Abstract
Gaining informed consent from people being researched is central to ethical research practice. There are, however, a number of factors that make the issue of informed consent problematic, especially in research involving members of vulnerable groups such as children and people with learning disabilities. This paper reports on an ESRC-funded project concerned to identify and disseminate best practice in relation to informed consent in research with six such groups. This project involves the analysis of researchers’ views and experiences of the meaning of informed consent; of what is regarded as good and bad practice in obtaining and managing informed consent; of the impact that different methods of gaining informed consent have on the research process; and of the impact ‘gatekeepers’ or ‘stakeholders’ have on the research that is undertaken. The paper will argue that although guidelines on good practice are an increasingly important part of the framework of research governance, they cannot completely resolve the difficulties inherent in the processes of gaining and maintaining informed consent. It highlights two issues in particular: first, that there is no consensus amongst researchers concerning how to know when enough has been done to achieve informed consent, and when the point of doing too much has been reached; and secondly, that there is no consensus about whether the same sets of principles and procedures are equally applicable to research among different groups and to research conducted within different disciplinary frameworks.

Paper to the annual conference of the British Sociological Association at the University of York, 22-24 March 2004: ‘Sociological Challenges: Conflict, Anxiety and Discontent’

Introduction
This paper reports on findings arising out of the project ‘Informed consent and the research process’ which is being funded under the ESRC’s Research Methods Programme. It is a report on work in progress, since the data-collection phase of the project is not yet complete, and we do not yet have the informed consent from all of our participants to use the material collected so far. This is because we have offered participants the opportunity to read through the transcript of the interview or focus group in which they have been involved and to comment on it before any part of it becomes more public. The material discussed in this paper comes from approximately one third of our participants. Rather than attempt to paraphrase their contributions we have chosen to quote directly from their remarks. The project involves 30 interviews (mainly conducted by telephone) and 6 focus groups designed to elicit information from researchers relating to their views on and practices around gaining informed consent from people being researched. The researchers contacted for this purpose are people who have experience of collecting, analysing and reporting on data relating to groups that are often constructed as potentially vulnerable within the research process: children, school-based research, young people, older people, people receiving palliative care, and people with learning disabilities and mental health problems. We are very conscious of the fact that a larger study would have enabled us to contact directly a broader range of people to solicit their views and experiences, but we are also inviting further contributions via our website:  www.sociologyandsocialpolicy.soton.ac.uk/proj/informed_consent/index.htm in response to the questions: What does informed consent mean to you? What is good and bad practice in obtaining and managing informed consent? How have you managed and obtained informed consent in your research? What impact have different methods of gaining informed consent had on your research? What impact have stakeholders or ‘gatekeepers’ had on your research? This avenue for feeding ideas and information into the project is still open, and we are happy to receive submissions via this route.

The rationale for undertaking the project is the increased attention that is being paid to the issue of informed consent in the research process. An important part of the background to this increased attention being paid to informed consent by researchers is the broad change that is taking place in research governance. We cannot consider here in this paper all of the issues that this raises, but rather seek to concentrate our attention on two particular difficulties inherent in the processes of gaining and maintaining informed consent. The first of these is that there is no consensus amongst researchers concerning how to know when enough has been done to achieve informed consent, and when the point of doing too much has been reached. This is discussed in section A of the paper, ‘Informed consent’. The second particular difficulty is that there is no consensus about whether the same sets of principles and procedures are equally applicable to research among different groups and to research conducted within different disciplinary frameworks. This is discussed in section B of the paper, ‘Potential vulnerability’. In these two sections of the paper we will draw on our findings to date to highlight the nature of these difficulties and some of our participants’ responses to them. These issues have relevance to wider debates about the role of guidelines for ethical practice in research, and we will conclude with some comments on this.

