Participant perspectives on data sharing in craniofacial clinical research: qualitative interviews with participants with craniofacial microsomia and their caregivers

Abstract

Objective:

To describe perspectives of participants and caregivers of pediatric participants on data sharing in clinical craniofacial research.

Design:

Qualitative interview study of research participants’ attitudes about data sharing.

Setting:

One-to-one interviews via web conferencing.

Patients, Participants:

Adults (n=7) and adolescents (n=8) with craniofacial microsomia and caregivers (n=20) of children with craniofacial microsomia who previously participated in clinical craniofacial research.

Interventions:

Semi-structured interview guide addressing four main topics: (1) expectations about data collection and sharing; (2) preferences and limitations for data sharing; (3) consent/assent language; and (4) gaps and future needs.

Main Outcome Measure(s):

We iteratively developed a qualitative codebook based on inductive interview transcript review and conducted a thematic analysis of coded data.

Results:

We identified five themes within participants’ descriptions of their attitudes about data sharing: (1) participants hope that research participation and data sharing will advance science for the benefit of the craniofacial community; (2) sharing images is broadly recognized as important for craniofacial research but raises discomfort for some; (3) participants generally view broad data sharing and use positively but raise concerns focused on harm to the craniofacial community; (4) trustworthy researchers and data protections provide reassurance for sharing data; and (5) decisions about pediatric data sharing are complex in the context of developing autonomy.

Conclusions:

These findings illustrate ethical complexities for data sharing in clinical craniofacial research related to balancing community-oriented benefits and risks, providing control over sharing images, ensuring researcher trustworthiness, and respecting a child’s future autonomy interests.

Background

Craniofacial microsomia (CFM) is an inclusive diagnosis referring to a range of birth conditions that are characterized by underdevelopment of one or more parts of the face, most often the ears (i.e., microtia) and/or mandible (i.e., mandibular hypoplasia).¹ Research to understand the etiology, impact, and treatment of CFM is essential so that children affected by this condition can achieve optimal health and quality of life outcomes.²

Data sharing is an increasingly important step in clinical research and can contribute to more rapid scientific advances, increased rigor and reproducibility, and greater data transparency and accessibility.^3,4 As of January 2023, new research project proposals to the National Institutes of Health are required to submit a Data Management and Sharing Plan delineating how study data will be shared.⁵ However, there are many unknowns in terms of how to balance the scientific benefits of data sharing with ethical obligations to participants.^6,7

In research on CFM, several factors intersect to heighten the potential impact of data sharing on participants. First, CFM is a pediatric condition, meaning that often parents or legally authorized representatives—as the best situated parties to make decisions on behalf of their children—are providing permission for their child’s data to be collected and shared. As children’s autonomy develops throughout childhood, adolescence, and adulthood, their ability to be involved in decision-making evolves, but many decisions about data sharing are made before they can engage in the process.^8–10 Second, CFM is a relatively rare condition, which may heighten privacy considerations while also increasing the potential benefits of data sharing.¹¹ Third, research in this context may include images, sometimes including photos or scans of the face, which may be seen as particularly sensitive and/or identifiable and raise challenges for deidentification.^12–14 Fourth, many individuals with CFM have lived experience with visual facial differences, including stigma and discrimination in some cases,^15–17 which may shape how the risks and benefits of data sharing and secondary research are experienced.

These considerations have not previously been explored in the context of research on CFM, leaving researchers without guidance on how to balance their obligations to advance science while respecting and protecting the participants in their research. The objective of this exploratory qualitative interview study was to describe participant and caregiver perspectives on data sharing in research on CFM.

Methods

Overview

We conducted a descriptive qualitative study¹⁸ using 1–1 virtual interviews with adult and adolescent participants and caregivers of children with CFM who had previously participated in a clinical study of craniofacial microsomia. We were guided by a social constructivist approach, meaning that we sought to characterize how individuals in the CFM community perceive the value and implications of data sharing with respect for how these perceptions reflect their lived realities and experiences.¹⁹ The research team included investigators and staff with expertise in empirical bioethics and clinical craniofacial research.

