Objective To document the legal guardian-related barriers to consent procurement, and their stated reasons for non-participation in a paediatric critical care research study.
Study design A multicentre, prospective, cohort study.
Participants Legal guardians of children who participated in a multicentre study on adrenal insufficiency in paediatric critical illness. Data were collected on all consent encounters in the main study.
Methods Screening data, reasons for consent not being obtained, paediatric risk of mortality (illness severity) scores and age were collected on all 1707 patients eligible for participation in the Adrenal Insufficiency Study.
Results The main barriers to approaching legal guardians for consent were lack of availability of the legal guardians (321/1707) and language barriers (84/1707). Legal guardians of 917 patients were approached with an overall consent rate of 42% (range 14–56% across the seven sites). 81% of the 528 legal guardians who declined consent provided an unsolicited reason for refusal. The three most commonly stated reasons were: being overwhelmed (117/429), not wanting anything else done to their child (63/429) and not wanting an additional medication (53/429). In addition, 14.2% cited research-related concerns as the reason for their non-participation.
Conclusions Barriers to consent procurement in a non-therapeutic paediatric critical care study appear to occur at many levels with lack of availability of legal guardians, and legal guardians feeling overwhelmed, being the most commonly recorded reasons. Further research into the impact of these findings on the validity and generalisability of the results of such studies is necessary prior to the development and study of future consent models.
- Research Ethics
- Research on Special Populations
- Informed Consent
- Minors/Parental Consent
Statistics from Altmetric.com
The performance of clinical research in vulnerable populations, such as critically ill children, represents a balance between the need to advance therapies and the obligation to prevent the exploitation of those who are not competent to consent for themselves. The Declaration of Helsinki1 states that ‘for a potential research subject who is incompetent, the physician must seek informed consent from the legally authorised representative’. In Canada, the Tri-Council Policy Guidelines follow the principles outlined by the Helsinki Declaration but also state that ‘those who are not competent to consent for themselves shall not be automatically excluded from research that is potentially beneficial to them as individuals or to the group they represent’. In Europe, the Oviedo Convention on Human Rights and Biomedicine states that advances obtained through biomedical research contribute to saving lives and improving quality of life prior to reiterating that ‘where a minor does not have the capacity to consent to an intervention, the intervention may only be carried out with the authorisation of his or her representative’. As a result, in order to conduct research and advance the care of critically ill children, the vast majority of consents must be obtained from legal guardians rather than the patients themselves.2 Therefore, the legal guardian-related factors that affect the consent process in the context of paediatric critical care research are especially important, and may involve unique considerations.
Many studies support the notion that the paediatric intensive care environment is more acutely stressful than the general paediatric ward,3 ,4 is a unique environment,5 ,6 and that paediatric intensive care unit (PICU) families have different needs than those in NICU.7 Other studies also suggest a difference between adult and paediatric informed consent processes.8 ,9 The above differences make it difficult to compare the consent process in other settings to the paediatric critical care environment which emphasizes the need to study consent issues in this unique group of vulnerable patients separately.
Although there are a few studies on the consent process in critical care research, no paediatric studies were found that systematically reported on the reasons for not obtaining consent in eligible but non-enrolled patients. In addition, actual studies on legal guardians’ reasons for agreeing to, or declining participation in, a paediatric critical care research study are limited. Only two studies documented unsolicited parental responses to participation in PICU research.2 ,10 However, one of these studies10 involved a combined NICU/PICU population and required a 6-month time commitment from families and, therefore, may not be truly reflective of parental opinions on consent for paediatric critical care research studies. The other study had a small sample size of only 45 consent refusals.2
The limited literature in this area, the importance of research in moving patient care forward, and the unique nature of the paediatric critical care environment support the need for more information on the consenting process in paediatric critical care research. The objectives of this specific study were to document legal guardian-related barriers to consent procurement, and their voluntarily stated reasons for non-participation in a non-therapeutic study in the paediatric critical care population.