A. Informed consent:

A.1. Information: how much information to give
The provision of information about a project is an important part of ensuring that prospective participants in research are informed, and a researcher can aspire, as one of our participants put it, to ‘making sure that the amount of information that they receive is sufficient for them to know exactly what that participation entails’ (I5: 1). But it is not always easy to be confident that the information that has been provided is sufficient for this purpose. One problem is that people may quite understandably wish to know more than it is possible for researchers to provide, because the results of research are not entirely predictable: ‘you can never quite know what the outcome is going to be at the outset’ (I15: 4). One cannot guarantee useful outcomes: ‘as researchers…. do we always know what’s going to happen with that piece of research?…. it’s quite difficult to explain to people that, “well I don’t know quite where this is going to end up or whether it will have a purposeful end or not”’ (FG2: 3). When presented by a prospective respondent with the question ‘what good will it do?’ one opinion was that ‘we couldn’t really cover all of that in an information sheet’ (FG1: 3). The experience of being involved in research was likewise seen as difficult to convey: ‘I’m not sure that it is possible to outline exactly what it’s going to be like for somebody before they’ve entered into the experience, especially if they’ve not been a participant in research previously’ (I11: 4)

Even if we leave aside how best to deal with unknowables such as the outcomes of research, there may still be a sense of there being too much information to convey. In a hospice setting, for example, a page and a half of A4 of information may be too much for prospective respondents to take in: ‘I was required to provide a lot of information, which actually in practice, for some people, proved too much for them to read, but was required by the Ethics Committee to be given out because of questions of data protection’ (FG1: 3). Another group for whom more information did not necessarily mean better research was children: ‘with children there’s always that balance between overloading them with information and giving them enough appropriate information’ (I4: 11). In a study of young people and drug use, the sheer amount of information about the several stages of the research made it inappropriate to describe that in detail in the club setting in which the initial interviews took place; rather than give 20-30 minutes of information about the whole project at the start, the research team of which one of our participants had been a part ‘decided…. that we would ask them for consent at each stage’ (I5: 2). This team thus opted to provide information sequentially in more manageable chunks.

Even this limited supply of information proved problematic: ‘one of the problems was that…. people would cut us off and say “yes I’m happy to participate” before we had chance to finish [reading out the introductory information]…. people often love to participate in research, specifically I think when they are being asked about their drug use, drug users can often be very keen to talk about their drug use’ (I5: 3). Eagerness to participate in research has been mentioned in several contexts: ‘we do what we can just to hold back young people’s enthusiasm for taking part because of the whole most young people are very keen to take part and to be listened to…. we as researchers can be sort of overwhelmed by young people’s enthusiasm and just think “yeah they understand, fine let’s get on”…. “that’s informed consent” and I, you know, I don’t think it is…. they just think “oh great, this sounds fun”’ (I6: 9-10). Palliative care researchers reported similar concerns about being overridden when providing information: ‘they were very, very keen and they said “oh we trust you, it’s fine, you don’t need to explain all that” …. People can be quite naïve about how far the information will travel. So my experience is that we have to keep reinforcing the information’ (FG1: 5); ‘people get to trust you and I think they do say “no, it’s fine”’ (FG1: 11).

One clear difference of view relates to how much of themselves researchers reveal. One participant described how the information that she provided was ‘usually one or two sides of A4, it’s the usual things, that they can pull out when they want to, that it won’t affect their treatment that their confidentiality is preserved and we also, because we are a user research organisation, we’ve, we’ve put in that we are users ourselves doing the research in mental health’ (I9: 1). Others were more circumspect about what they revealed about themselves. One focus group participant felt that giving fuller details of herself would have adversely affected her project’s response rate: ‘when I did my PhD it was, it took, supposedly a feminist piece of research but I didn’t tell the other women that my theoretical underpinning to my research was feminist…. I’m sure it would have influenced some of them about whether they spoke with me, if they thought I was a feminist’ (FG2: 8-9). One of our participants expressed the view that this attitude may mask researchers’ self-interest: ‘as a profession, we are slightly duplicitous…. the people who haven’t done research don’t really understand how their information’s going to be used and I think we do use their information…. in ways that they probably wouldn’t like’ (I4: 10). Some researchers reported making a point of maintaining contact with respondents up to the publication of results and beyond, partly with this issue in mind. But the view has also been expressed that in dealings with respondents ‘we never sort of say “well what’s your understanding of, you know, your role in this research then?”’ (FG2: 3).

Limitation of the information given to prospective participants was sometimes justified in terms of the need to avoid pointing out the obvious. One interviewee felt that respondents know that discussing certain topics such as bereavement may ‘be distressing and it looks quite patronising to warn them of that’. A further point was that ‘it’s very difficult to anticipate what might cause distress, sometimes it’s not the most obvious issue…. one can blunder into a field that, you know, doesn’t seem to be ethically challenging or, or delicate or sensitive’ (I15: 3). And more than one researcher has pointed out that providing information does not mean that it will necessarily be absorbed. Thus, one recalled that ‘half the people there kind of thought I was doing some therapeutic group, having explained to them at length that I wasn’t’ (FG2: 8). Another reported that despite having a ‘written information sheet that explained the study really carefully to people…. it was clear they thought I was a social worker or some sort of an appliance officer…. however much time and attention that you explain what research is, people interpret what you do in a number of different ways. And some people see it as a form of therapy, even if you don’t intend that to be the case’ (FG1: 6). Much depends, clearly, not only on what information is conveyed but also on how this is done.