Recruitment

We recruited previous participants in one of our two recent NIH-funded investigations of craniofacial microsomia. The first study, entitled “Craniofacial microsomia: Accelerating Research and Education (CARE)” was designed to investigate the psychological impacts and healthcare burden of craniofacial microsomia.²⁰ One hundred and eight English- and Spanish-speaking caregivers (n=62) and individuals with CFM (n=46) were recruited across the US and participated in the CARE study. Detailed description of the recruitment methods and participant characteristics have been published.^21,22 Participants completed a narrative ‘Life Story’ interview, which asked participants to divide their CFM story into “chapters” based on experiences that were meaningful to them with the goal of creating a conceptual thematic framework of psychological adjustment in caregivers and individuals with CFM.²² CARE study procedures also included collection of photographs and a detailed medical history interview. The second study, entitled “Craniofacial microsomia: Accelerating Understanding of the Significance and Etiology (CAUSE)” included n=351 participants from across the US and South America, along with their family members. Data collection included photographs, medical history, and a DNA sample.

We used purposive sampling to recruit interview participants; we aimed to include approximately 20 caregivers, 10 adult participants, and 10 adolescent participants, which we anticipated would allow us to reach thematic saturation within each subgroup.^23,24 Within each of these groups, we further stratified our sample to include approximately 25% Spanish speaking individuals, and we aimed to include a clinically and demographically diverse group to ensure we captured a breadth of perspectives.

We contacted potential interview participants by email and phone. Those who indicated interest in participating in an interview reviewed an information sheet, provided verbal informed consent (or parental permission and adolescent assent), and were scheduled for an interview.

Interviews

Interviews were conducted using a secure Zoom account and lasted approximately 45 minutes on average. We developed a semi-structured interview guide (see Supplemental Material A) divided into four sections that explored: (1) current expectations about research data collection and sharing; (2) attitudes about data sharing, including preferences for limitations on uses and users of data; (3) reflection on consent/assent language about data sharing from enrollment in CARE/CAUSE; and (4) remaining gaps in understanding and future needs. We also developed a brief slide show (see Supplemental Material B) describing the purpose and key features of data sharing, which we shared with interview participants before the questions in section 2 of the interview guide. The slide show also included excerpted language about data sharing from the CARE or CAUSE consent/assent form, which we shared before section 3.

Interviews were conducted by a trained interviewer (LS, EA, AB) in either English or Spanish, as preferred by the participant, between March and August 2024. Interviews were recorded in Zoom and transcribed verbatim by a professional transcription service. Spanish-language interviews were subsequently translated into English by a professional translator. Interviewers reviewed transcripts for accuracy and cleaned transcripts of potentially identifying information.

Analysis

We iteratively developed a codebook based on review of transcripts and domains from the interview guide. The qualitative analysis team (SK, LS, EA, AB) reviewed and conducted line-by-line initial coding on initial transcripts,²⁵ then through group discussion drafted an initial codebook. We then test-coded additional transcripts, stratified by participant age and language groups to ensure our codebook represented the breadth of our data, and refined the codes and definitions until the team’s coding was consistent (estimated >80% agreement) and additional codebook revisions minimal. The coding team (SK, LS, EA) then coded all transcripts using the qualitative analysis platform Dedoose.²⁶ We used regular group-based consensus meetings to maintain coding consistency and review independently coded transcripts; 15 of 35 transcripts (43%) were coded by two or more independent coders and then to consensus using this approach and the remainder coded by one coder. We began to identify potential themes through regular group discussion during this process. We continually evaluated for thematic saturation, including across subgroups of age and language, and we ended interview enrollment when our analysis team agreed we had reached saturation. To formally summarize our findings and facilitate identification of additional themes, we drafted summaries of coded excerpts, with attention to any differences between subgroups. We shared our preliminary themes with a multidisciplinary group including clinical craniofacial researchers and individuals with personal connections to the CFM community, who validated that the themes identified aligned with their experiences.

Results

Participant characteristics and overview

We interviewed 35 participants: 20 caregivers of children with CFM, 7 adults (age 18+ years) with CFM, and 8 adolescents (age 12–17 years) with CFM. Among caregiver participants, 95% were female, 70% were white, and 60% had a Bachelor’s degree or higher. Among individuals with CFM, 53% were male, 73% were white, and 47% had a Bachelor’s degree or higher. Table 1 shows detailed participant characteristics.

Table 1.

