For peer review only

BMJ Open

Public health policy, context and ethics: the example of unlinked anonymous testing of blood

Journal:

BMJ Open

rp Fo Manuscript ID: Article Type:

Date Submitted by the Author:

Complete List of Authors:

Kessel, Anthony; London School of Hygiene and Tropical Medicine, Social and environmental health research Datta, Jessica; London School of Hygiene and Tropical Medicine, Wellings, Kaye; London School of Hygiene and Tropical Medicine, Perman, Sarah; Health Protection Agency, Public Health Strategy Public health

Ethics, Infectious diseases

ev

Keywords:

28-May-2012

rr

Secondary Subject Heading:

Research

ee

Primary Subject Heading:

bmjopen-2012-001427

PUBLIC HEALTH, MEDICAL ETHICS, Health policy < HEALTH SERVICES ADMINISTRATION & MANAGEMENT

w

ie ly

on For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 1 of 26

Original paper

Public health policy, context and ethics: a qualitative study of unlinked anonymous testing of blood Unlinked anonymous testing of blood (running title) Anthony S Kessel (corresponding author and guarantor of study), Honorary Professor, London

rp Fo

School of Hygiene & Tropical Medicine, Faculty of Public Health and Policy, 15-17 Tavistock Place, London WC1H 9SH (Tel: 020 7759 2715; email: [email protected])

Jessica Datta, Research Fellow, London School of Hygiene & Tropical Medicine

ee

Kaye Wellings, Professor, London School of Hygiene & Tropical Medicine

rr

Sarah Perman, Health Protection Agency

Key words

w

ie

ev Unlinked, anonymous, blood, ethics, public health, policy, HIV

Word count

ly

6437 words

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

1

BMJ Open

Abstract In this study we explore the ethical issues around unlinked anonymous testing (UAT) of blood, a method of sero-prevalence surveillance for infectious diseases. Our study focused on UAT for HIV, although UAT can be used for other infectious diseases. Objectives The objectives of the research were to gain a better understanding of the views of key informants in

rp Fo

countries adopting different UAT testing strategies, and to use the findings of the research to inform health policy Design

Qualitative study using in-depth interviews and ethical analysis.

ee

Setting

rr

Four countries using different strategies around unlinked anonymous testing of blood for HIV (United Kingdom, United States, Netherlands and Norway). Participants

Twenty-three key informants in the four countries.

w

Results

ie

ev

Participants from the four countries have different views on UAT of blood, and that the approaches and policies on UAT adopted by different countries have been historically and culturally determined.

on

We use our findings to explore the relationship between public health policy and ethics, framing our discussion in relation to two important contemporary debates: informed consent for participation in

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 26

medical and public health research; and the balance between individual rights and the public good. Conclusions Qualitative research and ethical analysis of unlinked anonymous testing of blood in different countries has yielded important findings for consideration by policy makers. The policy of unlinked anonymous testing of blood for HIV and other diseases in the UK needs re-consideration in light of these findings.


2

Page 3 of 26

Public health policy, context and ethics: the example of unlinked anonymous testing of blood

In this paper we explore unlinked anonymous testing (UAT) of blood for HIV and other diseases, and examine the ethical issues associated with this method of sero-prevalence surveillance for infectious diseases. We report original results from a study funded by the UK Economic and Social Research Council (ESRC) – qualitative interviews with stakeholders in

rp Fo

four countries adopting different UAT policies – and use the subject matter, and our findings, to explore the relationship between public health policy, context and ethics.

BACKGROUND

ee

Unlike voluntary testing for clinical purposes, unlinked anonymous testing (UAT) of blood is

rr

carried out in order to provide information about the prevalence of particular infectious diseases in a population. The use of UAT began in the late 1980s for HIV, and remains most

ev

commonly associated with this infection, but can be applied to other infectious diseases such as hepatitis C.1 In this paper we use the term UAT synonymously with sero-prevalence surveys, or sero-prevalence surveillance.

w

ie

The principles behind UAT can be explained with reference to the approach used in the UK for HIV. Leftover blood from samples taken for diagnostic or routine screening purposes is

on

made anonymous, irreversibly unlinked from the source, and then tested for HIV.2 3 This means that individuals whose blood tests positive for HIV cannot be contacted or informed.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

The data derived from UAT are then used for monitoring purposes and also to inform the planning and evaluation of health services and health promotion activities.

Historically, UAT has developed in different ways in different settings, yet exploration of the moral dimensions of such developments has been relatively limited.4 5 6 Further, recent attention given to individual rights and autonomy – in the health and healthcare context – make it an appropriate time to reassess moral considerations around UAT. A brief summary


3

BMJ Open

of developments in UAT in the four countries used in our study is provided as a necessary introduction to the research project.

History of UAT in different countries A national sero-prevalence surveillance programme using UAT to estimate the prevalence of HIV has been in place in the UK since 1990.1 The programme was first developed in

rp Fo

response to concerns about the spread of HIV during the 1980s7 and, despite early opposition, the methodology eventually gained political support.8 The programme was funded initially as a pilot project by the Medical Research Council, and has since been funded by the Department of Health. [1] [2] The programme currently uses data collected from three ongoing surveys (newborn babies, [maternal antibodies], users of genitourinary medicine

ee

(GUM) clinics, and intravenous drug users [saliva]) to measure the proportion of individuals within the general population, and within subgroups of the population, infected with HIV.

rr

Since the programme’s inception, around ten million blood samples have been tested in this way.9

ev

A system of ‘prior notification’ operates in locations where testing goes on, involving the

ie

display of leaflets or posters explaining that blood may be tested anonymously. Consent to participation is implicit rather than explicit; provision is made to opt out of testing, although this naturally depends on awareness.9 10

w

on

In the USA, UAT was extensively used in the late 1980s and early 1990s using discarded blood samples. The largest survey was national and used newborn heel prick tests. Other

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 26

UAT surveys focused on population groups, for instance a survey of young people volunteering for job corps, although these were regional or local. In contrast to the UK the prior notification system, with the opportunity to opt out, was not used. UAT essentially stopped in the USA in 1996, as evidence emerged in relation to reducing vertical transmission of the HIV virus from a pregnant mother to the foetus through use of anti-viral medicine.11 The interplay at the time of clinical, social and political factors was complex.12 The main method for HIV surveillance in the Netherlands is named voluntary testing rather than UAT (these are ‘opt-out’ programmes, such as the antenatal testing programme and the


4

Page 5 of 26

testing of users of STI clinics, as well as the extensive ‘opt-in’ testing carried out by general practitioners). The UAT that does occur in the Netherlands is carried out using blood samples from intravenous drug users (IDUs) and saliva samples from other groups. Unlike the UK or USA, informed consent is required from those surveyed by UAT, either verbal or, more recently, written consent. From 2002 to 2006, other high risk groups haven been included in UAT research surveys in the Netherlands (e.g. migrants from the Cape Verde islands, Surinam and the Netherlands Antilles, as well as commercial sex workers), because of concerns of a higher proportion of undiagnosed HIV in such groups, who were also felt to be

rp Fo

accessing health services less than others. In Norway, public health efforts have focused on diagnosing all those infected, as it was thought (certainly in earlier stages) that those who were diagnosed would be less likely than those undiagnosed to transmit infection to others. Mandatory named AIDS reporting was

ee

introduced in Norway in 1983 and mandatory anonymous HIV case reporting in 1986, both including cases diagnosed before those dates.13 The Norwegian approach has therefore

rr

centred on individual testing and clinician reporting, and UAT has not been deemed epidemiologically necessary and has never been undertaken.

ie

ev AIMS AND OBJECTIVES OF THE RESEARCH STUDY

w

The ESRC-funded project had three components: a questionnaire survey of 471 patients

on

attending two inner-urban genito-urinary medicine (GUM) clinics in two English cities, exploring their views of, and attitudes towards, UAT;14 qualitative in-depth interviews with a sub-sample (20) of these patients; and in-depth interviews with key informants in four

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

countries chosen to represent different approaches towards sero-surveillance for HIV. In this paper, we present findings from this third component, the qualitative interviews with key informants.

The over-arching aim of this component of the research study was to use the subject of UAT, and our findings, to explore the relationship between public health policy, context and ethics. There were two more specific objectives of the research. The first was to gain a better understanding of how UAT has developed in different settings over time, with particular


5

BMJ Open

attention to the ethical considerations. The second was to use UAT as an instrument to gain empirical understanding of the ethical tensions between surveillance and patients’ rights, and to explore how these tensions may have developed and been handled in different contexts.

METHODS Four countries were selected purposively to participate in this study component (the key

rp Fo

informant interviews), based on their representing a range of approaches, or positions taken, with regard to UAT: the UK (specifically England), USA, the Netherlands and Norway. Purposive sampling was used to select the key informants from these countries, informed by the following criteria: range of stakeholder perspectives (e.g. epidemiology, clinical, public health policy, charity or pressure group, and academia (public health or medical ethics));

ee

involvement in UAT, historically and/or current; existing links with the research team or advisory group members; and practical issues (e.g. availability and likelihood of participation).

ev

rr

A topic guide was developed that allowed for a semi-structured interview. Elements within the topic guide included: introduction and background to the purpose and nature of the

ie

interview (prior information had been provided); knowledge and views on unlinked anonymous testing of blood (and national programme as appropriate), including advantages

w

and disadvantages; ethical considerations including consent, prior notification, confidentiality, anonymity, individual rights/liberties, social responsibilities, and any changes

on

in ethical position over time; the value to public health of different surveillance methods; blood as a tissue; legal considerations; and any other relevant issues the respondent wished to

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 26

discuss. The topic guide was initially tailored for the different countries based on our own understanding, but then applied flexibly during the interview.

A researcher, with a background and skills in qualitative interviewing, was employed for the duration of the project, and undertook all the interviews. Interviews were carried out either face-to-face or by telephone between January and September 2008. Interviews were recorded (except for two: the first due to unsuitable location and the second to mechanical failure) with the participant’s permission, and transcribed verbatim. Interviews lasted between 20 minutes and one hour. In total 21 interviews were completed with 23 respondents (two interviews


6

Page 7 of 26

were with two respondents together). Of these, 12 were with UK respondents (three clinicians in genito-urinary medicine/HIV; one person working in government health policy; two HIV epidemiologists; two ethics academics; two charity representatives [MedFash, Terence Higgins Trust]; one person working at the British Medical Association; one editor of a medical journal), four with respondents in the USA (one health policy academic; one academic in health law; one academic in public health; and one epidemiologist), three with respondents from the Netherlands (two epidemiologists; and one academic in medical ethics), and four with respondents in Norway (one epidemiologist; three academics in medical ethics

rp Fo

– one of whom is also a hospital clinician). There were three interview refusals, all in the UK. [3]

Interview data were coded by the researcher, and emergent themes assessed by the same researcher (in the context of the aims and objectives of the research); a second researcher

ee

assisted with coding and analysis. Inevitably some of the themes inter-link with each other (e.g. ‘value of UAT’ and ‘ethical issues’). Content was analysed for description of UAT

rr

developments in the country concerned, the context in which UAT may have changed, ethical tensions in UAT between surveillance and human rights, and broader connections between UAT and public health policy.

ly

on

UK

w

Value of UAT

ie

RESULTS

ev

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

The programme is largely seen by respondents as indispensable in providing an unbiased measure of HIV prevalence and risk factors in the UK population. The most important outcome from the surveys has been information on undiagnosed HIV infection which, respondents suggested, has led to changes in policy on named testing; for example, HIV testing in pregnant women has changed from opt in to opt out, and sexual health clinics have moved from targeted to universal voluntary testing. The neonatal dried blood spot survey is used as a check on the performance of local antenatal screening. The public health rationale for the programme has not changed since 1990, and respondents cited the continuing high infection rate, particularly among men who have sex with men, as justification:


7

BMJ Open

"There was far more feeling of a public health emergency (then) than there is today, although you could argue … that it's actually much worse today because of where we see the disease." (UK4) Several respondents suggested that a review of UAT would be timely, including consideration of extending the surveys to other health care settings and other diseases. UK6 discussed how advances in HIV testing, including self-testing, might alter the questions asked

rp Fo

by the programme. Some suggested that if named testing was normalised, there would be less need for anonymous testing.

A particular theme was whether the British public has a duty to participate in public health monitoring, possibly as part-and-parcel of receiving the benefits of a national health service

ee

(NHS):

rr

"There is a very good case for really exploring with the public, and then pushing very hard to get public agreement, that in general terms that sort of tissue, extra bits of

ev

tissue or blood, should be available for use." (UK4)

ie

USA

According to one respondent, the surveillance programme was valuable in terms of planning,

w

resource allocation, advocacy and interventions to prevent transmission. The heel prick survey provided accurate national data on prevalence in childbearing women and was seen as

on

extremely valuable, with nothing since that has given as accurate a picture of national prevalence (USA 2).

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 26

However, given the questionable ethics of UAT (see later), as well as the resultant lack of good quality prevalence data, USA3 argued that enough became known about prevalence to target the most vulnerable groups for health promotion. “And how much more detail do you want from that, so what would you do differently if you knew that in the next level of detail.” (USA3) Netherlands


8

Page 9 of 26

Unlinked anonymous surveys have been used in the Netherlands to measure HIV prevalence among high risk populations, but with informed consent (verbal of written). The populations surveyed by UAT have changed over time. Injecting Drug Users (IDUs) have been surveyed bi-annually since 1994 in different cities in the Netherlands, and are recruited via methadone care; sex workers and migrant groups have also been subject to UAT surveys.

