Should Feedback of Individual Results be Integrated into the Consent Process in African Genomics? Participants’ Views from an HIV-TB Genomics Research Project in Botswana

ABSTRACT

Background

Whilst informed consent is a key component of considering whether individual genomic research results could or should be fed back to research participants, little is known about the views of African research participants on its role.

Methods

We carried out a qualitative study to explore views of adolescents and parents or caregivers regarding informed consent for feedback of individual results from a genomics research project in Botswana. We conducted 24 deliberative focus group discussions with 93 participants (44 adolescents and 49 parents or caregivers) and 12 in-depth interviews (6 adolescents and 6 parents).

Results

Our findings revealed that most participants would like to be informed about the possibility of discovering individual genetic results during the consent process and that consent be obtained for feedback during the enrollment process. They further expressed that in cases where prior consent to feedback was not obtained, then participants should be re-contacted where life-saving genetic information is discovered. Participants emphasized the need for researchers to ensure that participants’ decisions regarding feedback of results are well-informed. Autonomy, transparency, and communication were identified as key values to uphold during the consent process.

Conclusion

In conclusion, obtaining participants’ consent for feedback of results is important to ensure that their rights and wellbeing are protected in research. This is critical in building trust relationships between participants and researchers.

KEYWORDS:

• Informed consent
• feedback
• individual genetic research results
• genomics research
• Botswana
• Africa

Introduction

Amidst a growing consensus in the research community that some individual genomic research results ought to be fed back, is growing emphasis on the importance of incorporating the possibility of feedback in the consent process. Yet the call to integrate information about the feedback of results in consent is set against concerns about the complexity of consent for genomics research and whether people effectively understand all elements of such research (Tindana and de Vries 2016). Elements that would need to be effectively conveyed in the consent process is that most genetic information present risks not certainties, that a single gene has the ability to influence multiple traits or conditions and variants can have opposite effects. In addition, genetic variants may be reclassified or gain (or lose) significance (Bredenoord et al. 2011). As a result, Rotimi and Marshall (2010) pointed out that uncertainties associated with the clinical relevance of genetic results has led to many genomic research projects conducted in the past not to feedback results to participants. This was also true for most projects in the Human Heredity for Health in Africa (H3Africa) consortium in 2016, which did not seek explicit consent for the return of results (Munung et al. 2016). However, this position is becoming more difficult to justify as many genomics studies are now likely to discover genetic information that is clinically relevant (Kengne Kamga et al. 2020; Rotimi and Marshall 2010; Wonkam and de Vries 2020). Many authors also now argue in favor of feedback of genetic results to participants. Affleck (2009) reasons that participants should decide for themselves whether they would like to take up risks that come with knowing their results, and that where participants attach personal value to results, they should be fed back out of respect for their autonomy. This view is supported by Budin-Ljøsne (2012) who indicated that participants have a right to know their health information regardless of its utility. Other key values that are discussed in this literature are respect, charity, mutuality, and reciprocity (Solberg 2012) as well as duty to rescue (Sullivan and Berkman 2018).

Therefore, it is important that this issue be attended to, and participants views be accounted for, to ensure that their rights and wellbeing are protected in research. This is particularly important as many studies have revealed that participants would like their individual genetic research results to be fed back (Christensen et al. 2017; Fernandez et al. 2009; Sanderson et al. 2016; Shahmirzadi et al. 2014). Since the ultimate decision about whether to receive individual genetic research results rests with the participant, this suggests that some type of informed consent process will be needed for these decisions (Appelbaum et al. 2015). Even so, feedback of genetic research results to participants requires customized consent documents that sensibly accounts for ethical and social obligations to participants as well as those related to them (Rotimi and Marshall 2010).

Our Africa-based study team was particularly interested in understanding how African genomics research participants view these issues. Specifically, we were interested to know a) whether or not to obtain consent for feedback of results and why, b) do participants have a right to know or not to know their genetic results? c) what should happen if consent was not initially obtained and d) what should happen when a preventable life-threatening gene is discovered for a participant who initially indicated not wanting to receive results. We carried out a qualitative study to explore views of adolescents and parents or caregivers regarding informed consent process for feedback of individual results from a genomics research project in Botswana.