A.2. Information: how to give it
Bearing in mind the above remarks about the potential for information to be misunderstood, several participants in our project have emphasised the importance of information being presented in a user-friendly way. One described how ‘It’s very simple information in a sort of accessible style and format with pictures from the picture bank…. members of our advisory group…. have been helping us out with designing the project information’ (I8: 3-4). Another spoke of providing ‘Normally up to about a side of A4…. a side of A4 which just tells young people what’s going on. Often we do them quite graphically as well so that it doesn’t just look like a huge side of text’ (I6: 2-3). A third aspired to convey information by ‘writing in a manner that would not be off putting but would incorporate a reasonable explanation of what I was trying to do…. and whether they wished to engage or not…. many of them couldn’t read….. it was always difficult to know to what extent their consent had been influenced by the way it was presented to them’ (FG2: 3).

Communication difficulties have been touched on by others. One questioned the value of being directed to follow a uniform format: ‘we have to remember around 13% of the UK population’s functionally illiterate, so these written documents which we are required to produce, even though they’re written trying to be as clear as possible, may not be useful for a small percentage of people’ (FG1: 8). Another researcher, working with children with learning disabilities, described her response to the issue of comprehension relating to people being researched having the right to stop an interview if they felt like it: ‘with one child…. her level of understanding was quite high although she was non-verbal, we had a stop sign and we practised at the beginning and we made it really fun so I was going “ok, go on, tell me to stop, tell me to stop” and she put it up and she loved it and she just thought it was really funny. And at one point in the interview she did do it and she just said she wanted something to drink and she was just tired and, but then she let me know when she wanted to start again’ (I4: 8)

We have also had relayed to us accounts of poor practice that our participants have come across. One researcher whose fieldwork included school settings declared that ‘I know some organisations are reasonably happy with sort of…. forcing young people essentially to participate in research…. so they give out a questionnaire in exam conditions’. Understandably unhappy with this, she has developed ways of moderating the impact of pupils being dragooned in this way: ‘we have had that quite often where a group of pupils have turned up in a room and, and have absolutely no idea, they haven’t been given the information sheets that, that we asked to be given out and haven’t been told what they’re there for…. I think they have a right at that stage just to go “well I didn’t realise that’s what it was” and they can head off, although actually that’s quite tricky in a school because we’ve also often been told to keep the group together, you know, can’t have pupils wandering around the school…. a researcher has to manage that and what we’ve done sometimes is just literally sort of divided a group up in class, in a classroom or the room that we’re in and those who don’t want to take part are over there and those who do are here and we take some wordsearch things for them to do, you know, that they can do so that they’re not wandering around the school…. it doesn’t always work but we try as hard as we can to do it’ (I6: 3-5).

The pressure to get consent can make it tempting to go along with poor practice. As one focus group member put it, ‘there’s masses of stuff out there where people are thinking, you know, “if we don’t get this kind of seventy-five by whenever then my contract is up the chute”…. And then you go and cry in the toilet because you’re not going to get it’ (FG1: 16). In a different focus group another contributor declared ‘I felt quite delighted when the first person said to me “I don’t want to do this, I’m going now”. The next day he came and found me and said “I do want to be interviewed today” so…. he’d understood the information and he was using it as he wanted to which was good’ (FG2: 6). Our participants have emphasised the importance of how information is presented. Information sheets that look official can be problematic, as they were for the researcher who used them only to find that ‘about 90% of people came back and said “oh I’ve spoken to my daughter and she said I mustn’t touch this because this is far too formal”…. they felt it was very threatening, the way it was written’ (FG1: 8). The converse, of researchers trying to be as informal as possible, was mentioned in the context of a study of  bereavement: ‘when people contacted us at the University, we responded by sending a hand-written card back, thanking them, we felt it was important that it looked kind of personal and not, not official…. what we got was letters back …. and those letters are in fact signed consent. They’re not a signed consent form’ (I15: 2, 7).