There are two reasons, indicated by respondents, why UAT in Netherlands has been of limited value. First, the uptake of voluntary testing is generally high in the Netherlands,

rp Fo

meaning the quality of that data are generally good. And second, UAT has only been deemed ethically permissible in the Netherlands with informed consent, making large-scale UAT surveys impracticable, so restricting UAT to specific population groups and/or the research context.

ee

Norway

The unknown, undiagnosed segment of the population is known as the ‘dark figure(s)’ in

rr

Norwegian, although neither the ‘dark figure’ nor the number of cases of HIV has been perceived as significant issues in Norway. Uptake of voluntary testing for HIV was described

ev

as very high in intravenous drug users and antenatal clinic users, meaning that UAT has never been perceived as to be of value epidemiologically. The system of surveillance for HIV in

ie

Norway instead covers all diagnosed cases (i.e. not undiagnosed cases as with UAT), and was reported as being simple, inexpensive and capable of providing a representative and timely

w

overview that guides prevention, while respecting confidentiality and adhering to ethical principles.

ly

Quality and nature of data

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

UK The epidemiological data yielded from unlinked anonymous testing are considered by respondents to be of unrivalled quality internationally. It was suggested, however, that the apparent high number of undiagnosed HIV infections from UAT surveys might be inflated by double-counting:


9

BMJ Open

"You can quite understand that if you've got gonorrhoea and you've been told by your doctor not to have unsafe sex, you might well decide to go to a different clinic to have your gonorrhoea treated." (UK1) Others defended the statistics and argued that they were explained by a resurgence of high risk behaviour, particularly among gay men. UK1 argued that people who refuse a named test and who are subsequently found to be HIV positive through UAT are likely to be people who know that they are positive.

USA

rp Fo

There was little comment from USA interviewees about the quality of UAT data, since UAT surveys stopped in 1996. However, a theme that emerged in relation to UAT data was around the difference between research and surveillance.

ee

The fact that there is no ‘right’ to health care in the US was felt to have an effect on people’s

rr

views on surveillance – in other words, the ‘disconnect’ between health care provision (not available to all) and surveillance (done to all) (USA 1). The differences between surveillance,

ev

research and practice, with perhaps different ethical standards required for each, emerged as a theme from the interviews (USA 4).

ie

One point of view was that surveillance (because it is named and part of routine care) is part

w

of clinical care rather than purely research (USA 2). However, while some of the surveillance undertakings at state level in the US were considered to be practice, at the level

on

of CDC they were treated as research (USA 1). The same respondent commented that cancer sufferers advocated in favour of surveillance for the purposes of research on the disease,

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 26

since “they were willing to trade privacy and consent for a good which was surveillance that they saw as serving their interests”. Norway and the Netherlands There were no comments from Norway respondents about the quality of UAT data since such surveys are not carried out in Norway, in part because of the good quality of data on HIV and AIDS from other sources.


10

Page 11 of 26

Two respondents from the Netherlands judged that large-scale UAT programmes were not necessary because named HIV testing is widespread. The quality of information obtained from named testing was felt to be robust and could be used to model estimates of HIV prevalence in the population. Coverage of pregnant women is very high and this group is considered a valid proxy for the heterosexual population. Named HIV testing among users of STI clinics was also presented as being a valid method for estimating prevalence. It was argued that, as long as the uptake of named testing by pregnant women and STI clinic users remains high (above 90%) then a national programme of UAT was not thought to be

rp Fo

necessary. However, notwithstanding the above point it was acknowledged that current methods of surveillance do not allow undiagnosed HIV infection to be measured accurately.

Ethical issues UK

rr

ee

Respondents felt that, on balance, no harm was caused by the programme. A recurrent argument for this was that, in settings where anonymous testing takes place, people are offered a named HIV test:

ie

ev

"I think the ethics are not in fact, are not a problem because everybody is now very strongly advised to have a test and therefore the residual rump who don't have a test,

w

it's important to know what their rate of HIV positivity is." (UK1)

on

Some respondents, however, said there would be ethical concerns if people in these settings were not routinely offered a named test. Several respondents considered the ethics of UAT to

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

be relatively unimportant, for instance in the context of more significant health-related problems in Africa:

"There are these enormous ethical, real ethical issues, and to me, this one … I just can't understand why one should put so much emphasis on it." (UK9) UK11 suggested that the question of ownership of the sample was relevant, but that this could be argued both ways:


11

BMJ Open

"Is it yours or is it somebody else's? ... If it's waste that's going to be chucked away then why shouldn't people use it if they find it useful? On the other hand, 'I gave permission for this test for my own benefit and I didn't give you permission to do things to help others.'" (UK11) There was universal agreement that changing to a system of informed consent would damage the programme. There was a fear that introducing information about UAT into a clinical consultation could confuse patients. Patients might think they were getting a named test:

rp Fo

"You would lose someone's confidence and trust because they might not understand the words or the reason behind why we're doing it, and they would think that you could get back to them and you did know." (UK12)

ee

Other respondents argued that it would reduce time for discussing important clinical matters:

rr

"When the midwife turns up, does she pull out a leaflet and say by the way, I'm putting this beside you here? … it would be a perversion of the consultation to divert it into discussing this ..." (UK5)

ie

ev

Further concerns were that a system of informed consent would increase refusals and lead to bias in the surveys, and also the time and cost of gaining consent from patients:

w

"You know we've got ten minute appointments. It could take you five minutes, half of

on

your time just to explain what this is all about. So doing it in a consultation is not realistic." (UK3) The respondent suggested that a system of clinic self-registration,

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

using computers, could be used to explain anonymous testing.

All the respondents thought that prior notification was the right theoretical approach. However, some were concerned that, in practice, it might not be working: "It's a reasonable thing in practical terms provided people are informed. So … if people don't know about their option for opting out and their option for finding out then the system hasn't been working." (UK11)


12

Page 13 of 26

Some suggested that the approach needed to be more "active" with information displayed visibly on clinic walls or in leaflets given out to each patient. One clinician, however, expressed his discomfort with his role in the programme. He described filling out the form for UAT blood testing in front of the patient, without explaining what he was doing, and sometimes after the patient had declined a named HIV test:

"You're doing something that sort of seems slightly underhand…you're not explaining it to the patient. Yeah, I'm not comfortable with that, I think one wants to be, to

rp Fo

communicate well and to be as open as possible with patients and that's slightly in conflict with that." (UK3) USA

Consent was not deemed necessary initially in the USA for participation in UAT as patients

ee

had consented to the blood being collected in the first place, and this was seen as adequate. USA1 suggested that surveillance is a public good, meaning there are strong reasons for

rr

making it a universal obligation on citizens – and that notification that surveillance is occurring is different from asking for consent.

ev

The blood spot survey continued until the publication of the first trials showing that vertical

ie

transmission of HIV could be dramatically reduced by treating mothers. When it became clear that infected women could be offered treatment, and therefore should know their HIV

w

status, “it became impossible for the US Centre for Communicable Disease Control and Prevention (CDC) to continue that programme and it was dropped in the early 1990s along

on

with all the other ones” (USA3). It had become difficult to reconcile public health benefits of UAT with individual rights. If CDC had not voluntarily withdrawn funding from UAT, the

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

same respondent indicated that it was likely that Congress would have mandated cessation, as there was a strong coalition in favour of this.

A related legal issue was brought by USA4, in that public health surveillance appears to go against the “fundamental legal principles” in the USA, and constitutional scholars are shocked when they hear what is done.


13

BMJ Open

‘Once you begin to argue that the government can take your personally identifiable information for any reason it finds useful without your consent you've opened up and blown a hole in much of the fundamental legal principles in the country.’ (USA4) The Tuskegee episode (a study between 1932 and 1972 in Alabama in which black syphilitic Americans were not given treatment although thought they were) was mentioned by USA1 as a reason for US citizens, particularly Afro-Americans, to be suspicious of surveillance practices. In the context of opposition to named case reporting and a consequent lack of trust

rp Fo

in surveillance, USA3 argued that there are probably cultural dimensions to issues such as consent, and the acceptability of particular interventions may vary from country to country. “It [UAT] became more of a civil rights issue than a gay specific one. It was, the government knows something that you don’t and the attitude in the States, at least at

ee

the time, towards the government was much less trusting than in this country [i.e. UK]. People did not assume and still don’t assume that the government necessarily is

rr

looking out for your well being. And I think there’s a much more paranoid attitude towards the government than there is in this country.” (USA3)

ev

Related to these cultural dimensions, a particular ethical concern was that the US (and, in

ie

particular, CDC) supports UAT in developing countries, with the inherent contradiction that it has been rejected in the US.

w

‘It seems to me the CDC can’t at the one time say it’s unethical to do it in the US and

on

say it’s ethical to do it abroad. Different public agencies may come to different decisions, but the same agency it seems to me should be bound by some rule of consistency.’ (USA1)

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 26

USA2 felt very strongly that arrangements for UAT in developing countries are not ethical where pregnant women do not have the opportunity to have a named test, counselling and treatment. Netherlands The respondents felt that more extensive UAT surveys would produce more accurate information for public health surveillance. However, inter-linked practical and ethical doubts


14

Page 15 of 26

about UAT were presented. One concern was that UAT might deter high risk groups from seeking testing and treatment. Another concern was that, if UAT focused only on high risk groups, such groups would risk being stigmatised. It was suggested that reporting of findings from UAT surveys of high risk groups would need to be carried out sensitively. A particular issue was consent and maintaining trust with patients. One interviewee expressed that staff in HIV treatment centres and STI clinics demand a more stringent consent process than required by the research ethics committees in the Netherlands in order to promote trust

rp Fo

among patients. Norway

A UAT system was never thought relevant, necessary or appropriate for Norway. In Norway it is not legal to test or analyse blood or biological material for infectious diseases for non-

ee

diagnostic purposes without consent unless the purpose is surveillance of an ‘epidemic’. One respondent felt this displayed a ‘moral grey zones’, as the law hinges on the definition of ‘epidemic’ [Norway 1].

ev

rr

There is very little concern or awareness in Norway about ethical issues around UAT. However, there is a general willingness to provide information for public health purposes in

ie

Norway, and people are used to being surveyed. It was reported that many Norwegians have given blood to the Biobank, and give broad consent for their tissue/blood to be used for a

w

wide range of research. A respondent indicated that many are willing to waive their consent in Norway because ‘there is significant trust’ in the health care system. However, another

on

respondent expressed that there was a possible danger that UAT blood could be used to identify someone via the Home Office DNA database, and that researchers might be forced to

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

share information under a court order (the respondent mentioned that the murderer of the Swedish politician Anna Lindh in 2003 was convicted on the basis of a match with DNA held in the national biobank).

One respondent suggested that informed consent could be different in different contexts. In a country like Norway, with low prevalence and easy access to health care, the current surveillance system (non-UAT) was described as acceptable and functioning well. However, in a country in Africa (where this respondent was carrying out research), with a high rate of


15

BMJ Open

HIV positive pregnant women and a low rate of testing, routine testing with opt-out consent procedures would be morally acceptable.

DISCUSSION Since its inception, unlinked anonymous testing (UAT) has been a subject of epidemiological, ethical and policy debate.15 16 17 Different approaches to sero-surveillance

rp Fo

were initially adopted in different countries in response to the developing HIV/AIDS epidemic.4 Over time some of those approaches have changed. The research findings presented in this paper are part of a larger project exploring the reasons behind such differences. The first findings, a quantitative study of the views towards UAT of users of users of two genito-urinary clinics in England, have recently been published.14

ee

In this paper we have presented the results of qualitative interviews with key informants in

rr

four countries pragmatically selected because of differences in approach to testing. It is important to note the limitations of the research.18 Only four settings (countries) were

ev

selected, but these were chosen deliberately as they represent a range of approaches to UAT. Although more data would have been generated by adding further countries, this was not practical for this relatively small study. A total of 23 respondents were interviewed, a

ie

reasonable number for a study of this kind, but they were differentially distributed between

w

countries: 12 in the UK, four in the USA, four in Norway and three in the Netherlands. This distribution (as well as the make-up of respondents) was chosen purposefully to gain useful

on

insights in a qualitative piece of research of this nature, and was not intended to provide a quantitatively representative sample.19 Future study could be undertaken with expanded numbers.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 26

First a summary of key findings will be presented, followed by a discussion of how the findings from this paper fit into broader contemporary debates about public health ethics and policy.

Summary of key findings from interviews


16

Page 17 of 26

In the United Kingdom (UK), a balance of epidemiological and ethical considerations in the late 1980s led to the implementation of the ‘prior notification’ approach to UAT; this involves leaflets and posters informing clients that UAT is occurring, allowing opting out but not necessitating the gaining of informed consent.7 10 Respondents stressed that the rationale remains apt today, and that this approach provides data unrivalled in terms of epidemiological quality,9 20 21 and which is of use for health service planning.22 Ethical concerns were generally felt to be minimal today since clients are (also) usually offered voluntary (named) testing in settings where UAT is occurring. However, there was mention of concern about the

rp Fo

lack of full explanation, whether people do actually know UAT is happening, and ‘ownership’ of blood.23 It was indicated that an approach involving informed consent would be cumbersome, time-consuming and epidemiologically damaging.24 In the United States (USA), UAT of blood for HIV was extensively used in the 1980s and

ee

early 1990s,25 but was stopped in the late 1990s as evidence emerged that anti-viral therapy in pregnant women could help reduce the vertical transmission of HIV from pregnant mother to

rr

child – named testing needed to be maximised so that HIV-positive pregnant women could be identified and offered advice and therapy.26 The interviews indicated that data from UAT

ev

were useful epidemiologically but, since UAT has stopped, there has been sufficient other data to plan services accordingly. Other issues raised include a general suspicion around

ie

epidemiological surveillance post-Tuskegee, cultural dimensions of UAT with respect to minority groups, a lack of trust in the Government, and the related disconnect between

w

surveillance (done to all) and health care provision (variable, and not done to all).6 11 There was also comment about the apparent contradiction in the contemporary support of the US

on

Centre for Communicable Disease Control and Prevention (CDC) for UAT in low-income countries.27

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Interviews in the Netherlands showed that named (voluntary) testing, which has generally been widely taken up, has provided the main vehicle for surveillance of HIV.4 There is acknowledgement that large-scale UAT would provide additional epidemiological data, but this has been balanced against concerns of this leading to diminished voluntary testing – especially in at risk groups – through loss of trust in the healthcare system.28 Smaller UAT surveys have been intermittently undertaken in such groups (e.g. intravenous drug users, sex workers, migrants) as research studies, but have required individual consent (initially oral, now written).


