BACKGROUND. Valid consent for research requires comprehensive and understandable information to be disclosed to participants. The way that information is shared varies, but regulatory bodies usually determine style. Some reports have suggested that although information may be all-inclusive, it does little to support understanding.
OBJECTIVE. To explore the impact of various information-sharing approaches on parents' understanding of a research study and the validity of their consent.
METHODS. This was a randomized, controlled trial. Parents of immature but well infants admitted to a large tertiary NICU in Edinburgh, Scotland, were randomly assigned within 72 hours of their infant's admission to receive 1 of 2 information leaflets, with or without a standardized verbal explanation, for a hypothetical intensive care research study. The leaflets differed in length and in the amount of detail in which the study process, risks, benefits, and patient rights were described. A questionnaire was used to elicit understanding about the purpose of the research, design of the study, procedures involved, and the consent process.
RESULTS. Forty-one parents participated in the study. Those who received the longer leaflet without verbal explanation gained only limited understanding of the purpose of the research. The procedures involved in the study were understood better by those who received the shorter leaflet. Issues relating to consent and study design were readily understood in all groups. Irrespective of documentation style, verbal explanation significantly improved understanding. Differences in understanding had little effect on whether a parent would enroll his or her infant into the study.
CONCLUSIONS. Verbal explanation significantly enhances understanding of the research process for participants regardless of the style of written documentation. However, shorter written information may lead to better understanding than lengthy, more complex documentation.
The principle of consent is founded in moral philosophy and the law and is expressed by individuals having the freedom to make choices about what happens to them. Historically, consent to medical treatment has been based in case law, whereas consent for research has been regulated by codes of practice.1 In recent years, research governance has become embedded in law in the United Kingdom and, as a consequence, transparency and accountability in the conduct of research has increased, particularly with respect to consent. There are 3 criteria by which the validity of consent may be assessed:
the person giving consent is deemed competent to do so;
the information given is comprehensible and likely to be relevant to an individual's decision whether to participate; and
participation is voluntary and free of coercion, and the participant is able to withdraw without penalty.
Influenced by national and international regulation,2–4 participant information about trials, and in particular drug trials, has resulted in complicated documentation that is highly detailed and lengthy. Beardsley et al5 showed that the numbers of pages in patient information and consent forms related to cancer trials have increased from a median of 7 to 11 between 2000 and 2005. More importantly, patients' knowledge about the trials was higher if the patient information and consent forms had ≤7 pages.
With newborns, the issue of consent for research is often compounded by parental stress, pain, and/or uncertainty around the time of delivery, which may compromise understanding of new or complex information. Ballard and colleagues6 questioned 64 parents who had consented to their infant participating in a randomized, controlled trial: 8% had no recollection of the trial, 68% understood the purpose of the trial, but only 5% of these parents understood potential risks. In a study by Stenson et al,7 12% of respondents could not remember being approached for a trial in which their infant had participated. Of those that remembered being approached, 89% felt an explanation had been given, but it was completely understood by only 27% of parents. Despite this lack of understanding, 62% of parents were completely happy with their decision to enroll their infant in the study.
Although systems are in place to safeguard potential subjects, outcomes from studies on the assessment of the consent process have given cause for concern. They suggest that, although the information required for consent is present, the quality of documents has not supported understanding.6,8–10
We tested parents' understanding of research and consent issues using information sources that had previously been used during the Neurologic Outcomes and Preemptive Analgesia in Neonates (NEOPAIN) trial.11
The NEOPAIN trial was an international multicenter trial that evaluated the effect of morphine infusion versus placebo on neurologic outcome in infants born between 23 and 32 weeks' gestation. At the time of enrollment in the NEOPAIN trial, infants were <72 hours old and required mechanical ventilation. The supporting documentation was structured to meet legal and ethical requirements of the various host countries.
The UK arm consisted of a single sheet of information, and the US arm consisted of 5 sheets of information. The documents also differed in the amount of detail in which the study process, risks, benefits, and patient rights were described. Statements about compensation, costs, and liability were included in the US leaflet but not the UK leaflet, reflecting international differences in regulations and requirements.
