Banking the Future: Adolescent Capacity to Consent to Biobank Research

Abstract

Adolescents are an important population to represent in biobanks. Inclusion of biospecimens from adolescents advances our understanding of the long-term consequences of pediatric disease and allows the discovery of methods to prevent adult diseases during childhood. Consent for biobanking is complex, especially when considering adolescent participation, as it brings up issues that are not present with general clinical research. The development and successful implementation of an adolescent capacity assessment tool applied specifically to biobanking can potentially provide researchers and clinicians with contextualized information on participants’ understanding, appreciation, reasoning, and voluntary choice for biobanks. This tool would enhance current studies looking at the role of shared decision-making in biobanking, as well as provide a formal measurement when considering decisions around pediatric and adolescent biobanking participation. This study adapted the MacCAT-CR for use with a hypothetical adolescent biobank study and examines predictors of MacCAT-CR scores on healthy and chronically ill adolescents.

Adolescents are an important population to represent in biobanks because research with their biospecimens can advance understanding of the long-term consequences of pediatric disease and allow the discovery of methods to prevent adult diseases during childhood. A 2012 national survey of biobanks found that 44% stored biospecimens collected from participants under the age of 18.¹ Biobanks contain various types of biospecimens, including blood, tissue, feces, urine, cell cultures, and saliva.² Biobanking involves the long-term storage of biospecimens, some that may have a code attached linking them to identifiable or deidentified personal and health information of the individual donors.

Because, at the time that biospecimens are collected, it is often impossible to know about the type of studies that will be conducted with them and what scientific methods will be used, consent for biobanking raises a host of ethical issues,³ especially when biospecimens are obtained from legal minors whose biospecimens may not be used until after they become legal adults. Biobanking brings up unique issues in consent that are not present with general clinical research. For example, while it is easy to categorize the risks of collecting biospecimens for biobanks, it is impossible to know the risks to individuals’ privacy and confidentiality, particularly if their genetic information will be analyzed and shared with researchers. These concerns are even more important in adolescent populations due to the evolving nature of adolescents’ capacity to provide valid informed consent⁴ and to questions about whether, how, and when they should be contacted for consent for use of their biospecimens after they become legal adults.⁵

“Capacity to consent” refers to an individual’s ability to make an informed decision about participation in research or treatment.⁶ We draw from Applebaum and Grisso’s⁷ framework for capacity to consent, which lays out four elements that define the capacity. To consent, an individual must be capable of (1) understanding the information relevant to make a decision, (2) appreciating how the decision will impact them personally, (3) manipulating the information rationally and reasoning, and (4) communicating a voluntary choice. Because of the importance of consent in biobanking, it will be important to systematically assess capacity for populations with emerging capacity. Capacity is a continuous construct, and having “decision-making competence” is typically decision specific. The MacArthur Competency Assessment Tool for Clinical Research (MacCAT-CR) is a well-known comprehensive assessment tool initially designed for adults with impaired capacity.⁸ Cognition and brain research suggest that adolescents make adult-style medical decisions by approximately 14 years of age, and a recent study of pediatric capacity to consent using a simplified MacCAT-CR showed that by 12 to 13 years of age, adolescents have capacity to make decisions about general clinical research.⁹

The principle of respect for persons suggests that those capable of self-determination should be directly involved in the consent for their participation in research. Yet according to federal human research regulations, regardless of cognitive capacity to provide valid informed consent, individuals who have not reached the legal age of majority (18 in most states) cannot give consent to participate in research. Instead, their parents or guardians must provide permission for them to participate in it. When appropriate, assent from legal minors may also be required.¹⁰ The regulations, however, do not take into account adolescents’ evolving capacity to consent to research participation,¹¹ nor do the regulations reflect studies showing that adolescents themselves have expressed interest in having a direct say in the long-term banking of their biospecimens with attendant linkages to their personal health information.¹² Moreover, a recent study by Murad and colleagues found that, of the adolescents they interviewed, 89% thought that pediatric biobank participants who did not provide assent for participation should be recontacted once they reach the age of majority.¹³

For our study, we used a modified version of the MacCAT-CR to investigate adolescent capacity to consent to biobank participation. The development and successful implementation of an adolescent capacity assessment tool applied specifically to biobanking can provide researchers and clinicians with contextualized information on adolescents’ understanding about, appreciation of, reasoning about, and voluntary choice for biobank participation. This tool would build on previous studies that examined the role of shared decision-making in biobanking,¹⁴ as well as provide a formal measurement to use when considering decisions about adolescent participation in biobanks. In our study, we hypothesized that chronic disease may increase adolescents’ capacity to consent, because of increased familiarity with medical procedures. Additionally, health literacy has been identified as one of the most important emerging constructs in health communication and is currently viewed as a key factor in determinants of health care utilization and outcomes. Cognitive characteristics, such as health literacy, as well as contextual factors, such as socioeconomic status and age, were also evaluated.