A.3. Information: when and how often to give it
Good practice is seen by many of the researchers to whom we have spoken as building in a time gap between when the information about the project is conveyed and when consent is gained or confirmed. Thus, ‘it is good practice to try and get as much information to somebody in advance because…. it gives them an opportunity to think it through but also obviously just to talk it through’ (I6: 2). Another reported how ‘we do give time between…. leaving time between a person signing a consent form and you doing the interview, I don’t think that should be done on the same day…. Good practice here would dictate that you leave it at least twenty four hours for people to think about things and have time to, you know, say “no” if they don’t want to go ahead…. it’s like when there’s a salesman on the doorstep’. This researcher, who worked with people with learning disabilities, recounted how ‘on that first visit we would only talk about the project with them and ask them to sign a consent form…. And that consent form is witnessed as well…. it doesn’t have to be a signature it can be a mark on the consent form. But we do, we prefer to get something even if it’s, you know, a scribble on the page’ (I8: 9, 1). For another interviewee, ‘one of the central issues is giving people enough time to absorb the information and make up their minds about whether…. they want to participate in this study’ (I9: 4). A focus group member echoed this, by highlighting  the importance of ‘how long you give people to digest the information…. it just seems important that people have got, you know, long enough to go away, think about it and talk to other people about it before agreeing’ (FG2: 5). A researcher reported how she herself had been made to feel manipulated by being pressurised to take part in an interview, ‘I was kind of processed into saying “yes”’ (FG1: 16), and was conscious of how allowing a period of time to pass before data are collected increases the opportunity for respondents to decline to take part.

The significance of informed consent as a process has been highlighted in the literature (Ramcharan and Cutliffe 2001) and this was echoed by our informants. Rather than undertaking a one-off box-ticking exercise, researchers are conscious of  ‘the need to reiterate quite frequently that they [research participants] are able to sort of opt out… at any point, at any stage. Sometimes I think it’s very easy to forget that once you’re in the process of data collection’ (FG2: 5). Another focus group member remarked that ‘there does come a point at which people forget that’s what you’re doing [research] and then that’s another issue for informed consent…. you can…. go into the “Do you remember I’m a researcher?” [mode, but]…. that changes it…. You’re working, I think, in grey territory at that point’ (FG1: 9). It can also be risky to do this too often: ‘I sometimes felt that they, they kind of forgot that this was, you know, this was research and I, I made a point of saying “is it ok for me to speak to you to-day?”…. and people were getting irritated with “well yes, you asked me that before” ’ (FG2: 5). Asking too frequently can carry with it another risk: ‘You don’t really want to give them the chance to opt out sometimes’ (FG2: 5).

Reminding research participants that their involvement is not compulsory may be prompted by visual cues. Thus, ‘it’s not just what people say, it’s how they are, whether they’re agitated, whether they were kind of not wanting to sit down, wanting to go out, so there’s all those sort of levels of consent and assent’ (FG2: 5). Researchers should be aware that ‘if the children look distracted, you know, bored’ that should be taken as an indication of withdrawing informed consent, although this participant mentioned one researcher who disregarded this on the grounds that ‘it just buggers up your sample size’. Her own view expressed later in the interview was that ‘If a child is crying in front of you they are not consenting to talking to you’. Another thing that she looked out for was if members of a group being interviewed were not showing interest, perhaps by talking among themselves. In such circumstances she reported that ‘I would say “do you want to stop?”…. I can think of one instance where it happened’ (I4: 3, 13, 4). Another participant was confident about her research participants knowing that they were present voluntarily: ‘I think that most of the young people felt…. quite empowered just to say “Oh I think I’ve had enough of that thanks”’ (I6: 9).

A.4. Informed consent: how to gain it:
Whether or not it is feasible and desirable to gain a signature as evidence of informed consent is a vexed question. On one side is the position whereby ‘I don’t think you should have somebody as a research subject who hasn’t got, where there is no document that shows that they are consenting to participate in the research’ (I9: 8). On the other is the standpoint that this is problematic: ‘it’s usually verbal consent that we get rather than written consent…. very rarely do I try and go for written consent…. When you’re researching sort of very excluded groups…. it’s very threatening to ask someone to sign a form. Like the young refugee project…. it would just be incredibly threatening and it would be counter-productive’ (I4: 2). Some people go further than simply gaining a signature: ‘not only did they [research participants] have to sign the form saying it was ok they had literally to contact me to say “look I’m interested, I want to be part of this”’ The rationale for this was that ‘it should be opt in rather than “I’ll opt out”…. Otherwise people can definitely feel pressurised’ (FG2: 4-5).