Materials and methods
This substudy was part of the Adrenal Insufficiency in paediatric Critical Illness (AIP) study11 which took place in the PICU of seven academic teaching centres across Canada between 2005 and 2008. The AIP study was an observational cohort study that involved blood sampling of ∼5 ml (one teaspoon) per patient from already existing invasive lines, performing an adrenocorticotrophic hormone (ACTH) stimulation test, and required recruitment within 26 h of admission (see table 1), and again 24 h later if the patient was still in the intensive care unit. The ACTH stimulation test is a well-established procedure that is sometimes used in critically ill children as part of routine management. It involves taking a small blood sample to test for cortisol levels immediately before and 30 min after the administration of one microgram of ACTH (a naturally occurring hormone in our bodies). Our inclusion criteria were patients aged from newborn to 17 years with existing invasive lines. We excluded premature infants, patients with known or suspected adrenal, pituitary or hypothalamic disease, those who received systemic steroids for more than 10 days in the previous month or had more than one dose of systemic steroids within the 24 h prior to admission, and those who were expected to have care withdrawn, or were transferred from another intensive care unit.
The following information was recorded by the research assistants immediately following the consent encounter: all patients who met inclusion criteria, reasons for exclusion, whether consent was attempted, reasons for consent not being obtained on eligible patients, verbatim spontaneous comments of legal guardians on reasons for declining consent, and qualitative comments on the consent process from the research assistants. It is important to note that this substudy simply recorded consent-related procedures as per each centre but did not attempt to change or influence the process in any way. The age and paediatric risk of mortality (PRISM) score12 for those who did and did not consent was collected from the patients’ charts for the last 6 months of the study. Research ethics board approval was obtained for this substudy from all participating centres. Any identifying information on guardian gender, race, socioeconomic status or postal code was not permitted to be collected.
Consent rate was defined as the number of legal guardians who consented to participation over the total number of legal guardians who were asked for consent. Descriptive statistics were used to summarise the study sample as well as for consent rates. A Student t test was used to compare the mean age and PRISM scores of those who did and did not consent. Homogeneity of consent rates between and across participating centres was tested using a Breslow–Day test.13 The stated reasons for non-participation were grouped into predetermined categories as judged by two independent reviewers who were not involved in the study.
There were 1707 patients who met eligibility criteria for the adrenal insufficiency study. The research assistants were unable to approach the legal guardians of 790 patients for a variety of reasons. In 321 (18.8%) of these eligible patients, a legal guardian was not available within the 26 h time limit required by the study, and in 84 (4.9%) cases, the legal guardians did not speak French or English and a translator for their preferred language was not available. Thirty-three patients were under the care of child protective services from whom consent was denied in all of the four cases for which it was requested. In the remaining 352 families who were not approached for consent, the two most common reasons were lack of agreement of the treating physician (74 legal guardians, 4%) or prior enrolment of the patient in another study which precluded their participation in the current study (60 legal guardians, 3.5%).
A total of 917 legal guardians and/or patients were approached for consent for participation, of whom 528 declined giving an overall consent rate for this study of 42% (range from 14% to 56% in the seven sites—see table 2). The consent rate was not significantly different between sites (p=0.72). There was no difference in the PRISM scores (6 (IQR 3–10) vs 6 (IQR 3–9); p=0.81) or ages (p=0.72) of those in whom consent was and was not obtained.
Of the 528 consent refusals, 429 legal guardians provided a reason for declining participation. The stated reasons for consent refusal are shown in figure 1. The most common reason for non-consent was that the legal guardians were under too much stress (27.3%, 117/429) with the next most common category being wanting nothing else done to their child (15.6%, 67/429). Research-related concerns were voiced in 17.3% (72/429, see table 3).
None of the seven sites reported a standardised approach or formal ethics board policy for assent, therefore, the decision to ask for assent was left to the discretion of the research assistant. There were 239 patients aged ≥12 years. Patient assent was requested in 16/239 patients for whom the legal guardian had either already provided consent for participation or deferred participation to the child. Fifteen of these patients refused, and one agreed but was subsequently over-ruled by the legal guardian (patient had attempted suicide).
Multiple factors including lack of availability of the legal guardians, language issues and child protective services involvement resulted in only half the eligible patients being approached for consent. It is difficult to determine if this is consistent with other studies as it has not been explicitly reported in previous paediatric critical care studies.5 ,14 ,15 However, this has important practical and ethical implications for both the interpretation of the results of this study as well as for the design, budget and implementation of future studies.
