Parental Engagement in Consent Processes for Enrollment in Biomedical HIV Prevention Trials: Implications for Minor Adolescents' Willingness to Participate

Abstract

Purpose:

Minor adolescents are often excluded from HIV prevention clinical trials due to unresolved ethical issues. Their underrepresentation in research leads to delayed access to new HIV prevention approaches. We examine the relationship between consent procedures, trial features, demographic and social characteristics, and minor adolescents’ willingness to participate (WTP) in biomedical HIV prevention research.

Methods:

We recruited 14–17 year-olds at risk of HIV for this quasi-experimental study. Adolescents were randomly assigned to: 1) self-consent, 2) adult permission required, or 3) parental permission required and underwent simulated consent procedures for two types of HIV prevention trials. They rated likelihood of participating in each study if offered the opportunity, and completed a survey with demographic, social and behavioral measures.

Results:

129 adolescents with diverse identities and socioeconomic status enrolled. Among the 58% of participants who identified as lesbian, gay, bisexual, transgender, or queer (LGBTQ), 76% were out to at least one parent/guardian (outness). Mean WTP was 3.6 (out of 5; 5=definitely would participate) across all participants and both trial types. We found no evidence of an association between WTP and consent condition, LGBTQ identity, or outness. However, medical mistrust, communication with parents, and concern about HIV were associated with WTP.

Conclusions:

Our results suggest adolescents are willing to participate in HIV prevention trials, and parental involvement in the consent process may not be the most important deciding factor. However, variation in WTP within consent groups, and variation in other significant variables, underscores the need for individualized approaches to recruitment and consent for these trials.

Background

Children and minor adolescents (aged 12–17 years) are underrepresented in clinical trials,^1,2 especially those that focus on sensitive and stigmatizing health problems, like HIV. Approximately 20% of incident HIV infections in the US occur in youth ages 13–24 years old. A disproportionate share of those infections occur in Black men who have sex with men,³ who have an estimated 1 in 2 lifetime risk of an HIV diagnosis.⁴ These data underscore the importance of making new HIV prevention interventions—like pre-exposure prophylaxis (PrEP) and vaccines—accessible to adolescents, which requires including them in HIV prevention trials.

PrEP, which involves taking an antiretroviral medication to prevent HIV, is highly effective when taken consistently and correctly.⁵ In 2012, the US Food and Drug Administration approved oral tenofovir disoproxil fumerate with emtricitabine (TDF/FTC) for PrEP in adults. However, due to lack of data on its safety and tolerability in minors, it was not labeled for use in minors until 2018. The six-year delay in PrEP for adolescents created a missed opportunity to prevent some of the 50,000 HIV infections that occurred in youth during that period of time.⁶ This is a pattern observed in other areas of health research, in which the under enrollment of minors raises concerns about equity in access to therapies that may prevent or treat health problems with lifelong implications for individuals, their partners and families, and the health care system.²

Ethical issues are a key barrier to engaging minors in research on sensitive and stigmatizing health problems.^7,8 Consent to enrollment is one of the most critical complexities. Minors are considered a vulnerable population requiring additional protection from research-related harms that extend beyond that afforded to all human subjects.⁹ Federal regulations identify the minor’s parent or guardian (parents, henceforth) as an additional protection from harm through parent participation in the research consent process.¹⁰ However, parental engagement in consent processes may create risk of harm. Consent forms typically disclose the study’s purpose, eligibility criteria, and procedures. Thus, when parental permission is required for minor participation in clinical trials about sensitive health problems, like HIV, a minor’s previously concealed identities or sexual behaviors may be revealed. Such disclosures can place a minor adolescent at risk of harms ranging from loss of social and financial support to violence.¹¹

