
Questionnaire
The study population was selected from databases of three acute SCI clinical trials conducted at the University of Miami: “Systemic Hypothermia in Acute Cervical Spinal Cord Injury—A Prospective Case Controlled Study” (1), “North American Clinical Trials Network (NACTN) Registry” (2), and “Systemic Hypothermia in Acute Cervical Spinal Cord Injury—A Prospective, Multi-center Case Controlled Study” (3). Of the three trials, the first was initiated in 2009 and concluded in 2016; the second began in 2008 and remains ongoing; and the third started enrollment in 2017 and is also ongoing. The first and third trials evaluate the same therapeutic intervention, whereas the second is conducted as an observational study. Collectively, these three studies have provided the largest cohort available to us of patients enrolled in acute spinal cord injury studies. These databases provided baseline information on participants’ AIS grades, neurological levels, time of injury, and treatment group assignment (interventional vs. observational), allowing us to compare their recollections against their actual study records. They also provided the Injury Severity Scores (ISS), a standardized anatomical scoring system used to assess overall trauma severity, and we selected nonparametric Kruskal–Wallis test to evaluate the difference between the consented and enrolled groups. Additionally, a Cochran–Armitage test for linear trend was performed to assess the relationship between age groups and consent recollection; this test was selected because it specifically evaluates whether a binary outcome (viz., recollection) changes linearly across ordered categorical groups (age strata).
None of the enrolled subjects had sustained traumatic brain injury (TBI) during the initial SCI. While TBI may be as frequent as 60% in the SCI population, 19 these studies were initially screened to exclude subjects who had TBI. The study was approved by the University of Miami Institutional Review Board.
Results
Participant recruitment and demographics
From an initial pool of 119 subjects previously enrolled in acute SCI clinical trials, we successfully contacted 40 individuals by phone. Of these, we learned that two subjects were deceased. Eight individuals declined to participate, resulting in a final cohort of 32 subjects for the current survey study (Fig. 1).
Total potential candidates.
Although 2 of the 32 subjects did not complete the entire questionnaire, all subjects provided their age. The participants were categorized into 6 age groups, with the largest group being 30–39 years old (32.3%, n = 10). Other age distributions were: >59 years (25.8%, n = 8), 40–49 years (22.6%, n = 7), 50–59 years (12.9%, n = 4), and 20–29 years (9.7%, n = 3). There were no participants in the < 20 age group. The self-reported year of injury for participants ranged from 2010 to 2022, with the highest number of injuries occurring in 2018 (19%, n = 6) (Fig. 2).
Year of injury.
Statistical analysis was performed via nonparametric Kruskal–Wallis test and showed no statistically significant difference in ISS scores between either the consented groups or the enrolled groups (consented group: χ2 (2) = 2.916, p = 0.233, and enrolled group: χ2 (2) = 0.448, p = 0.799). Similarly, the anatomical region of the injury (cervical vs. thoracic/lumbar) did not affect subjects’ responses, as shown in Figures 3 and 4.
Distribution of consent recollection by injury region.
Distribution of enrollment recollection by injury region.
However, the time elapsed from injury to survey and the participant’s age did appear to be associated with differences in recollection, as shown in Table 2. For instance, 100% of participants in the 20–29 age group recalled being consented, compared to 63% of the 30–39 group, 50% of the 40–49 group, and 0% of the 50–59 group. A Cochran–Armitage test for linear trend was performed to assess the relationship between age group and recollection of consent. The analysis revealed a statistically significant negative trend (p = 0.007), indicating that increasing participant age is significantly associated with a lower likelihood of recalling the consent process.
Year of Injury Per Subject
Color signifies the difference in time.
