Background More effective recruitment strategies like alternative approaches to consent are needed to facilitate adequately powered trials. Witholding Enteral feeds Around Transfusion was a multicentre, randomised, pilot trial that compared withholding and continuing feeds around transfusion. The primary clinical outcome was necrotising enterocolitis. The trial used simplified opt-out consent with concise parent information and no consent form.
Objective To explore the views and experiences of parents and health professionals on the acceptability and feasibility of opt-out consent in randomised comparative effectiveness trials.
Methods A qualitative, descriptive interview-based study nested within a randomised trial. Semistructured interview transcripts were analysed using inductive thematic analysis.
Setting Eleven neonatal units in England.
Participants Eleven parents and ten health professionals with experience of simplified consent.
Results Five themes emerged: ‘opt-out consent operationalised as verbal opt-in consent’, ‘opt-out consent normalises participation while preserving parental choice’, ‘opt-out consent as an ongoing process of informed choice’, ‘consent without a consent form’ and ‘choosing to opt out of a comparative effectiveness trial’, with two subthemes: ‘wanting “normal care”’ and ‘a belief that feeding is better’.
Conclusion Introducing a novel form of consent proved challenging in practice. The principle of a simplified, opt-out approach to consent was generally considered feasible and acceptable by health professionals for a neonatal comparative effectiveness trial. The priority for parents was having the right to decide about trial participation, and they did not see opt-out consent as undermining this. Describing a study as ‘opt-out’ can help to normalise participation and emphasise that parents can withdraw consent.
- qualitative research
This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.
Statistics from Altmetric.com
What is already known on this topic?
Much neonatal and paediatric care is not based on robust evidence as definitive randomised trials are lacking.
Effective and efficient recruitment strategies, such as opt-out consent, have been proposed as a way to increase participation and make trials more generalisable.
Little is known about parents’ and health professionals’ views and experiences of opt-out consent in the context of an individually randomised neonatal trial.
What this study adds?
The principle of opt-out consent was generally considered feasible and acceptable by health professionals for use in a neonatal comparative effectiveness trial, but was challenging to implement.
Parents prioritised the right to decide about trial participation for their baby, and they did not see the principle of opt-out consent as interfering with this.
Describing a study as ‘opt-out’ can help to normalise participation and emphasise that consent is an ongoing process.
High-quality evidence is often lacking in neonatal and paediatric care1 because of a paucity of appropriately powered and methodologically robust trials.2 Effective recruitment into randomised trials is needed to facilitate definitive studies.3 4 Alternative approaches to consent, such as verbal or delayed consent, are effective and acceptable in particular situations.5–8 Opt-out consent, where participants are given full information about a trial and enrolled unless they actively opt out, has been proposed as another way to increase participation9 and reduce injurious misconception10 and selection bias.11 Although studies have asked patients and health professionals for their views in theory about opt-out consent in an individually randomised trial,12 13 little is known about views in practice.
Prevention of necrotising enterocolitis (NEC) is a neonatal research priority.14 Red-cell transfusions have been identified as a potential risk factor for NEC,15 and observational data suggest that withholding feeds around red cell transfusions may reduce NEC risk; this has not been tested in adequately powered trials.16 17 Standard practice on some neonatal units in England is to continue feeds during transfusion, while on others feeds are withheld.18
The Witholding Enteral feeds Around Transfusion (WHEAT) pilot was a multicentre, randomised pilot trial carried out in 11 neonatal units in England19 that enrolled babies born before 30 gestational weeks and evaluated the feasibility of comparing two care pathways, withholding and continuing feeds around transfusion, on the incidence of NEC. A simplified opt-out consent process developed with parents (box 1) was intended. This paper reports a nested qualitative study exploring the views and experiences of parents and health professionals of the consent process.
Witholding Enteral feeds Around Transfusion pilot trial intended opt-out consent process
Parents were approached by a neonatal health professional after their baby was admitted to the neonatal unit, usually within the first 48 hours.