In surveys and focus groups, lesbian, gay, bisexual, transgender and queer-identified (LGBTQ) adolescents indicate they would not enroll in a study if it would “out” them to their parents.^12,13 While federal regulations allow minor consent for biomedical research in limited situations, state minor consent laws vary widely¹⁴ and federal regulations have been applied inconsistently in adolescent HIV biomedical prevention research.⁸ Tensions between minors’^11,13 and parents’ interests,¹⁵ and those of investigators and their institutions⁹ have led ethicists to call for a reconsideration of research vulnerability.¹⁶ They argue for a shift from group membership (e.g. age group) as an indicator of vulnerability to consideration of an individual’s circumstances and capacity to consent.^16,17 Starting from approximately age 12, adolescents consistently demonstrate capacity to understand research concepts and make decisions about participation similar to those made by adults.¹⁸

Considering the evidence cited above, HIV clinical trials networks developed protocols with different consent models for minors. For example, the Adolescent Medicine Trials Network for HIV/AIDS Treatment and Prevention (ATN) allowed minor adolescents to self-consent for enrollment in its open-label safety and tolerability study of TDF/FTC as PrEP (ATN 113).¹⁹ The Microbicide Trials Network, however, required minor adolescents to have parental permission to enroll in its Phase IIa trial of a vaginal ring impregnated with the antiretroviral dapivirine (MTN 023).²⁰ The trials enrolled different populations and tested drugs in different stages of development that had different delivery mechanisms, which may account for differences in the approach to minor consent. There were challenges to recruitment in both protocols, although the ATN 113 trial ultimately provided the data to support FDA approval of TDF/FTC as PrEP for minors.

In this study, we examine the relationship between minors’ WTP in biomedical HIV prevention research and several factors discussed above, including: consent conditions, gender, sexual orientation, and clinical trial features. We also explore the relationship between WTP and several sociobehavioral factors we hypothesized may be relevant, including: communication with parents about sex and relationships, perceptions of HIV risk, disclosure of sexual orientation to parent(s), and medical mistrust.

Methods

Study Design

The study approach is described in depth elsewhere.²¹ Briefly, we used a quasi-experimental design to examine WTP from the perspective of minor adolescents at risk of sexual acquisition of HIV and from the perspectives of parents of minor adolescents. This paper focuses on adolescent participants only.

Human Subjects

All research activities were reviewed and approved by the Indiana University IRB (Protocol # 1608179327), which served as the single IRB of record. It granted a waiver of parental permission for minors.

Setting

Participants were recruited from four research sites: Baltimore, Maryland; Chicago, Illinois; Denver, Colorado; and Tampa, Florida. The sites were chosen for their history of participating in HIV biomedical trials that engaged minor adolescents and young adults, their demographic diversity, and the prevalence of HIV among adolescents and young adults.

Population

Adolescents were eligible if they were aged 14–17 years, inclusive, spoke English, were HIV negative or unsure of HIV status, and had engaged in at least one sexual behavior associated with risk of acquiring HIV in the 12 months prior to enrollment. Sexual risk behavior criteria were: unprotected vaginal or anal sex with a male; protected anal or vaginal sex with three or more males; sex with a male for money, gifts, shelter, or drugs; sex with a male and history of sexually transmitted infection; and/or sex with someone living with HIV.

Recruitment

We recruited participants using printed materials posted at clinics and distributed at events (e.g., Pride festivals), social media, and face to face recruitment in adolescent health clinics and community organizations. Participants received $50 in cash or gift cards for completion of the study visit and were offered help with transportation.

Study Procedures

Study visits took place at clinical and community facilities. Figure 1 provides a summary of procedures, which are described elsewhere.²¹ Briefly, participants were randomized to one of three consent conditions (CC): CC1 - minor self-consent (ability to consent on own), CC2 - adult permission required (ability to choose between asking parent/guardian for permission or having a study appointed ombudsperson help with decision making), and CC3 - parent permission model. After randomization, they underwent a simulated consent process for each of two hypothetical trials. One trial was modeled after ATN 113 (see above), the other was modeled after HPTN 077, a Phase IIa trial of long-acting injectable cabotegravir (CAB) as PrEP. We selected HPTN 077 because it had features similar to MTN 023 (described above) but was applicable to natal males as well as natal females. Consent procedures for both hypothetical trials were tailored to the adolescent’s assigned consent condition. After the simulated consent process for each study, the adolescent completed a computer assisted self-interview to: assess their willingness to participate in each study if offered the opportunity and collect demographic, attitudinal, and behavioral measures (see next section).