Recollection of injury and study details
A significant portion of participants, 68.8% (n = 22), were unsure of their ISNCSCI exam ASIA Impairment Scale (AIS) grade. 20 Among the 10 subjects who did provide a grade, 60% (n = 6) reported it correctly according to our records. However, three subjects under-reported their AIS grade compared with their last follow-up exam. Regarding the level of injury (LOI), only two subjects were unsure. Of the remaining 29 subjects, 14 described their LOI as a range of spinal cord segments rather than a single neurological level.
Out of 30 subjects who answered questions about the consent process, 47% (n = 14) recalled undergoing it within a week of their SCI (Table 3). In contrast, 23% (n = 7) did not recall the consent process at all, and 30% (n = 9) were unsure. When asked about enrollment, 60% (n = 18) of subjects correctly recalled being enrolled in an acute clinical trial, while 33% (n = 10) reported they were not, and 7% (n = 2) were unsure (Table 3).
Summary of Responses
Understanding and perception of study participation
There was notable confusion regarding the nature of study participation. Thirty-seven percent (n = 11) of subjects believed they had received a study-related treatment. However, according to our records, only 4 of these 11 subjects (36%) were in interventional trials; the rest were in observational studies. Conversely, of the 14 subjects (47%) who reported not receiving any treatment, three (21%) had in fact received an intervention. All 18 subjects who remembered being enrolled in a study also recalled being visited by study personnel within the first week, with the number of remembered visits ranging from one to six. All participants who completed the questionnaire confirmed they were able to contact study personnel when needed. Regarding their understanding of the research, 79% (n = 15) of those who remembered being enrolled felt they understood the study’s purpose, and 78% (n = 14) understood what was asked of them as participants.
Interest in future research
Of the 30 subjects who responded, 50% (n = 15) reported having looked for other clinical trials. Of this group, 73% (n = 11) had never used the ClinicalTrials.gov website during or after their participation in the original studies, and 60% (n = 9) had participated in at least one subsequent trial. Of the 30 subjects who answered item 11c of the survey (Table 1), 50% (n = 15) said they were actively looking for another trial to participate in, 37% (n = 11) said they were not, and 13% (n = 4) were not sure.
When presented with hypothetical scenarios, 67% (n = 20) of participants stated they would participate in trials requiring nonreimbursed, in-person visits, whereas 27% (n = 8) said they would not, and 7% (n = 2) were not sure. When asked about a trial with a small chance of functional improvement but a risk of additional pain, 40% (n = 12) were willing to participate, 43% (n = 13) were not, and 17% (n = 5) were unsure. Finally, a vast majority, 87% (n = 26), affirmed they would be willing to complete a similar questionnaire in the future.
When a Legally Authorized Representative (LAR) was used, only 43% (n = 3) of subjects later remembered being consented, and just 28.5% (n = 2) remembered being enrolled. For the eight subjects who were physically unable to sign their own consent form, witnesses had to be used. Only half (n = 4) of these subjects remembered the consenting process, three did not (37.5%), and one (12.5%) was unsure. Seven (87.5%) of them remembered being enrolled in the study, and one (12.5%) did not.
Discussion
Our survey revealed that many participants in acute SCI trials have poor or inaccurate recollection of the informed consent process and key aspects of their study participation. This finding is critical for the ethical conduct of research in this vulnerable population.
Factors influencing participant recollection
Interestingly, the severity of the injury did not appear to significantly influence recall, as shown in Table 4. The average ISS of our population was 17, which was slightly higher than reported in other SCI studies.21,22 This could be explained by the inclusion criteria of the two initial trials, for which these subjects were selected, as they focused on more severe injuries. Subjects who remembered being enrolled had an average ISS score of 17.1, those who didn’t remember had 16.5, and those not sure had 16. Subjects who remembered being consented had an average ISS of 16.9, while those who did not have a score of 19, and those who were not sure had an average score of 15.3.
AIS Grade Per Subject
Color signifies the difference in time.
