Article Text

Download PDFPDF

Improved pregnant women’s understanding of research information by an enhanced informed consent form: a randomised controlled study nested in neonatal research
  1. Nut Koonrungsesomboon1,2,
  2. Chanchai Traivaree3,
  3. Sangkae Chamnanvanakij3,
  4. Pimchitr Rungtragoolchai3,
  5. Yawana Thanapat4,5,
  6. Juntra Karbwang2
  1. 1 Department of Pharmacology, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand
  2. 2 Department of Clinical Product Development, Institute of Tropical Medicine, Nagasaki University, Nagasaki, Japan
  3. 3 Department of Pediatrics, Phramongkutklao Hospital and Phramongkutklao College of Medicine, Bangkok, Thailand
  4. 4 Department of Royal Thai Army Medical, Phramongkutklao College of Medicine, Bangkok, Thailand
  5. 5 Bumrungrad International Hospital, Bangkok, Thailand
  1. Correspondence to Dr Juntra Karbwang, Department of Clinical Product Development, Institute of Tropical Medicine, Nagasaki University, 1-12-4 Sakamoto, Nagasaki 852-8523, Japan; karbwangj{at}nagasaki-u.ac.jp

Abstract

Objective This study aimed to test the applicability and effectiveness of the enhanced informed consent form (ICF) methodology, proposed by the Strategic Initiative for Developing Capacity in Ethical Review (SIDCER), in neonatal research requiring maternal consent.

Design A single-centre open-label randomised controlled study.

Setting Antenatal care clinics at Phramongkutklao Hospital, Thailand.

Patients 234 pregnant women who were at risk of preterm labour were enrolled; 232 individuals completed the study.

Interventions The participants were randomly assigned to read either the SIDCER ICF or the conventional ICF.

Main outcome measures The participants’ understanding of essential trial-related information was assessed using 25 closed-ended questions. The primary endpoint was the proportion of the participants who obtained the satisfactory level of understanding at 80% (score of ≥20/25).

Results 72.5% (87/120) of the participants in the SIDCER ICF group and 59.8% (67/112) of the conventional ICF group achieved the primary endpoint (relative risk (RR)=1.212, 95% CI 1.005 to 1.462, p=0.041). The superiority of the SIDCER ICF over the conventional ICF was significant, particularly among the participants whose education was at the high school level or below (63.5% vs 44.1%, RR=1.441, 95% CI 1.022 to 2.030, p=0.031).

Conclusions The SIDCER ICF methodology is applicable to neonatal research requiring maternal consent. The SIDCER ICF significantly improved the understanding of pregnant women, particularly among those with lower levels of education. The present study confirms the value of the SIDCER ICF methodology in research involving individuals with a limited academic background.

  • parental consent
  • consent forms
  • newborn
  • neonatal research
  • pregnant women
View Full Text

Statistics from Altmetric.com

What is already known on this topic?

  • Despite the ethical and legal obligations of parental consent in neonatal research, several challenges exist in an attempt to reconcile practice with the theoretical ideal. Informed consent forms (ICFs) currently being used in a number of trials are lengthy and complex.

  • A limited understanding of trial-related information among the parents of newborn subjects is commonly evidenced in the literature.

What this study adds?

  • The present study addresses the enduring issue of a limited understanding of information among surrogates, particularly those with limited academic background.

  • The SIDCER ICF methodology could assist neonatologists in the development of ICFs to enhance the validity of parental consent.

Introduction

The inclusion of newborn babies in clinical research during childbirth requires permission from the parent(s) prior to delivery. Under this situation, the validity of parental consent may be compromised by maternal conditions, stress of labour or the sense of responsibility that accompanies a decision about a trial.1 Any of these factors could render pregnant women (and perhaps even their husbands/partners) vulnerable.2 Despite the ethical and legal imperatives of informed consent, some parents may be unaware of the fact that they have given proxy consent on behalf of their baby for enrolment in a trial.3 Empirical studies show that several parents were not cognisant of the risks involved and the fact that they could have chosen to refuse to let their baby participate in or withdraw their baby from the trial.4 5 Inadequate understanding of parents could jeopardise their vulnerable newborn’s best interests, and this would be intensified if there are complications in the infant’s health.

The quality of information disclosure and the adequacy of parental understanding are of critical importance in neonatal research.6 An informed consent form (ICF) must contain complete and accurate information and be presented in a simple, concise and comprehensible format that could facilitate the parents' understanding of the given information.7 A well-written ICF could guarantee that adequate information for decision-making is provided, while informed consent discussion with investigators may vary from one parent to another. However, empirical observation of the current practice indicates that many ICFs are excessively long, incomprehensible or incomplete, making them less likely to be read and understood.8–11 There is therefore a need to improve the quality of ICFs in order to ensure parental understanding and the validity of the consent obtained.