Methods

In this qualitative study, we purposively selected 93 participants: 44 adolescents who had previously participated in an HIV-TB genomics study in Botswana and 49 parents and caregivers of children and adolescents who participated in the same study, which is a part of the Collaborative African Genomics Network (CAfGEN). We decided to link our study to this larger genomic study mainly because genomics research is relatively new in Botswana and most people would likely not be very familiar with genomics concepts. The CAfGEN study had a quite extensive community engagement process in place, coupled with a robust consent process. This sensitized participants in the CAfGEN study to key concepts of genomics research. It also gave them a more direct interest in questions relating to what should be done with individual genetic research results. For these reasons, we specifically recruited participants who had enrolled in CAfGEN. For the purpose of this study, we recruited adolescents aged 15 − 18 years, involved in the CAfGEN study, as well as parents and caregivers of children involved in the same study, aged 2 − 18 years.

We used deliberative focus group discussions (dFGDs) (Rothwell et al. 2016) and in-depth interviews (IDIs) to explore a broad range of issues relating to the return of individual genetic research results including participants’ views regarding the consent process. A dFGD manual developed for this study was piloted with 2 groups of parents and caregivers and 1 group of adolescents, while the IDI tool was piloted with 3 parents/caregivers and 2 adolescents. The dFGDs and IDIs for parents and caregivers were conducted in Setswana using translated tools, while a combination of Setswana and English was used in the dFGDs and IDIs with adolescents. The data was collected from February 2019 to March 2020.

Deliberative focus group discussions involve a combination of the conventional FGD methodology with expanded opportunities for learning and discussion, in an effort to enable participants to engage with the research topic in greater depth. The dFGD technique adopted for this study involved holding two dFGD meetings with each group of participants a week apart. The initial dFGD meeting involved an interactive information-sharing session with participants. Case studies elaborating particular ethical considerations in the return of individual genetic results were used for information sharing during the first meeting. This allowed participants to learn about genes, their functions and impact on health. This information session took about 45 − 60 minutes. Participants were then given a 30 minutes break which was followed by a dFGD to get participants’ initial or immediate thoughts about feedback of findings from genomics research. This dFGD took about 60 − 90 minutes. Overall, the first session of the dFGD took about 3 hours. Participants were then provided with information covering a summary of the content that was discussed in the workshop to take home to be used as a reference to help participants reflect on or revisit the information and questions interrogated during the first session. They were then invited for a second dFGD meeting a week later, where their opinions about feedback of findings was interrogated further. This second dFGD took about 30 − 60 minutes.

DFGD meetings were held with 12 groups of participants (6 groups of adolescents and 6 groups of parents and caregivers), resulting in 24 dFGD meetings in total. The initial dFGD meetings for parents and caregivers comprised of about 4 − 11 participants, while adolescents initial dFGDs meetings were attended by about 5 − 12 participants. Overall, 34 adolescents and 38 parents and caregivers were retained to participate in the follow-up dFGD meetings. The other 21 participants (10 adolescents and 11 parents and caregivers) had competing interests and were unable to attend the follow-up dFGD meetings. For these follow-up meetings, parents and caregivers’ groups varied from 4 to 8 participants, and from 2 to 11 for adolescents. The dFGDs were complemented with 12 in-depth interviews (6 adolescents and 6 parents/caregivers) with people who had also participated in the dFGDs. The IDIs focused on clarifying and probing further some of the early insights from the study. These were conducted with participants from the dFGDs who either engaged more in the discussions or held uncommon views, as well as a few others who were less interactive in the dFGDs so as to give them an opportunity to share their views in a more private conversation. However, no strict criteria were used to assess participants’ level of engagement in the dFGDs. IDI sessions took 30 − 80 minutes (for more details on the methodology, see Ralefala et al. (2020)).

All discussions were audio-recorded, transcribed, translated into English and imported into NVivo qualitative data analysis software (QSR International Pty Ltd. Version 12, 2018) for coding and data analysis. The Framework Method for data analysis was used to analyze the qualitative data resulting from this study (Gale et al. 2013). This method involves producing a matrix in which summarized data are displayed in rows (cases) and columns (codes) to create a layout in which data can be systematically presented and analyzed both by theme and by participant or event. This process involves five key phases which are similar to the analytical process of thematic analysis, which are: (a) familiarization with the data (b) establishing a thematic framework (c) indexing (d) charting and (e) mapping and interpretation of data (Ritchie et al. 1994).

Ethical review

This study was reviewed and approved for ethics compliance by Institutional Review Boards in South Africa and Botswana. Written assent was provided for by all adolescent participants and permission by their parents or caregivers. Parents and caregivers also provided signed informed consent for study participation.