17

BMJ Open

Interviews in Norway revealed that UAT was never felt to be necessary epidemiologically, as HIV has not been a significant issue and the uptake of voluntary testing has been high.13 Surveillance has thus tended to focus on the monitoring of the diagnosed population (anonymised diagnosed cases), which has been supported by the gay community; the undiagnosed population is interestingly referred to as ‘dark figure(s)’. In Norway it is not legal to test without consent unless there is an epidemic, although respondents reported few ethical concerns since people generally have trust in the health care system, and are willing to

rp Fo

provide information for public health purposes (many give blood to the Biobank for a range of research purposes).29

The findings of this research on the ethics of UAT of blood need consideration in the context of two broader contemporary debates within medical and public health ethics. The first

ee

debate centres on the boundaries of informed consent in health research, particularly in relation to the use of human tissue. The second debate focuses on the balance between

rr

individuals’ rights and the appropriate role of government3 in infringing on those rights for broader public health interests or concerns. These debates are inter-connected, although will be presented in turn.

Informed consent and ethics: a changing debate

w

ie

ev

Respect for autonomy is the bioethical principle that is most likely to be jeopardised by a

on

programme of unlinked anonymous blood testing, since informed consent is not usually deemed necessary for an individual’s blood to be included; this has tended to be the basis for

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 26

opposition to UAT.5 30 The findings from our research show, however, that such positions differ between countries, and have also changed with time.

Debates in the late 1980s and early 1990s demonstrated the opposing views of those who were concerned about patient consent,8 25 and those who felt that the priority was to control what was seen as an epidemic.15 UAT programmes did go ahead in the USA and UK, and other high income countries: there was no requirement for consent in the USA, whereas in the UK an approach of ‘prior notification’ was adopted through leaflets and posters in clinics – consent is presumed but individuals have the opportunity to opt out.


18

Page 19 of 26

Over the past two decades informed consent has emerged as ‘the ethical touchstone of medical research, and is now enshrined in practice and a range of international guidelines as one of the main preconditions of medical research’.24 31 Some commentators argue that even anonymised tissue should not be used without consent. Trouet, for example, believes that the use of such material is in contravention of Article 22 of the Convention on Human Rights and Biomedicine. This states that ‘when in the course of an intervention any part of a human body is removed, it may be stored and used for a purpose other than that for which it was removed,

rp Fo

only if this is done in conformity with appropriate information and consent procedures’, and in accordance with Article 8 of the European Convention on Human Rights which protects the right to respect for private life.32 Our research indicates that, rather than the universality of human rights discourse, a form of

ee

weak moral relativism operates in policy-making and practice around UAT.33 In the USA, the change in social context and debates about individual rights, along with evidence of reduction

rr

in mother-to-child vertical transmission, resulted in the abandonment of UAT in 1996.12 Interviews in the UK showed that the prior notification approach that remains in use today

ev

still holds strong support – based on perception that there is minimal harm to the individual, and the value of high quality, unbiased epidemiological data on HIV (held to be the most

ie

robust and accurate worldwide) (Buehler et al., 1994).34 Respondents felt that any change to UAT requiring informed consent would be cumbersome, time-consuming, and costly.

w

According to the Nuffield Council on Bioethics it may also be ethically unnecessary: ‘... it is acceptable to collect and use anonymised data for assessing and predicting trends in

on

infectious disease without consent, as long as any invasion of privacy is reduced as far as possible.’.35

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

This position is not, however, cross-culturally held. In the Netherlands, UAT has not been felt to be acceptable without informed consent, initially verbal and more recently written. The societal emphasis on individual autonomy, including in the health research environment, has been the paramount consideration.5 Northern European and Scandinavian countries are often held up as having a strong community ethos, with an expectation on individual involvement in endeavours for the public good.36 This was not reflected in UAT practice in the Netherlands, although was indicated to be important in Norway, albeit in a setting where UAT has never been taken up.


19

BMJ Open

These findings indicate the importance of social and political context in relation to medicoethical debates, and how the context can change with time and place.37 Differences exist between the moral evaluations and resultant UAT approaches between countries such as England and the Netherlands. Within the USA, in contrast, the practice of UAT has been reevaluated with time. In the late 1980s and early 1990s the feared AIDS epidemic was seen as a public health emergency, there was no treatment available, transmission from mother to unborn child was not preventable, and the stigma and discrimination associated with the

rp Fo

infection were more overt than today. As these factors shifted, so did the policy on UAT in the USA. The situation in low income countries today has created a perception of the acceptability of UAT akin to that in western countries two decades ago.

ee

Individual rights and public good

rr

There has been criticism in recent times of government for its ‘nanny state’ approach to interventions in, for example, the areas of smoking, alcohol and diet. More recently the

ev

notion of ‘stewardship’ in public health rejects these negative associations,38 and advocates that government should have a role to play in protecting the health of its citizens, preventing

ie

them from harming themselves or others, and promoting healthy behaviour.[4] In this view, public health interventions should not be thought of as unwanted restrictions imposed by a

w

‘nanny’ on powerless children, but as actions that favour both individuals and society more widely as prescribed by a ‘steward’ administering and protecting citizens’ interests.[5] The

on

Nuffield Council on Bioethics, for example, argues that the state should act as steward both to individuals and to the population as a whole, but that individuals’ personal choices should also be taken into account.35

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 26

The subject of UAT sits squarely within the debate about public health stewardship for social good. Recently, commentators have argued that patients of a publicly supported health care system like the NHS have a moral duty to participate in scientific research,39 40 although most do not go so far as to articulate that such participation should be mandatory in practice.41 Nevertheless, UAT could be understood as something that individuals should feel they have a moral obligation to participate in, regardless of ethical concerns about individual rights and personally held views. Our own survey of over 400 users of two genitor-urinary medicine


20

Page 21 of 26

clinics (a parallel part of this research project) has found that a large majority of respondents (89%) agree that people should have a responsibility to take part in research studies, and the same proportion would agree to the use of their blood in unlinked anonymous seroprevalence testing. However, when asked about the question of consent, 74% indicated that they should be asked to consent before their blood is used for public health surveillance.14 Notwithstanding this important latter point, such support for use of blood samples for prevalence surveys or for future research has not, however, been demonstrated in an earlier study in the UK42 or more recently in Egypt.43

rp Fo

The way forwards perhaps is to ensure that patients – and the public more widely – are informed of the existence, purposes and value of disease surveillance arrangements, as well as the importance of contributing to future health research, while investigating arrangements for gaining consent which are not unnecessarily cumbersome or likely to create bias amongst

ee

those who agree that their blood be used.44

rr

The issue of consent in seroprevalence surveillance, such as UAT, remains an important instrument to gauge the moral direction of public health policies more broadly.45 Debate

ev

continues around how public health programmes can embrace a Kantian stewardship relationship, rather than a nanny paternalistic one, while respecting autonomy and advocating

ie

social responsibilities. Further examination of why different countries have adopted different positions on seroprevalence surveillance will help illuminate our understanding of how social

w

morality is embedded in public health policies and practice.

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open


21

BMJ Open

Acknowledgments The authors would like to thank the participants in the study. We also acknowledge the valuable advice given by Vivian Hope, Will Nutland, Cicely Marston, and Graham Hart.

Conflicts of interest AK is also Director of Public Health Strategy and Medical Director of the Health Protection Agency. The views expressed in this paper are those of the authors, and are not intended to represent the views of the Health Protection Agency.

Funding

rp Fo

This research was supported by the Economic and Social Research Council (ESRC Grant Reference: RES-000-22-2096).

Ethical approval

ee

The study was approved by the South West Research Ethics Committee on August 16th 2007

rr

(reference 07/H0206/46), and by the London School of Hygiene & Tropical Medicine ethics committee on 3rd September 2007 (reference 5174).

ev

Notes

The England and Wales programme is currently managed by the Health Protection

w

[1]

ie

Agency (HPA) and the Institute of Child Health, University of London. The programme was previously managed by the Communicable Disease Surveillance Centre, part of the Public

on

Health Laboratory Service, which was subsumed into the Health Protection Agency in 2005. [2]

Health Protection Scotland (HPS) co-ordinates data collection in Scotland.

[3]

Those chosen had particular involvement or interest in UAT and ethics, but further

details cannot be provided to maintain their confidentiality. [4]

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 26

By government, the meaning here is health and public health policy makers within

government. [5]

It is interesting that in this discourse ‘nanny’ (female) is seen as negative and

‘steward’ (male) as positive, when both are servants employed to manage and protect.


22

Page 23 of 26

References

1

Nicoll A, Gill ON, Peckham C et al. The public health applications of unlinked

anonymous seroprevalence monitoring of HIV in the United Kingdom. International Journal of Epidemiology 2000;29:1-10. 2

Department of Health. Prevalence of HIV in England and Wales in 1997: Annual

Report of the Unlinked Anonymous Prevalence Monitoring Programme. London: Department

rp Fo

of Health, 1998. 3

UK Collaborative Group for HIV and STI Surveillance. Testing Times: HIV and

Other Sexually Transmitted Infections in the United Kingdom. London: Health Protection Agency, 2007. 4

ee

Bayer R, Lumey LH, Wan L. The American, British and Dutch responses to unlinked

anonymous HIV seroprevalence studies: an international comparison. AIDS 1990;4(4):283-

rr

290. 5

ev

Kopelman LM. Informed consent and anonymous tissue samples: the case of HIV

seroprevalence studies. Journal of Medicine and Philosophy 1994;19:525-52. 6

ie

Fairchild AL, Bayer R. Ethics and the conduct of public health surveillance. Science

2004;303:631-632. 7

w

Gill ON., Adler MW, Day NE. Monitoring the prevalence of HIV: foundations for a

on

Programme of unlinked anonymous testing in England and Wales. British Medical Journal 1989:299:1295-7. 8

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Berridge V. AIDS in the UK: the Making of Policy 1981-1994. Oxford: Oxford

University Press, 1996. 9

Health Protection Agency. Overall HIV prevalence. Health Protection Agency 2008.

http://www.hpa.org.uk/web/HPAweb&Page&HPAwebAutoListName/Page/1201094588821. 10

Gillon R. Testing for HIV without permission. British Medical Journal 1987;294:821-

823.


23

BMJ Open

11

Fairchild AL, Bayer R, Colgrove J. Searching Eyes: Privacy, the State, and Disease

Surveillance in America. Berkeley: University of California Press, 2007. 12

Stoto MA. Almario DA, McCormick MC. Reducing the Odds: Preventing Perinatal Transmission of HIV in the United States. Washington: National Academy Press, 1999.

13

Aavitsland P, Nilsen Ø, Lystad A. Anonymous reporting of HIV infection: an

evaluation of the HIV/AIDS surveillance system in Norway 1983-2000. European Journal of Epidemiology 2001;17:479-89. 14

rp Fo

Datta J, Kessel AS, Wellings K, Nanchahal K, Marks D, Kinghorn G. The views of

genitourinary medicine (GUM) clinic users on unlinked anonymous testing of blood for HIV: evidence from a pilot study of clinics in two English cities. Journal of Medical Ethics 2011 doi:10.1136/jme.2011.042705. 15

Doll R. A proposal for doing prevalence studies of AIDS. British Medical Journal

rr

1987;294:443-444. 16

ee

Avins AL, Lo B. To tell or not to tell: the ethical dilemmas of HIV test notification in

ev

epidemiologic research. American Journal of Public Health 1989;79/11:1544-7. 17

Datta J, Kessel AS. Unlinked anonymous testing for blood purposes: an ethical

ie

dilemma? In: Peckham S, Hann A, eds. Public Health Ethics and Practice. Bristol: Policy Press, 2010:101-116.

w

18

Bowling A. Research Methods in Health. Buckingham: Open University Press, 2000.

19

Green J, Thorogood N. Qualitative Methods for Health Research. London: Sage,

2005.

ly

20

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 26

Duong T, Ades AE, Rogers P, et al. Non-participation bias in unlinked anonymous

HIV-prevalence surveys in England and Wales. Epidemiology and Infection 1999;122:26772. 21

Pulver WP, Glebatis D, Wade N, et al. Trends from an HIV seroprevalence study

among childbearing women in New York State from 1988 through 2000: a valuable epidemiologic tool. Archives of Pediatric and Adolescent Medicine 2004;158(5):443-8.


24

Page 25 of 26

22

Kessel AS, Watts CJ. Usefulness of information from the Unlinked Anonymous

Prevalence Monitoring Programme for HIV in England and Wales: survey of planners of HIV/AIDS services. International Journal of STD and AIDS 1999;10:808-11. 23

Pfeffer N, Laws S. 'It's only a blood test': what people know and think about

venepuncture and blood. Social Science & Medicine 2006;62:3011-3023. 24

Cassell J, Young A. Why we should not seek individual informed consent for

rp Fo

participation in health services research. Journal of Medical Ethics 2002;23:313-317. 25

Bayer R. The ethics of blinded HIV surveillance testing. American Journal of Public

Health 1993;83/4:496-7. 26

de Zulueta P. The ethics of anonymised HIV testing of pregnant women: a

ee

reappraisal. Journal of Medical Ethics 2000;26:16-21. 27

Rennie S, Norris Turner A, Mupenda B, et al. Conducting unlinked anonymous HIV

rr

surveillance in developing countries: ethical, epidemiological, and public health concerns. PLoS Medicine 2009;6(1):e1000004. 28

ev

Postema EJ, Willems PWJM, de Ridder MAJ, et al. Comparison of patients refusing

ie

with patients accepting unlinked anonymous HIV testing in an outpatient STD department in the Netherlands. International Journal of STD and AIDS 1997;8:368-72. 29

w

Skolbekken J-A, Ursin LO, Solberg B, et al. Not worth the paper it’s written on?

on

Informed consent and Biobank research in a Norwegian context. Critical Public Health 2005;15(4):335-347. 30

Beauchamp TL and Childress JF. Principles of Biomedical Ethics. New York: Oxford


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Singleton P, Wadsworth M. Consent for the use of personal medical data in research.

British Medical Journal 2006;333:255-8. 32

Trouet C. New European guidelines for the use of stored human biological materials

in biomedical research. Journal of Medical Ethics 2004;30:99-103. 33

Rachels J. The Elements of Moral Philosophy. New York: McGraw-Hill, 1993.