The health professional gave parents a simple two-page information leaflet about the trial and explained the trial verbally.
Health professionals explained that babies would be randomly assigned to one of two care pathways around any blood transfusions their baby might need, and that the risks of participation were minimal.
The health professional explained that all eligible babies were in the trial unless parents opted out, and they could opt out at any time by telling any member of staff.
There was no consent form.
The health professional recorded on the electronic patient record that parents had received information and had not opted their baby out of the trial.
The criteria for inclusion in the qualitative study were age ≥18 years, a parent whose baby had been recruited into the WHEAT pilot or a health professional involved in recruitment. Sampling for health professionals involved recruiting one from each neonatal unit, nominated by the WHEAT lead clinician. Due to resource constraints, parents were recruited from four neonatal units comprising a single neonatal network. Any parent with recent experience of WHEAT was invited by the local clinical team; recruitment continued until data saturation was reached within a demographically varied sample.21 The interviewer (JMcL) was independent of the WHEAT trial team and had no prior contact with participants.
Data were collected through semistructured qualitative interviews from September to November 2019. Participant information and consent forms were emailed to participants in advance; written or oral informed consent was obtained at the beginning of the interview. Interviews were audio-recorded and professionally transcribed. Interview topic guides are available (online supplementary files A and B).
Health professionals’ and parents’ transcripts were analysed as separate data sets using inductive thematic analysis,22 in parallel with ongoing data collection. Transcripts were checked against audio recordings and reread for familiarity. Codes were identified inductively and recorded using NVIVO. Codes were refined and combined as data collection continued, and themes were identified. JMcL analysed all transcripts and FA analysed a subset; codes and themes were discussed and agreed. Constant comparison23 was used to reconsider earlier codes and themes in the light of subsequent interviews. Themes from both data sets were integrated into an overall thematic analysis.
Ten health professionals from the 11 sites took part—one doctor worked across two sites. At seven sites, normal practice was to feed during transfusion; at two sites, feeds were withheld; and at two sites, there was no unit policy. Interviews lasted 19–37 min (mean 25.5); nine were by telephone; and one was by face-to-face in the hospital. Participants’ occupations and normal feeding practice at their unit are shown in table 1.
Eleven parents of nine singletons and two sets of twins took part in telephone interviews lasting 12–23 min (mean 19); 2 other parents who were invited declined to participate. Five parents identified as white–British, three as British–Asian, one as black–African, one as Asian–other and one as white–other; their ages ranged from 22 to 42 years; and occupation group ranged from semi-routine to higher managerial/professional using the National Statistics socioeconomic classification24 (24). Seven had babies who were allocated to the continue feeds pathway, three to the withhold feeds pathway, and one could not remember.
Five themes were identified relating to opt-out consent and the comparative effectiveness trial. Table 2 shows illustrative quotations.
Opt-out consent operationalised as verbal opt-in consent
Despite the intention to use opt-out consent, many health professionals described how in practice they had sought consent using a verbal opt-in approach. Usually, there were at least two conversations with parents to explain WHEAT and then inquiry about participation. This was reflected in the experiences of parents, 10 of whom said they were given information and then asked to confirm an explicit choice to participate or not, while one could not remember any discussion of WHEAT.
Health professionals described three factors that had shaped this approach: first, consent was actively recorded in the electronic patient record, and they did not feel they could affirm this unless the parents had positively consented. Second, some health professionals did not feel confident explaining opt-out consent despite training, and staff turnover meant not all were trained. Third, one did not feel it was ethical to proceed without opt-in consent.
Opt-out consent normalises participation while preserving parental choice
The phrase ‘opting out’ was used in verbal explanations and the information leaflet. This framing was felt by health professionals to normalise participation in the trial, tapping into parents’ desire for normality in an abnormal situation.
Some parents had not noticed the mention of opting out, while others had understood it and found it acceptable as they did not see it as compromising their right to consent. Several parents confirmed health professionals’ belief that opt-out consent could be less stressful, because they preferred not to have to make an active choice.