STUDY METHODS

Our analysis of adolescents’ capacity to consent to biobanking research draws from a larger study that examined adolescent capacity to consent to different types of research.¹⁵ Inclusion criteria included being between the ages of 12 and 24 years old and having the ability to read and speak English. Exclusion criteria included being obviously cognitively altered due to use of alcohol or illicit drugs and having a known severe cognitive impairment, as reported by a parent.

The age range for our inclusion criteria reflects the variety of different age ranges and names ascribed to the developmental period we are interested in. The specific names and age ranges for this group differ based on profession, organization, and geographical location. The sands frequently shift as new names and ranges are ascribed to this group. As our study focuses on cognitive capacity, we needed to include the broadest of these time frames to better understand differences within this group. Our definition of “adolescence” fits within a number of overlapping frameworks—some starting as early as age 10 and ending as late as the early 30s. For instance, people 10 to 24 are now referred to as “young people” by the World Health Organization, a relatively new overlapping or nested definition that organization has created. With regard to the Society for Adolescent Health and Medicine, there is still much internal debate on what numbers should be used in research settings to delineate adolescents, emerging adults, young adults, and adults. Specifically, they now use “young adults” to denote people 18 to 25. Because we are concerned less with these definitional issues and more with identifying issues related to capacity to consent, we focused on the original meaning and intent of the term “adolescence” borne out by the literature on brain development. Literature in this area generally accepts that there are no hard cutoffs to distinguish the stage of adolescence, a sharp contrast from the legal conception in which adulthood is conferred by age. For our research and this study, we utilize the broadest sense of the term that focuses on the transitional phase from puberty into typical adult roles.

We used a broad community-based strategy to recruit participants with a broad range of social and life experiences. Using snowball sampling, we contacted teachers and administrators, youth programs, and medical clinics serving adolescents and young adults for permission to recruit participants from these sites. The institutional review board at Indiana University—Purdue University Indianapolis approved the research protocol. For participants under 18 years of age, parents or guardians provided permission, and the adolescents provided their assent.

For each participant, structured information on demographic, developmental, medical history, and health literacy was collected. Study participants then participated in a random ordered simulated consent process for the Indiana Biobank, a sexual health screening study, and a pharmaceutical study investigating a new migraine medication. The biobanking protocol was presented, as if participants would be contributing their biospecimens and allowing their medical records to be accessed, using the actual Indiana Biobank forms.¹⁶ After the informed consent process, the interviewer administered a revised version of the MacCAT-CR specifically designed for this study to assess capacity to consent to a biobank. Participants were given a $20 gift card.

Outcome measure.

While there are data supporting the use of a simplified MacCAT-CR with children from middle childhood through adolescence,¹⁷ there were no adolescent-specific studies. The MacCAT-CR is a semistructured tool that has four subscales: understanding of the research, appreciation of the individual’s own situation, rational manipulation of information, and evidence of choice.¹⁸ Use of the MacCAT-CR with chronically ill adults has been shown to be a reliable and effective way to assess patient capacity for clinical trials.¹⁹ Effectiveness with chronically ill participants was important to assess, as this is the pediatric population most likely to participate in clinical trials research.²⁰ The tool uses semistructured, open-ended questions to assess participants’ ability to comprehend study-specific information based across four subscales mapping onto the four domains of consent: understanding, appreciation, reasoning, and voluntariness. Each question is scored on a 0–2 scale where 0 indicates that the respondent did not correctly answer any portion of the question, 1 indicates that they were able to identify some portion of the correct answer, and 2 indicates that they were able to provide a response that captured all ideal aspects of a response. The MacCAT-CR is continuous, rather than binary (consent/no consent), leaving it up to the researchers to set the cutoff based on the risks, benefits, population, and contexts of the research decision. Since each version of the MacCAT-CR needs to be study specific in order to assess capacity, we consulted multiple individuals with topic expertise in developing the rubric for scoring answers on the 0–2 scale. We adapted the MacCAT-CR based upon pilot testing with adolescents, reducing repetition, integrating the tool into the consent process, and asking adolescents to explain terms repeated verbatim from the consent form to assess understanding²¹ (see table 1).

Table 1.