The purpose of the current study was to explore whether the differences in various information-sharing approaches contributed to parents' understanding and willingness to enroll their infant in a nonexistent, but previously conducted, research study (the NEOPAIN trial). A favorable ethical opinion and local National Health Service research and development approval were obtained for this study.
PATIENTS AND METHODS
The study period was between November 2003 and October 2005. A convenience sample of parents whose immature infant(s) were admitted to a large tertiary NICU in Edinburgh, but who did not require intensive care (ie, not requiring mechanical ventilation or continuous observation), were identified as potential participants. Parents were approached within 72 hours of their infant's admission, and written consent was obtained by the researchers (Drs Freer and Boyle). Demographic data on parent age, socioeconomic status, and previous experience with neonatal research were collected. The deprivation categories (DEPCAT) score was used as a composite measure of socioeconomic status. This is a postcode area measure of deprivation based on 4 variables: home occupancy, male employment status, social class, and car ownership.12 Scores are categorized from very high (DEPCAT 7) to very low levels of deprivation (DEPCAT 1). High levels of social and economic deprivation are associated with low educational attainment13; therefore, the DEPCAT score was used as a surrogate measurement of educational achievement and ability to process information.
Research Study Information
We were interested in the effect of different styles of documentation, but were aware that although the addition of a verbal explanation is common to the consent process, it is not necessarily a requirement. Thus, to evaluate the effect of different documentation style and the delivery of a standardized explanation, parents were randomly assigned to 1 of 4 arms: group 1, US leaflet alone; group 2, US leaflet and explanation; group 3, UK leaflet alone; and group 4, UK leaflet and explanation.
Because some spelling and terminology differs between British and American use of the English language, the US documentation was “translated” for the purpose of this study to avoid confusion for participants. The 2 leaflets were assessed using a readily available “readability” tool, the Flesch Reading Ease and Flesch-Kincaid Grade Level tools. These tools indicate how difficult or easy a document is to read and the reading age one is expected to have to readily understand the content.
Randomization was conducted using sequential, sealed, opaque envelopes, prepared by an independent person. The researcher gave the standardized explanation at the time parents received the written information; although the explanation was standardized, the process allowed for questions by the parents and additional clarification and explanation by the researcher.
Parents were then asked to complete a questionnaire developed specifically for the study. This consisted of questions about parental sociodemographic status and 16 true and false statements relating to the study and research participation. The statements covered 4 themes necessary for consent: background and purpose of the study, design of the study, procedures involved in the study, and information specific to consent procedures (for example voluntariness and ability to withdraw). To elicit understanding, parents were asked to identify the correctness of the statements. When completing the questionnaire, they were asked not to refer back to the information leaflet. The time taken for parents to return the questionnaire varied depending on how soon they were able to find time to complete it. All completed questionnaires were retrieved within 24 hours, and the same researcher reviewed all responses.
The primary outcomes were differences between:
the US leaflet group and the UK leaflet group in summary scores of understanding;
the group receiving a verbal explanation and the group without an explanation in summary scores of understanding; and
the 4 groups in willingness to allow their infant to be entered into the study.
The secondary outcomes were differences in the appreciation of various elements of information between the 4 groups.
This was a pragmatic exploratory study. We knew of no previous data on which to base a power calculation and, therefore, a calculation could not be conducted. It was decided to try to recruit 10 parents to each of the groups.
A single score was generated for each participant on the basis of the proportion of correct answers given. A mean score for each randomized study group was then derived. Mean scores were compared by using 1-way analysis of variance and Tukey's honestly significant differences test. Data were analyzed first for all questionnaire responses in total and then further analyzed by statement theme.