Figure 1.

Study Procedures

Sample size and power calculation

Sample size was calculated based on a linear regression model with consent condition as the independent variable of interest. To reach an 80% power for detecting a small-medium effect size (f²) of 0.084 using an across-participant comparison, target enrollment was approximately 144 adolescents.

Measures

Participants completed demographic, behavioral, and attitudinal measures described in detail elsewhere.²¹ Briefly, measures included: demographic characteristics (sex, gender identity, racial and ethnic identity, education, employment); disclosure of LGBTQ identities to a parent (“Outness”); socioeconomic status (Family Affluence Scale-III; “FAS-III”);²² social support (Multidimensional Scale of Perceived Social Support, family subscale “Family Social Support”);²³ medical mistrust (The Group-Based Medical Mistrust Scale, modified, “Medical Mistrust”);²⁴ parent-teen communication about relationships and sexuality (“Communication”);²⁵ and concern about acquiring HIV in the future (“Concerned about acquiring HIV”).

Data Collection & Management

Participants were screened online or in person through a survey programmed in Qualtrics, a Web-based data collection system appropriate for use with sensitive data. Eligible and enrolled participants completed quantitative survey measures in the Qualtrics system.

Data Analyses

The primary outcome of interest was mean WTP (across the two study types) with categorical consent condition as the independent term, adjusting for site and sex assigned at birth. WTP was also analyzed using a repeated measures model with similar terms listed above and with a term for study type.

Secondary analyses of associations with mean total WTP were conducted for variables of interest by first fitting bivariable regression models of mean total WTP with each variable of interest. For the 121 subjects with completely non-missing data, the best fit multivariable model was obtained by selecting the model with smallest Akaike’s Information Criteria (AIC) from among models which included variables that were statistically significant (p<0.05) in bivariate models (site, medical mistrust and LGBTQ with disclosure to parent) with all possible combinations of other subject characteristics. This variable approach has been shown to be consistent, where consistency is defined in terms of model selection^.26 All analyses were performed in SAS Software version 9.4 (copyright 2106), SAS Institute, Cary NC. USA.

Results

Demographic & Sociobehavioral Characteristics

It was challenging to locate and enroll minor adolescents who met our inclusion criteria for risk of sexually acquired HIV infection. This is consistent with the very few biomedical HIV prevention trials that included minors. For instance, the target enrollment for the ATN 113 trial was approximately 100, but the final sample enrolled from 6 U.S. cities was 78.¹⁹ Despite general challenges enrolling minors with self-reported risk for sexually acquired HIV infection, more than 1,100 adolescents completed the study screening survey; 20% were eligible, and 58% of eligible adolescents enrolled in the study for a total sample size of 129, which was sufficient to power planned analyses. Participants were racially and ethnically diverse (Table 1). Slightly more than half of the sample (54%) identified as LGBTQ, with 9 identifying as transgender or gender diverse. Among LGBTQ adolescents, 73% were out to at least one parent. Socioeconomic status was varied, with Family Affluence Scale III scores ranging from 0.036–0.929, with a sample mean of 0.568 (standard deviation = 0.205), indicating medium socioeconomic status.

Table 1.