MacCAT-CR Items: Comprehensive Breakdown of the Modified MacCAT-CR

Understanding
What is the purpose of the research project I described to you?
How long will the research project last?
What sorts of things will be done with people who agree to be in the study?
What sorts of things will people have to do if they agree to be in the study?
Do you believe this study is primarily for research or primarily for treatment?
What might doctors learn about diseases if people decide to be in this research project?
In what way might people who volunteer be better off by being in this research project?
Is it possible that being in this study will not have any benefit to you?
What unpleasant side effects might people experience in this study?
What uncomfortable things are done to people in the study?
Who will pay for your medical care if you are injured as a direct result of participating in this study?
What will happen if a person refuses to be in the research project?
If you withdraw from this study, will you still be able to receive regular treatment?
Appreciation
Do you believe that you have been asked to be in this study primarily for your personal benefit?
What makes you believe that this was/wasn’t the reason you were asked?
What do you believe would happen if you were to decide not to be in this study?
What makes you believe that this would happen?
Reasoning
What makes you want to consider participating in this study?
How might [risk or benefit] affect your daily life?
Expressing a choice
Now that we’ve discussed everything, do you think you want to participate?

Predictor measures.

Health literacy has been identified as one of the most important emerging constructs in health care communication²² and is currently viewed as a key factor in determinates of health care utilization and outcomes.²³ The Rapid Estimate of Adult Literacy in Medicine (REALM) is one of the most-used tools to assess health literacy.²⁴ The REALM consists of 66 items and produces a raw score than can be quickly converted into a reading-grade equivalency.²⁵ Socioeconomic status was measured by the Family Affluence Scale II (FASII), a self-reported measure validated for use in adolescents.²⁶ The FASII asks questions that average adolescents would know the answer to, such as whether they have their own bedroom, the number of cars and computers in their household, and how many family vacations were taken last year. The scale ranges from 0–9, with 0–2 corresponding with low affluence, 3–5 middle affluence, and 6–9 high affluence, using international comparators.²⁷ We measured chronic illness using participants’ responses to three items that asked if they had a chronic illness, took daily medicine, and had ever spent the night in the hospital.

We first examined bivariate predictors of capacity using correlation, t-tests, and ANOVA in R. We then used regression analysis to evaluate predictors of both overall MacCAT-CR scores and subscales for understanding, appreciation, and reasoning. Predictors significant in bivariate analysis were entered into the multivariate models.

STUDY RESULTS

We enrolled and interviewed 92 individuals (with a mean age of 17 +/−3 years) who were mostly female (66%). Fifty-four percent were white, and 42% African American. Participants’ health literacy was approximately eighth- to ninth-grade reading level (REALM mean = 60, standard deviation [SD] = 6), similar to levels found in studies of adult health literacy.²⁸ The mean family affluence score was 6.5 (SD = 1.7), corresponding to upper middle class on this global scale ranging from 1–9. Twenty-four percent of participants reported a chronic illness, half reported taking daily medicine other than vitamins or supplements, and 36% had stayed overnight in the hospital due to being sick (see table 2).

Table 2.

Population Characteristics and Capacity to Consent to Biobank Research

Characteristics	Observed range	Mean (SD) or n (%)
Age	12–24	16.9 (2.7)
Female gender		65.9%
Ethnicity
white		54.3%
African American		42.4%
Latinx		2.2%
other (mixed race, Asian, etc.)		1.1%
Family affluence	1–9	6.5 (1.7)
Health literacy (REALM)	39–66	60.3 (6.3)
Chronic illness		24.2%
Takes a daily medicine		50.5%
Ever stayed in a hospital		36.3%
Capacity to consent to research
Overall MacCAT-CR score	12–36	30.1 (5.0)
Understanding subscore	6–20	16.3 (3.1)
Appreciation subscore	2–6	5.1 (1.1)
Reasoning subscore	1–8	6.7 (1.6)
Voluntariness subscore	2–2	2 (0.0)

Participants had a mean of 16.3/20 (SD = 3.0) on the understanding subscale, 5.1 (SD = 1.1) on the appreciation subscale, and 6.7 (SD = 1.5) on the reasoning subscale. All participants said their participation would be voluntary, answering affirmatively to the questions, “Do you have a choice to participate in the study?” and then, “Do you think you would want to participate in the study?” (see table 2).