Fifty-four parents whose infant(s) were admitted to the NICU but who did not require intensive care agreed to participate. Of these, 41 (76%) returned the questionnaires. Reasons for parents not completing the questionnaires were earlier-than-expected discharge from hospital or lack of time. Minor differences existed in baseline characteristics of participating parents (Table 1); overall, the mean maternal age was 31.5 years (SD: 6.2) and the median DEPCAT score was 4 (range: 1–6). Those not returning their questionnaire were slightly younger (29.7 years [SD: 7.69]) and more deprived (median DEPCAT score: 5 [range: 4–6]) than those returning their questionnaire. All those participating were able to read. Consent was given by the mother on 63% and by both parents on 10% of occasions. Five parents had previously participated in research but not in neonatal research.
When applying the Flesch Reading Ease tools, the UK leaflet had a higher reading score than the US leaflet (56.7% vs 43.3%, respectively), meaning it was easier to read. The Flesch-Kincaid Grade Level tool gave a grade level score of 11 for the UK leaflet and 12 for the US leaflet. These are equivalent to reading ages of 15 to 16 years and 16 to 18 years for the UK and US leaflets, respectively.
Mean scores for the randomized groups were 0.66 (SD: 0.18) for group 1, 0.75 (SD: 0.16) for group 2, 0.69 (SD: 0.17) for group 3, and 0.85 (SD: 0.09) for group 4. Analysis of variance revealed a significant difference between the 4 groups (P = .036); Tukey's test showed this difference to lie between the US (no explanation) and UK (with explanation) groups.
Irrespective of documentation style, the addition of a standardized explanation improved understanding (P = .013). For those receiving the US leaflet, the addition of an explanation increased the mean score, but this did not reach statistical significance.
There was, however, a significant difference between the 2 UK groups with the explanation supporting understanding (P = .015).
To establish whether there was better understanding within the various elements of documentation, group scores were further analyzed by statement theme (Table 2). Analysis of variance and Tukey's test were used to explore differences between groups. There was limited understanding about the background and purpose of the study, particularly in those receiving the US leaflet without verbal explanation. There were significant differences in scores between groups 1 (US leaflet without explanation) and 2 (US leaflet with explanation) (P = .022), and between groups 1 (US leaflet without explanation) and 4 (UK leaflet with explanation (P = .012). There was no significant difference in parents' perceptions about the concepts of study design (eg, blinding and randomization) between groups, although the addition of an explanation enhanced understanding. Those receiving the UK leaflet showed significantly greater understanding of the procedures involved in the study when compared with those receiving the US leaflet (P = .034). Mean scores for groups 1, 2, and 3 were 0.60 (SD: 0.29), 0.60 (SD: 0.27), and 0.66 (SD: 0.17), respectively; parents who received the UK leaflet and explanation scored 0.87 (SD: 0.13), indicating significantly greater understanding of the procedures than parents in either group 1 or 2 (P = .045). Issues relating to consent procedures were readily understood irrespective of whether the leaflet was the UK or US version or accompanied by the explanation.
Parents' perceptions about the amount of information received and its influence on enrollment were evaluated. Of the parents receiving the US leaflet, 58% (11 of 19) thought the amount of information was optimal, 37% (7 of 19) thought the amount was too much, and 5% (1 of 19) thought the amount was too little. Of the parents receiving the UK leaflet, 70% (14 of 20) thought the amount was optimal, and 30% (6 of 20) thought the amount was too little. The perceived amount of information received had no effect on parents' decisions to enroll their infant into the study (US versus UK information, 47% vs 55%, respectively, would enroll their infant).
The centrality of autonomy lies in an individual's freedom of choice. Before seeking voluntary agreement of the participant to enter a research study, all relevant information must be disclosed. Furthermore, current interpretation of these 2 criteria also requires comprehension of the information on part of the participant.14 Literature on validity of the consent process is based on these criteria and in the main, these studies have concluded that failure in understanding is a fault of study design, documentation, and/or process.6,8–10,15 To overcome these shortcomings, investigators have assessed the value of different styles of information format and delivery on retention and comprehension of the message.16–18 It is clear from this work that the standardized format required by regulatory and review bodies is not helpful.