The Strategic Initiative for Developing Capacity in Ethical Review (SIDCER) has recently proposed ICF development methodology to address such a need.12 The SIDCER ICF methodology has been shown to be effective in improving participants’ understanding of research information in previous studies, especially among those with lower levels of education.13 14 The objective of the present study was to test the applicability and effectiveness of this methodology in neonatal research requiring maternal consent.

Methods

Study design, setting and research context

This study was an open-label randomised controlled study of two different ICF interventions (1:1)—the SIDCER ICF and the conventional ICF—using the post-test questionnaire as an assessment tool. This ICF study was conducted at Phramongkutklao Hospital, Bangkok, Thailand. The study protocol and related documents received ethical approval from the Institutional Review Board (IRB) of the Royal Thai Army Medical Department.

The study was nested in two neonatal trials which assessed the effect of delayed cord clamping versus cord milking in preterm infants born at <34 weeks of gestation (TCTR20150106001) and at 34(+1)–36(+6) weeks of gestation (TCTR20150107001). These two trials shared the same informed consent process and document and had been approved by the IRB. Pregnant women who were at risk of preterm labour were recruited at antenatal care clinics at Phramongkutklao Hospital. Maternal informed consent for the enrolment of their baby in the neonatal trials was obtained during a prenatal visit.

ICF interventions: SIDCER ICF and conventional ICF

The original ICF (in Thai) for the neonatal trials was considered as the conventional ICF and acted as a control in this study. It had seven pages with 3237 words.

The SIDCER ICF (in Thai) was developed following the SIDCER ICF methodology extensively described elsewhere.12 Study-specific information that is essential and relevant to prospective pregnant women’s decision-making on whether to let their baby participate in research was incorporated into the ICF template (available from: http://ijme.in/pdf/appendix1.pdf?v=1). The SIDCER ICF had five pages with 1957 words (see online supplementary figure 1)

Supplementary file 1

Assessment tool: the post-test questionnaire

An assessment tool was modified from the post-test questionnaire used in previous studies13–15 to include study-specific information of the neonatal trials. The questionnaire (in Thai) consisted of 25 scenarios followed by closed-ended questions addressing 25 required elements of the ICF content: 5 general items, 4 rights of the participant, 8 scientific aspects and 8 ethical aspects.12 Of three answer choices provided in each question, there was only one correct answer with a score of 1, making the maximum score 25.

Study endpoints

The primary endpoint was the proportion of the participants who had a post-test score of ≥20/25 (≥80%). The secondary endpoints included the total score, the score of each category of the ICF elements and the time spent reading a given ICF and completing the questionnaire.

Sample size determination

Using G*power 3.1.9.2 for Windows, the z-test was applied to detect the 20% difference between two independent proportions of the primary endpoint (p 1=0.8 and p 2=0.6), with the precision and confidence level of 95%, 80% power, allocation ratio of 1 and a continuity correction.16 Given an estimate of 10% for missing data, a sample size of at least 101 participants was adequate for each group.

Randomisation and allocation concealment

Stratified randomisation with a block size of four was applied for balancing the participants' educational level between the two groups. A randomisation code was generated and packed in an opaque sealed envelope. The envelope was opened only after a participant’s demographic information was collected.

Study procedure

During the process of informed consent of the neonatal trials, pregnant women were informed about and invited to participate in this ICF study. Individuals who declined to take part in the study or could not read and write Thai were excluded from the study procedure.

Participants were randomly assigned to read either the SIDCER ICF or the conventional ICF. While reading, they could ask any questions until they achieved satisfaction; the questionnaire was then distributed. Individuals were allowed to keep and consult the ICF while completing the questionnaire, but they could not ask any questions during this process. Time spent reading the ICF and doing the questionnaire was recorded. After the post-test, the research team collected the questionnaire and reviewed the answers. Any inaccurate understanding of information was explained before the participants signed consent for their baby’s participation in the neonatal trials.

Data analysis

Descriptive data were analysed and presented as frequency, median or mean, as appropriate. The proportion of the participants in the SIDCER ICF group who achieved the outcome divided by that of the conventional ICF group was presented as relative risk (RR). The χ2 test was applied for the comparison of nominal data between the two groups. After testing for normality with the Shapiro–Wilk test, the Wilcoxon–Mann–Whitney test was applied for the comparison of interval data between the two groups since none of the data were normally distributed. Subgroup analyses were performed by applying the χ2 test to determine the impact of age and education. Any association between demographic variables and the primary outcome was examined using a multivariable logistic regression model.

Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 22.0. A statistical significance was set at α=0.05 for all tests.