Results

Demographic data was collected from the study population including gender, age, educational level and residential location. Most adolescents (61 percent) in our study were female. Adolescents’ age ranged from 15–18 years, with most of them aged 16–17 years (61 percent), consequently most of them (70 percent) were either in junior secondary school or had completed, while 23 percent were either in senior secondary school or had completed. Half of the adolescents resided in villages nearby the city of Gaborone where this research was conducted, 39 percent resided in the city, while 11 percent did not indicate their place of residence. An overwhelming majority of parents and caregivers in our study were female (92 percent). One possible explanation for this could be that women are normally the main caregivers in Botswana (Kang’ethe 2011; Kang’ethe 2013; Maundeni et al. 2009), owing to social, cultural and legal factors (Jorosi-Tshiamo et al. 2013). A significant number of parents and caregivers (39 percent) were between 41 − 50 years of age and 94 percent had educational qualifications varying from primary to tertiary education. Like the adolescent population, most parents and caregivers (57 percent) resided in villages nearby the city and 43 percent resided in the city of Gaborone.

Our data revealed a range of important themes regarding the return of individual genetic research results, including (for instance) the importance of reciprocity in guiding expectations of feedback, the reasons why our participants overwhelmingly wanted to receive results and practical considerations around the return of results. These findings were the subject of other manuscripts submitted for publication elsewhere. This paper specifically focuses on exploring participants’ views on the role of the consent process in the feedback process.

Prior consent for feedback of results

Our findings revealed that most adolescents expressed that it is important to obtain their informed consent regarding possible individual genetic results at enrollment, so that they could know their options and stay mentally prepared for whatever will be revealed by the genetic test. However, a few expressed that being informed about the possibility of finding individual genetic results upfront may not be ideal as it could cause emotional anxiety. People of this view indicated that it may be better to discuss and obtain consent from the participant for feedback of results when they are available. Most parents also emphasized the need to be told about the possibility of discovering individual genetic results at the beginning of the study and having their consent obtained for feedback. One parent said:

I think before I take part in the study, everything should be explained to me before any tests as you have explained these pamphlets to us. They should let me know of the possible results that might come up, so maybe I have positive results I would have already counselled myself and that I can prepare myself for any results that may come up, that way it becomes easy for me to receive the results because I have already been taught at the beginning. I would be ready for anything, negative or positive results, that way it becomes easy, when the doctor releases the results, I will tell him that I accept any result, this will make me not to collapse so being taught first is important to ready my heart (P7G4).

Transparency in consent for feedback of results

In addition, both adolescents and parents also emphasized the importance of transparency in research. Adolescents shared that it is important for researchers to clearly inform them about the possibility of finding incidental results instead of providing information about the disease under study only. They worried that if there is no transparency in the consent process then participants could later interpret the discovery of incidental results to think that researchers have gone overboard to test them for other things to pursue a different research agenda. Others pointed out that some participants may not receive results well if they were not initially informed about the possibility of discovering incidental results. Likewise, parents also emphasized the importance of transparency between researchers and participants and shared that researchers should tell them all the information they need to know to make informed decisions so that they could prepare themselves on what might come out. To highlight the importance of transparency in the consent process, one parent had this to say:

…imagine we are 10 and none of us was told anything, what would you do as the researcher when you find a condition that you didn’t expect? Will you hide it? If you are going to share it, how will you share it without having notified any of us who could be affected? (P11G5).

The above view was also supported by other participants who also emphasized that "it is wrong for researchers to withhold information from participants".

Nonetheless, some participants observed that there could be limitations to informed consent in genomics research. One parent shared that genomics research is wide and complex and that researchers may not always be able to predict upfront what they may discover hence may not be able to inform the participants of what is expected at the beginning of the study but emphasized that what is key is for participants to be counseled when results become available so that they can decide whether to receive them or not.

I think research is not as straightforward as it seems; however, if I take it in good faith then we shouldn’t be worried that we were not told of the possibilities of extra results. Genomics studies are complex and it’s difficult to focus on one thing. When you are looking at genes in a blood sample, you are looking at a vast sample, there are many different illnesses that can be found in the blood like cancer and many other different illnesses, so it could be difficult for them to explain it. Unless we did not understand what a genomics study entails, but if it has been explained like the way it has been explained to us and we understand it then we should expect anything. But as the ladies have mentioned, it is important however that there be counselling at a time when one is told about results that were not expected, to try and help her receive the unexpected results and avoid extreme reactions. What I know about genomics is that it is a wide net, if it was a particular condition maybe that’s when I will worry that one has gone overboard without telling me (P11G5).