In our study, all participants were able to demonstrate knowledge and understanding of some of the essential components of the consent process for a clinical trial. Understanding was better in those presented with a concise rather than lengthy leaflet. The use of additional verbal explanation optimized understanding irrespective of the type of written information. A number of factors affecting retention and comprehension of research-related information are explored to explain our findings.
Because spelling and syntax differ between written use of the English language in the 2 countries, the US leaflet was “translated” to make a number of words more familiar for British parents. Although these alterations might have had an impact on the interpretation of the information provided in the leaflet, there was no change in content, and we believed the alterations would enhance understanding in this setting.
In the most recent advice from the National Research Ethics Service,3 it is suggested that the language level used in participant information leaflets should be no more difficult than that used in information leaflets of medicines for the general public or in tabloid newspapers. We used a readability formula to determine if there was a difference in ease of reading. The UK leaflet had a reading score slightly higher than the US leaflet, meaning it was easier to read, but as it was still categorized as being fairly difficult, the difference is thought not to contribute greatly to understanding. Although readability formulas are often used to determine the level of ease/difficulty, they have their limitations and do not reflect other factors affecting comprehension (eg, medical terminology, writing style).19 Parents receiving the US leaflet were more likely to report the amount of information as being too much; it may be that, as shown by Beardsley et al,5 volume of information rather than content was the influential factor on understanding. However, no matter how well written the participant information leaflet, if the participant has limited literacy, the information will be incomprehensible. With the parents in our study, we were able to assess level of deprivation using the DEPCAT score and from that inferred their ability to read, write, and reason.13 There was little difference in the DEPCAT scores between the groups of parents, and we postulate that it is unlikely the differences seen in our findings can be attributed to aptitude. Having shown there was little difference between the leaflets in reading score and between the groups of parents in the DEPCAT score, it was interesting to find that the verbal explanation supported recall and understanding in both groups but more so in the group with the concise leaflet.20–22 It is possible that the briefness of the UK leaflet made it easier to read but left the reader wanting more information, whereas the length of the US leaflet may have made it more difficult to read and, therefore, necessitated clarification.
Another potential confounding factor is the parents' knowledge that this was a hypothetical study. It is possible that this may have led them to read the leaflets less thoroughly than if they were truly being asked to give consent for their infant to enter a real study. None of the infants whose parents were approached would have been eligible to enter the NEOPAIN study, as they were not sick enough. Although this may have affected our results, we considered it unethical to approach parents of infants who were extremely sick for this type of research. However, in including parents of infants who were less sick, but required neonatal unit admission, we were able to replicate the parental anxiety arising from separation and uncertainty.
It has been argued that the current “text-assessment” approach to ascertaining participants' recall and understanding of the meaning of medical information ignores interaction between the text itself and the participant.23 Dixon-Woods et al23 discuss how individual background knowledge, level of anxiety, personal beliefs, and experiences invariably influence conceptual understanding irrespective of the format of the information leaflet or message. These authors contend that although it is right that information should be presented in a form that is not misleading and that allows participants to arrive at decisions that conform to their beliefs, understanding of the complexities and possible risks within a message or study protocol may be of little consequence in that decision making.
This exploratory work was limited by the small sample size, which necessitates caution in attributing differences in understanding exclusively to leaflet type and provision of a verbal explanation. In addition, because our sample represents people consenting on behalf of their child rather than for themselves, it limits their representativeness and, therefore, the generalization of our findings to the wider population of research participants. Nevertheless, our results suggest that in this group the shorter leaflet was effective at improving understanding of consent and research.
The consent process is founded on basic moral principles of autonomy, with regulatory processes in place to prevent participant exploitation. We have shown that a concise information leaflet and verbal explanation leads to better understanding of various elements of a study that are prerequisites to gaining valid consent. We have also shown that additional verbal explanation is a crucial part of the process, allowing important interaction between investigator and participant. However, there were still aspects where parents demonstrated incomplete understanding. Given that participants agree to take part in studies for a variety of reasons: societal benefit, personal benefit, and perceived risk/harm, it is time to reconsider “comprehension” as being a fundamental of valid consent.24 Given the significant findings of our small study, we recommend continuing consideration of these and other important issues in neonatal consent in an additional study involving larger numbers of parents.