Comparison to other clinical settings
Our findings contrast sharply with studies in other populations. In a study of cancer patients, 90% of 207 respondents reported being well-informed and satisfied with the consent process. 23 This differs significantly from our study’s 27% response rate, out of which only 47% recalled being consented at all. The primary difference is likely the context; in acute SCI, consent occurs amidst extreme physical and psychological distress, concurrent with life-saving interventions that can overwhelm a person’s ability to process and retain complex information. Thus, while it is important for ethical and regulatory purposes to uphold the tenets of the consent process regardless of study design and target population, optimization of the informed consents conducted in the early timeframe after SCI merits further attention.24,25
Our results are, however, consistent with findings from other acute procedural settings. A 2022 study of patients undergoing cardiac catheterization found that 38% did not recall giving written consent when surveyed shortly after the procedure. Furthermore, of those who did recall giving consent, only 17.5% could name even one of the most frequent risks involved. 17 This highlights a crucial distinction. Even when the act of consenting is remembered, the comprehension of vital information, which is a cornerstone of truly informed consent, is often profoundly lacking. This is mirrored in our own data, where participants inaccurately recalled whether they were in an interventional or observational study arm.
Challenges to informed consent in acute trauma
Applying the Belmont Report’s 26 core principles of respect for persons, beneficence, and justice is exceptionally challenging in the acute SCI setting. Regarding information and comprehension, the consent process involves communicating complex details about neurobiology, critical care, and experimental interventions. It is difficult to distill this information into simple concepts, especially when the injured person is experiencing emotional, psychological, and physiological distress that can overwhelm their ability to comprehend and retain information. Furthermore, voluntariness is compromised by the high probability of “therapeutic misconception,” where patients may hold unrealistic expectations that lead them to consent without sufficient understanding. 27 The patient’s dependence 28 and vulnerability in this state can also make them susceptible to coercive influence, potentially leading them to accept higher-risk interventions than they might otherwise.
Alternative methods of consent in acute trauma
Utilizing an Internal Review Board (IRB)-approved verbal consent method may be one way to address the comprehension/retention difficulties and the time constraints that affect the informed consent process in an acute setting. Kashur et al. compared patient acceptance and comprehension between verbal and written consents in acute myocardial infarction patients. 29 They reported that patients understood verbal consenting adequately as compared with written consents. Additionally, subjects who consented verbally reported feeling “less pressure” during the consent process as compared with the subjects provided with the written consent. Considering the delicate condition of patients in acute settings and the time urgency, this is an important possibility to consider in this population.
Barriers to effective informed consent
Language barriers have been known to pose a potential obstacle to informed consent. 30 In our study, the two non-English speaking participants were consented using a Spanish form with professional interpreters. While both remembered being consented, only one remembered being enrolled in the study after consent, and the other was unsure. This suggests that language-concordant methods can be effective in mitigating barriers to effective informed consent.
The value of continuous and culturally considerate consent
Given these challenges, a single-event informed consent may be insufficient. In our survey, there was a modest difference between the number of subjects who remembered being consented (47%) versus being enrolled (60%). We believe this difference reflects the effect of multiple follow-up visits and communications, which likely reinforced the memory of being enrolled in the study.
This finding strongly supports implementing a continuous consent process. This approach, which involves revisiting consent information and verifying understanding at multiple points, could significantly improve a participant’s comprehension and sustained awareness of their research involvement. 31
Furthermore, it is critical to address the importance of cultural considerations. Attention to the nuances of culture and differences in perception among cultural groups is essential to achieving truly informed consent, particularly in a highly multicultural metropolitan area such as that served by the University of Miami and Jackson Memorial Hospital.
Recollection rates correlate with future research interest
Our results suggest a link between a participant’s ability to recall the initial consent and enrollment process and their subsequent interest in future research.
Of the 14 subjects who remembered being consented for the original studies, 11 (78.6%) said they have later looked for more clinical trials. In contrast, of the 16 subjects who didn’t remember being enrolled or were not sure, only 4 (25%) expressed the same interest. This suggests that participants who remembered being enrolled were significantly more likely to have looked for other clinical trials (78.6%) compared with those who did not remember or were unsure (25%).