Concern about people’s motives for participating in research related not only to coercion but also to inducement, and again there was no clear consensus about payments being offered to research participants. One view was that ‘I just think it’s good practice to recompense people for their time’ (I9: 7). Concern about whether this might lead potentially vulnerable groups to participate for the wrong reasons led to the suggestion that rewards might come at the end of the process: ‘what we try to do is where possible include a thank you for young people a financial thank you, like five or ten pound voucher normally, but that’s only after a young person’s gone through the whole process of agreeing…. or just don’t mention it at all and do it at the end so that then it isn’t an incentive’ (I6: 12). Echoing this, another participant’s position was that ‘I do think that payments should never be used as an inducement to take part in research…. [if a small token is given] good practice should dictate that it is given at the end as a complete surprise to them, the person taking part in the research’ (I8: 11). Doubt was cast on whether this will be possible for much longer as word gets around. One participant’s experience was that ‘practice is now so common that young people are, now that’s their expectation, that they will almost ask you about it before you would get to the point of talking to them about it’ (I3: 9).

Of course, the benefits of participating do not have to take the form of money or vouchers. Researchers researching drug taking in clubs provided participants in their study with what was in effect a chill-out area, ‘And they really liked it. We often found that…. we were often asked for cigarettes as well…. it seemed fair [to have cigarettes to give to them] although, looking back on it, providing people with, you know, cancer causing agents isn’t probably the best thing to do’ (I5: 6). Sometimes participation in a project requires altruism. The same participant reported how participation in another project with which she had been involved meant that receipt of a service was delayed, whereas people who declined to take part got that service straight away. As a result the study included only ‘people who wanted to do it out of the goodness of their heart…. because it was against their own self-interest to be in the study’ (I5: 10). Further reassurance about people’s voluntary participation can be gleaned from the use of particular research techniques. One participant felt that ‘focus groups are a good way because that gives you the option of people to get up and walk out or not participate…. If someone doesn’t want to take part in it, they don’t have to, even if they’re sitting in the group’ (I7: 6-7).

B. Potential vulnerability:

B.1. Do some groups of people need special treatment?
People in receipt of services are a good example of a group that can be regarded as particularly vulnerable to experiencing as coercive requests to take part in research. The comment was made in the focus group concentrating on palliative care that ‘the patient…. might feel obliged to do that for you’; people may be ‘afraid “I might not get the best treatment if I…. don’t take part”’ (FG1: 2). The point also came up in another focus group, the possibility that people might ask themselves, ‘well if I don’t give consent is that going to affect any future treatment I have?’ (FG2: 4). In the same focus group an example of how offenders might come to have parallel understandings was given: ‘I was kind of troubled on occasion about their motives for consent…. one left after two minutes, he said “I don’t want to do this any more but I was told it would affect my parole”. Now, you know, I’d been very clear in all the literature I’d sent out that [it would not] but clearly it’s, you know, whoever was in charge of him had, had said that and that was disappointing because for me, also I’d travelled for three quarters of an hour for two minutes of nothing so it was an hour and a half out of your day’ (FG2: 5-6). From what we have been told so far it is abundantly clear that service providers in educational, medical and  penal establishments do not always convey information about informed consent as researchers would like it to be conveyed.

This influence can work both ways, sometimes providing academics with ‘research fodder’, and sometimes denying access to prospective research participants who show indications of preparedness to participate but are denied the opportunity by paternalistic gatekeepers. One focus group member contrasted this with what ideally would happen: ‘the whole philosophy behind informed consent is that people make rational choices and they make them on their own…. I’ve had that experience two or three times where actually a respondent has indicated that they might like to take part and then they’ve been overruled either by a care-worker or by a spouse’ (FG1: 7-8). Another participant reported similar experiences: ‘working through social workers, there’s many reasons why it can be hard to get the children sometimes.… they can be very paternalistic and very protective of the children…. much more so than the child would want them to be’. This participant went on to elaborate on this theme: ‘there are huge issues around children with disabilities…. listening to children with disabilities…. we’re only just beginning to feel our way through…. For instance, children with learning disabilities…. there is an awful lot of assumptions that those children.... can’t have good informed consent because they can’t think it through well enough…. we would dispute that’ (I3: 6, 7). Another participant agreed: ‘a lot of children don’t even get to give informed consent ’cause somebody’s already denied them consent’ (I4: 9). And a participant in a focus group commented, ‘we’re insisting that people with dementia have got a right to decide for themselves whether they take part or not but in a couple of cases where they have agreed to be interviewed then relatives have stepped in and prevented access’ (FG2: 2).