Participant characteristics

Caregivers	N=20	%	Individuals with CFM	N=15	%
Sex			Age: mean (SD), range
Male	1	5%	Overall	19.2 (4.8), 14.5–27.7
Female	19	95%	Adults only (n=7)	23.5 (3.5), 18.8–27.7
Race			Adolescents only (n=8)	15.4 (0.9), 14.5–16.9
White	14	70%	Sex
Another race	5	25%	Male	8	53%
Multi-racial	1	5%	Female	7	47%
Ethnicity			Race
Hispanic/Latinx	6	30%	Asian	1	7%
Not Hispanic/Latinx	14	70%	Black	1	7%
Preferred language			White	11	73%
English	15	75%	Asian	2	13%
Spanish	5	25%	Ethnicity
Education			Hispanic/Latinx	5	33%
No high school diploma/GED	2	10%	Not Hispanic/Latinx	10	67%
High school or equivalent	2	10%	Preferred language
Some college/associate’s degree	4	20%	English	13	87%
Bachelor's degree or higher	12	60%	Spanish	2	13%
Proband phenotype category			Phenotype category
Microtia only	3	15%	Microtia only	1	7%
Microtia + mandibular hypoplasia	13	65%	Microtia + mandibular hypoplasia	13	87%
Other	4	20%	Other	1	7%

We identified 5 broad themes within participants’ attitudes about data sharing: (1) participants hope that research participation and data sharing will advance science for the benefit of the craniofacial community; (2) sharing images is broadly recognized as important for craniofacial research but raises discomfort for some; (3) participants generally view broad data sharing and use positively but raise concerns focused on harm to the craniofacial community; (4) trustworthy researchers and data protections provide reassurance for sharing data; and (5) decisions about pediatric data sharing are complex in the context of developing autonomy. We compared how themes arose between language (English or Spanish) and age (caregiver, adult, or adolescent) subgroups. We did not identify any substantial differences between the English and Spanish language subgroups. Where we identified differences between age groups, we highlight those below.

Theme 1: Participants hope that research participation and data sharing will advance science for the benefit of the craniofacial community.

Most interview participants expressed a desire to give back and help others navigate the medical and support system. Some described this explicitly as a motivation for participating in research:

“Anything that I can do to potentially help other families, to help bring awareness and knowledge and information to make the path easier for other families, really, that’s my motivation.” (Caregiver 065)

“I’d like to help other people who are in my same experience.” (Adolescent 069)

Some participants mentioned that, as part of a community of people with a relatively rare condition, they felt a particularly strong motivation to contribute:

“My condition, I wouldn’t say it’s rare, but I guess it kind of is. And I’ve never really seen people doing research about people with hemifacial microsomia or similar conditions. And so I think I wanted to have a voice in what future research might be or what future implications could be done. And I think, especially for me, growing up and having so many surgeries and looking different and all this stuff was really hard. And so I really wanted to be an advocate for kids who are younger than me in being able to [be] like, ‘Hey, is there something that we can change?’” (Adult 041)

“If anything ever comes up to improve anything related to my son’s cranial facial syndrome and helping researchers and families, I’m always willing to do that because it’s such a small community of us. It’s not like there’s a lot of people with these syndromes.” (Caregiver 054)

Among adolescents and adults with CFM, some expressed that participating in research was an opportunity to feel like they were part of something bigger than themselves:

“I think it’s helpful in one respect for research and furthering research in our community. But also, just personally, it was really helpful to just have a conversation with people who are interested in not just my medical diagnoses for once. Just not listing out all the things that everybody says is wrong, but holistically, what’s going on.” (Adult 012)

Most participants said that the fact that their data would be shared had no impact on their decision to participate, although some added that they hoped sharing their data would increase the impact of their contribution:

“It did not really affect [my decision] because I had a hunch that that would happen. And personally, I think it’s pretty fine by me because whatever it takes for knowledge to— I mean, the knowledge is more shared among others or people, and so it does not affect my decision, but it just gives more attention to my diagnosis.” (Adolescent 076)

Others pointed out that if they had had concerns that their data might be misused, that could have changed their decision:

“If it wasn’t going to be shared, that wasn’t going to stop me from trying to help. But if it was going to be shared to a point where it was exploited and could potentially be harmful to me, then yeah, that would have affected my decisions.” (Adult 033)

When discussing the likely benefits of data sharing, most participants described how it would promote general advancements in knowledge, without specifics about what benefits they expected:

“I think it’s very likely that the more the data is shared and the more researchers look at the data, the more answers we will get.” (Caregiver 020)

“I mean, it’s hard to say [how likely the benefits would be to materialize]. I mean, I guess I like to think so, but I mean, it’s hard to say.” (Adult 024)

Several interview participants expressed a desire to learn more specifics about how their data contributions were being used and how they were helping others:

“I kind of want to know if anyone— is anyone researching it right now? Is anyone requesting it right now? Who is requesting it? Which doctor? Which hospital? I want to know who’s requesting it right now, or is anyone at all requesting it right now? Is this a futuristic like, ‘Gee whiz, one day down the road, someone might in the future ask for this certain set of data points or something,’ or is there more specific, like, ‘Hey, they’re researching this right now, and this is the third study they’re doing,’ or, ‘This is the very’ - I don’t know – ‘well-thought-out grant,’ kind of thing, ‘being researched.’” (Caregiver 050)

Theme 2: Sharing images is broadly recognized as important for craniofacial research but raises discomfort for some.