Adolescent Characteristics

	Overall N=129	Consent Condition 1 Minor self-consent N=43	Consent Condition 2 Adult permission N=45	Consent Condition 3 Parental Permission N=41
	N (%)	N (%)	N (%)	N (%)
Demographics
Age
14–15	17 (13.2)	8 (18.6)	4 (8.9)	5 (12.2)
16–17	112 (86.8)	35 (81.4)	41 (91.1)	36 (87.8)
Sex
Female	69 (53.5)	23(53.5)	24(53.3)	22 (53.7)
Male	60 (46.5)	20 (46.5)	21 (46.7)	19 (46.3)
Gender Identity
Female	69 (53.5)	22 (51.2)	24 (53.3)	23 (56.1)
Male	47 (36.4)	15 (34.9)	18 (40.0)	14 (34.1)
Transgender, gender queer, other	13 (10.1)	6 (14.0)	3 (6.7)	4 (9.8)
Sexual Orientation
Straight or heterosexual	54 (41.9)	20 (40.8)	22 (48.9)	12 (34.3)
Gay, queer, same gender loving	36 (27.9)	12 (24.5)	11 (24.4)	13 (37.1)
Bisexual	33 (25.6)	14 (28.6)	10 (22.2)	9 (25.7)
Other	6 (4.7)	3 (4.1)	2 (4.4)	1 (2.9)
Race
Black or African American	67 (51.9)	19 (44.2)	27 (60.0)	21 (51.2)
White	30 (23.3)	9 (20.9)	7 (15.6)	14 (34.2)
Other	32 (24.8)	15 (34.9)	11 (24.4)	6 (14.6)
Ethnicity
Hispanic or Latino	30 (24.0)	11 (26.8)	11(25.0)	8(20.0)
Lives with Legal Guardian
Yes	112 (86.8)	38 (88.4)	39 (86.7)	35 (85.4)
No	11 (8.5)	4 (9.3)	3 (6.7)	4 (9.8)
N/A^	6 (6.5)	1 (2.3)	3 (6.7)	2 (4.9)
Employed *
Yes	51 (39.8)	13 (30.2)	24 (54.6)	14 (34.2)
Socioeconomic Status 1
Family Affluence Scale III index score	0.57 (0.21)	0.54 (0.21)	0.59 (0.20)	0.57 (0.21)
Sociobehavioral Scales
	N (%) or Mean (SD)	N (%) or Mean (SD)	N (%) or Mean (SD)	N (%) or Mean (SD)
Family Social Support
Low (1–2)	15 (11.6)	4 (9.3)	5 (11.1)	6 (14.6)
Moderate (3–5)	44 (34.1)	15 (34.9)	19 (42.2)	10 (24.4)
High (6–7 )	70 (54.3)	24 (55.8)	21 (46.7)	25 (61.0)
Communication 2
Communication index score	1.9 (0.6)	2.0 (0.6)	1.8 (0.6)	2.0 (0.6)
Concern About Acquiring HIV
Disagree	21 (16.3)	10 (20.4)	7 (15.6)	4 (11.4)
Neutral	22 (17.1)	12 (24.5)	8 (17.8)	2 (5.7)
Agree	86 (66.7)	27 (55.1)	30 (66.7)	29 (82.9)
Medical Mistrust 3
Group Based Medical Mistrust Scale index score	2.3 (0.9)	2.3 (1.0)	2.3 (0.9)	2.3 (0.9)

^*p<0.05

^{^}N/A = Does not have a legal parent or guardian, by self-report

¹Score 0–1, where 0-.2000=low SES; .2001-.8000 (medium SES); .8001–1 (high SES)

²Score 1–3, averaged across 5 items measuring frequency of communication with parent over prior 6 months. 1=never, 2=once or twice, 3=many times

³Scale 1–5, with higher scores indicative of greater mistrust of medical research/researchers.

Most participants reported moderate (34%) or high (55%) social support from their families. Mean sexual communication scale score was 1.9 (standard deviation = 0.6), corresponding with discussions about relationships and sexuality with parents “once or twice” in the prior 6 months. Half of participants reported concern about the possibility of acquiring HIV in the future. Approximately 81% of adolescent medical mistrust scores were in the low to neutral level of mistrust range (summary score <3.1 out of 5).