How such problems arise can be readily appreciated in the contexts of particular groups. As one focus group member reported, ‘a lot of the people I’m spending time with can’t write their name so actually, you know, any signature for me is kind of meaningless…. we’ve had to look at more creative ways of getting people to consent…. people being able to say yes or no without having to go via somebody else has been a real issue’ (FG2: 2). Discussion in the same focus group returned to this subject when the question was raised about ‘people with dementia, can you be sure that they really understand what you are doing?’ (FG2: 7), and again later: ‘working with learning disabled people feels different to working with young people, you know, there’s issues of comprehension and attention span perhaps’ (FG2: 11). An even more serious problem was referred to in another focus group: ‘When I was working in intensive care…. clearly I couldn’t get consent from the patient because they were unconscious…. I got a surrogate consent, or proxy consent…. And I used to wear a big badge saying “researcher”’ (FG1: 10). And a participant identified the extreme case, ‘the question of [where] does responsibility for consent lie, to utilising records of people who have died…. Because clearly the people themselves couldn’t give consent because they had died, under data protection laws the records do not belong to the patients’ carers or relatives, they belong to the health authority and there were questions around what harm or benefit might result from looking at the records of people who had died’ (I11: 8). One might agree that procedurally ‘the ideal would be to have informed consent and relative or carer assent as well’ (I7: 5), but be aware of all sorts of situations in which in practice this ideal is impossible to achieve.

Difficulties surrounding other dimensions of potential vulnerability were also mentioned. Some of our informants have returned in this context to the theme of the importance of thinking about informed consent as a process. One example was ‘specific to people with dementia… there’s…. the issue of continued consent, because obviously they might forget later’ (FG2: 2). A participant working in a different field made the related point that ‘it’s difficult for anyone to stop an interview mid-way…. but, you know, it’s particularly difficult for children and young people when there’s the sort of power issue’ (I4: 4). In a focus group the general point developing this theme was made, that ‘there’s that blanket assumption that this is the ethics procedure and we’ll just apply it to everybody and maybe it doesn’t quite work for some people’ (FG2: 3). This general theme was picked up in another focus group: ‘we make an assumption that these are independent, rational people, but actually consent is enacted as an interaction between at least two people who themselves have all sorts of hidden agendas and pressures on them’ (FG1: 16).

The assumption of independence is frequently found to be erroneous in the context of research with groups constructed by others as vulnerable. In the field of dementia research, for example, relatives may be protective of prospective respondents: ‘we have had problems around getting them to sign to consent, I think because people often fear that signing a piece of paper you’re selling them insurance or something and they’ve been told by a relative not to, you know, not to sign anything unless they’re present’ (FG2: 2). Other personal agendas hidden to researchers may work in favour of agreement to be interviewed, as was the case for the informant who had found people wanting to present their account because they were unused to being listened to: ‘most of the carers I speak to, it’s the first time that they’ve actually had the opportunity to tell their story…. as opposed to…. the patient’s story’ (FG1: 13). Taboos around certain subjects can also work to predispose people to talk to researchers. In the same focus group the point was made that ‘there are all sorts of motivations for people taking part, but one of the biggest is that there is such a lack of concern in our society with, you know, death and dying and bereavement’ (FG1: 13). This followed on from the comment of another participant in the same focus group, that ‘bereaved people…. are adults, just being bereaved does not render you less of an adult, these are adult people. So I think sometimes there’s a paternalism practised by Ethics Committees and maybe by us, and we as researchers collude with that, and I think people should be much more empowered to make their own decisions’ (FG1: 12).