Many interview participants highlighted the importance of images for both clinical and research purposes, expressing that images are uniquely able to illustrate craniofacial conditions and their change over time:

“With our condition, it is an external thing. …. So I think seeing pictures is super helpful and seeing how it affects everyone differently. …. You can explain that all you want in words. But I think pictures do a much better job of showing like, ‘Oh, this is what this actually looks like,’ and just to compare and contrast and have a better— I think you can have a better understanding of the condition with pictures.” (Adult 041)

However, some participants specified that images can only tell part of the story and must be contextualized alongside other data, including changes over time:

“Craniofacial, you absolutely need to see it so that you understand what we’re talking about. You need to see it. You need to see the asymmetry. You need to see the malformation of things and the procedures and the scars and stuff like that. But then also, you need to gather, obviously, dates and just the history of someone. You need to talk to someone and say, ‘Tell me from the beginning, when did you notice this?’ or whatever. So I feel like you can’t see some of that stuff in a picture. I think some pictures are going to be very helpful for some researchers, and some people are just going to not need the images and just need more of the family history.” (Caregiver 050)

“Photos may not always tell the whole story. Yeah. Every day, you kind of wake up a little bit different, and it would be kind of hard to pinpoint what exactly is going on from a single photo. But overall, I just think it’s a good way to kind of see progression in something.” (Adult 055)

Participants also spoke to the uniqueness of images as a data type. Many said that photos are more personal than other data types, citing their identifiability and linkage with identity:

“The biggest concern is being traced. It’s not even the information. Well, yeah, it’s really mostly photos. Honestly, it’s really personally identifiable, searchable information.” (Caregiver 013)

“Saying my name or having a picture of me, that would obviously make me feel weird. But I think in general, as long as it’s words or just information I’m sharing, I think that’s fine, as long as I’m anonymous.” (Adolescent 063)

However, other participants expressed a lack of concern about the identifiability of their photos:

“I mean, there are photos involved, but without a name, it’s just a face at the end of the day. So I mean, for me, personally, it wouldn’t bother me because at the end of the day, it’s not a personal thing when it comes down to that. It’s definitely more scientific, just more analytical based.” (Adult 043)

“Other people, I guess, might not feel comfortable with everyone looking at their face. But I mean, I don’t really think that’s too big of a concern for me.” (Adolescent 051)

Interview participants also reported varying levels of comfort with sharing images. Some expressed self-consciousness about sharing photos, even though ultimately they had decided they were comfortable contributing photos to research:

“At the time that we first started participating …, the one thing that [my child] was uncomfortable with was her photo being used and being seen by other people. And it was not because— it was not because she didn’t want that used for research purposes. It was because she was feeling very self-conscious about her microsomia and atresia microtia, those kinds of things. That was something she was trying to hide a lot when she was younger. She feels completely differently now. …. She feels very differently now and has a lot of self-confidence. And so I know that now she would say, ‘I don’t care if people see my picture.’ But back then, I think she was a little bit uncomfortable with the fact that people would see her photo, not because of the research part but just because she was very self-conscious at that time. But she did agree to it. I mean, we did talk about it before we participated in the study, and she said yes, they could take her picture. So it wasn’t that she was uncomfortable enough to say no. She just felt a little bit of discomfort about that.” (Caregiver 020)

“There is a little aspect, especially when I was younger, of self-conscious and not loving that photos of myself are out there where I don’t know where they are. But overall, I’m okay with it.” (Adult 043)

Other participants expressed a desire to share, citing the visibility of their or their child’s craniofacial condition in daily life:

“I have no problem sharing my story. It’s not like I’m ashamed or afraid, or I want to cover that up. I’m not embarrassed at all. It’s part of what we have to live. That’s what we shared. And photos of [my child] too, she has no problems either. She has learned, she isn’t self-conscious. We have no problem with that.” (Caregiver 076 (translated from Spanish))

Some interview participants described the value of educational use of photos for parents of children with CFM or medical trainees:

“As a parent, if I was doing research and I came across an article or study and there were pictures in there of individuals that had the same condition as my child, that would be really reassuring and like, ‘Oh, yeah, this article or information that I’m reading is going to be helpful because it’s about the same thing.’” (Caregiver 038)

