Objective To establish the empirical evidence base for the information that participants want to know about medical research and to assess how this relates to current guidance from the National Research Ethics Service (NRES).
Data sources Medline, Web of Science, Applied Social Sciences Index and Abstracts, Sociological abstracts, Health Management Information Consortium, Cochrane Library, thesis index's, grey literature databases, reference and cited article lists, key journals, Google Scholar and correspondence with expert authors.
Study selection Original research studies published between 1950 and October 2010 that asked potential participants to indicate how much or what types of information they wanted to be told about a research study or asked them to rate the importance of a specific piece of information were included.
Study appraisal and synthesis methods Studies were appraised based on the generalisability of results to the UK potential research participant population. A metadata analysis using basic thematic analysis was used to split results from papers into themes based on the sections of information that NRES recommends should be included in a participant information sheet.
Results 14 studies were included. Of the 20 pieces of information that NRES recommend should be included in patient information sheets for research pooled proportions could be calculated for seven themes. Results showed that potential participants wanted to be offered information about result dissemination (91% (95% CI 85% to 95%)), investigator conflicts of interest (48% (95% CI 27% to 69%)), the purpose of the study (76% (95% CI 27% to 100%)), voluntariness (39% (95% CI 2% to 100%)), how long the research would last (61% (95% CI 16% to 97%)), potential benefits (57% (95% CI 7% to 98%)) and confidentiality (44% (95% CI 10% to 82%)). The level of detail participants wanted to know was not explored comprehensively in the studies. There was no empirical evidence to support the level of information provision required by participants on the remaining seven items.
Conclusions There is limited empirical evidence on what potential participants want to know about research. The existing empirical evidence suggests that individuals may have very different needs and a more tailored evidence-based approach may be necessary.
Statistics from Altmetric.com
What information do potential participants want to know when they are deciding whether to take part in research?
What is the established empirical evidence base?
How does the current empirical evidence base relate to current guidance from the NRES?
There is little empirical evidence of what information potential participants want to know about research when they are making the decision to take part.
The limited empirical evidence available suggests that potential participants may have very different information needs.
Further research is required to determine what potential participants really want to know about research and how this can be delivered in a way that takes into account their different informational needs.
Strengths and limitations of this study
An extensive search strategy ensured that the review was systematic in capturing all available empirical evidence.
Papers included in the review differed in their methodologies and presentation of results, making comparisons between papers extremely difficult.
Medical research is central to the advancement of treatments, services and technology.1–3 Potential participants have the right to choose whether they participate in medical research,4 5 and individuals must give their consent prior to participating in research. As part of this ongoing process, potential participants must be provided with sufficient information to make a voluntary and informed decision.2 6–11 In research settings, study information is usually conveyed to potential participants in the form of a written participant information sheet (PIS), which is later reinforced by a verbal consent interview with a member of the research team.12
In the UK, the National Research Ethics Service (NRES) provides extensive guidance on how a PIS should be written and presented. The guidance suggests that a PIS should be split into two parts where part one provides a brief and clear explanation of the essential elements of the specific study and allows participants to make an initial choice of whether the study is of interest. Part two should then contain additional information on matters such as confidentiality, indemnity and publication intentions.
There is some concern that PIS have become increasingly lengthy over recent years.10 13 14 Complex studies, for example, where the potential participant might, for example, on the basis of test results be invited to participate in a further phase of the study, often use detailed and lengthy PISs. This can lead to poor understanding by participants15–17 and a corresponding concern that consent criteria are not always met. The NRES guidance is not explicit in the level of detail to be included in a PIS, and there is disagreement among experts about how much information to include.18 If PISs become so complex that only the most confident and educated participants are able to digest all the information, this may result in selection bias meaning that research is less generalisable.19 Furthermore, there is a risk that healthcare researchers are becoming increasingly paternalistic in their information provision without recognising individual participant needs. In order to help address the problem of how much information to include in PIS, we conducted a systematic review that aimed to establish the empirical evidence base for the information that potential participants want to know when they are deciding about participation.
Selection criteria and literature search
This systematic review included all studies that asked participants to indicate how much or what type of information they wanted to be told about a research study or asked them to rate the importance of a specific piece of information. We included studies published between 1950 and 27 October 2010 with no limit to language or participant group. We only included studies of participant opinion and excluded studies of healthcare professional or other expert opinion.
We combined Mesh terms Patient, Research Subjects, Consent forms, Informed Consent and Research ethics with terms relating to information provision (online appendix 1). We conducted searches in Medline, Web of Science, Applied Social Sciences Index and Abstracts, Sociological abstracts, Health Management Information Consortium and the Cochrane Library electronic databases. We also searched thesis index's, grey Literature databases, reference and cited article lists, key journals and Google Scholar and we asked expert authors to identify relevant studies.
We did not conduct a formal quality assessment of included literature because there were both quantitative and qualitative studies, widely varied study methods and different types of results that were often not comparable between papers. Instead, we conducted a critical appraisal of each paper using five quality indicators (response rate, sample size, demographics, participant characteristics and strengths and limitations of study methods). The strengths and limitations of each study are presented in table 1.
Data extraction and synthesis
One researcher (HMK) extracted data from papers using a pre-defined data extraction sheet and a second researcher (TK) checked it for accuracy with disagreements resolved by discussion between these two authors (table 1). A metadata analysis using basic thematic analysis was used to analyse the data from the 14 papers. Themes were based on the sections of information that NRES recommends should be included in a PIS (table 2).10 Each paper was assessed to identify any further themes relating to what information research participants may want to know. A metadata analysis coded individual results based on their relevance to each theme and then themes were collated to report overall results. For themes where more than one quantitative study reported a proportion of participants wanting to know the information, pooled proportions with random effects were calculated using StatsDirect statistical software (StatsDirect Ltd).