Recontacting participants for feedback of results

Additionally, adolescents expressed that in cases where prior consent to feedback individual genetic results was not obtained, researchers should re-contact the participant to get consent to feedback when information is discovered about genetic risk for a preventable harmful condition, because the participant could benefit from receiving that information. Similarly, most parents also pointed out that should such information be found, then the participant should be recontacted to get their consent on whether or not to feedback the result because doing so might save the participant’s life. But one parent shared that:

… I always consider the person’s life as the key issue. The information available is for the quality of the participant’s life that is why I say that irrespective of the consent, one needs to know because knowledge is power, having consented or not at the end you will say because I value the participant’s life just as I value that which you have given me for me to get these findings and on top of my findings I have seen that you are under threat. Really those who want to consent are those who are saying we hear that South Africa has launched missiles but keep quiet or somebody hears about the elephants at the cattle post, they keep quiet while you go out and look for your goats, really you need the information (IDI-P004).

Participants also expressed that even in cases where a participant had indicated that they would not want to receive their results, when a harmful variant for a preventable condition is discovered, then the participant should be re-contacted to check if they would reconsider their decision to feedback. They noted that “…this finding might not have been expected …” (P12G5) or “…maybe when they signed the agreement, they did not have enough information or didn’t think it through” (IDI-A002). They emphasized that researchers have an obligation to warn the participant. One parent said, “to me, this issue is about ethical considerations, we are human. I don’t think someone would just sit on information that could potentially save a life” (P11G5).

Information-sharing and effective communication

Parents and caregivers also noted that some participants may refuse to receive results because they are scared or that they are not well informed. As a result, they pointed out that it is important for researchers to establish why an individual would not want to know their results, and that such individuals should be provided with information about the pros and cons of knowing their results so that they could make informed decisions. See a quote below from one of the parents:

This is why I was saying that you could label those who want to know the results and those who don’t want to know the results, then you could ask why they don’t want to know the results and tell them the benefits of knowing their results and that you can take action but if you don’t know your results then you won’t have information […] For those who don’t want their results then you should find out why they don’t want their results, how they are feeling and why they feel that way and give them the benefits and risks of knowing the information and not knowing the information (PIDI-P002).

The importance of communication in the consent process for feedback of genetic results was also emphasized. One parent expressed that:

…what is important is communication, in that case you have to call me again and revisit the issue of returning the results with me, make me aware of the implications of the study and what was found in relation to my health, because it is important for me to know since I am going to live with that condition or could get it addressed. I don’t think I will reject the results if you tell me that I have a cancer gene even if I had said I shouldn’t be given the results, because it is important for me to know so that I can prevent it or take care of myself. So, it is important to make me aware (P4G5).

Right to know or not to know

Inherent to a drive to include a discussion about feedback of results in the consent process is a question about whether people have a right to know or not to know, particularly for harmful, preventable conditions. As a result, we engaged participants on these discussions and found that most adolescents believed that they have a right to know their individual genetic results because these results could affect their lives and receiving results could empower them to act toward their health. Similarly, parents and caregivers shared that it is their right to know their results because researchers used their blood. As a result, they should know any health information that researchers have about them, so that they could take care of themselves. Others limited their right to know to genetic conditions that are actionable or those that are life-threatening. However, one parent expressed that their right to know applies only if there was prior agreement to feedback or if the discovered gene is related to the disease under study.