There is no consensus about whether it is appropriate to conclude from these observations that some groups of people need special treatment in relation to informed consent. One informant took the view that everyone deserved respectful treatment, commenting that ‘doing research with children is not in theory different than doing it with adults because it’s all about respect whoever it is…. there’s no point in talking to adults in language they don’t understand either’ (I3: 10). This is consistent with Christiansen and Prout’s (2002) argument that research with children does not necessarily raise unique methodological and ethical issues, and Casarett and Karlawish’s (2000) discussion of the supposed uniqueness of palliative care research. For a focus group member the matter was straightforward: ‘it’s people, of course they’re vulnerable’ (FG1: 18). One researcher with young people commented on how identifying a particular age as an indicator of competence to give informed consent appears arbitrary: ‘I can’t see how we can do research into young people’s rights…. and then say, you know, if they’re seventeen they can choose to participate, if they’re sixteen they can’t’ (I6: 13). (This echoes David et al’s (2001) comments on how competence is not necessarily age-related.) Another informant pointed out that research into youth gangs revealed the problematic requirement of parental consent to young people being involved in research, because parents may not know that their children are in gangs and researchers ought not be put into situations where such information is revealed to them (I5: 12). Concern was expressed by several participants that insistence that research into ‘vulnerable’ groups involve the consent of someone other than the person being studied might lead to the exclusion of those groups, and in this context one focus group member remarked that ‘some people are more or less capable than other people but still it’s, it’s at least a minefield…. how does one determine the capacity of somebody who consents?’ (FG2: 13, 15). In another focus group a participant who had grappled with the same problem of being required to gain consent from relatives reported that she had ‘had to debate the Human Rights Act with my Ethics Committee’ (FG1: 9).

B. 2. Do some issues need special treatment?

Running alongside debates about whether some categories of people needs special treatment in relation to consenting to involvement in research are debates about whether some issues need to be approached differently. The question of whether informed consent requires a signature has been mentioned already in the context of whether people have the capacity to sign, but the issue was also raised in relation to the type of data that researchers seek to collect. One participant in a focus group commented that ‘in certain methodologies…. requiring written informed consent seriously damages the method that you’re going to pursue…. Any recording or observation that requires spontaneously occurring behaviours or speech, and I’m thinking particularly things like conversation analysis and ethnography…. both of those are highly problematic. I think that you can get informed consent and that it can be explicit but I think signing a form is actually changing the whole nature of the social interaction’ (FG1: 8-9). Coomber (2002) has made a similar case in relation to research involving criminal populations, that seeking respondents’ signatures simply is not feasible with certain research topics.

Various participants referred to the need to tread particularly carefully when undertaking research in the field of health and illness. Information provided needs to give reassurance for those whose first question on being invited to participate might be ‘you’re coming to talk to me about end of life care and serious illness, well are you choosing me because I’m about to die?’ (FG1: 3). Another participant echoed these sentiments, observing that ‘in a lot of health-related research…. it’s enormously important for people to know why they’ve been selected otherwise they might assume that they have a particular, you know, sort of particularly virulent, you know, manifestation of an illness’ (I15: 2). And in relation to mental health, the problems of gaining informed consent from respondents when they are ill led one participant to suggest that ‘people should be asked if they would like to participate in research when they’re well so it would be kind of like an advanced directive’ (I9: 5). Their long association with procedures requiring the approval of medical research ethics committees means that social scientists working in the area of health and illness are particularly familiar with research governance as it relates to formalised procedures of gaining informed consent. It is clear from our other participants, however, that these procedures have moved well beyond this particular field.

Conclusion
These various issues have relevance to wider debates about the role of guidelines for and regulation of ethical practice in research, and we will conclude with some comments on this. The first is that a number of our informants have adopted a pragmatic stance. Thus one commented that ‘We start out with very good procedures but sometimes you just have to be pragmatic [laughs] and just get on with it basically’ (I3: 9). Pragmatism also underlay opposition to standardised procedures was voiced in terms of the undesirable effect of making some issues unresearchable. One focus group member, echoing the concerns discussed in Wendler (2000), asked ‘what about people who aren’t able to give their consent in a…. standard way. Do we always exclude them then?’ (FG2: 13). Another informant framed this point about the unduly constraining nature of regulation more in terms of types of data: ‘my involvement with ethics committees is that they haven’t completely understood qualitative research’ (I8: 6). This was echoed by the informant who felt that ‘if there’s going to be ethics committees they have to be very broad thinking about methodology and a bit more open-minded to new ideas’ (I4: 18), but a more downbeat conclusion was that ‘in these days of clinical governance…. there are maybe certain types of research it may not longer be possible to do’ (FG1: 11). Covert research fell into this category for one focus group member: ‘maybe some research just can’t be done…. If you can’t get informed consent then there’s maybe a good case for saying it can’t be done…. I can’t think of a piece of covert research I’d be happy to [do]’ (FG2: 13). This is not to suggest that concerns over the ethics of research styles are anything new. It is nearly two decades since Punch wrote that ‘there is simply no consensus on the key ethical questions raised by our research. There is no hard and fast way to calculate the costs and benefits of social scientific research’ (1986: 81).