In bivariate analysis, the overall MacCAT-CR score was associated with age (r = 0.41, p <. 01), family affluence (r = .36, p < .01), and health literacy (r = .69, p < .01). We found no associations with gender or chronic illness. We observed similar results for the other subscales. Understanding was associated with age (r = 0.37, p < .01), family affluence (r = .28, p < .01), and health literacy (r = .65, p < .01). Similarly, appreciation was associated with age (r = 0.39, p < .01), family affluence (r = .32, p < .01), and health literacy (r = .57, p < .01). Reasoning was associated with age (r = 0.27, p < .01), family affluence (r = .36, p < .01), and health literacy (r = .48, p < .01).

In multivariate analysis for overall capacity (see table 3), health literacy and family affluence were significant predictors. Age was no longer associated with capacity after controlling for health literacy and family affluence in any of our models. Our final model using family affluence and health literacy accounted for 56% of the variance in overall capacity (R²).

Table 3.

Multiple Linear Regression Models for Capacity to Consent to Biobanking

Predictor		B (SE)	β	t
Overall MacCAT score
Health literacy (REALM)		.52 (.06)	.66**	9.36
Family affluence (FAS)		.85 (.21)	.30**	4.17
	R2			.56
	F			56.7
Understanding MacCAT subscore
Health literacy (REALM)		.31 (.04)	.63**	8.18
Family affluence (FAS)		.39 (.14)	.22*	2.82
	R2			.47
	F			39.8
Appreciation MacCAT subscore
Health literacy (REALM)		.10 (.01)	.55**	6.66
Family affluence (FAS)		.18 (.05)	.27**	3.30
	R2			.40
	F			29.9
Reasoning MacCAT subscore
Health literacy (REALM)		.11 (.02)	.45**	5.17
Family affluence (FAS)		.30 (.08)	.32**	3.70
	R2			.33
	F			22.0

^*p < .01

^**p < .001

ADOLESCENTS AND BIOBANKS—TIME FOR A NEW CONSENT PARADIGM

Adolescents demonstrated a high capacity to consent to biobanking research, with capacity assessments similar to those reported for normal adults in similar studies using the MacCAT-CR.²⁹ This finding suggests that adolescents probably have the capacity to give valid informed consent to provide their biospecimens to a research biobank. There may be no need to contact them for consent once they become legal adults, since our study findings demonstrate adolescents’ adequate understanding of biobanking information, appreciation for how biobanking would affect them personally, and ability to apply reasoning skills such as weighing risks and benefits and creating a logical argument for their choice.

A second key finding is that adolescents in our study, regardless of age, all felt that they were capable of making a voluntary choice for themselves. None felt that they needed a parent or other adult to provide permission or make the decision for them. This finding supports the idea that we should respect their self-determination.

The two strongest predictors of all aspects of capacity were family affluence and health literacy—in fact, when these two are entered into the model for overall capacity, age is no longer significant. The significance of family affluence points to the importance of environmental influences on adolescent capacity, and we hypothesize that it is an important marker for broader exposures to scientific developments and institutions (in other words, the concepts of research and biobanks may be less foreign to adolescents from wealthier families). Health literacy, unlike age, is a modifiable factor, and lower health literacy can be addressed in the consent process and communication about research.

Having a chronic illness, taking daily medicine, and having stayed overnight in the hospital one or more times due to illness were not found to be significant factors in capacity to consent. However, this may have been an issue of statistical power—only 21 participants (23%) reported having a chronic illness. It could also be a function of measurement, as all chronic illnesses are not the same. Some illnesses (such as childhood cancer) require a large amount of contact with the health system; and others (like mild intermittent asthma), less contact. Most of the chronic illness in our sample was asthma or diabetes. Future studies will need to assess severity of illness and health care utilization.

This study has limitations. By setting 12 years as the minimum age for participation, we were unable to identify the age at which legal minors were unlikely to have capacity to consent to biobank participation. The MacCAT-CR was developed for adults with waning capacity and is sensitive to the cognitive difficulties of aging adults (such as memory loss) or adults with severe mental illness (such as a diminished sense of reality). It is likely less sensitive to the cognitive challenges of the developing adolescent brain, such as lack of decision-making experience, difficulties with perspective taking, different risk assessment and cognitive control, and difficulties with distraction and emotion.³⁰

The potential health and scientific benefits of adolescent biobanking are significant. The concept of appropriate access in pediatric research highlights the need for investigators to design supportive consent processes and study procedures tailored to the needs of minors.³¹ The Society for Adolescent Health and Medicine position statement on research with adolescents provides practical, developmentally tailored recommendations around consent, assuming adolescent capacity in lower risk research, and considering the use of a capacity assessment for higher risk research.³² Our findings are consistent with this developmental approach.