Primary outcomes

When asked how likely they were to join the hypothetical trial if invited to do so at the time of study visit, the combined mean score across all consent conditions was 3.6 (SD = 1.0) for both trials. For each trial separately, the combined mean scores across all consent conditions were 3.6 (SD = 1.1) and 3.6 (SD = 1.1), for ATN 113 and HPTN 077 trials, respectively. For a categorical breakdown of responses for each of the WTP variables by trial and consent condition group, see Table 2. Between consent condition groups, mean scores ranged from 3.6–3.7 for the ATN 113 trial and 3.4–3.8 for the HPTN 077 trial. Neither the model for mean WTP nor the repeated measures model of WTP for each study type showed significant differences between consent conditions (Table 2).

Table 2.

Willingness to Participate by Consent Condition

		Overall	CC1 Minor self-consent	CC2 Adult permission	CC3 Parent permission	P-value
		N=129	N=43	N=45	N=41
WTP, both trials	Mean ± SD	3.6 ± 1.0	3.7 ± 0.8	3.7 ± 1.2	3.5 ± 0.8	0.43
	Median (Min, Max)	3.5 (1.5, 5.0)	3.5 (2.0, 5.0)	4.0 (1.5, 5.0)	3.5 (2.0, 5.0)
WTP for ATN 113 (distribution)	Definitely would not participate	4 (3.1)	2 (4.7)	1 (2.2)	1 (2.4)
	Probably would not participate	16 (12.4)	3 (7.0)	11 (24.4)	2 (4.9)
	Might or might not participate	35 (27.1)	15 (34.9)	5 (11.1)	15 (36.6)
	Probably would participate	41 (31.8)	12 (27.9)	12 (26.7)	17 (41.5)
	Definitely would participate	33 (25.6)	11 (25.6)	16 (35.6)	6 (14.6)
WTP for ATN 113 (mean, median)	Mean ± SD	3.6 ± 1.1	3.6 ± 1.1	3.7 ± 1.3	3.6 ± 0.9	0.79
	Median (Min, Max)	4.0 (1.0, 5.0)	4.0 (1.0, 5.0)	4.0 (1.0, 5.0)	4.0 (1.0, 5.0)
WTP for HPTN 077 (distribution)	Definitely would not participate	3 (2.3)	0 (0.0)	2 (4.4)	1 (2.4)
	Probably would not participate	17 (13.2)	2 (4.7)	6 (13.3)	9 (22.0)
	Might or might not participate	38 (29.5)	17 (39.5)	11 (24.4)	10 (24.4)
	Probably would participate	36 (27.9)	12 (27.9)	8 (17.8)	16 (39.0)
	Definitely would participate	35 (27.1)	12 (27.9)	18 (40.0)	5 (12.2)
WTP for HPTN 077 (mean, median)	Mean ± SD	3.6 ± 1.1	3.8 ± 0.9	3.8 ± 1.2	3.4 ± 1.0	0.14
	Median (Min, Max)	4.0 (1.0, 5.0)	4.0 (2.0, 5.0)	4.0 (1.0, 5.0)	4.0 (1.0, 5.0)

In unadjusted bivariate analyses, several variables were associated with mean total WTP (Table 3). Neutral concern about infection with HIV was associated with higher mean total WTP (p < 0.05). Higher medical mistrust index score was associated with lower mean total WTP (p<0.05). Finally, higher communication index score was associated with higher mean total WTP (p<0.001). Of note, consent condition was not significantly associated with WTP, and therefore was not included in the multivariable analysis.

Table 3.