Whether or not the research agenda has been narrowed by the growth of research governance, there is widespread recognition of how different things now are. One informant said that ‘I’ve…. noticed…. over…. twenty years…. it being more and more difficult to actually carry out research because of certain kinds of ethical guidelines…. things have changed’ (I5: 8). Another looked back at remarked that ‘thirty years ago you didn’t need ethics committee approval for anything’ (F7: 10). The changed environment has produced some interesting comments. One informant’s view was that ‘one has to be pragmatic about getting things through committee’ (I15: 9). Another remarked that things are still a long way from a uniform set of practices having been adopted: ‘we had a conference a couple of years ago on ethics and law in relation to young people and that was really good for just getting, for enabling people to say, you know, “but I thought everybody did X this way”’ This same informant had earlier pointed out that guidelines can be adopted or substituted for:  ‘if you disagree with a part of a professional guideline then it’s very difficult to adopt them for your organisation, so…. that’s why we created our own…. Some of the other professional guidelines, you know, you could interpret them so broadly that actually they’re not much use’ (I6: 20, 14-15).

This remark suggests that the area is likely to see significant further developments as debates about the balance between consistency and flexibility in guidelines are worked through. An additional comment was made to highlight the view that however far research governance extends, regulation cannot fully determine what happens in the research process, not least because ‘I don’t think, however well-prepared any researcher is in terms of…. having read every code of ethical practice and filled in however many, you know, research ethics proposal forms there are, anything can prepare you for the emotions of the experience’. This echoed a point made earlier in the interview that ‘it’s only once you’re managing the field that you find out that the, the issues are more complex’  (I15: 13, 1) than what research training prepares people for. The dilemma of what to put into research guidelines was summed up by the focus group member who said, ‘often the problem with guidelines is in their interpretation, if they’re too flexible then they become valueless, if they’re too rigid they can interfere badly’ (FG2: 11). The remark of his focus group co-member that he did not ‘have any sort of hard and fast rules’ (FG2: 10) about the particular issue of how much information to give illustrated the viewpoint nicely. Flexibility of this sort needs to be distinguished from an ‘anything goes’ philosophy. Our informants have had little difficulty in identifying examples of bad practice in relation to informed consent. This reflects among other things an awareness that the need for guidelines and rules arises from the likely consequence of poor practice, that it will make research increasingly difficult for all researchers, not only the bad practitioners. As one informant said, ‘where people feel they’ve been badly treated or, or just that, you know, we’ve used slightly underhand methods I don’t think it helps anybody in the long run and I certainly have people who won’t engage in research because they’ve had very bad experiences’ (I6: 19). Discussing the difficulties surrounding informed consent in the research process and sharing information about how best to handle them is thus motivated by pragmatic as well as ethical considerations.

References
Casarett, D. and Karlawish, J. (2000) ‘Are special ethical guidelines needed for palliative care research?’, Journal of Pain and Symptom Management 20(2), 130-39.
Christiansen, P. and Prout, A. (2002) ‘Working with ethical symmetry in social research with children’, Childhood 9(4), 477-97.
Coomber, R. (2002) ‘Signing your life away? Why Research Ethics Committees (REC) shouldn’t always require written confirmation that participants in research have been informed of the aims of a study and their rights – the case of criminal populations’ Sociological Research Online 7(1), http://www.socresonline.org,uk/7/1/coomber.html
David, M., Edwards, R. and Aldred, P. (2001) ‘Children and school-based research: “informed consent” or “educated consent”?’, British Educational Research Journal 27(3), 347-65.
Punch, M. (1986) The Politics and Ethics of Fieldwork. Beverly Hills: Sage.
Ramcharan, P. and Cutliffe, J. (2001) ‘Judging the ethics of qualitative research: considering the “ethics ass process” model’, Health and Social Care in the Community 9(6), 358-66.
Wendler, D. (2000) ‘Informed consent, exploitation and whether it is possible to conduct human subjects research without either one’, Bioethics 14(4), 310-39.