Opt-out consent as an ongoing process of informed choice
Health professionals were clear that their responsibility to ensure parents made an informed decision was not affected by opt-out consent. Most parents felt that they had been given sufficient information and appreciated that the information leaflet was short and proportionate. They particularly valued explanations from staff. All parents described feeling overwhelmed with their situation following very preterm birth; some parents and health professionals said that any consent given in the first days might not be truly informed because of this. Some parents suggested that consent should be revisited later, and some health professionals noted that they actively reconfirmed ongoing consent when a blood transfusion became necessary. This was felt to be particularly important if there was a long period between the original consent and transfusion. Some parents interpreted opting out as emphasising their right to change their mind about trial participation, and some used this to defer substantive decision making until their baby needed a blood transfusion.
Consent without a consent form
Most health professionals thought the absence of a paper consent form was positive and made decision making less stressful for parents. Two were concerned this could potentially leave them exposed to complaints if parents did not remember the consent conversation.
Most parents made no comment on the absence of a consent form. One who had experience of clinical trials was surprised but satisfied with the explanation given.
Choosing to opt out of a comparative effectiveness trial
The expectation had been that opt-out consent would simplify recruitment. Health professionals reported that some parents declined participation for reasons that might affect any neonatal trial: they were traumatised and unable to engage with research, were opposed to any intervention not of proven benefit, or the person who approached them lacked the necessary communication skills. However, additional reasons for declining related to parents’ understanding of comparative effectiveness and normal care.
Wanting ‘normal care’
Health professionals said some parents were not persuaded by the argument that both trial arms represented normal care in England, but rather focused on normal care at the individual unit. Parents may have prioritised the clinical judgement of clinicians they had met over unknown professionals elsewhere.
Wanting local normal care could also be connected with parents not wanting to be responsible for taking a decision with unknown outcomes; this occurred even at a neonatal unit where both arms of the trial were considered normal care.
Belief that feeding is better
Another reason for declining was connected to parents’ perceptions of risk and benefit. While parents said they had specifically asked about risks before deciding, no parents mentioned NEC, despite it being the primary outcome of the trial and being clearly discussed in the parent information sheet, which stated ‘We want to know if feeding babies or not feeding babies while they have a blood transfusion has an effect on the number of babies that get NEC’. Instead, most had a general sense that the trial was about improving care and was perhaps not particularly important. Some health professionals described having difficulty explaining uncertainty around feeding during blood transfusions without undermining parents’ confidence. This may have been linked to the fact that despite clinical equipoise, some had clear personal views about the topic of the trial.
Some parents reported basing their decision to opt out on an intuitive sense that feeding was better than not feeding and that WHEAT required an interruption in feeding which could affect weight gain and hunger. All parents interviewed had consented for their babies to be in WHEAT, but some had shared these concerns. Some parents from the ‘continue feeds’ group said that they might have opted out if their baby had been allocated to the ‘withhold feeds’ group, representing another form of deferred decision making.
Methodologically robust trials are essential to advance clinical care. In the neonatal setting, this means asking new parents to process complex information about uncertainty and risk, and to make consent decisions when their baby is critically unwell. It is perhaps unsurprising that previous studies have questioned the validity of consent given in these circumstances.25–27 Despite this, parents overwhelmingly support neonatal research and want to be involved in decisions about their babies’ participation.8 28 29 Efficient and effective approaches to consent are therefore key to informed involvement by parents. Opt-out consent in an individually randomised comparative effectiveness trial is novel and is the focus of this study.