25

BMJ Open

34

Beuhler JW, Petersen LR, Ward JW, et al. Defending HIV seroprevalence surveys.

American Journal of Public Health 1994;84:319-320. 35

Nuffield Council on Bioethics. Public Health: Ethical Issues. London: Nuffield

Council on Bioethics, 2007 36

Cappelen AW, Norheim OF. Responsibility in health care: a liberal egalitarian

approach. Journal of Medical Ethics 2005;31:476-80. 37

rp Fo

Mariner WK. Mission creep: public health surveillance and medical privacy. Boston

University Law Review 2007;87/2:347-95. 38

Jochelson K. Nanny or Steward? The Role of Government in Public Health. King's

ee

Fund Working Paper. London: King's Fund 2005. 39

Harris J. Scientific research is a moral duty. Journal of Medical Ethics

2005;31(4):242-248.

rr

40

Evans HM. Do patients have duties? Journal of Medical Ethics 2007;33(12):689-694.

41

Shapshay S, Pimple KD. Participation in biomedical research is an imperfect moral

ie

ev

duty: a response to John Harris. Journal of Medical Ethics 2007;33:414-417. 42

w

Kessel AS, Watts C, Weiss HA. Bad blood? Survey of public's views on unlinked

anonymous testing of blood for HIV and other diseases. British Medical Journal 2000;320:

on

90-91. 43

Abou-Zeid A, Silverman H, Shehata M, et al. Collection, storage and use of blood

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 26

samples for future research: views of Egyptian patients expressed in a cross-sectional survey. Journal of Medical Ethics 2010;36:539-547. 44

Nicholls SG. Knowledge or understanding? Informed choice in the context of

newborn bloodspot screening. Public Health Ethics 2010;3:128-136. 45

Haimes E. What can the social sciences contribute to the study of ethics? Theoretical,

empirical and substantive considerations. Bioethics 2002;16/2:89-113.


26

BMJ Open

The ethics of unlinked anonymous testing of blood: views from in-depth interviews with key informants in four countries

rp Fo Journal:

Manuscript ID: Article Type:



BMJ Open bmjopen-2012-001427.R1 Research 15-Aug-2012


Keywords:

Public health


ev


rr

ee

Kessel, Anthony; London School of Hygiene and Tropical Medicine, Social and environmental health research Datta, Jessica; London School of Hygiene and Tropical Medicine, Wellings, Kaye; London School of Hygiene and Tropical Medicine, Perman, Sarah; Health Protection Agency, Public Health Strategy


w

ie ly


Page 1 of 53

Original paper

The ethics of unlinked anonymous testing of blood: views from in-depth interviews with key informants in four countries Unlinked anonymous testing of blood (running title)

Anthony S Kessel (corresponding author and guarantor of study), Honorary Professor, London

rp Fo



ee


rr


Key words

w

ie


Word count

ly

6437 words

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open


1

BMJ Open

Abstract In this study we explore the ethical issues around unlinked anonymous testing (UAT) of blood, a method of sero-prevalence surveillance for infectious diseases. Our study focused on UAT for HIV, although UAT can be used for other infectious diseases.

Objectives The objectives of the research were to gain a better understanding of the views of key informants in countries adopting different UAT testing strategies, and to use the findings of the research to inform

rp Fo

health policy

Design


ee

Setting

Four countries using different strategies around unlinked anonymous testing of blood for HIV (United

rr

Kingdom, United States, Netherlands and Norway).

Participants

ev


ie

Results

Participants from the four countries have different views on UAT of blood, and the approaches and

w

policies on UAT adopted by different countries have been historically and culturally determined. We use our findings to explore the relationship between public health policy and ethics, framing our

on

discussion in relation to two important contemporary debates: informed consent for participation in medical and public health research; and the balance between individual good and the public good.

Conclusions

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 53

Qualitative research and ethical analysis of unlinked anonymous testing of blood in different countries has yielded important findings for consideration by policy makers. The policy of unlinked anonymous testing of blood for HIV and other diseases in the UK needs re-consideration in light of these findings.


2

Page 3 of 53



rp Fo


BACKGROUND

ee


rr


ev


w

ie


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open


Historically, UAT has developed in different ways in different settings, yet exploration of the moral dimensions of such developments has been relatively limited.4 5 6 Further, dominant attention given to individual rights and autonomy – in the health and healthcare context – make it an appropriate time to reassess moral considerations around UAT. A brief summary


3

BMJ Open


History of UAT in different countries

A national sero-prevalence surveillance programme using UAT to estimate the prevalence of HIV has been in place in the UK since 1990.11 The programme was first developed in

Formatted: Endn

rp Fo


ee


rr


ev


ie


w

Formatted: Endn

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 53

UAT surveys focused on population groups, for instance a survey of young people volunteering for job corps, although these were regional or local. In contrast to the UK the prior notification system, with the opportunity to opt out, was not used. UAT essentially stopped in the USA in 1996, as evidence emerged in relation to reducing vertical transmission of the HIV virus from a pregnant mother to the foetus through use of anti-viral medicine.11 The interplay at the time of clinical, social and political factors was complex.12

The main method for HIV surveillance in the Netherlands is named voluntary testing rather than UAT (these are ‘opt-out’ programmes, such as the antenatal testing programme and the


4

Page 5 of 53


rp Fo

accessing health services less than others.

In Norway, public health efforts have focused on diagnosing all those infected, as it was thought (certainly in earlier stages) that those who were diagnosed would be less likely than those undiagnosed to transmit infection to others. Mandatory named AIDS reporting was

ee


rr


ie


w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open




5

BMJ Open


METHODS

Four countries were selected purposively to participate in this study component (the key

rp Fo


ee

involvement in UAT and with appropriate knowledge or expertise, historically and/or current; existing links with the research team or advisory group members; and practical issues (e.g.

rr

availability and likelihood of participation). A pragmatic approach meant that more informants were interviewed in the UK than the other countries.

ev

A topic guide was developed that allowed for a semi-structured interview. Elements within

ie

the topic guide included: introduction and background to the purpose and nature of the interview (prior information had been provided); knowledge and views on unlinked

w

anonymous testing of blood (and national programme as appropriate), including advantages and disadvantages; ethical considerations including consent, prior notification,

on

confidentiality, anonymity, individual rights/liberties, social responsibilities, and any changes in ethical position over time; the value to public health of different surveillance methods;

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 53

blood as a tissue; legal considerations; and any other relevant issues the respondent wished to discuss. The topic guide was initially tailored for the different countries based on our own understanding, but then applied flexibly during the interview.

A researcher, with a background and skills in qualitative interviewing, was employed for the duration of the project, and undertook all the interviews. Interviews were carried out, in English, either face-to-face or by telephone between January and September 2008. All nonUK informants spoke good English, without evidence to the research team of any obvious language barrier issues. Interviews were recorded (except for two, when notes were taken: the


6

Page 7 of 53

first due to unsuitable location and the second to mechanical failure) with the participant’s permission, and transcribed verbatim. Interviews lasted between 20 minutes and one hour, the length directed by the contributions of the informants. In total 21 interviews were completed with 23 respondents (two interviews were with two respondents together). Of these, 12 were with UK respondents (three clinicians in genito-urinary medicine/HIV; one person working in government health policy; two HIV epidemiologists; two ethics academics; two charity representatives [MedFash, Terence Higgins Trust]; one person working at the British Medical Association; one editor of a medical journal), four with respondents in the USA (one health

rp Fo

policy academic; one academic in health law; one academic in public health; and one epidemiologist), three with respondents from the Netherlands (two epidemiologists; and one academic in medical ethics), and four with respondents in Norway (one epidemiologist; three academics in medical ethics – one of whom is also a hospital clinician). There were three interview refusals, all in the UK. [3]

ee

Two researchers (primary researcher, JD, and a second researcher, SP) coded the interview

rr

data using a thematic content analysis approach, looking for emergent themes (open coding in the context of the aims and objectives of the research). There were no discrepancies between

ev

the two researchers that required further discussion and consensus. Content was analysed for description of UAT developments in the country concerned, the context in which UAT may

ie

have changed, ethical tensions in UAT between surveillance and human rights, and broader connections between UAT and public health policy. Inevitably some of the themes inter-link

w

with each other (e.g. ‘value of UAT’ and ‘ethical issues’). Coding frames and analyses were shared with the wider research team.

on

Interview data pertaining to the history and state of sero-surveillance in different countries

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

were triangulated with literature sources. It was not appropriate to triangulate informants’ views on sero-surveillance.

RESULTS

The results, which represent the views and perceptions of informants, are presented by themes and by country. The labels describe the themes that emerged, in the context of informants’ awareness of the aims and objectives of the study.


7

BMJ Open

Value of UAT

UK The programme is largely seen by respondents as indispensable in providing an unbiased measure of HIV prevalence and risk factors in the UK population. The most important outcome from the surveys has been information on undiagnosed HIV infection which, respondents suggested, has led to changes in policy on named testing; for example, HIV

rp Fo

testing in pregnant women has changed from opt in to opt out, and sexual health clinics have moved from targeted to universal voluntary testing. The neonatal dried blood spot survey is used as a check on the performance of local antenatal screening. The public health rationale for the programme has not changed since 1990, and respondents cited the continuing high infection rate, particularly among men who have sex with men, as justification:

ee

"There was far more feeling of a public health emergency (then) than there is today,

rr

although you could argue … that it's actually much worse today because of where we see the disease." (UK4)

ev

Several respondents suggested that a review of UAT would be timely, including

ie

consideration of extending the surveys to other health care settings and other diseases. UK6 discussed how advances in HIV testing, including self-testing, might alter the questions asked

w


on

A particular theme was whether the British public has a duty to participate in public health

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 53

monitoring, possibly as part-and-parcel of receiving the benefits of a national health service (NHS):

"There is a very good case for really exploring with the public, and then pushing very hard to get public agreement, that in general terms that sort of tissue, extra bits of tissue or blood, should be available for use." (UK4)

USA


8

Page 9 of 53

According to one respondent, the surveillance programme was valuable in terms of planning, resource allocation, advocacy and interventions to prevent transmission. The heel prick survey provided accurate national data on prevalence in childbearing women and was seen as extremely valuable, with nothing since that has given as accurate a picture of national prevalence (USA 2).

However, given the questionable ethics of UAT (see later), as well as the resultant lack of good quality prevalence data, USA3 argued that enough became known about prevalence to

rp Fo

target the most vulnerable groups for health promotion.

“And how much more detail do you want from that, so what would you do differently if you knew that in the next level of detail.” (USA3)

Netherlands

ee

Unlinked anonymous surveys have been used in the Netherlands to measure HIV prevalence

rr

among high risk populations, but with informed consent (verbal or written). The populations surveyed by UAT have changed over time. Injecting Drug Users (IDUs) have been surveyed

ev

bi-annually since 1994 in different cities in the Netherlands, and are recruited via methadone care; sex workers and migrant groups have also been subject to UAT surveys.

ie

There are two reasons, indicated by respondents, why UAT in Netherlands has been of

w

limited value. First, the uptake of voluntary testing is generally high in the Netherlands, meaning the quality of that data are generally good. And second, UAT has only been deemed

on

ethically permissible in the Netherlands with informed consent, making large-scale UAT surveys impracticable, so restricting UAT to specific population groups and/or the research context.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Norway The unknown, undiagnosed segment of the population is known as the ‘dark figure(s)’ in Norwegian, although neither the ‘dark figure’ nor the number of cases of HIV has been perceived as significant issues in Norway. Uptake of voluntary testing for HIV was described as very high in intravenous drug users and antenatal clinic users, meaning that UAT has never been perceived as to be of value epidemiologically. The system of surveillance for HIV in Norway instead covers all diagnosed cases (i.e. not undiagnosed cases as with UAT), and was


9

BMJ Open

reported as being simple, inexpensive and capable of providing a representative and timely overview that guides prevention, while respecting confidentiality and adhering to ethical principles.


UK

rp Fo

The epidemiological data yielded from unlinked anonymous testing are considered by respondents to be of unrivalled quality internationally. It was suggested, however, that the apparent high number of undiagnosed HIV infections from UAT surveys might be inflated by double-counting:

ee

"You can quite understand that if you've got gonorrhoea and you've been told by your doctor not to have unsafe sex, you might well decide to go to a different clinic to have

rr

your gonorrhoea treated." (UK1)

ev

Others defended the statistics and argued that they were explained by a resurgence of high risk behaviour, particularly among gay men. UK1 argued that people who refuse a named test

ie

and who are subsequently found to be HIV positive through UAT are likely to be people who know that they are positive.

w

USA

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 53

The fact that there is no ‘right’ to health care in the USA was felt to have an effect on people’s views on surveillance – in other words, the ‘disconnect’ between health care provision (not available to all) and surveillance (done to all) (USA 1). The differences between surveillance, research and practice, with perhaps different ethical standards required for each, emerged as a theme from the interviews (USA 4).


10

Page 11 of 53

One point of view was that surveillance (because it is named and part of routine care) is part of clinical care rather than purely research (USA 2). However, while some of the surveillance undertakings at state level in the USA were considered to be practice, at the level of CDC they were treated as research (USA 1). The same respondent commented that cancer sufferers advocated in favour of surveillance for the purposes of research on the disease, since “they were willing to trade privacy and consent for a good which was surveillance that they saw as serving their interests”.

rp Fo

Norway and the Netherlands

There were no comments from Norway respondents about the quality of UAT data since such surveys are not carried out in Norway, in part because of the good quality of data on HIV and AIDS from other sources.

ee

Two respondents from the Netherlands judged that large-scale UAT programmes were not necessary because named HIV testing is widespread. The quality of information obtained

rr

from named testing was felt to be robust and could be used to model estimates of HIV prevalence in the population. Coverage of pregnant women is very high and this group is

ev

considered a valid proxy for the heterosexual population. Named HIV testing among users of STI clinics was also presented as being a valid method for estimating prevalence. It was

ie

argued that, as long as the uptake of named testing by pregnant women and STI clinic users remains high (above 90%) then a national programme of UAT was not thought to be

w

necessary. However, notwithstanding the above point it was acknowledged that current methods of surveillance do not allow undiagnosed HIV infection to be measured accurately.