Results

A total of 234 pregnant women were enrolled between March 2015 and February 2016; 120 were randomly assigned to the SIDCER ICF group and 114 to the conventional ICF group (figure 1). Two pregnant women in the conventional ICF group were excluded from analysis because they did not complete the questionnaire. The demographic characteristics of the 232 participants are described in table 1.

Table 1

Demographic data of the participants (n=232)

Figure 1

Study procedure. *Two individuals withdrew from the study after reading the conventional informed consent form (ICF); they did not perform the post-test questionnaire so they were excluded from the analysis.

Eighty-seven of the 120 participants (72.5%) in the SIDCER ICF group achieved the primary endpoint, which was significantly higher than 59.8% (67/112) of the conventional ICF group (RR=1.212, 95% CI 1.005 to 1.462, p=0.041; figure 2). Subgroup analyses on the primary endpoint showed that participants aged 30 years or younger and those whose education was at the high school level or below statistically significantly favoured the SIDCER ICF (figure 2). Multivariable logistic regression demonstrated that the educational level of the participants was significantly associated with the primary endpoint (OR 3.003, 95% CI 1.626 to 5.545, p<0.001) whereas age was not (OR 1.031, 95% CI 0.986 to 1.079, p=0.177).

Figure 2

Proportion of participants whose post-test score was satisfied according to the 80% pass level (primary endpoint). *Relative risk is the ratio derived from the proportion of participants in the SIDCER ICF group with a satisfactory score according to the 80% pass level divided by that of the conventional ICF group. ICF, informed consent form.

The assessment of the secondary endpoints showed that the total score of the post-test and the score of ethical aspects in the SIDCER ICF group were significantly higher than those of the conventional ICF group (total score: 21/25 vs 20/25, p=0.005; ethical aspects: 7/8 vs 6/8, p<0.001). However, there was no statistical difference between the two groups in the scores of the other aspects (general items: 4/5 vs 4/5, p=0.222; rights of the participant: 4/4 vs 4/4, p=0.156; and scientific aspects: 7/8 vs 6/8, p=0.055). The participants in the SIDCER ICF group spent less time reading a given ICF than those in the conventional ICF group (11.5 min vs 15.0 min, p=0.040), while time spent completing the questionnaire did not differ between the two groups (15.0 min vs 15.0 min, p=0.285).

The proportion of the participants who correctly answered each element was compared between the two groups. Of 25 elements, the results significantly favoured the SIDCER ICF in five elements—namely, foreseeable risks (RR=2.352, p<0.001), compensation for injury (RR=1.324, p=0.014), identification of any experimental procedures (RR=1.313, p=0.010), confidentiality of records (RR=1.223, p=0.018) and number of subjects required (RR=1.118, p=0.007) (figure 3). However, the conventional ICF was significantly favoured in one element—namely, trial treatment and random assignment (RR=0.814, p=0.041) (figure 3).

Figure 3

Proportion of participants who correctly answered each item in the post-test questionnaire. *Relative risk is the ratio derived from the proportion of participants in the SIDCER informed consent form (ICF) group who correctly answered each item in the questionnaire divided by that of the conventional ICF group.

Discussion

This is the first study to validate the applicability and effectiveness of the SIDCER ICF methodology on surrogate consent. It demonstrates that the SIDCER ICF methodology was applicable and effective in improving maternal understanding of research information in the real setting. While the nature of informed consent is generally context-specific, the results of our previous and present studies consistently support the applicability and effectiveness of the general features of the SIDCER ICF methodology in various settings.13 14 This could be attributable to the fact that the SIDCER ICF is concise, organises all the required elements in a simple format, and focuses on relevant information for the participants’ decision-making.12 In addition, the SIDCER ICF makes use of several presentational means including a narrative format, summary boxes, highlight keywords and illustrations, as appropriate. The evidence suggests that shortening the length of ICFs alone seems to be inadequate to produce significant improvement in comprehension.17–19 Pictorial aids and enhanced processability partly or synergistically contribute to an improved understandability of written information.20–23

Individuals’ educational background was found to be a major determinant of maternal understanding as well as the impact of the SIDCER ICF on the improved understanding. The present study showed that educational attainment was independently associated with the level of understanding regardless of the type of ICF, as demonstrated by multivariable analysis. This finding supports the current dogma in the field that individuals with higher levels of education have a better understanding of the consent information compared with those with lower levels of education.24 25 Furthermore, it has been shown that the SIDCER ICF was particularly effective among participants whose education was at the high school level or below, but it had little effect on those who had a bachelor degree or above (see figure 2 and Koonrungsesomboon et al 14). This indicates the value of the SIDCER ICF in clinical research involving participants with low levels of education who are commonly found to have a limited understanding of information in ICFs.24 25