Most health professionals believed that opt-out consent was ethically acceptable and advantageous for comparative effectiveness research, and helped to normalise trial participation. This is consistent with patient and health professional support for opt-out consent previously identified in a hypothetical comparative effectiveness trial.12 Opt-out consent in WHEAT was, however, operationalised by health professionals as opt-in verbal consent from parents, contrary to the intended process. In part, this was because documentation required active recording of consent, and in part because of what health professionals were used to for clinical trials. Active consent in research contrasts with clinical practice where parents are usually not consulted about comparative effectiveness decisions at all.30 As a result of how consent was operationalised by health professionals, this study was not able to explore the views of parents following completely opt-out consent. In keeping with previous work,9 concise participant information was acceptable to parents, who found it easy to understand and read. The principle of default participation in comparative effectiveness trials was also acceptable to those parents who understood it. This contrasts with parents’ their views on a hypothetical trial of novel treatments,31 but is consistent with views reported in Euricon26 and other studies,9 13 where what was most important for parents was the right to decide about participation. Parents did not see opt-out consent as infringing on this right.
Most health professionals in this study supported not using written consent forms and this was acceptable to parents, as in the Cord Pilot Trial.32 It has, however, been shown in neonatal trials that even when using conventional, written, opt-in consent a proportion of parents do not recollect having consented.25 This risk might be amplified by the absence of signed paperwork verifying consent. This troubled some health professionals who felt it might leave them vulnerable to complaints, echoing concerns raised about verbal opt-in consent in perinatal trials.27 33 It is not clear whether absence of paperwork in an opt-out trial could have adverse effects on the emotional well-being or trust of parents who did not remember consenting, over and above those that may exist in neonatal trials generally.
The description of the trial as ‘opt-out’ was felt to emphasise that the consent process was ongoing and included the right to change a decision to participate at any time, which has not always been well understood by parents in neonatal trials.26 29 Health professionals checked ongoing consent by reminding parents about their right to opt out. This was empowering for parents, several of whom used it to defer a substantive decision until they knew whether their baby needed a blood transfusion, or knew to which trial arm their baby had been randomised. This could result in differential opt-out rates by intervention arm and hence selection bias, impacting the validity of trial findings. Future trials using opt-out consent or delays between consent and the trial intervention should ensure such potential bias is minimised and clearly reported to quantify the magnitude of this potential problem . It is important to highlight that consent for any research should be an ongoing process rather than a one-off event.
This study illustrates the challenges of explaining comparative effectiveness trials, particularly that ‘normal practice’ is based on care across the wider health service rather than an individual unit. It was additionally challenging because parents had intuitive beliefs about the topic of the trial (feeding) and which arm was better for their baby, as is often the case in neonatal and paediatric trials.32 34 35 However, unlike parents in the CANDA trial10 who had an exaggerated perception of additional risks, parents in WHEAT were not worried about the adverse outcome NEC but were rather focused on the effect of the intervention on their baby being hungry or not gaining weight. Moreover, some health professionals described reservations about the trial, including rejection of equipoise, which may have affected how they communicated the trial to parents. This has implications for training health professionals to explain uncertainty and risk without causing additional anxiety.
Strengths and limitations
Key strengths of this study are that it included the views and experiences of parents from a wide range of backgrounds randomised into a comparative effectiveness trial, and of health professionals from all trial sites with varied experiences of trial recruitment.
Limitations include that all participating parents had babies in neonatal units where normal practice was to continue enteral feeds during transfusion. Parental reactions from units that withheld feeds or had no standard practice were reported by health professionals, but parents at those units may have had more complex views which would have been missed. A further limitation is that, unlike previous studies,10 26 we did not have ethical approval to interview parents who opted out. Instead we describe health professionals’ reports of what those parents had told them. Interviews with parents were carried out up to 12 months after birth, which may have affected parents’ recollections. Finally, bias may have been introduced through reliance on the neonatal units themselves to nominate or invite participation in the qualitative study.
It is important to recognise that this study was undertaken within a comparative effectiveness trial which compared two care pathways that were (and continue to be) standard of care in the UK, and hence, there was considered to be minimal additional risk from randomisation within a trial setting. Therefore, these findings are not necessarily generalisable to trials comparing more high-risk interventions.