“Any sort of university or hospital, any sort of learning environment where people are trying to educate themselves and others about cranial facial syndromes, I’m fine with the use of pictures for that.” (Caregiver 054)

However, others raised concerns about sharing photos of themselves for educational purposes, highlighting the importance of clarifying what the purpose of sharing is:

“Just the fact of people, I don’t know, looking at me and I don’t know, just imagining how they would describe the picture or use it as a teaching tool. And say like, ‘Oh, what’s wrong with them?’ Or whatever, however people use it, so I don’t know. It doesn’t feel like a positive experience for me.” (Adult 012)

To address concerns about identifiability, some participants suggested specific approaches to preserve privacy, such as blocking out eyes, limiting photos to affected parts of the face, or sharing medical images like x-rays rather than photos:

“I’ve noticed sometimes with pictures that I’ve seen where they kind of hide the eyes, things that make a person easily identifiable, just something blurring it out so that it’s a little bit more fuzzy, not exact. I could see that being beneficial, especially for young kids because [] they’re not providing consent. …. Blocking the eyes [] could still show what the study’s trying to show, but giving [] a little bit more privacy.” (Caregiver 065)

“How my bone structure on my face has changed after having surgeries, I will share that. But that’s x-rays. Those are different. Those do not show my face. They show my face, but not distinguishable features of my face. I feel better about sharing that than actual photos.” (Adolescent 066)

Theme 3: Participants generally view broad data sharing and use positively but raise concerns focused on harm to the craniofacial community.

Most interview participants expressed support for data sharing for the purpose of advancing knowledge about craniofacial microsomia and other craniofacial conditions:

“I think I want it to be used for patients to feel safe or to gain knowledge in just living with craniofacial [conditions]. But in general, I think it should be used for research and to give information to other patients.” (Adolescent 063)

Many participants also expressed general support for sharing data to advance medical research outside of the craniofacial context, although the value of their data for other research was not always clear and some wanted more information about the purpose of other uses before they agreed:

“If the craniofacial [data] can help do other things, then that would be fine.” (Adult 176)

“I would be kind of fine with it being used for anything relevant to [my child]. So anything cranial facial related or anything with children that have to go through multiple procedures and surgeries because I’m sure there’s other ailments and diseases and syndromes that could potentially fit that category, or just kids that have been through a lot. So there’s probably a few other things outside of the craniofacial-specific part of him, and I would be okay with that.” (Caregiver 054)

When asked about desired limitations on uses of shared data, interview participants raised concerns that focused around uses that would undermine the goals of research. For example, one person said it would be inappropriate to exercise exclusive rights over data:

“I’m okay with people, I guess, using the data to create things. But where they then try to assert some kind of ownership, then, in the data and an exclusion of others, I’m not great with, right? So if it’s an exclusive where, then, nobody else is going to be allowed to have access to use that data for research, then I would not be happy about that. Because again, it would go back to the broad dissemination of the data that the more hands, the more eyes, the more possibility for innovation.” (Caregiver 013)

Others called out uses that could harm the craniofacial community, for example by contributing to stigmatizing attitudes or devaluing the lives of individuals with craniofacial conditions:

“If it’s in the wrong hands, they could kind of use it to paint a negative perspective or kind of spew the data in a way that’s not only misrepresentative of the data that we collected, but also misrepresent the patients and kind of how they want to— how their healthcare should be dealt with.” (Adult 055)

“I guess my only reservation might be if it was to try to prevent the birth of children with that same condition because I wouldn’t agree with that morally.” (Caregiver 066)

A few participants also highlighted the possibility of shared data contributing to discriminatory insurance practices:

“Obviously, I’m concerned about someone making money off of it or insurance companies maybe increasing the price or premiums or something or whatever and using that to harm us or to make it more difficult because our kids happened to be born with this condition or something. I would be concerned like that, someone making it harder, using our data to make it harder for us.” (Caregiver 050)

Some participants discussed the importance of secondary data uses being focused on the value to children with craniofacial conditions and their families, rather than on financial profit to the researchers:

“I think mostly if it’s coming or reaching to the wrong hands or people are using it badly. ….[Interviewer: Who would that be?] I haven’t thought about it thoroughly, but I will say mostly to people that will take advantage of how they actually make more money out of that instead of actually helping families.” (Caregiver 047)

“I would want parameters if the data was going to be sold in any way for profit unless that profit was specifically geared towards funding, services, research for craniofacial issues with kids.” (Caregiver 065)