Bivariable and Multivariable Associations with Mean Total WTP

	Mean Total WTP
	Bivariable Associations		Multivariable linear model
	Mean difference (std error)	95% CI	Mean difference (std error)	95% CI
Age	-0.04 (0.11)	(−0.26, 0.18)
Sex assigned at birth
Male	Ref
Female	-0.21 (0.17)	(−0.55, 0.12)
Site
Baltimore	0.46 (0.26)	(−0.06, 0.97)	1.00 (0.25) ***	(0.51, 1.49)
Chicago	0.43 (0.24)	(−0.03, 0.90)	0.69 (0.26)**	(0.18, 1.21)
Denver	0.04 (0.22)	(−0.40, 0.49)	0.28 (0.21)	(−0.14, 0.69)
Tampa	Ref		Ref
Outness
Not out to parent	0.41 (0.28)	(−0.15, 0.96)	0.55 (0.26) *	(0.04, 1.06)
Out to parent	0.14 (0.18)	(−0.22, 0.51)	-0.19 (0.17)	(−0.54, 0.15)
N/A: Not a sexual minority	Ref		Ref
Socioeconomic Status
FAS-III index Score	-0.20 (0.41)	(−1.01, 0.62)	-0.24(0.47)	(−1.18, 0.69)
Concerned about acquiring HIV
Disagree	0.23 (0.28)	(−0.34, 0.79)	0.22 (0.26)	(−0.30, 0.74)
Neutral	0.57 (0.22) *	(0.13, 1.01)	0.45 (0.22)	(0.03, 0.88)
Agree	Ref
Medical Mistrust
GBMMS index score	-0.20 (0.09)*	(−0.38, −0.02)	-0.38 (0.09)***	(−0.57, −0.20)
Family Social Support
MSPSS Family Subscale score	0.03 (0.05)	(−0.08, 0.14)
Communication
Communication index score	0.45 (0.13) ***	(0.19, 0.72)	0.53 (0.13) ***	(0.26, 0.79)

^*p<0.05

p<0.01

^***p<0.001

Multivariable regression analyses are reported in Table 3, where we present the final best fitting model for mean WTP across both trials combined. In the final models, site was associated with mean WTP, with participants in Baltimore (mean difference=1.00; 95% CI [0.51, 1.49]) and Chicago (mean difference=0.69; 95% CI [0.18, 1.21]) reporting higher WTP on average, compared to those in Tampa. Outness was significantly associated with WTP scores—LGBTQ adolescents who were not out to a parent were more willing to participate in the trials (mean difference=0.55; 95% CI [0.04, 1.06]) compared to those who identified as straight/heterosexual. In a separate analysis of the relationship between outness and WTP (not depicted in Table 3), WTP scores were significantly higher among LGBTQ youth who were not out to a parent compared to those who were (mean difference 0.746; 95% CI [0.23, 01.26]; p=0.005), and we did not find evidence of a difference in WTP scores between LGBTQ youth who were out to parents compared to straight/heterosexual youth. Greater medical mistrust scores were associated with lower WTP (mean difference=−0.38; 95% CI [−0.57, −0.20]). Finally, more frequent communication with parents about relationships and sexuality was associated with higher WTP (mean difference=0.53; 95% CI [0.26, 0.79]).

Discussion

We examined minor adolescents’ willingness to participate in two types of biomedical HIV prevention trials. Our study’s enrollment criteria matched those of two HIV research network biomedical HIV prevention trials, resulting in a sample of minor adolescents at risk of HIV. Overall, adolescents were moderately interested in participation, though there was individual variation in WTP such that a small minority of adolescents indicated they definitely would not participate and approximately 25% indicated they definitely would. We found no evidence for an association between parental or adult involvement in the consent process and WTP across both trials. Our findings are different than those reflected in a large body of prior research which indicates requiring parental permission is a significant barrier to minor adolescents’ participation in sensitive and stigmatizing research. ^27,28 Below we discuss several possible explanations for our results.

Associations between consent condition and WTP: Potential role of study design, disclosure, and imagined benefits of participation

Study design.