The principle of opt-out consent for a neonatal comparative effectiveness trial was generally considered feasible and acceptable by health professionals. Obtaining opt-out consent was, however, countercultural for some health professionals, and introducing it proved challenging. Beyond the background risks to validity of consent that may affect any neonatal research, no additional risks from the simplified consent approach used in this trial were identified. Parents’ priority was having the right to decide about trial participation. Although parents did not experience a full opt-out consent process, those who understood that the intention was default participation in a comparative effectiveness trial did not see opt-out consent as undermining this right. Describing a study as opt-out can additionally help to normalise participation and emphasise that parents can withdraw consent at any time. Processes should be adapted for opt-out consent, including training health professionals to explain the opt-out process and the concepts of equipoise and comparative effectiveness as they apply to clinical care.
Many thanks to all the parents and health professionals who took part in this study and to Iza Andrzejewska, Elizabeth Lek and Kirupalini Mariampillai who facilitated parents’ involvement. Thanks to James Barnaby for comments on an earlier draft of this article.
Collaborators Members of the WHEAT trial development group: MJ (National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford), AF (National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford), RH (Parent of preterm twins, c/o Bliss – The National Charity for the Newborn, Chapter House, 18-20 Crucifix Lane, London, SE1 3JW), CC (National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford), DJ (Division of Neonatal-Perinatal Medicine, Faculty of Medicine, Dalhousie University, IWK Health Centre, 5850/5890 University Avenue, Halifax, Nova Scotia, B3K 6R8, Canada), GC (Neonatal Medicine, School of Public Health, Faculty of Medicine, Imperial College London, Chelsea and Westminster Hospital campus, London, SW10 9NH), CL (Bliss – The National Charity for the Newborn, Maya House, 134-138 Borough High Street, London, SE1 1LB), Jawad S (Neonatal Medicine, School of Public Health, Faculty of Medicine, Imperial College London, Chelsea and Westminster Hospital campus, London, SW10 9NH), Bowler U (National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford), Forster A, King A (National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford), Linsell L (National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford), Turner M (Women’s and Children’s Health, Institute of Translational Medicine, University of Liverpool, Liverpool, L69 3BX), Stanbury K (National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford), van Staa T (Centre for Health Informatics, Division of Informatics, Imaging and Data Science, School of Health Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester Academic Health Science Centre, Manchester, United Kingdom), Modi N, Juszczak E (National Perinatal Epidemiology Unit, Nuffield Department of Population Health, University of Oxford).
Contributors Authorship is on behalf of the WHEAT trial development group. The research question for this study was conceived by CG and developed by CG, HR, LC, CC, NM, UB, AF, AK, LL, KS, TvS, JD and ED. The study takes place within the WHEAT pilot trial, which was developed by CG, HR, LC, NM, UB, AF, AK, LL, MT, KS, TvS, JD and ED. JMcL undertook the data collection and JMcL and FA took part in data analysis. JMcL drafted the manuscript with input from CG and FA. All authors were involved in interpretation, review and revision of the draft manuscript and approval of the final version.
Funding This study was funded by the UK Medical Research Council: MR/N008405/1.
Competing interests CG reports grants from Medical Research Council and the National Institute for Health Research during the conduct of the study; grants from National Institute for Health Research, Mason Medical Research Foundation, Rosetrees Foundation and from Canadian Institute for Health Research outside the submitted work. He reports grants and personal fees from Chiesi Pharmaceuticals outside of the submitted work; the grant is for a research study and the personal fee was to support attendance at an educational meeting. He was an unremunerated member of the Neonatal Data Analysis Unit Board, which oversees the National Neonatal Research Database.
Patient consent for publication Not required.
Ethics approval The London Bloomsbury Research Ethics Committee (reference 18/LO/0900) approved the study. An information leaflet was provided and informed consent to participate was obtained.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available upon reasonable request. Following the consent process, the individual qualitative interview transcripts will not be made publicly available.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.