“I wouldn’t want my data sold. …. I feel like at that point, it stops being my story, my words, and it just starts being an item for sale.” (Adolescent 035)

Participants also objected to the use of data to market products to families or unauthorized use of participant images for advertisements:

“If they used it to market to us, I would not like that. …. Like, ‘Oh, that little girl needs a prosthetic ear. I know that because she doesn’t have an ear,’ and then they would come like trying to sell us a prosthetic ear.” (Caregiver 069)

“I think if there was some form of advertisement where [my child’s] image was used or direct quotes were used without our awareness previously, would be a little jarring.” (Caregiver 038)

Theme 4: Trustworthy researchers and data protections provide reassurance for sharing data.

Interview participants spoke about the importance of trustworthiness in terms of both the research team using the data they contributed and the data protections in place. Some participants spoke about how their decision to participate and share their data stemmed in part from their prior experiences with a trusted research and/or clinical team, highlighting the link between those trusting relationships and their trust in data sharing:

“I feel like I trusted [organization], which is where we initially learned about the study. And since I trusted them, I would trust someone that they were affiliated with. So I did not worry about it that much. I didn’t think about it that much, because I felt like it was one trustworthy source that would have another trustworthy source.” (Caregiver 069)

“I think being part of the [research] study, like meeting all the researchers that have been involved, it seems very intentional and very well-done for our community. And I think very, again, like a holistic approach versus a lot of medical approaches just don’t feel the same. So I think part of my participation and continued participation is really because of the team and the way that the study is being done. And I think it’s still good that my data could be used for larger studies and be helpful in different ways.” (Adult 012)

Others spoke generally about the importance of being able to trust researchers to treat their data with respect:

“I think I would like researchers to just have an understanding that these are real people with real stories and that they deserve privacy. And as long as I know that the researchers understand that, I feel pretty comfortable.” (Caregiver 020)

“So that’s a concern that I do have because they didn’t get the consent from me personally. …. They don’t really know who they’re sending it towards, but it’s essential that they know it’s from me, and being confidential is also important too.” (Adolescent 076)

Many participants supported the use of approval processes or other access limitations to ensure that only researchers who were trustworthy and had appropriate research-oriented goals would have access to data:

“I like how they double-check the people who are going to get the information.” (Adolescent 051)

“They will only use it for researchers, no one else, and they have it in a safe place, so that no one would have access to this information. So, it gave me the confidence to participate.” (Caregiver 073 (translated from Spanish))

“All of the precautions that you take, it’s definitely reassuring in deciding to join. I know that [my child’s] information isn’t just going to be frivolously thrown around to just anybody.” (Caregiver 054)

Deidentification was a particularly important data protection, especially for facial images, and most interview participants discussed the importance of removing names and other identifiers:

“Just don’t show my face and keep my name anonymous.” (Adolescent 069)

“For sure, the anonymity, taking out identifying information. I think that that’s kind of the biggest one is just taking out identifying information so that then future researchers, they may come across the information, but they’re not going to be talking to friends and family and colleagues of, ‘Oh, yeah, this person with this name and this situation.’” (Caregiver 038)

However, one adolescent participant acknowledged challenges with deidentification in the context of emerging facial recognition technology:

“Well, facial recognition. You’re keeping a face. Like, what’s the point of removing the name and date of birth if you find the face? …. Like, you could find name, like the face. Do a picture search, something like that.” (Adolescent 066)

Many participants also brought up security and encryption as key protections. These topics were discussed without much detail, instead reflecting participants’ trust in the researchers and their institutions:

“I think it’s mostly on the IT and the software, on that side, the engineers to make it safe for people not to break into your computers and steal information there.” (Caregiver 047)

“I don’t understand how all of it works, and I don’t need to, but just the complexity of that process. But yeah. I mean, I trust the system, I guess.” (Caregiver 039)

One participant added that the level of security should be adjusted depending on the accessibility of the data:

“I feel like [the approach to data protection] depends on the amount of the potential exposure for that data. So, for example, if my data is going to be public, right, where anybody can access it, I would want it to be super secure where no one can ever track me down.” (Adult 024)

Finally, when interview participants were asked about the language from the informed consent process describing data sharing, most initially expressed that they did not remember the language. Upon reviewing the consent language, however, many recalled the section and pointed out that it had provided them with reassurance that their data would be deidentified and that they would be able to withdraw their data should they later change their mind:

“I do remember reading this [consent language], and I do remember that it was very clear, and I felt comfortable with the fact that if at any time either of us felt uncomfortable, we could contact the [study team] and say, ‘We want to be done now,’ and that our information would be destroyed at that time. So I think it’s very clearly written and very easy to understand, and I think that it made us feel very comfortable about how our data was going to be used.” (Caregiver 020)

“If they no longer want to participate, it’s stated very clearly that you don’t have to participate anymore, that it would be destroyed, but what was collected, was collected.” (Caregiver 076 (translated from Spanish))

However, at least one participant expressed confusion about when data could no longer be withdrawn after having been shared:

“I find it interesting that it says the information could be used for future research after you say that you no longer want your data to be shared. Is that my misunderstanding? So it says identifiers might be removed, which is that coding, I’m going to assume, we were talking about, from the identifiable private information and that after such removal, the information could be used for future research studies, distributed— okay. So no, I understood that incorrectly. Sometimes I got to read it to understand it. Okay. So basically, if I say out of the blue, ‘Hey, I no longer want my information being used,’ you guys destroy it.” (Caregiver 005)

Theme 5: Decisions about pediatric data sharing are complex in the context of developing autonomy.

When asked what they thought their child’s perspective on data sharing would be, most caregivers said they thought their child would be fine with it based on what they knew of their child’s values. Some described discussing research participation with their child:

“I would be more concerned [if my child were] younger, if I was kind of just— I was kind of just doing it on my own, not including his opinion, whether he wanted to do something like this or not. But he was older, and he really wanted to do the study. So that’s why I went ahead with it. And as they’re older, they’re able to voice more concerns or problems or anything like that that they might feel.” (Caregiver 063)

Some caregivers of younger children reflected on what their child’s future self might think and discussed the challenges of making decisions that had implications for their child’s privacy. Several expressed hope that their child would agree with their decision but appreciated having the option to withdraw data later if their child were to express a different perspective:

“The fact that [my child] can’t really provide feedback or doesn’t really understand probably makes it feel like it’s more our decision. I mean, we’re taking into account what we think she would want, but she doesn’t really have strong opinions. So, if she got older and had stronger opinions and said, ‘I don’t want to do this,’ obviously we would honor that.’” (Caregiver 068)

“At this point, I mean, [my child] is very open to it. Down the road, if he would object to us doing this, I would certainly want to honor his wishes as he gets older and teenager and doesn’t want— I don’t see that happening because, like I said from the beginning, we’ve always been very, very open, very honest, and it was nothing we ever felt like we had to hide. So I don’t see that changing, but that would be— the only time is if he started showing some objections to it.” (Caregiver 039)

Some adolescent and adult interview participants similarly discussed how their views were changing or had changed over time, describing an evolving comfort level with data sharing throughout adolescence and beyond:

“The more I grow, the clearer things are, what I shouldn’t do more of, what I should do more of, what may help, what may not help. …. I did know that there was a risk of that data ending up where it shouldn’t be, but I’d say, at the time I didn’t care, because it could help people, and that’s the only thing I thought about.” (Adolescent 070 (translated from Spanish))

“I think with being older now and kind of getting a better perspective of things, I think the study came at a good time for me. Not that I would have— I don’t think I would have rejected it 10 years ago, but I think I would have been more hesitant to just speak about and let people have some of that information and data other than my doctors at the time. And especially going through school too, I didn’t get a science-related degree or anything, but I did take a few classes here and there. And I think the perspective of this is being used kind of for the greater good. And I think eventually, hopefully, this will help people down the road. And just the maturity in that, I think, makes a difference for sure. As a teenager, I’m not sure. I just wouldn’t have been thinking that far down the road. And so I would have kind of been selfishly thinking like, ‘Well, I don’t want to— I don’t know if I want to tell them this about myself,’ or I wouldn’t have been thinking the bigger purpose than me.” (Adult 043)

Discussion

This qualitative interview study provides novel insights into the perspectives of individuals with craniofacial microsomia and their families on data sharing in clinical research. These findings illustrate the complexities of advancing science through data sharing while adequately protecting participant data in the context of craniofacial research. Here, we discuss four sets of ethical considerations that these findings highlight: (1) balancing the benefits and risks of data sharing through a community-oriented lens, (2) navigating variation in comfort levels for sharing facial images, (3) developing research practices that promote trustworthiness, and (4) respecting pediatric participants’ developing autonomy given practical constraints.