When asked on whether participants have a right not to know, many adolescents expressed that they do not have a right not to know as receiving genetic information could be lifesaving. This view was also expressed when participants were asked whether they have a right to change their mind and decide not to receive results after having signed a consent form indicating their desire for feedback. In this case they reasoned that having signed a consent form could have encouraged researchers to go an extra mile to look for help for them before delivering the results hence participants seemed to view changing their mind as not being considerate or appreciative of the researchers’ efforts. Participants’ views on this issue could be influenced by the Setswana culture where being courteous is considered an indication of good manners. A refusal to receive results or to withdraw from disclosure of results may be viewed as being rude and is something that most participants may not feel comfortable to do. This is grounded in the African philosophy of Ubuntu which requires members of the community to treat other people with kindness, compassion, respect and care (Murove 2012). Nonetheless, other adolescents pointed out that their right not to know exists even where consent to feedback was initially given and that participants have the right to change their minds. They also shared that their right not to know is especially relevant if there is no help for the participant. Likewise, most parents and caregivers expressed that participants do not have a right not to know genetic information that could be lifesaving. One parent expressed that "a person who refuses to receive results doesn’t have responsibility toward themselves (P4G3)". Their reasoning seems to have been influenced by their HIV experience as demonstrated by this expression from one of the parents; “I don’t think if you have accepted your HIV status you could be scared to receive any results at this point (P3G3)". They shared that it is through knowing their HIV results that they were able to get treatment for themselves and their children, and now they are doing well. Some participants also felt that refusal to receive feedback of individual genetic results is not right as not only the participant’s life may be in danger, but those of their family members too. They also noted that refusal of participants to receive the results could be “…tough on the researchers as [they would] have seen the results and believe that the participant needs to know, but if he or she had said no, then [researchers] cannot force them…” (IDI-P002). They suggested that researchers should put an effort to ensure that participants are making informed decisions on whether or not to receive their genetic results. Nonetheless, some parents maintained that participants do have the right not to know even if a consent to feedback was provided and trying to force results on those who have expressed that they do not want to know could attract lawsuits against the research team.

Discussion

Our findings revealed that most participants would like to be informed about the possibility of discovering individual genetic results during the consent process and that consent be obtained regarding whether to feedback results before enrolling in genomics research or not. This desire to be informed and to be given an opportunity to decide on feedback of results is consistent with the principle of autonomy which states that participants have a right to self-determination, to decide what should happen to their bodies, samples or information about them (CIOMS 2002).

Participants also expressed that there should be transparency in research and that it is generally wrong for researchers to withhold information from participants. They further expressed that in cases where prior consent to feedback was not obtained, then participants should be re-contacted when life-saving genetic information is discovered. Re-contacting participants with lifesaving information was recommended even where a participant had initially indicated not to receive results. Similarly, participants in Appelbaum et al. (2014) supported having consent for potential re-contact obtained in the initial informed consent. Our participants argued that sometimes participants may have been ill-informed at the time of making a decision not to receive results and that participants could also change their minds over time. This suggests that our participants seemed to view consent as an ongoing process and not a one-time event. This view is consistent with the CIOMS guidelines which states that, “obtaining informed consent is a process that is begun when initial contact is made with a prospective subject and continues throughout the course of the study” (p. 33). These guidelines further state that investigators in long-term studies should periodically seek informed consent from their study participants (CIOMS 2002). In this case, researchers could make provision for re-consenting participants for feedback of results as the study progresses and not necessarily only when a preventable, life-threatening gene is discovered as suggested by our participants, but this can be done for all study participants at certain intervals as determined by the study team. This will provide an opportunity for participants to reflect on their previous decisions and allow those who would like to change their mind about previous choices to do so. This is particularly important in feedback of findings from genomics research where it might be impossible for researchers to anticipate which results will be found because of the complexity of genomic studies (Horton and Lucassen 2019)

Participants’ strong desire to be re-contacted with lifesaving information seems to be founded on the principle of duty to rescue and could be influenced by participants’ HIV experience where knowledge of their HIV status has empowered them to get relevant help and survive the HIV virus. This finding is consistent with observations by Bredenoord et al. (2011, who shared that there is often a strong case for researchers to feedback life-saving genetic information of immediate clinical utility. According to Bovenberg et al. (2009, this obligation cannot easily be overruled by other moral considerations—except, for example, when the participant has expressed a prior wish not to be informed. Similarly, while our participants generally wanted to be re-contacted with lifesaving genetic information, some of them cautioned that re-contacting participants who had initially indicated that they would not want to receive results should be handled with sensitivity and that counseling should be used when returning results.

In addition, participants pointed out that researchers should ensure that participants’ decisions regarding feedback of results are well-informed as some genetic information may be lifesaving. This view resonates with the H3Africa recommendations for feedback which describe that in cases where participants are given an opportunity to decide on their preferences for feedback, researchers need to make sure that participants are accurately informed about the possible information they may or may not receive, and how that information would affect their lives. If participants are not properly informed, they would make misinformed decisions which could have real consequences for their lives and wellbeing in future (H3Africa 2018).