Ethical issues

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

UK Respondents felt that, on balance, no harm was caused by the programme. A recurrent argument for this was that, in settings where anonymous testing takes place, people are offered a named HIV test:


11

BMJ Open

"I think the ethics are not in fact, are not a problem because everybody is now very strongly advised to have a test and therefore the residual rump who don't have a test, it's important to know what their rate of HIV positivity is." (UK1)

Some respondents, however, said there would be ethical concerns if people in these settings were not routinely offered a named test. Several respondents considered the ethics of UAT to be relatively unimportant, for instance in the context of more significant health-related problems in Africa:

rp Fo

"There are these enormous ethical, real ethical issues, and to me, this one … I just can't understand why one should put so much emphasis on it." (UK9)

UK11 suggested that the question of ownership of the sample was relevant, but that this could be argued both ways:

rr

ee

"Is it yours or is it somebody else's? ... If it's waste that's going to be chucked away then why shouldn't people use it if they find it useful? On the other hand, 'I gave

ev

permission for this test for my own benefit and I didn't give you permission to do things to help others.'" (UK11)

ie

There was universal agreement that changing to a system of informed consent would damage

w

the programme. There was a fear that introducing information about UAT into a clinical consultation could confuse patients. Patients might think they were getting a named test:

on

"You would lose someone's confidence and trust because they might not understand

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 53

the words or the reason behind why we're doing it, and they would think that you could get back to them and you did know." (UK12)




12

Page 13 of 53


"You know we've got ten minute appointments. It could take you five minutes, half of your time just to explain what this is all about. So doing it in a consultation is not realistic." (UK3) The respondent suggested that a system of clinic self-registration, using computers, could be used to explain anonymous testing.

rp Fo

All the respondents thought that prior notification was the right theoretical approach. However, some were concerned that, in practice, it might not be working:

"It's a reasonable thing in practical terms provided people are informed. So … if people don't know about their option for opting out and their option for finding out

ee

then the system hasn't been working." (UK11)

rr

Some suggested that the approach needed to be more "active" with information displayed visibly on clinic walls or in leaflets given out to each patient. One clinician, however,

ev

expressed his discomfort with his role in the programme. He described filling out the form for UAT blood testing in front of the patient, without explaining what he was doing, and

ie

sometimes after the patient had declined a named HIV test:

w


on

communicate well and to be as open as possible with patients and that's slightly in conflict with that." (UK3)

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

USA

Consent was not deemed necessary initially in the USA for participation in UAT as patients had consented to the blood being collected in the first place, and this was seen as adequate. USA1 suggested that surveillance is a public good, meaning there are strong reasons for making it a universal obligation on citizens – and that notification that surveillance is occurring is different from asking for consent.


13

BMJ Open

The blood spot survey continued until the publication of the first trials showing that vertical transmission of HIV could be dramatically reduced by treating mothers. When it became clear that infected women could be offered treatment, and therefore should know their HIV status, “it became impossible for the US Centre for Communicable Disease Control and Prevention (CDC) to continue that programme and it was dropped in the early 1990s along with all the other ones” (USA3). It had become difficult to reconcile public health benefits of UAT with individual rights. If CDC had not voluntarily withdrawn funding from UAT, the same respondent indicated that it was likely that Congress would have mandated cessation, as

rp Fo

there was a strong coalition in favour of this.

A related legal issue was brought by USA4, in that public health surveillance appears to go against “fundamental legal principles” in the USA, and constitutional scholars are shocked when they hear what is done.

ee

‘Once you begin to argue that the government can take your personally identifiable

rr

information for any reason it finds useful without your consent you've opened up and blown a hole in much of the fundamental legal principles in the country.’ (USA4)

ev

The Tuskegee episode (a study between 1932 and 1972 in Alabama in which black syphilitic

ie

Americans were not given treatment although thought they were) was mentioned by USA1 as a reason for USA citizens, particularly Afro-Americans, to be suspicious of surveillance

w

practices. In the context of opposition to named case reporting and a consequent lack of trust in surveillance, USA3 argued that there are probably cultural dimensions to issues such as

on

consent, and the acceptability of particular interventions may vary from country to country.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 53

“It [UAT] became more of a civil rights issue than a gay specific one. It was, the government knows something that you don’t and the attitude in the States, at least at the time, towards the government was much less trusting than in this country [i.e. UK]. People did not assume and still don’t assume that the government necessarily is looking out for your well being. And I think there’s a much more paranoid attitude towards the government than there is in this country.” (USA3)


14

Page 15 of 53

Related to these cultural dimensions, a particular ethical concern was that the US (and, in particular, CDC) supports UAT in developing countries, with the inherent contradiction that it has been rejected in the USA.

‘It seems to me the CDC can’t at the one time say it’s unethical to do it in the US and say it’s ethical to do it abroad. Different public agencies may come to different decisions, but the same agency it seems to me should be bound by some rule of consistency.’ (USA1)

rp Fo

USA2 felt very strongly that arrangements for UAT in developing countries are not ethical where pregnant women do not have the opportunity to have a named test, counselling and treatment.

Netherlands

ee

The respondents felt that more extensive UAT surveys would produce more accurate

rr

information for public health surveillance. However, inter-linked practical and ethical doubts about UAT were presented. One concern was that UAT might deter high risk groups from

ev

seeking testing and treatment. Another concern was that, if UAT focused only on high risk groups, such groups would risk being stigmatised. It was suggested that reporting of findings

ie

from UAT surveys of high risk groups would need to be carried out sensitively.

w

A particular issue was consent and maintaining trust with patients. One interviewee expressed that staff in HIV treatment centres and STI clinics demand a more stringent consent process

on

than required by the research ethics committees in the Netherlands in order to promote trust among patients.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Norway

A UAT system was never thought relevant, necessary or appropriate for Norway. In Norway it is not legal to test or analyse blood or biological material for infectious diseases for nondiagnostic purposes without consent unless the purpose is surveillance of an ‘epidemic’. One respondent felt this displayed a ‘moral grey zones’, as the law hinges on the definition of ‘epidemic’ [Norway 1].


15

BMJ Open

There is very little concern or awareness in Norway about ethical issues around UAT. However, there is a general willingness to provide information for public health purposes in Norway, and people are used to being surveyed. It was reported that many Norwegians have given blood to the Biobank, and give broad consent for their tissue/blood to be used for a wide range of research. A respondent indicated that many are willing to waive their consent in Norway because ‘there is significant trust’ in the health care system. However, another respondent expressed that there was a possible danger that UAT blood could be used to identify someone via the Home Office DNA database, and that researchers might be forced to

rp Fo


One respondent suggested that informed consent could be different in different contexts. In a

ee

country like Norway, with low prevalence and easy access to health care, the current surveillance system (non-UAT) was described as acceptable and functioning well. However,

rr

in a country in Africa (where this respondent was carrying out research), with a high rate of HIV positive pregnant women and a low rate of testing, routine testing with opt-out consent procedures would be morally acceptable.

w

DISCUSSION

ie

ev

Since its inception, unlinked anonymous testing (UAT) has been a subject of

on

epidemiological, ethical and policy debate.15 16 17 In this paper we have presented the results of qualitative interviews with key informants in four countries pragmatically selected because they represent a range of approaches to UAT. 18 A total of 23 respondents were interviewed,

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 53

differentially distributed between countries: 12 in the UK, four in the USA, four in Norway and three in the Netherlands. First a summary of key findings is presented (in the context of historical development of UAT in different countries, as well as what is already know of the subject) , followed by a discussion of how the findings from this paper fit into broader contemporary debates about public health ethics and policy.



16

Page 17 of 53

In the United Kingdom (UK), a balance of epidemiological and ethical considerations in the late 1980s led to the implementation of the ‘prior notification’ approach to UAT; this involves leaflets and posters informing clients that UAT is occurring, allowing opting out but not necessitating the gaining of informed consent.77 1010 Respondents stressed that the

Formatted: Endn

Formatted: Endn

rationale remains apt today, and that this approach provides data unrivalled in terms of epidemiological quality,99 19 20 and which is of use for health service planning.21 Ethical

Formatted: Endn

concerns were generally felt to be minimal today since clients are (also) usually offered

rp Fo

voluntary (named) testing in settings where UAT is occurring. However, there was mention of concern about the lack of full explanation, whether people do actually know UAT is happening, and ‘ownership’ of blood.22 It was indicated that an approach involving informed consent would be cumbersome, time-consuming and epidemiologically damaging.23

ee

In the United States (USA), UAT of blood for HIV was extensively used in the 1980s and early 1990s,24 but was stopped in the late 1990s as evidence emerged that anti-viral therapy in

rr

pregnant women could help reduce the vertical transmission of HIV from pregnant mother to child – named testing needed to be maximised so that HIV-positive pregnant women could be

ev

identified and offered advice and therapy.25 The interviews indicated that data from UAT were useful epidemiologically but, since UAT has stopped, there has been sufficient other

ie

data to plan services accordingly. Other issues raised include a general suspicion around epidemiological surveillance post-Tuskegee, cultural dimensions of UAT with respect to

w

minority groups, a lack of trust in the Government, and the related disconnect between surveillance (done to all) and health care provision (variable, and not done to all).66 1111 There

on

Formatted: Endn

Formatted: Endn

was also comment about the apparent contradiction in the contemporary support of the US Centre for Communicable Disease Control and Prevention (CDC) for UAT in low-income countries.26

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Interviews in the Netherlands showed that named (voluntary) testing, which has generally been widely taken up, has provided the main vehicle for surveillance of HIV.44 There is

Formatted: Endn

acknowledgement that large-scale UAT would provide additional epidemiological data, but this has been balanced against concerns of this leading to diminished voluntary testing – especially in at risk groups – through loss of trust in the healthcare system.27 Smaller UAT surveys have been intermittently undertaken in such groups (e.g. intravenous drug users, sex


17

BMJ Open

workers, migrants) as research studies, but have required individual consent (initially oral, now written).

Interviews in Norway revealed that UAT was never felt to be necessary epidemiologically, as HIV has not been a significant issue and the uptake of voluntary testing has been high.1313

Formatted: Endn

Surveillance has thus tended to focus on the monitoring of the diagnosed population (anonymised diagnosed cases), which has been supported by the gay community; the undiagnosed population is interestingly referred to as ‘dark figure(s)’. In Norway it is not

rp Fo

legal to test without consent unless there is an epidemic, although respondents reported few ethical concerns since people generally have trust in the health care system, and are willing to provide information for public health purposes (many give blood to the Biobank for a range of research purposes).28

ee


rr


ev

individuals’ rights and the appropriate role of government3 in infringing on those rights for broader public health interests or concerns. These debates are inter-connected, although will

ie

be presented in turn.

w Informed consent and ethics: a changing debate

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 53

programme of unlinked anonymous blood testing, since informed consent is not usually deemed necessary for an individual’s blood to be included; this has tended to be the basis for opposition to UAT.55 29 The findings from our research show, however, that such positions

Formatted: Endn

differ between countries, and have also changed with time.

Debates in the late 1980s and early 1990s demonstrated the opposing views of those who were concerned about patient consent,88 2425 and those who felt that the priority was to control what was seen as an epidemic.

1515

Formatted: Endn

Formatted: Endn

UAT programmes did go ahead in the USA and UK, and

Formatted: Endn

other high income countries: there was no requirement for consent in the USA, whereas in the


18

Page 19 of 53

UK an approach of ‘prior notification’ was adopted through leaflets and posters in clinics – consent is presumed but individuals have the opportunity to opt out.

Over the past two decades informed consent has emerged as ‘the ethical touchstone of medical research, and is now enshrined in practice and a range of international guidelines as one of the main preconditions of medical research’.2324 30 Some commentators argue that

Formatted: Endn

even anonymised tissue should not be used without consent. Trouet, for example, believes that the use of such material is in contravention of Article 22 of the Convention on Human

rp Fo

Rights and Biomedicine. This states that ‘when in the course of an intervention any part of a human body is removed, it may be stored and used for a purpose other than that for which it was removed, only if this is done in conformity with appropriate information and consent procedures’, and in accordance with Article 8 of the European Convention on Human Rights which protects the right to respect for private life.31

ee

Our research indicates that, rather than the universality of human rights discourse, a form of

rr

weak moral relativism operates in policy-making and practice around UAT.32 In the USA, the change in social context and debates about individual rights, along with evidence of reduction

ev

in mother-to-child vertical transmission, resulted in the abandonment of UAT in 1996.1212

Formatted: Endn

The fact that the USA continued to support UAT in international research overseas illustrates

ie

further an attitude of moral relativism, although it could be argued that differential health needs justified different approaches.

w

Interviews in the UK showed that the prior notification approach that remains in use today

on

still holds strong support – based on perception that there is minimal harm to the individual, and the value of high quality, unbiased epidemiological data on HIV (held to be the most

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

robust and accurate worldwide).33 Respondents felt that any change to UAT requiring informed consent would be cumbersome, time-consuming, and costly. According to the Nuffield Council on Bioethics it may also be ethically unnecessary: ‘... it is acceptable to collect and use anonymised data for assessing and predicting trends in infectious disease without consent, as long as any invasion of privacy is reduced as far as possible.’.34

This position is not, however, cross-culturally held. In the Netherlands, UAT has not been felt to be acceptable without informed consent, initially verbal and more recently written. The societal emphasis on individual autonomy, including in the health research environment, has


19

BMJ Open

been the paramount consideration.55 Northern European and Scandinavian countries are often

Formatted: Endn

held up as having a strong community ethos, with an expectation on individual involvement in endeavours for the public good.35 This was not reflected in UAT practice in the Netherlands, although was indicated to be important in Norway, albeit in a setting where UAT has never been taken up.