First, participants highlighted benefits to the craniofacial community as a critical foundation for sharing data. Advancing science for other children and families was a key motivation for many interview participants to both enroll in research and share their data, which aligns with motivations expressed by participants in other settings, including pediatric clinical research^27,28 and research on other rare conditions.^11,29 Similarly, when discussing potential harmful uses of data, many of the potential harms identified centered on the implications for the craniofacial community. Others have previously discussed the potential for secondary research to contribute to group harms,^6,30 which may have a greater impact on participants with marginalized identities. Our findings illustrate that it is essential to assess the benefits and risks of data sharing through a community-oriented lens.³¹ That is, to the extent that participants and the broader craniofacial community take on risks associated with data sharing, those risks must be balanced by meaningful benefits to the craniofacial community. While there is no guarantee that any given use of data will produce widespread medical advances, secondary researchers should ensure their study questions are aligned with craniofacial community values, their methods are scientifically sound to produce valid results, and their uses of data are aligned with their stated research purpose.³² Data repositories can implement infrastructure and access requirements to facilitate high-quality, community-informed secondary research.

Second, while many participants expressed overall support for broad data sharing and use, participants expressed varying levels of concern about sharing facial images. For some participants, understanding and having control over how their images would be used was important. Because of the unique privacy implications of image data, broad sharing of images collected in research may warrant explicit attention and discussion at the time of informed consent. Transparency about how images may or may not be used (e.g., for research vs. teaching or advertising)¹³ and methods of deidentification,¹⁴ as well as the ability to opt out of sharing images, may be warranted. Such transparency would align with the American Cleft Palate Craniofacial Association’s ethical rule that patients “should be made aware of all possible uses of their image and likeness,”³³ as well as with ethical codes from other professional organizations generally emphasizing the importance of preserving research participant confidentiality.^32,34

Third, trustworthiness—of both researchers themselves and the protections in place for participant data—was central to participants’ comfort with contributing and sharing data. Because children with CFM and their families often have frequent and longitudinal contact with medical providers, they may develop close relationships with their doctors and/or medical institutions. Prior work in other contexts has shown that trusting relationships in medicine can contribute to trusting relationships with researchers,³⁵ particularly when clinical providers and researchers are part of a singular institution or team. This trust can provide a foundation for decisions to enroll in research, but it raises complexities for both participants and primary researchers when unknown secondary researchers access data. Governance procedures can help to ensure that access for secondary research does not undermine the trusting relationships between participants and the primary research teams with whom they directly interacted. This may be particularly important for preserving relationships with participating families who may have personal or historical reasons for being cautious about trusting the research team.³⁶

Finally, developing autonomy in the pediatric context raised complexities for both caregivers and individuals with CFM. Several adult and adolescent participants described how their views on data sharing had evolved over time, illustrating the gradual formation of preferences and attitudes that might be relevant to research. Tailoring consent and assent conversations to an individual’s needs and capacity, while involving caregivers to support minors’ developing autonomy, may enhance these conversations.³⁷ Caregiver participants in this study were very thoughtful about how to align their decisions with their children’s values and acknowledged the challenge of sharing data that reflected a child who was too young to engage in the consent, and sometimes assent, process. Caregivers greatly valued the future option to withdraw data in case their child later objected. However, an important practical tension is that data that have been fully deidentified and/or shared cannot be withdrawn. Biobanks have previously navigated similar challenges with long-term storage of pediatric data and specimens.³⁸ Clear communication upfront about these challenges, as well as the limitations to future control, will be essential to ensure patients and families can make informed decisions.

This study has some limitations. All interview participants had previously participated in craniofacial research and agreed to participate in a second study with overlapping team members. Those who agreed to participate in both studies may have been particularly engaged in research and had a positive experience with the research team and/or clinical team, which may have shaped their responses and comfort level with data sharing. Views of those who declined research participation or could not access a research study are not represented in our sample. Further, while interview participants roughly mirrored the diversity of CARE study participants, there may be different perspectives among groups not well represented in our interview sample, particularly individuals from racialized identity groups, those who speak languages other than English or Spanish, and families with limited educational attainment. While we did not identify differences between the perspectives of those who completed interviews in English versus Spanish, a broader sample of participants may reveal important communication or cultural considerations. Further exploration of these issues—including potential benefits and harms of research, privacy preferences, and optimal communication approaches during informed consent—among those with intersecting identities will be important as data sharing practices move forward.

Conclusion

Interviews with participants and caregivers of pediatric participants in clinical craniofacial research revealed several important ethical considerations for data sharing. Researchers and owners of data repositories should incorporate participant perspectives as they develop practices and policies for data collection, sharing, and use.

Data availability:

Deidentified data are available upon reasonable written request to the corresponding author.