First, our sample was restricted to adolescents who were at risk of sexual acquisition of HIV, and they indicated moderate concern about becoming infected with HIV in the future. These adolescents may have been more WTP in prevention research—regardless of parental involvement in the consent process—than a sample that included adolescents with no self-reported risk factors. This may be especially relevant when comparing our results to those of Liu et al.’s meta-analysis in which 13 of the 14 studies included occurred in school-based settings where the adolescents likely had a wider range of behavioral risks (e.g. from no risk or low risk to behavioral risks similar to our participants).²⁹

Second, our participants attended an in-person study visit at a clinical site. This may have influenced our findings in two ways. First, research site itself was significantly associated with WTP, which may reflect regional differences in perspectives about or HIV prevention trials that affected our outcome. Second, it may have been difficult for minors to attend the study visit without parents’ awareness. Thus, our sample may have included a significant proportion of adolescents who were not concerned about inadvertent disclosure of their participation in this study to a parent or guardian, which would have potentially important effects on the relationship between WTP and parental involvement in the consent process.

Disclosure.

There are several possible explanations for our unexpected finding that LGBTQ adolescents who were not out to a parent had higher mean WTP scores than those who were out, and those who identified as straight/heterosexual. First, the aforementioned studies^27–29 had different inclusion criteria and study settings than ours. Most were conducted online, and some didn’t require face-to-face interaction with a researcher or other participants. This may have captured a more socially vulnerable population of sexual and gender minorities with greater disclosure related concerns compared to our study which required an in-person study visit. Second, some other studies did not have specific sexual risk criteria, and therefore it is possible that their participants were less concerned about HIV acquisition compared to our participants,³⁰ which may have affected their overall interest and WTP in HIV prevention trials. Finally, prior research indicates disclosure of sexual orientation to parents is associated with lower sexual risk of HIV³¹ and lower odds of HIV infection. Thus, it is plausible that adolescents in our sample who were not out to parents had a higher risk (or perceived risk) of HIV acquisition compared to those who were, and thus may have had a greater interest in participating in HIV prevention trials.

Imagined benefits.

Adolescents who are not out to a parent may anticipate benefits from research participation that extend beyond the health benefits of the study. Fisher et al. found anticipation of a positive relationship with research staff was significantly associated with willingness to participate in an HIV prevention study, and youth who were not out to a parent or guardian were significantly more likely to identify getting access to PrEP as a study benefit.³² These findings are similar to those from our own study of the experiences of participants in the ATN 113 trial,¹² who reported substantial social benefits from research participation, including: being part of something important, having a positive relationship with study staff, and receiving education about safer sex. It is possible, therefore, that adolescents who are not out to their parents may imagine different benefits from participation in research than those who are out.

WTP and the roles of communication, medical mistrust, and concern about acquiring HIV

Although not a primary objective, our findings highlight the role that some sociobehavioral characteristics play in WTP in biomedical HIV prevention trials.

Communication.

Open communication about sexual orientation, and other indicators of parental acceptance, are associated with greater self-esteem, social support, and general health among LGBTQ youth.³³ Our finding that more frequent communication with parent(s) about relationships and sexuality is associated with greater WTP in biomedical HIV prevention research is consistent with this and other research suggesting family communication plays a significant role in adolescents’ access to and use of sexual and reproductive health interventions.³¹ A recent meta-analysis of the relationship between family communication about sex and protective sexual behaviors (condom and contraceptive use) indicates there a significant positive effect of communication that is moderated by the sex of the adolescent and the sex of the parent, with more robust effects for girls (compared to boys) and comunication with mothers compared to fathers^.34 This suggests a need for improving communication between fathers and adolescents, and for boys in particular. Sexual and gender minority adolescents have unique communication needs. In particular, they identify a need for sexual communication scripts that are more inclusive, rather than heteronormative. Some key needs for these scripts include: safer sex for specific sexual activities (e.g. anal sex), navigating harrassment, sexual orientations and identities, and sexual safety.³⁵

Medical Mistrust.