- Accepted August 27, 2008.
- Address correspondence to Elaine M. Boyle, MB, ChB, MD, MSc, University of Leicester, Department of Health Sciences, Leicester LE1 6TP, United Kingdom. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
Dr Freer collected, analyzed, and interpreted the data, drafted the article, and jointly approved the final version of the manuscript; Dr Boyle conceived and designed the study and questionnaire, collected the data, revised the manuscript critically, and approved the final version of the manuscript; and Drs Teunisse, Anand, and McIntosh conceived and designed the study and questionnaire, revised the manuscript critically, and jointly approved the final version of the manuscript.
What's Known on This Subject
Informed consent is an essential component of clinical research and relies on comprehensive and understandable information being given to participants. In neonatal research, the consent process is complicated by parental stress around delivery, which may compromise understanding of complex information.
What This Study Adds
Short and concise written information enables better participant understanding of the research process than lengthy documentation. Understanding is significantly improved by the addition of a verbal explanation.
- ↵del Carmen MG, Joffe S. Informed consent for medical treatment and research: a review. Oncologist.2005;10 (8):636– 641
- ↵Office of Human Subjects Research. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research. Bethesda, MD: National Institutes of Health; 1979. Available at: http://ohsr.od.nih.gov/guidelines/belmont.html#goc1. Accessed October 15, 2007
- ↵National Research Ethics Service. Information Sheets & Consent Forms Guidance for Researchers & Reviewers. London, United Kingdom: National Patient Safety Agency; 2007. Available at: www.nres.npsa.nhs.uk/rec-community/guidance/#PIS. Accessed March 11, 2009
- ↵US Federal Government. General requirements for informed consent. In: Code of Federal Regulations. 45 CFR 46.116. Washington, DC: Department of Health and Human Services; 2005. Available at: www.hhs.gov/ohrp/documents/OHRPRegulations.pdf. Accessed September 29, 2007
- ↵Beardsley E, Jefford M, Mileshkin L. Longer consent forms for clinical trials compromise patient understanding: so why are they lengthening? J Clin Oncol.2007;25 (9):e13– e14
- ↵Stenson BJ, Becher JC, McIntosh N. Neonatal research: the parental perspective. Arch Dis Child Fetal Neonatal Ed.2004;89 (4):F321– F324
- ↵Burgess E, Singhal N, Amin H, McMillan DD, Devrome H. Consent for clinical research in the neonatal intensive care unit: a retrospective survey and a prospective study. Arch Dis Child Fetal Neonatal Ed.2003;88 (4):F280– F286
- ↵Carstairs V, Morris R. Deprivation and Health in Scotland. Aberdeen, Scotland: Aberdeen University Press; 1991
- ↵Sparkes J. Schools, education and social exclusion. CASE article 29. Available at: http://sticerd.lse.ac.uk/dps/case/cp/CASEpaper29.pdf. Accessed October 13, 2005
- ↵Beauchamp TL, Childress JF. Principals of Biomedical Ethics. 5th ed. New York, NY: Oxford University Press; 2001:77–93
- ↵Coyne CA, Xu R, Raich P, et al. Randomized, controlled trial of an easy-to-read informed consent statement for clinical trial participation: a study of the Eastern Cooperative Oncology Group. J Clin Oncol.2003;21 (5):836– 842
- Greenley RN, Drotar D, Zyzanski SJ, Kodish E. Stability of parental understanding of random assignment in childhood leukemia trials: an empirical examination of informed consent. J Clin Oncol.2006;24 (6):891– 897
- ↵Hoehn KS, Wernovsky G, Rychik J, et al. What factors are important to parents making decisions about neonatal research? Arch Dis Child Fetal Neonatal Ed.2005;90 (3):F267– F269
- Copyright © 2009 by the American Academy of Pediatrics