The search yielded 11 943 unique references. We discarded 11 291 after reviewing the title, 620 after reviewing the abstract and a further 18 after reviewing the full paper (figure 1). HMK conducted the citation screening and TK independently validated approximately 10% of the references identified from electronic databases (96.0% κ agreement rate). All 14 included studies were identified from searches of Medline and Applied Social Sciences Index and Abstracts. Expert authors identified 37 unique references; 13 were duplicates from the electronic searches and 24 did not meet the inclusion criteria.
Of the 14 studies included in the review, three specifically considered the return of research results to participants and six considered only investigator conflicts of interest. Five studies looked broadly at what information potential research participants wanted to know.
Of the 20 sections of information NRES suggest should be included in a PIS, there were seven categories where no empirical evidence was identified that suggested what information research participants wanted to know (table 2). No further themes, beyond the NRES categories, were identified. We were able to calculate pooled proportions for seven themes. Participants wanted to be told about dissemination of study results (91% (95% CI 85% to 95%)), investigator conflicts of interest (48% (95% CI 27% to 69%)), the purpose of the study (76% (95% CI 27% to 100%)), voluntariness (39% (95% CI 2% to 100%)), how long the research would last (61% (95% CI 16% to 97%)), benefits (57% (95% CI 7% to 98%)) and confidentiality (44% (95% CI 10% to 82%)). Although the majority of participants appeared to want information for most of these themes, some participants did not and the level of detail that participants wanted was not explored comprehensively.
Of the 14 papers that met inclusion criteria, five looked broadly at what information research participants wanted to know. These studies focused on the category of information required rather than how much detail participants wanted. All 14 studies had substantial limitations to generalisability when applied to the wider research population because, for example, they focused on specific subsections of the population, for example, six studies included only cancer patients23 24 26 28 30 31 and only one study conducted in the UK.30 A number of studies included only women21 26 28 30 and participants that were mostly Caucasian23 26 and well educated.23–25
In the absence of empirical evidence to suggest what information potential research participants want, the NRES have based their guidance on expert opinion. It does, however, mean that current information provision for research may not adequately address the informational needs of the general population or ‘hard to reach’ groups such as socially deprived or African–American and minority ethnic groups. While the NRES recognise that one size does not fit all and that low-risk studies with little or no intervention may need shorter information sheets, there is little empirical evidence to identify what level of information provision should be made.32 A potential difficulty in conducting research to determine what should be included in a PIS is that an individual's information preferences may change as they move from being a potential to actual participant.35 36
Responding to individuals' information needs may prove challenging, but the provision of high-quality appropriate information in a timely manner is crucial to the consent process. Electronic information provision may be one way to address different information needs. Recent research by Antoniou et al37 that allowed participants to access three increasingly detailed levels of information electronically found that the basic level of information was accessed by 70%–82% of participants, but only 9%–18% accessed the level of information currently recommended in NRES guidance and only 3%–12% accessed all three levels of information. Interestingly, 20% (93/552) participants that said they wanted more information even though fewer than this (3%–12%) read all the information available to them.
The study by Antoniou et al37 is an important first step in determining what information potential research participants really want to know when they agree to take part in a study. Further research is required to assess the feasibility and acceptability of unfolding electronic information sheets.
Ideally, differences in informational requirements for subgroups of the population would have been explored but the small numbers of studies identified and limited data extracted from papers meant this was not feasible.
There is limited empirical evidence as to what information potential participants want to know at the time they are deciding whether or not to participate in research. Real-time studies need to be conducted to explore what information potential participants access when given a choice. This will enable us to determine exactly what information research participants want to know and could, in addition to other sources such as expert opinion, help tailor PIS towards specific population subgroups and enable appropriate high-quality information to be provided to meet individual needs.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online appendix 1
To cite: Kirkby HM, Calvert M, Draper H, et al. What potential research participants want to know about research: a systematic review. BMJ Open 2012;2:e000509. doi:10.1136/bmjopen-2011-000509
Contributors HMK, MC, SW and HD conceived and designed the research. HMK and TK collected, validated and extracted the data. All authors made substantial contribution to the analysis and interpretation of the data. HMK drafted the manuscript and SW, HD, MC and TK revised it.
Funding The study was funded by the Medical Research Council Midland Hub for Trials Methodology Research (Medical Research Council Grant ID G0800808). The study sponsor had no role in study design, collection, analysis or interpretation of the data, in the writing of the report or in the decision to submit the article for publication. HMK and TK are PhD students funded by Medical Research Council Midland Hub for Trials Research Methodology and MC is Education Lead for the Medical Research Council Midland Hub for Trials Research Methodology.
Competing interests All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that (1) HMK, MC, HD, TK and SW have support from the University of Birmingham for the submitted work; (2) HMK, MC, HD, TK and SW have no relationships with any companies that might have an interest in the submitted work in the previous 3 years; (3) their spouses, partners or children have no financial relationships that may be relevant to the submitted work and (4) HMK, MC, HD, TK and SW have no non-financial interests that may be relevant to the submitted work. HD is an author of one of the papers included discussion.37 SW was also acknowledged in this paper for comments on an early draft.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement All authors had full access to all the data in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis. Technical appendix and data set available from the corresponding author at . Referenced Manager (Version 12) was used to analyse data. Stats Direct was used to calculate pooled proportions with random effects.
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.