The term “medical mistrust” refers to the belief(s) that clinicians and researchers: deceive people or experiment on them without their knowledge or consent, provide substandard care or therapies to minoritized populations, or ask people to participate in research that would not benefit them.^36,37 Several systematic reviews have identified medical mistrust as a barrier to research participation,^36–38 but very few studies – and none of the systematic reviews – examine the relationship between medical mistrust and minors’ WTP in research. We found greater medical mistrust scores were associated with lower WTP among minor adolescents. This finding is consistent with research indicating an association between mistrust, research participation and acceptance of biomedical innovations.

The relationships between structural racism, historic human rights violations in research, medical mistreatment, and health inequities are complex and therefore difficult to address. A critical starting point requires shifting the onus from research participants to researchers and their institutions—requiring the latter to demonstrate trustworthiness rather than asking the former to increase trust.³⁹ A recent report from the National Academies of Science, Engineering, and Medicine⁴⁰ examined the issue of underrepresentation of women and persons of color in clinical trials, and identifies concrete short and long-term actions for multiple elements of the research enterprise. Some of these recommendations include improving workforce diversity, engaging with community stakeholders at each phase of research design and implementation, and providing fair compensation and supports for research participation.

Concern about HIV.

The relationship between sexual behavior, perceived risk of acquiring HIV, and HIV prevention behaviors is complex. While several studies indicate risk perception is an important predicator of PrEP interest or adherence,^30,41 the evidence of an association between risk perception and other protective behaviors, such as HIV testing and condom use, is mixed.⁴² In this study, we found an association between concern about future infection with HIV and WTP in biomedical HIV prevention trials.

Limitations

There are several limitations to this study. First, we assessed WTP in biomedical HIV prevention trials among youth who were willing to enroll in our study, which may have led to sample selection bias in favor of youth were particularly inclined to participate in research. However, adolescents who are inclined to participate in research are likely representative (in terms of WTP) of those who would enroll in a biomedical HIV prevention trial. Second, we measured hypothetical WTP, and it is possible that adolescents’ anticipated WTP would differ from their final decision to enroll in a clinical trial.⁴³ To counter this limitation our study was structured to mimic actual trial enrollment processes as closely as possible—study visits were convened in person at a research site and consent processes were simulated exactly as if participants were enrolling in study. Third we required in-person study visits, which may have limited the sample to adolescents who were either unconcerned about a parent discovering their participation in a research study or who have sufficient freedom of movement to do so without risk of discovery. However, this limitation is also a strength of the study because biomedical trials require in person clinical visits, which renders our sample similar to those adolescents who would enroll in biomedical clinical trials. Finally, the smaller than anticipated sample size resulted in small cell sizes for certain categories, particularly for those who indicated that they definitely would not participate (see Table 2). As a result, it is possible that we did not have adequate power to detect real differences across groups. In a future study with a larger sample size, it may be possible to identify patterns of results that differ by consent condition and study type.

Conclusion

To the best of our knowledge, this is the first study to quantitatively examine the relationship between consent processes, clinical trial structure, and WTP in biomedical HIV prevention trials among minor adolescents who meet criteria for biomedical HIV prevention trials. Our results indicate adolescents are willing to take part in biomedical HIV prevention research trials, and parental involvement in the consent processes may not be the most important deciding factor for many of them. While prior research underscores the particular importance of parental involvement as a barrier to research participation for LGBTQ adolescents, we found adolescents who are not out to a parent or guardian had higher WTP scores than those who were out. There was some variation in WTP within consent groups, which, combined with variation in WTP by other variables such as outness, medical mistrust, and concern about HIV infection indicates recruitment and consent procedures must be designed to maximize each adolescent’s autonomy and safety, and there may not be a singular approach to consent that achieves that goal.

Implications and contribution.

Prior research indicates teens are reluctant to participate in HIV and sexual health research if a parent/guardian must provide permission. We found no relationship between parental permission and teens’ willingness to participate in biomedical HIV prevention trials, but other factors (e.g. medical mistrust, communication with parents) were important.