These findings indicate the importance of social and political context in relation to medicoethical debates, and how attitudes and policies can change in time within any given

rp Fo

jurisdiction.36 Differences exist between the moral evaluations and resultant UAT approaches between countries such as England and the Netherlands. Within the USA, in contrast, the practice of UAT has been re-evaluated with time. In the late 1980s and early 1990s the feared AIDS epidemic was seen as a public health emergency, there was no treatment available, transmission from mother to unborn child was not preventable, and the stigma and

ee

discrimination associated with the infection were more overt than today. As these factors shifted, so did the policy on UAT in the USA. The situation in low income countries today

rr

has created a perception of the acceptability of UAT akin to that in western countries two decades ago.

ie

ev Individual good and community (public) good

w

There has been criticism in recent times of government for its ‘nanny state’ approach to interventions in, for example, the areas of smoking, alcohol and diet. More recently the

on

notion of ‘stewardship’ in public health rejects these negative associations,37 and advocates that government should have a role to play in protecting the health of its citizens, preventing

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 53

them from harming themselves or others, and promoting healthy behaviour.[4] In this view, public health interventions should not be thought of as unwanted restrictions imposed by a ‘nanny’ on powerless children, but as actions that favour both individuals and society more widely as prescribed by a ‘steward’ administering and protecting citizens’ interests. The Nuffield Council on Bioethics, for example, argues that the state should act as steward both to individuals and to the population as a whole, but that individuals’ personal choices should also be taken into account.3435


Formatted: Endn

20

Page 21 of 53

The subject of UAT sits squarely within the debate about public health stewardship for social good. Recently, commentators have argued that patients of a publicly supported health care system like the NHS have a moral duty to participate in scientific research,38 39 although most do not go so far as to articulate that such participation should be mandatory in practice.40 Nevertheless, UAT could be understood as something that individuals should feel they have a moral obligation to participate in, regardless of ethical concerns about individual rights and personally held views. Our own survey of over 400 users of two genitor-urinary medicine clinics (a parallel part of this research project) has found that a large majority of respondents

rp Fo

(89%) agree that people should have a responsibility to take part in research studies, and the same proportion would agree to the use of their blood in unlinked anonymous seroprevalence testing. However, when asked about the question of consent, 74% indicated that they should be asked to consent before their blood is used for public health surveillance.1414

Formatted: Endn

Notwithstanding this important latter point, such support for use of blood samples for

ee

prevalence surveys or for future research has not, however, been demonstrated in an earlier study in the UK41 or more recently in Egypt.42

rr

The way forwards perhaps is to ensure that patients – and the public more widely – are

ev

informed of the existence, purposes and value of disease surveillance arrangements, as well as the importance of contributing to future health research, while investigating arrangements

ie

for gaining consent which are not unnecessarily cumbersome or likely to create bias amongst those who agree that their blood be used.43

w

The issue of consent in seroprevalence surveillance, such as UAT, remains an important

on

instrument to gauge the moral direction of public health policies more broadly.44 Debate continues around how public health programmes can embrace a stewardship relationship,

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

rather than a nanny paternalistic one, while respecting autonomy and advocating social responsibilities. Further examination of why different countries have adopted different positions on seroprevalence surveillance will help illuminate our understanding of how social morality is embedded in public health policies and practice.


21

BMJ Open

Acknowledgments The authors would like to thank the participants in the study. We also acknowledge the valuable advice given by Vivian Hope, Will Nutland, Cicely Marston, and Graham Hart.

Conflicts of interest AK is also Director of Public Health Strategy and Medical Director of the Health Protection Agency. The views expressed in this paper are those of the authors, and are not intended to

rp Fo

represent the views of the Health Protection Agency.

Funding


rr

Ethical approval

ee

The study was approved by the South West Research Ethics Committee on August 16th 2007 (reference 07/H0206/46), and by the London School of Hygiene & Tropical Medicine ethics

ev

committee on 3rd September 2007 (reference 5174).

ie

Notes [1]

w


Agency (HPA) and the Institute of Child Health, University of London. The programme was

on

previously managed by the Communicable Disease Surveillance Centre, part of the Public Health Laboratory Service, which was subsumed into the Health Protection Agency in 2005.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 53

[2]


[3]




government.


22

Page 23 of 53

References

1





rp Fo

of Health, 1998. 3



ee



rr

290. 5

ev



ie


2004;303:631-632. 7

w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open






823.


23

BMJ Open

11




13



rp Fo




rr

1987;294:443-444. 16

ee


ev



ie


w

18


19


ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 53



among childbearing women in New York State from 1988 through 2000: a valuable epidemiologic tool. Archives of Pediatric and Adolescent Medicine 2004;158(5):443-8. 21


Prevalence Monitoring Programme for HIV in England and Wales: survey of planners of HIV/AIDS services. International Journal of STD and AIDS 1999;10:808-11.


24

Page 25 of 53

22






Health 1993;83/4:496-7. 25

rp Fo




ee


rr


with patients accepting unlinked anonymous HIV testing in an outpatient STD department in

ev

the Netherlands. International Journal of STD and AIDS 1997;8:368-72. 28


ie


on


University Press, 1994.



ly

30

w

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open




33


American Journal of Public Health 1994;84:319-320.


25

BMJ Open

34







rp Fo




2005;31(4):242-248.

ee

39


40


rr


ev


ie

anonymous testing of blood for HIV and other diseases. British Medical Journal 2000;320: 90-91. 42

w


on




ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 53




26

Page 27 of 53

Original paper


rp Fo

Anthony S Kessel (corresponding author and guarantor of study), Honorary Professor, London School of Hygiene & Tropical Medicine, Faculty of Public Health and Policy, 15-17 Tavistock Place, London WC1H 9SH (Tel: 020 7759 2715; email: [email protected])


ee



Key words

6437 words

ly

on

Word count

w

Unlinked, anonymous, blood, ethics, public health, policy, HIV

ie

ev

rr

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

1


BMJ Open


Objectives

rp Fo

The objectives of the research were to gain a better understanding of the views of key informants in countries adopting different UAT testing strategies, and to use the findings of the research to inform health policy

Design

ee


Setting

rr

Four countries using different strategies around unlinked anonymous testing of blood for HIV (United Kingdom, United States, Netherlands and Norway).

Participants Twenty-three key informants in the four countries.

w

ie

Results

ev

Participants from the four countries have different views on UAT of blood, and that the approaches and policies on UAT adopted by different countries have been historically and culturally determined. We use our findings to explore the relationship between public health policy and ethics, framing our

on

discussion in relation to two important contemporary debates: informed consent for participation in medical and public health research; and the balance between individual rights good and the public good.

Conclusions

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 53


2


Page 29 of 53

Public health policy, context and ethics: the example ofThe ethics of unlinked anonymous testing of blood: views from in-depth interviews with key informants in four countries unlinked anonymous testing of blood

In this paper we explore unlinked anonymous testing (UAT) of blood for HIV and other

rp Fo

diseases, and examine the ethical issues associated with this method of sero-prevalence surveillance for infectious diseases. We report original results from a study funded by the UK Economic and Social Research Council (ESRC) – qualitative interviews with stakeholders in four countries adopting different UAT policies – and use the subject matter, and our findings, to explore the relationship between public health policy, context and ethics.

BACKGROUND

rr

ee

Unlike voluntary testing for clinical purposes, unlinked anonymous testing (UAT) of blood is carried out in order to provide information about the prevalence of particular infectious

ev

diseases in a population. The use of UAT began in the late 1980s for HIV, and remains most commonly associated with this infection, but can be applied to other infectious diseases such

ie

as hepatitis C.1 In this paper we use the term UAT synonymously with sero-prevalence surveys, or sero-prevalence surveillance.

w

The principles behind UAT can be explained with reference to the approach used in the UK

on

for HIV. Leftover blood from samples taken for diagnostic or routine screening purposes is made anonymous, irreversibly unlinked from the source, and then tested for HIV.2 3 This

means that individuals whose blood tests positive for HIV cannot be contacted or informed.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open


Historically, UAT has developed in different ways in different settings, yet exploration of the moral dimensions of such developments has been relatively limited.4 5 6 Further, recent dominant attention given to individual rights and autonomy – in the health and healthcare context – make it an appropriate time to reassess moral considerations around UAT. A brief

3


BMJ Open

summary of developments in UAT in the four countries used in our study is provided as a necessary introduction to the research project.


A national sero-prevalence surveillance programme using UAT to estimate the prevalence of

rp Fo

HIV has been in place in the UK since 1990.111 The programme was first developed in

Formatted: Endnote Reference

7

response to concerns about the spread of HIV during the 1980s and, despite early opposition,


the methodology eventually gained political support.8 The programme was funded initially as a pilot project by the Medical Research Council, and has since been funded by the Department of Health. [1] [2] The programme currently uses data collected from three

ee

ongoing surveys (newborn babies, [maternal antibodies], users of genitourinary medicine (GUM) clinics, and intravenous drug users [saliva]) to measure the proportion of individuals within the general population, and within subgroups of the population, infected with HIV.

rr


ev

A system of ‘prior notification’ operates in locations where testing goes on, involving the display of leaflets or posters explaining that blood may be tested anonymously. Consent to

ie

participation is implicit rather than explicit; provision is made to opt out of testing, although this naturally depends on awareness.999 10

w

Formatted: Endnote Reference Formatted: Endnote Reference

In the USA, UAT was extensively used in the late 1980s and early 1990s using discarded

on

blood samples. The largest survey was national and used newborn heel prick tests. Other UAT surveys focused on population groups, for instance a survey of young people

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 53

volunteering for job corps, although these were regional or local. In contrast to the UK the prior notification system, with the opportunity to opt out, was not used. UAT essentially stopped in the USA in 1996, as evidence emerged in relation to reducing vertical

transmission of the HIV virus from a pregnant mother to the foetus through use of anti-viral medicine.11 The interplay at the time of clinical, social and political factors was complex.12

The main method for HIV surveillance in the Netherlands is named voluntary testing rather than UAT (these are ‘opt-out’ programmes, such as the antenatal testing programme and the 4


Page 31 of 53

testing of users of STI clinics, as well as the extensive ‘opt-in’ testing carried out by general practitioners). The UAT that does occur in the Netherlands is carried out using blood samples from intravenous drug users (IDUs) and saliva samples from other groups. Unlike the UK or USA, informed consent is required from those surveyed by UAT, either verbal or, more recently, written consent. From 2002 to 2006, other high risk groups haven been included in UAT research surveys in the Netherlands (e.g. migrants from the Cape Verde islands, Surinam and the Netherlands Antilles, as well as commercial sex workers), because of

rp Fo

concerns of a higher proportion of undiagnosed HIV in such groups, who were also felt to be accessing health services less than others.

In Norway, public health efforts have focused on diagnosing all those infected, as it was thought (certainly in earlier stages) that those who were diagnosed would be less likely than

ee

those undiagnosed to transmit infection to others. Mandatory named AIDS reporting was introduced in Norway in 1983 and mandatory anonymous HIV case reporting in 1986, both including cases diagnosed before those dates.13 The Norwegian approach has therefore

rr


AIMS AND OBJECTIVES OF THE RESEARCH STUDY

w

ie

ev The ESRC-funded project had three components: a questionnaire survey of 471 patients attending two inner-urban genito-urinary medicine (GUM) clinics in two English cities,

on

exploring their views of, and attitudes towards, UAT;14 qualitative in-depth interviews with a sub-sample (20) of these patients; and in-depth interviews with key informants in four

countries chosen to represent different approaches towards sero-surveillance for HIV. In this

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

paper, we present findings from this third component, the qualitative interviews with key informants.


5


BMJ Open


METHODS

rp Fo

Four countries were selected purposively to participate in this study component (the key informant interviews), based on their representing a range of approaches, or positions taken, with regard to UAT: the UK (specifically England), USA, the Netherlands and Norway. Purposive sampling was used to select the key informants from these countries, informed by the following criteria: range of stakeholder perspectives (e.g. epidemiology, clinical, public

ee

health policy, charity or pressure group, and academia (public health or medical ethics)); involvement in UAT and with appropriate knowledge or expertise, historically and/or current; existing links with the research team or advisory group members; and practical issues (e.g.

rr


ev


ie

interview (prior information had been provided); knowledge and views on unlinked

w


confidentiality, anonymity, individual rights/liberties, social responsibilities, and any changes

on

in ethical position over time; the value to public health of different surveillance methods;

blood as a tissue; legal considerations; and any other relevant issues the respondent wished to

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 53


A researcher, with a background and skills in qualitative interviewing, was employed for the duration of the project, and undertook all the interviews. Interviews were carried out, in English, either face-to-face or by telephone between January and September 2008. All nonUK informants spoke good English, without evidence to the research team of any obvious language barrier issues. Interviews were recorded (except for two, when notes were taken: the 6


Page 33 of 53

first due to unsuitable location and the second to mechanical failure) with the participant’s permission, and transcribed verbatim. Interviews lasted between 20 minutes and one hour, the length directed by the contributions of the informants. In total 21 interviews were completed with 23 respondents (two interviews were with two respondents together). Of these, 12 were with UK respondents (three clinicians in genito-urinary medicine/HIV; one person working in government health policy; two HIV epidemiologists; two ethics academics; two charity representatives [MedFash, Terence Higgins Trust]; one person working at the British Medical

rp Fo

Association; one editor of a medical journal), four with respondents in the USA (one health policy academic; one academic in health law; one academic in public health; and one epidemiologist), three with respondents from the Netherlands (two epidemiologists; and one academic in medical ethics), and four with respondents in Norway (one epidemiologist; three academics in medical ethics – one of whom is also a hospital clinician). There were three interview refusals, all in the UK. [3]

ee

Interview data were coded by Two the researchers (primary researcher, JD, and a second

rr

researcher, SP) coded the interview data using a thematic content analysis approach, and looking for emergent themes emergent themes assessed by the same researcher (open coding

ev

in the context of the aims and objectives of the research); a second researcher assisted with coding and analysis. There were no discrepancies between the two researchers that required further discussion and consensus. Inevitably some of the themes inter-link with each other

ie

(e.g. ‘value of UAT’ and ‘ethical issues’). Content was analysed for description of UAT

w

developments in the country concerned, the context in which UAT may have changed, ethical tensions in UAT between surveillance and human rights, and broader connections between UAT and public health policy. Inevitably some of the themes inter-link with each other (e.g.

on

‘value of UAT’ and ‘ethical issues’). Coding frames and analyses were shared with the wider research team.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Interview data pertaining to the history and state of sero-surveillance in different countries

Formatted: Font: Not Bold

were triangulated with literature sources. It was not appropriate to triangulate informants’ views on sero-surveillance.