Following the SIDCER ICF methodology ensures that the ICF for a particular study is complete with respect to compliance with internationally recognised regulatory ICF requirements.12 The use of the SIDCER ICF methodology in neonatal research could thus ensure or enhance the validity of parental consent obtained. Although an interactive dialogue during the whole process of informed consent may substantially contribute to parental understanding,26 it could be variable. The evidence demonstrates that most investigators do not cover all content elements critical to fully informed consent in their discussions with parents.27 28 Well-written ICFs can ensure that parents receive sufficient and standardised information that enables them to arrive at informed decisions. The clarity of disclosure and ease of reading written materials could improve comprehension.29

In the present study the process of obtaining maternal consent took place during a prenatal visit to minimise factors that could interfere with maternal decision-making during the stage of active labour. In neonatal research where interventions are administered during childbirth, investigators are required to provide information to pregnant mothers at the right time, in the right manner, and in the right setting.30 Obtaining maternal consent on behalf of unborn babies during an intrapartum period may result in an invalid consent. Stress of labour and time limit can potentially lead a pregnant mother to consent to something that she would not have consented to in other environments or situations.31 In addition, pregnant women must recognise that surrogate decisions should be tailored to and be responsible for the best interests of newborn subjects.32 Their role as a proxy decision maker is one of the protective measures for their vulnerable and dependent newborn baby, and it is very different from their own consent whose role is to serve their own rights and interests.33

It was noticed that, among the essential ICF elements, the issue of foreseeable risks was least understood by the participants in both groups (figure 3). This finding is consistent with a recent integrative review on ICFs which showed that the section describing the risks of research participation is generally least understood.23 Inadequate understanding of risks is commonly evidenced as a major ethical challenge in conducting neonatal or paediatric research requiring parental consent.4 34 35 Although the SIDCER ICF was found to be superior to the conventional ICF on the risk issue (52.5% vs 22.3%, p<0.001), there is a need to consider additional means that can enhance visualisation (eg, using more graphic displays) or simplifying such information and tailoring it to better suit the target population.36 37

In the present study the conventional ICF conveyed the information related to treatment allocation better than the SIDCER ICF (figure 3). This might be attributable to better clarity or familiarity of the terms used to describe the concept of randomisation in the conventional ICF. Randomisation is one of the most difficult aspects in research and a lack of understanding of random allocation is commonly found in the literature.38–40 In this study, the term ‘at random like a coin flip’ (in Thai) was used in the conventional ICF, while the SIDCER ICF used ‘by random assignment with an equal chance of being assigned to either group’ (in Thai). It is important to take note of this weakness in the SIDCER ICF; using more colloquial terms could greatly improve the ICF.

Conclusion

The present study demonstrates the applicability and effectiveness of the SIDCER ICF methodology in neonatal research requiring maternal consent. The SIDCER ICF was effective in improving understanding of research information among pregnant women in this study, particularly those with lower levels of education. The SIDCER ICF methodology could be beneficial in research involving individuals with limited academic background.

Acknowledgments

We thank Dr Prisana Panichkul, Dr Benjamaporn Puangpaka, and Dr Tanai Trongkamontham, the investigators of the two neonatal trials, for their collaboration. Thanks are extended to Ms Chotimanee Kaewserm and her colleagues for their assistance in reviewing the SIDCER ICF from laypersons’ perspectives.

Acknowledgments

We thank Dr Prisana Panichkul, Dr Benjamaporn Puangpaka and Dr Tanai Trongkamontham, the investigators of the two neonatal trials, for their collaboration. Thanks are extended to Ms Chotimanee Kaewserm and her colleagues for their assistance in reviewing the SIDCER ICF from laypersons’ perspectives.

References

View Abstract

Footnotes

  • Contributor The studywas designed by NK, CT, SC and JK and conducted by CT, SC, PR and YT. Dataanalysis was done by NK and JK. The manuscript was written by NK and JK withcontributions from all authors. All authors approved the final versionsubmitted.

  • Contributors The study was designed by NK, CT, SC and JK and conducted by CT, SC, PR and YT. Data analysis was done by NK and JK. The manuscript was written by NK and JK with contributions from all authors. All authors approved the final version submitted.

  • Funding This study was partially supported by Phramongkutklao Hospital and Phramongkutklao College of Medicine and TDR, the Special Programme for Research and Training in Tropical Diseases, co-sponsored by UNICEF, UNDP, the World Bank and WHO through the Forum for Ethical Review Committees in the Asian and Western Pacific region (FERCAP). The funding sources had no role in the analysis and interpretation of the data; preparation, review or approval of the manuscript; and decision to submit this manuscript for publication.

  • Competing interests None declared

  • Ethics approval Institutional Review Board of Royal Thai Army Medical Department.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

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.

Linked Articles

  • Editorial
    Elizabeth E Foglia
  • Fantoms
    Ben Stenson