These results indicate that ensuring a durable memory of research participation may not only be ethically important but could also foster a community of individuals more engaged with the research enterprise.
Limitations
This study has several important limitations. First, as a single-institution survey with a small sample size and a low response rate (27%), the generalizability of our findings is limited. The low response rate also reflects the difficulties in maintaining long-term contact with this population. The challenges of recruiting for acute SCI trials are well-documented.32–34 Major challenges include short enrollment time windows in acute studies, the 24/7 study team coverage needed to identify all potential subjects, and the difficulty of obtaining proper consent amidst the multiple converging events that characterize the early acute care setting. A review of the regulatory challenges in emergency settings by Villarreal et al. highlights the unique realities of conducting research in acute patients. 32 Our acute SCI studies have encountered nearly all of the 15 regulatory challenge areas addressed, in particular: enrollment/recruitment, patient perception, Human Subjects Protection, LAR, and informed consent. Mitigating these obstacles requires a high level of logistical organization and cooperation, but even when studies are conducted at a high professional level, the distress of acute and severe injury may affect subjects’ level of reminiscence, which we aimed to assess.
Second, and most importantly, the study design contains significant confounding variables. The wide range in time from injury to survey (2010–2022) is a major confounder, as recall naturally degrades over time. This makes it difficult to directly compare the recollection of a participant who consented 12 years ago to one who consented 2 years ago. Similarly, the broad age range of participants (20 to > 59 years) is another significant confounder, as cognitive and memory functions can vary with age. These factors prevent us from drawing definitive conclusions about the causes of poor recall but highlight that recall is indeed inconsistent across different patient groups and timeframes. While our findings are therefore exploratory, they strongly suggest that the traditional, single-point-in-time consent model is fragile and warrants re-evaluation in the acute SCI population.
Conclusion
Clinical research in acute SCI is essential for advancing patient care. However, our findings indicate that the informed consent process requires improvement. A substantial portion of participants in our study did not recall being consented or enrolled. While this may be an unavoidable consequence of the physical and psychological trauma of the injury, it highlights a critical vulnerability in the ethical oversight of research. Practices such as continuous consent, which involve repeated engagement with the participant, may increase understanding and retention, better upholding the principle of respect for persons and potentially fostering greater long-term engagement in research.
Transparency, Rigor, and Reproducibility Statement
The study was submitted and approved by University of Miami, Miller School of Medicine IRB under study number: 20230243, effective date March 14, 2023. Sample size was 119 subjects enrolled in 3 clinical trials previously conducted at the University of Miami, department of Miami Project to Cure Paralysis: (1) Systemic Hypothermia in Acute Cervical Spinal Cord Injury—A Prospective Case Controlled Study; (2) NACTN Registry; and (3) Systemic Hypothermia in Acute Cervical Spinal Cord Injury—A Prospective, Multi-center Case Controlled Study. Data regarding subjects’ LOI, AIS grade, and whether subjects were enrolled or received treatment were used from databases of these three studies. All subjects have been called at least three times, and the questionnaire was administered. Thirty-two subjects agreed to complete the questionnaire. Statistical analysis was performed via nonparametric Kruskal–Wallis test. Deidentified data are available upon request.
Authors’ Contributions
G.J., P.G., and S.S.B. designed the protocol. G.J. and P.G. designed the questionnaire. G.J. composed the initial draft. J.T.Y. conducted the survey and assisted with statistical analysis. A.R.B., D.M., P.G., A.D.L., and S.S.B. made revisions. J.D.G. provided critical revisions and guidance on article development.
Footnotes
Acknowledgment
The authors thank Dr. Rizaldi Ahmad Fadli for assisting with statistical analysis, conducted together with G.J. and J.T.Y.
Author Disclosure Statement
The authors have no competing interests to disclose.
Funding Information
There was no funding provided for this research.