Consent for this non-therapeutic study was only 42% which is significantly lower than the consent rates seen with previous therapeutic studies (66–89%) with similar requirements for early recruitment.5 ,14 ,15 These studies also involved critically ill children and many of the same sites as this current study, but interestingly, also all provided the potential for direct benefit to the participating child. The AIP study, however, did not, and therefore, willingness to participate would have been based primarily on altruism. Only 1% of non-consenters in the AIP study stated that they did not wish to participate as there was no direct benefit to their child, but it is entirely possible that parents’ spontaneously expressed comments do not accurately reflect their true reasons for non-participation in research. In survey-based studies of theoretical scenarios, parents have stated a variable willingness to participate in research for altruistic reasons ranging from 32.4%16 to 72%.17 In the only two studies to spontaneously record parents’ stated reasons for consent refusal, lack of benefit to their child was not volunteered as a reason.2 ,10
The most commonly volunteered reason for consent refusal was that parents felt too overwhelmed. This finding is supported by others who have documented that parents suffer significant stress and anxiety when their children are admitted to intensive care.18 This raises the question of whether researchers can truly obtain informed consent in an intensive care environment, and the literature is divided on this issue. Two studies have suggested that despite a high level of anxiety, 90% of parents felt they had made an informed decision and understood the elements of a neonatal intensive care study,19 and 93% of parents demonstrated fair or excellent comprehension of their child's medical issues within the first 24 h following admission to the PICU.20 However, two other studies found that between 7.8%21 and 12%22 of parents had no recollection of the study they had agreed to participate in or of signing the consent form. Given the difficulties in obtaining informed consent from parents in a stressful environment, one potential approach might be to limit research in the PICU. However, both the Tri-Council Guidelines and the Oveida Convention emphasise the need for research to advance medical care, and for researchers to not automatically exclude those who cannot provide consent for themselves. The literature supports the fact that despite these recommendations, children are less likely to be enrolled in acute care clinical research studies,9 and that there is a lack of adequate medical research in children.23
Given the paucity of high-quality research in children, and the difficulties in obtaining informed consent in paediatric critical care studies, it behooves the research community (researchers, ethics boards, clinicians and the public) to explore options to improve this. Parents themselves have provided several suggestions for ways to decrease their stress levels in the intensive care unit including a visit from a pastor, consistency of nursing and the ability to talk about the possibility of their child's death.7 It has also been suggested that introduction of the research assistant to the parents by a member of the child's healthcare team may decrease parental stress and increase the likelihood of consent,2 and would be a simple procedure to implement. Another (more complicated) option is the use of deferred consent for critical care studies which is covered under the sections governing Research in Emergency Health Situations in the Tri-Council Guidelines and the Food and Drug Administration Guidelines in the USA. Deferred consent permits research ethics boards to allow research in health emergencies without the free and informed consent of the subject or their authorised third party, as long as several carefully laid out principles are followed. Deferred consent has rarely been used in paediatric critical care despite evidence suggesting that parents would support this model in certain cases.24
Strengths of this are that it is the largest study to systematically document the reasons for non-participation of eligible patients, and to record the stated reasons why legal guardians declined participation in a clinical research study within the PICU. Limitations of this study include that the stated reasons for declining consent may not accurately reflect the real reasons why legal guardians did not wish to participate. However, this is true of all survey studies which still yield useful information if interpreted within the context of this limitation. Another limitation of this study was that information on why legal guardians agreed to participate was not collected which would be an interesting area for future research. Finally, this study was conducted in paediatric critical care units in Canada and, therefore, may not be generalisable to other patient care settings or countries.
Barriers to consent procurement in a non-therapeutic paediatric critical care study appear to occur at many levels with lack of availability of legal guardians, and legal guardians feeling overwhelmed, being among the most commonly recorded reasons. Further research into the impact of these findings on the validity and generalizability of the results of such studies, as well as potential alternative approaches to consent in the paediatric critical care environment are necessary in order to advance the care of critically ill children in a valid and ethical manner.
Contributors Both authors made substantial contributions in the areas of conception and design, analysis and interpretation of data, drafting the article and revising it critically for important intellectual content and final approval of the version to be published.
Funding This study and the original AIP study were funded by the Canadian Institutes for Health Research which had no role in the design, implementation, analysis or write-up of this manuscript.
Competing interests None.
Ethics approval Children's Hospital of Eastern Ontario Research Ethics Board and the Ethics Boards of all the participating centres.
Provenance and peer review Not commissioned; externally peer reviewed.
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.