7


BMJ Open

RESULTS Formatted: Font: Not Bold

The results, which represent the views and perceptions of informants, are presented by

Formatted: Font: Not Bold

themes and by country. The labels describe the themes that emerged, in the context of informants’ awareness of the aims and objectives of the study.

Value of UAT

UK

rp Fo

The programme is largely seen by respondents as indispensable in providing an unbiased measure of HIV prevalence and risk factors in the UK population. The most important outcome from the surveys has been information on undiagnosed HIV infection which,

ee

respondents suggested, has led to changes in policy on named testing; for example, HIV testing in pregnant women has changed from opt in to opt out, and sexual health clinics have moved from targeted to universal voluntary testing. The neonatal dried blood spot survey is

rr

used as a check on the performance of local antenatal screening. The public health rationale for the programme has not changed since 1990, and respondents cited the continuing high

ev

infection rate, particularly among men who have sex with men, as justification:

"There was far more feeling of a public health emergency (then) than there is today,

ie

although you could argue … that it's actually much worse today because of where we see the disease." (UK4)

w


on

consideration of extending the surveys to other health care settings and other diseases. UK6

discussed how advances in HIV testing, including self-testing, might alter the questions asked

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 53


A particular theme was whether the British public has a duty to participate in public health monitoring, possibly as part-and-parcel of receiving the benefits of a national health service (NHS):

8


Page 35 of 53

"There is a very good case for really exploring with the public, and then pushing very hard to get public agreement, that in general terms that sort of tissue, extra bits of tissue or blood, should be available for use." (UK4)

USA According to one respondent, the surveillance programme was valuable in terms of planning, resource allocation, advocacy and interventions to prevent transmission. The heel prick

rp Fo

survey provided accurate national data on prevalence in childbearing women and was seen as extremely valuable, with nothing since that has given as accurate a picture of national prevalence (USA 2).

However, given the questionable ethics of UAT (see later), as well as the resultant lack of

ee

good quality prevalence data, USA3 argued that enough became known about prevalence to target the most vulnerable groups for health promotion.

rr


Netherlands

ev


ie

among high risk populations, but with informed consent (verbal orf written). The populations

w

surveyed by UAT have changed over time. Injecting Drug Users (IDUs) have been surveyed bi-annually since 1994 in different cities in the Netherlands, and are recruited via methadone care; sex workers and migrant groups have also been subject to UAT surveys.

on


limited value. First, the uptake of voluntary testing is generally high in the Netherlands,

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

meaning the quality of that data are generally good. And second, UAT has only been deemed ethically permissible in the Netherlands with informed consent, making large-scale UAT surveys impracticable, so restricting UAT to specific population groups and/or the research context.

Norway

9


BMJ Open

The unknown, undiagnosed segment of the population is known as the ‘dark figure(s)’ in Norwegian, although neither the ‘dark figure’ nor the number of cases of HIV has been perceived as significant issues in Norway. Uptake of voluntary testing for HIV was described as very high in intravenous drug users and antenatal clinic users, meaning that UAT has never been perceived as to be of value epidemiologically. The system of surveillance for HIV in Norway instead covers all diagnosed cases (i.e. not undiagnosed cases as with UAT), and was reported as being simple, inexpensive and capable of providing a representative and timely

rp Fo

overview that guides prevention, while respecting confidentiality and adhering to ethical principles.


ee

UK

The epidemiological data yielded from unlinked anonymous testing are considered by

rr

respondents to be of unrivalled quality internationally. It was suggested, however, that the apparent high number of undiagnosed HIV infections from UAT surveys might be inflated by

ev

double-counting:

"You can quite understand that if you've got gonorrhoea and you've been told by your

ie

doctor not to have unsafe sex, you might well decide to go to a different clinic to have your gonorrhoea treated." (UK1)

w

Others defended the statistics and argued that they were explained by a resurgence of high

on

risk behaviour, particularly among gay men. UK1 argued that people who refuse a named test and who are subsequently found to be HIV positive through UAT are likely to be people who know that they are positive.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 53

USA There was little comment from USA interviewees about the quality of UAT data, since UAT surveys stopped in 1996. However, a theme that emerged in relation to UAT data was around the difference between research and surveillance.

10


Page 37 of 53


One point of view was that surveillance (because it is named and part of routine care) is part

rp Fo

of clinical care rather than purely research (USA 2). However, while some of the surveillance undertakings at state level in the USUSA were considered to be practice, at the level of CDC they were treated as research (USA 1). The same respondent commented that cancer sufferers advocated in favour of surveillance for the purposes of research on the disease, since “they were willing to trade privacy and consent for a good which was

ee

surveillance that they saw as serving their interests”.


rr


ev

Two respondents from the Netherlands judged that large-scale UAT programmes were not

ie

necessary because named HIV testing is widespread. The quality of information obtained

w

from named testing was felt to be robust and could be used to model estimates of HIV prevalence in the population. Coverage of pregnant women is very high and this group is considered a valid proxy for the heterosexual population. Named HIV testing among users of

on

STI clinics was also presented as being a valid method for estimating prevalence. It was

argued that, as long as the uptake of named testing by pregnant women and STI clinic users remains high (above 90%) then a national programme of UAT was not thought to be necessary. However, notwithstanding the above point it was acknowledged that current

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

methods of surveillance do not allow undiagnosed HIV infection to be measured accurately.

Ethical issues

UK 11


BMJ Open

Respondents felt that, on balance, no harm was caused by the programme. A recurrent argument for this was that, in settings where anonymous testing takes place, people are offered a named HIV test:


rp Fo


ee


rr


ev

"Is it yours or is it somebody else's? ... If it's waste that's going to be chucked away

ie

then why shouldn't people use it if they find it useful? On the other hand, 'I gave

w


on

There was universal agreement that changing to a system of informed consent would damage the programme. There was a fear that introducing information about UAT into a clinical

consultation could confuse patients. Patients might think they were getting a named test:

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 53



12


Page 39 of 53



rp Fo


ee


rr

"It's a reasonable thing in practical terms provided people are informed. So … if people don't know about their option for opting out and their option for finding out then the system hasn't been working." (UK11)

ev

Some suggested that the approach needed to be more "active" with information displayed

ie

visibly on clinic walls or in leaflets given out to each patient. One clinician, however,

w

expressed his discomfort with his role in the programme. He described filling out the form for UAT blood testing in front of the patient, without explaining what he was doing, and sometimes after the patient had declined a named HIV test:

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open


USA Consent was not deemed necessary initially in the USA for participation in UAT as patients had consented to the blood being collected in the first place, and this was seen as adequate. USA1 suggested that surveillance is a public good, meaning there are strong reasons for 13


BMJ Open

making it a universal obligation on citizens – and that notification that surveillance is occurring is different from asking for consent.

The blood spot survey continued until the publication of the first trials showing that vertical transmission of HIV could be dramatically reduced by treating mothers. When it became clear that infected women could be offered treatment, and therefore should know their HIV status, “it became impossible for the US Centre for Communicable Disease Control and

rp Fo

Prevention (CDC) to continue that programme and it was dropped in the early 1990s along with all the other ones” (USA3). It had become difficult to reconcile public health benefits of UAT with individual rights. If CDC had not voluntarily withdrawn funding from UAT, the same respondent indicated that it was likely that Congress would have mandated cessation, as there was a strong coalition in favour of this.

ee

A related legal issue was brought by USA4, in that public health surveillance appears to go against the “fundamental legal principles” in the USA, and constitutional scholars are shocked when they hear what is done.

ev

rr

‘Once you begin to argue that the government can take your personally identifiable information for any reason it finds useful without your consent you've opened up and blown a hole in much of the fundamental legal principles in the country.’ (USA4)

w

ie

The Tuskegee episode (a study between 1932 and 1972 in Alabama in which black syphilitic Americans were not given treatment although thought they were) was mentioned by USA1 as a reason for USA citizens, particularly Afro-Americans, to be suspicious of surveillance

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 53


“It [UAT] became more of a civil rights issue than a gay specific one. It was, the government knows something that you don’t and the attitude in the States, at least at the time, towards the government was much less trusting than in this country [i.e. UK]. People did not assume and still don’t assume that the government necessarily is looking out for your well being. And I think there’s a much more paranoid attitude towards the government than there is in this country.” (USA3) 14


Page 41 of 53

Related to these cultural dimensions, a particular ethical concern was that the US (and, in particular, CDC) supports UAT in developing countries, with the inherent contradiction that it has been rejected in the USUSA.

‘It seems to me the CDC can’t at the one time say it’s unethical to do it in the US and say it’s ethical to do it abroad. Different public agencies may come to different

rp Fo

decisions, but the same agency it seems to me should be bound by some rule of consistency.’ (USA1)

USA2 felt very strongly that arrangements for UAT in developing countries are not ethical where pregnant women do not have the opportunity to have a named test, counselling and

ee

treatment.

Netherlands

rr

The respondents felt that more extensive UAT surveys would produce more accurate information for public health surveillance. However, inter-linked practical and ethical doubts

ev

about UAT were presented. One concern was that UAT might deter high risk groups from seeking testing and treatment. Another concern was that, if UAT focused only on high risk groups, such groups would risk being stigmatised. It was suggested that reporting of findings

ie


w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Norway


15


BMJ Open

There is very little concern or awareness in Norway about ethical issues around UAT. However, there is a general willingness to provide information for public health purposes in Norway, and people are used to being surveyed. It was reported that many Norwegians have given blood to the Biobank, and give broad consent for their tissue/blood to be used for a wide range of research. A respondent indicated that many are willing to waive their consent in Norway because ‘there is significant trust’ in the health care system. However, another respondent expressed that there was a possible danger that UAT blood could be used to

rp Fo

identify someone via the Home Office DNA database, and that researchers might be forced to share information under a court order (the respondent mentioned that the murderer of the Swedish politician Anna Lindh in 2003 was convicted on the basis of a match with DNA held in the national biobank).

ee

One respondent suggested that informed consent could be different in different contexts. In a country like Norway, with low prevalence and easy access to health care, the current surveillance system (non-UAT) was described as acceptable and functioning well. However,

rr


w

ie

DISCUSSION

ev


epidemiological, ethical and policy debate.15 16 17 Different approaches to sero-surveillance

on

were initially adopted in different countries in response to the developing HIV/AIDS

epidemic.4 Over time some of those approaches have changed. The research findings presented in this paper are part of a larger project exploring the reasons behind such

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 53

differences. The first findings, a quantitative study of the views towards UAT of users of users of two genito-urinary clinics in England, have recently been published.14

In this paper we have presented the results of qualitative interviews with key informants in four countries pragmatically selected because they represent a range of approaches to UATof differences in approach to testing. It is important to note the limitations of the research.18 Only four settings (countries) were selected, but these were chosen deliberately as they

16


Page 43 of 53

represent a range of approaches to UAT. Although more data would have been generated by adding further countries, this was not practical for this relatively small study. A total of 23 respondents were interviewed, a reasonable number for a study of this kind, but they were differentially distributed between countries: 12 in the UK, four in the USA, four in Norway and three in the Netherlands. This distribution (as well as the make-up of respondents) was chosen purposefully to gain useful insights in a qualitative piece of research of this nature, and was not intended to provide a quantitatively representative sample.19 Future study could

rp Fo

be undertaken with expanded numbers.

First a summary of key findings willis be presented (in the context of historical development of UAT in different countries, as well as what is already know of the subject) , followed by a discussion of how the findings from this paper fit into broader contemporary debates about public health ethics and policy.


ev

rr

ee

In the United Kingdom (UK), a balance of epidemiological and ethical considerations in the late 1980s led to the implementation of the ‘prior notification’ approach to UAT; this involves leaflets and posters informing clients that UAT is occurring, allowing opting out but

ie

not necessitating the gaining of informed consent.777 101010 Respondents stressed that the


w

rationale remains apt today, and that this approach provides data unrivalled in terms of


epidemiological quality,999 20 21 and which is of use for health service planning.22 Ethical



on

voluntary (named) testing in settings where UAT is occurring. However, there was mention


of concern about the lack of full explanation, whether people do actually know UAT is

happening, and ‘ownership’ of blood.23 It was indicated that an approach involving informed consent would be cumbersome, time-consuming and epidemiologically damaging.24

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

In the United States (USA), UAT of blood for HIV was extensively used in the 1980s and early 1990s,25 but was stopped in the late 1990s as evidence emerged that anti-viral therapy in pregnant women could help reduce the vertical transmission of HIV from pregnant mother to child – named testing needed to be maximised so that HIV-positive pregnant women could be identified and offered advice and therapy.26 The interviews indicated that data from UAT 17


BMJ Open

were useful epidemiologically but, since UAT has stopped, there has been sufficient other data to plan services accordingly. Other issues raised include a general suspicion around epidemiological surveillance post-Tuskegee, cultural dimensions of UAT with respect to minority groups, a lack of trust in the Government, and the related disconnect between surveillance (done to all) and health care provision (variable, and not done to all).666 111111


There was also comment about the apparent contradiction in the contemporary support of the


US Centre for Communicable Disease Control and Prevention (CDC) for UAT in lowincome countries.27

rp Fo




acknowledgement that large-scale UAT would provide additional epidemiological data, but


ee

this has been balanced against concerns of this leading to diminished voluntary testing – especially in at risk groups – through loss of trust in the healthcare system.28 Smaller UAT surveys have been intermittently undertaken in such groups (e.g. intravenous drug users, sex

rr

workers, migrants) as research studies, but have required individual consent (initially oral, now written).

ev


ie


Surveillance has thus tended to focus on the monitoring of the diagnosed population

w

(anonymised diagnosed cases), which has been supported by the gay community; the undiagnosed population is interestingly referred to as ‘dark figure(s)’. In Norway it is not legal to test without consent unless there is an epidemic, although respondents reported few

on

ethical concerns since people generally have trust in the health care system, and are willing to provide information for public health purposes (many give blood to the Biobank for a range of research purposes).29

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 44 of 53

The findings of this research on the ethics of UAT of blood need consideration in the context of two broader contemporary debates within medical and public health ethics. The first debate centres on the boundaries of informed consent in health research, particularly in relation to the use of human tissue. The second debate focuses on the balance between individuals’ rights and the appropriate role of government3 in infringing on those rights for

18


Page 45 of 53

broader public health interests or concerns. These debates are inter-connected, although will be presented in turn.



rp Fo

programme of unlinked anonymous blood testing, since informed consent is not usually deemed necessary for an individual’s blood to be included; this has tended to be the basis for opposition to UAT.555 30 The findings from our research show, however, that such positions



ee

Debates in the late 1980s and early 1990s demonstrated the opposing views of those who were concerned about patient consent,888 252525 and those who felt that the priority was to control what was seen as an epidemic.