The importance of providing participants with information is also identified as a critical component of the informed consent process by international guidelines including The Belmont Report (1979) and the CIOMS guidelines (CIOMS 2002). According to these guidelines, participants should be provided with pertinent information that could empower them to make an informed decision regarding their participation in research. Therefore, providing participants with information about the possibility of discovering genetic information as well as its possible implications could empower them to decide whether they would want to receive such results or not. This approach is also preferred by Bredenoord et al. (2011) who expressed that there should be an effort by the researcher (or another professional) to help participants understand genetic information as this will enable them to express their preferences regarding feedback of results. Although our study did not discuss what sort of information participants would like to be provided with during the consent process, a study by Appelbaum et al. (2014) found that researchers and participants alike embraced disclosure of a wide range of benefits, risks, and ancillary information, including potential impact on family members, protections for confidentiality, and how genetic results will be dealt with in the event of death or incapacity. The H3Africa guidelines also emphasize that value and validity of results should be well articulated to participants during the consent process as well as participants’ right to decline receiving their results (H3Africa 2018).

Furthermore, our participants also emphasized the importance of effective communication in handling consent issues. Kohane and Taylor (2010), also highlighted that effective communication between researchers and participants is imperative in understanding participants’ preferences and the significance of results. They further advocated for an approach in which participants’ ability to process appropriately communicated, complex, contingent information, is respected, rather than assuming that they will be unable to understand genomic complexities that are actually less intellectually taxing than some other issues they might encounter in clinical care (Kohane and Taylor 2010).

Conclusion

In conclusion, obtaining participants’ informed consent for feedback of results is important to ensure that their rights and wellbeing are protected in research. Therefore, this could mean that African genomics researchers should inform participants about the possibility of discovering individual genetic results in research and make provision for participants to decide whether they would like to receive such results or not. This could be done using a tiered consent model, where separate informed consent forms are developed for participants to consent, first for study participation and then for feedback of results. Doing this in stages, could address concerns about overloading participants with too many decisions to make during the consent process and create an opportunity for participants to be provided with relevant information that could empower them to make informed decisions. Autonomy, transparency and communication were identified as key values to uphold during the consent process. This is critical in building trust relationships between participants and researchers.

Limitations

Although these findings are important to the literature on feedback of results in genomics research in Botswana, they may not be generalized to other contexts. Firstly, most of our parents and caregivers (92%) were female, probably because women are generally the main caregivers in Botswana (Kang’ethe 2011; Kang’ethe 2013; Maundeni et al. 2009), because of social, cultural and legal factors (Jorosi-Tshiamo et al. 2013). As a result, this means that the views of male participants were not well represented in this study. Secondly, our participants might have been more knowledgeable about informed consent procedures than the general population because they are part of a genetic study that has a strong community engagement aspect. Thirdly, although we took care during the dFGD process to educate people about the complexity of genetic results, the application of these teachings may still be quite vague, as the feedback of individual genetic research results was presented as a hypothetical situation. The dFGD method that we used is novel and allows participants to be educated, engaged for longer time which means they become comfortable to share their views as compared to normal FGDs. This could be an effective way to engage participants on complex concepts such as genetics, but there is a risk that the material presented in the course of the discussion bias participants to articulate or support particular views.

Authors’ contributions

All the authors contributed to the conceptualization of this work. DR, JDV and MK designed the study. DR collected the data. DR, MK and JDV developed the data analysis strategy. DR prepared the first draft of the manuscript. JDV, MK, MM and AW commented. Revisions were made by DR and all the other authors (JDV, MK, MM and AW) read and approved the final manuscript. JDV, MK and MM supervised the overall work.

Acknowledgments

The authors would like to thank the participants and staff at BBCCCE especially Keofentse Mathuba, Keboletse Mokete and Pinkie Gobuamang for assisting with participants recruitment, as well as Refilwe Monnakgosi and Ludo Bitsang who were both Research Assistants in this study. The authors additionally thank Vicky Marsh and the broader IFGENERA study team, particularly Elsie Breet, Marlyn Faure and Olivia Matshabane for developing the dFGD manuals and for discussions about project findings. We would also like to thank anonymous reviewers whose comments and suggestions helped improve and clarify this paper.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declaration of interest

The authors declare that they have no conflict of interests.

Additional information

Funding

Research reported in this publication was supported by the National Human Genome Research Institute of the National Institutes of Health under Award Number U54HG009790. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.