151515


UAT programmes did go ahead in the USA and

rr


UK, and other high income countries: there was no requirement for consent in the USA,


whereas in the UK an approach of ‘prior notification’ was adopted through leaflets and


ev


posters in clinics – consent is presumed but individuals have the opportunity to opt out.

Over the past two decades informed consent has emerged as ‘the ethical touchstone of

ie

medical research, and is now enshrined in practice and a range of international guidelines as one of the main preconditions of medical research’.242424 31 Some commentators argue that

w


even anonymised tissue should not be used without consent. Trouet, for example, believes that the use of such material is in contravention of Article 22 of the Convention on Human

on

Rights and Biomedicine. This states that ‘when in the course of an intervention any part of a human body is removed, it may be stored and used for a purpose other than that for which it

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

was removed, only if this is done in conformity with appropriate information and consent

procedures’, and in accordance with Article 8 of the European Convention on Human Rights which protects the right to respect for private life.32

Our research indicates that, rather than the universality of human rights discourse, a form of weak moral relativism operates in policy-making and practice around UAT.33 In the USA, the change in social context and debates about individual rights, along with evidence of reduction in mother-to-child vertical transmission, resulted in the abandonment of UAT in 1996.121212


19


BMJ Open

The fact that the USA continued to support UAT in international research overseas illustrates further an attitude of moral relativism, although it could be argued that differential health needs justified different approaches.

Interviews in the UK showed that the prior notification approach that remains in use today still holds strong support – based on perception that there is minimal harm to the individual, and the value of high quality, unbiased epidemiological data on HIV (held to be the most

rp Fo

robust and accurate worldwide) (Buehler et al., 1994).34 Respondents felt that any change to UAT requiring informed consent would be cumbersome, time-consuming, and costly. According to the Nuffield Council on Bioethics it may also be ethically unnecessary: ‘... it is acceptable to collect and use anonymised data for assessing and predicting trends in infectious disease without consent, as long as any invasion of privacy is reduced as far as

ee

possible.’.35

This position is not, however, cross-culturally held. In the Netherlands, UAT has not been felt

rr

to be acceptable without informed consent, initially verbal and more recently written. The societal emphasis on individual autonomy, including in the health research environment, has

ev

been the paramount consideration.555 Northern European and Scandinavian countries are


often held up as having a strong community ethos, with an expectation on individual involvement in endeavours for the public good.36 This was not reflected in UAT practice in

ie

the Netherlands, although was indicated to be important in Norway, albeit in a setting where UAT has never been taken up.

w

These findings indicate the importance of social and political context in relation to medico-

on

ethical debates, and how attitudes and policies can change in time within any given

jurisdiction the context can change with time and place.37 Differences exist between the

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 46 of 53

moral evaluations and resultant UAT approaches between countries such as England and the Netherlands. Within the USA, in contrast, the practice of UAT has been re-evaluated with

time. In the late 1980s and early 1990s the feared AIDS epidemic was seen as a public health emergency, there was no treatment available, transmission from mother to unborn child was not preventable, and the stigma and discrimination associated with the infection were more overt than today. As these factors shifted, so did the policy on UAT in the USA. The situation in low income countries today has created a perception of the acceptability of UAT akin to that in western countries two decades ago. 20


Page 47 of 53

Individual rights good and public community (public) good

There has been criticism in recent times of government for its ‘nanny state’ approach to interventions in, for example, the areas of smoking, alcohol and diet. More recently the notion of ‘stewardship’ in public health rejects these negative associations,38 and advocates

rp Fo

that government should have a role to play in protecting the health of its citizens, preventing them from harming themselves or others, and promoting healthy behaviour.[4] In this view, public health interventions should not be thought of as unwanted restrictions imposed by a ‘nanny’ on powerless children, but as actions that favour both individuals and society more widely as prescribed by a ‘steward’ administering and protecting citizens’ interests.[5] The

ee

Nuffield Council on Bioethics, for example, argues that the state should act as steward both to individuals and to the population as a whole, but that individuals’ personal choices should also be taken into account.353535


rr


The subject of UAT sits squarely within the debate about public health stewardship for social

ev

good. Recently, commentators have argued that patients of a publicly supported health care system like the NHS have a moral duty to participate in scientific research,39 40 although most do not go so far as to articulate that such participation should be mandatory in practice.41

ie

Nevertheless, UAT could be understood as something that individuals should feel they have a

w

moral obligation to participate in, regardless of ethical concerns about individual rights and personally held views. Our own survey of over 400 users of two genitor-urinary medicine clinics (a parallel part of this research project) has found that a large majority of respondents

on

(89%) agree that people should have a responsibility to take part in research studies, and the

same proportion would agree to the use of their blood in unlinked anonymous seroprevalence

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

testing. However, when asked about the question of consent, 74% indicated that they should be asked to consent before their blood is used for public health surveillance.141414


Notwithstanding this important latter point, such support for use of blood samples for prevalence surveys or for future research has not, however, been demonstrated in an earlier study in the UK42 or more recently in Egypt.43

The way forwards perhaps is to ensure that patients – and the public more widely – are informed of the existence, purposes and value of disease surveillance arrangements, as well 21


BMJ Open

as the importance of contributing to future health research, while investigating arrangements for gaining consent which are not unnecessarily cumbersome or likely to create bias amongst those who agree that their blood be used.44

The issue of consent in seroprevalence surveillance, such as UAT, remains an important instrument to gauge the moral direction of public health policies more broadly.45 Debate continues around how public health programmes can embrace a Kantian stewardship

rp Fo

relationship, rather than a nanny paternalistic one, while respecting autonomy and advocating social responsibilities. Further examination of why different countries have adopted different positions on seroprevalence surveillance will help illuminate our understanding of how social morality is embedded in public health policies and practice.

Acknowledgments

w

ie

ev

rr

ee The authors would like to thank the participants in the study. We also acknowledge the

on

valuable advice given by Vivian Hope, Will Nutland, Cicely Marston, and Graham Hart.

Conflicts of interest

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 48 of 53

AK is also Director of Public Health Strategy and Medical Director of the Health Protection Agency. The views expressed in this paper are those of the authors, and are not intended to represent the views of the Health Protection Agency.

Funding This research was supported by the Economic and Social Research Council (ESRC Grant Reference: RES-000-22-2096).

22


Page 49 of 53

Ethical approval The study was approved by the South West Research Ethics Committee on August 16th 2007 (reference 07/H0206/46), and by the London School of Hygiene & Tropical Medicine ethics committee on 3rd September 2007 (reference 5174).

rp Fo

Notes [1]


Agency (HPA) and the Institute of Child Health, University of London. The programme was previously managed by the Communicable Disease Surveillance Centre, part of the Public Health Laboratory Service, which was subsumed into the Health Protection Agency in 2005.

ee

[2]


[3]


rr



government. [5]

ev

It is interesting that in this discourse ‘nanny’ (female) is seen as negative and

‘steward’ (male) as positive, when both are servants employed to manage and protect.

w

ie

References

1

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open



Report of the Unlinked Anonymous Prevalence Monitoring Programme. London: Department of Health, 1998.

23


BMJ Open

3




anonymous HIV seroprevalence studies: an international comparison. AIDS 1990;4(4):283290. 5

rp Fo




2004;303:631-632. 7

ee


Programme of unlinked anonymous testing in England and Wales. British Medical Journal

rr

1989:299:1295-7. 8




ie

9

ev


w


823. 11

on


Surveillance in America. Berkeley: University of California Press, 2007.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 50 of 53

12


13


evaluation of the HIV/AIDS surveillance system in Norway 1983-2000. European Journal of Epidemiology 2001;17:479-89.

24


Page 51 of 53

14




1987;294:443-444. 16

rp Fo





ee

18


19

Green J, Thorogood N. Qualitative Methods for Health Research. London: Sage,

2005.

ev

20

rr


HIV-prevalence surveys in England and Wales. Epidemiology and Infection 1999;122:26772.


w

21

ie

among childbearing women in New York State from 1988 through 2000: a valuable epidemiologic tool. Archives of Pediatric and Adolescent Medicine 2004;158(5):443-8. 22

on



ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open




participation in health services research. Journal of Medical Ethics 2002;23:313-317.

25


BMJ Open

25


Health 1993;83/4:496-7. 26




rp Fo




ee


Informed consent and Biobank research in a Norwegian context. Critical Public Health 2005;15(4):335-347.




British Medical Journal 2006;333:255-8.

w

32

ie

31

ev

30

rr


in biomedical research. Journal of Medical Ethics 2004;30:99-103.

on

33


34



ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 52 of 53




approach. Journal of Medical Ethics 2005;31:476-80.

26


Page 53 of 53

37






rp Fo

2005;31(4):242-248. 40


41



ee



rr

90-91. 43


ev


ie



w



ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

27


BMJ Open


rp Fo Journal:

Manuscript ID: Article Type:



BMJ Open bmjopen-2012-001427.R2 Research 08-Nov-2012


Keywords:

Public health


ev


rr

ee

Kessel, Anthony; London School of Hygiene and Tropical Medicine, Social and environmental health research Datta, Jessica; London School of Hygiene and Tropical Medicine, Wellings, Kaye; London School of Hygiene and Tropical Medicine, Perman, Sarah; Health Protection Agency, Public Health Strategy


w

ie ly


Page 1 of 53

Original paper

The ethics of unlinked anonymous testing of blood: views from in-depth interviews with key informants in four countries Unlinked anonymous testing of blood (running title) Anthony S Kessel (corresponding author and guarantor of study), Honorary Professor, London

rp Fo



ee


rr


Key words

w

ie


Word count

ly

6608 words

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open


1

BMJ Open

Abstract In this study we explore the ethical issues around unlinked anonymous testing (UAT) of blood, a method of sero-prevalence surveillance for infectious diseases. Our study focused on UAT for HIV, although UAT can be used for other infectious diseases. Objectives The objectives of the research were to gain a better understanding of the views of key informants in countries adopting different UAT testing strategies, and to use the findings of the research to inform

rp Fo

health policy Design


ee

Setting

Four countries using different strategies around unlinked anonymous testing of blood for HIV (United

rr

Kingdom, United States, Netherlands and Norway). Participants

ev


ie

Results


w

policies on UAT adopted by different countries have been historically and culturally determined. We use our findings to explore the relationship between public health policy and ethics, framing our

on

discussion in relation to two important contemporary debates: informed consent for participation in medical and public health research; and the balance between individual good and the public good. Conclusions

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 53



2

Page 3 of 53



rp Fo


BACKGROUND

ee


rr


ev


w

ie


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open




3

BMJ Open


History of UAT in different countries A national sero-prevalence surveillance programme using UAT to estimate the prevalence of HIV has been in place in the UK since 1990.1 The programme was first developed in

rp Fo


ee


rr


ev


ie


w

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 53

UAT surveys focused on population groups, for instance a survey of young people volunteering for job corps, although these were regional or local. In contrast to the UK the prior notification system, with the opportunity to opt out, was not used. UAT essentially stopped in the USA in 1996, as evidence emerged in relation to reducing vertical transmission of the HIV virus from a pregnant mother to the foetus through use of anti-viral medicine.11 The interplay at the time of clinical, social and political factors was complex.12 The main method for HIV surveillance in the Netherlands is named voluntary testing rather than UAT (these are ‘opt-out’ programmes, such as the antenatal testing programme and the


4

Page 5 of 53


rp Fo

accessing health services less than others. In Norway, public health efforts have focused on diagnosing all those infected, as it was thought (certainly in earlier stages) that those who were diagnosed would be less likely than those undiagnosed to transmit infection to others. Mandatory named AIDS reporting was

ee


rr


ie


w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open




5

BMJ Open


METHODS Four countries were selected purposively to participate in this study component (the key

rp Fo


ee

involvement in UAT and with appropriate knowledge or expertise, historically and/or current; existing links with the research team or advisory group members; and practical issues (e.g.

rr


ev

A topic guide was developed that allowed for a semi-structured interview. Elements within

ie

the topic guide included: introduction and background to the purpose and nature of the interview (prior information had been provided); knowledge and views on unlinked

w


on

confidentiality, anonymity, individual rights/liberties, social responsibilities, and any changes in ethical position over time; the value to public health of different surveillance methods;

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 53

blood as a tissue; legal considerations; and any other relevant issues the respondent wished to discuss. The topic guide was initially tailored for the different countries based on our own understanding, but then applied flexibly during the interview.

A researcher, with a background and skills in qualitative interviewing, was employed for the duration of the project, and undertook all the interviews. Interviews were carried out, in English, either face-to-face or by telephone between January and September 2008. All nonUK informants spoke good English, without evidence to the research team of any obvious language barrier issues. Interviews were recorded (except for two, when notes were taken: the


6

Page 7 of 53

first due to unsuitable location and the second to mechanical failure) with the participant’s permission, and transcribed verbatim. Interviews lasted between 20 minutes and one hour, the length directed by the contributions of the informants. In total 21 interviews were completed with 23 respondents (two interviews were with two respondents together). Of these, 12 were with UK respondents (three clinicians in genito-urinary medicine/HIV; one person working in government health policy; two HIV epidemiologists; two ethics academics; two charity representatives [MedFash, Terence Higgins Trust]; one person working at the British Medical Association; one editor of a medical journal), four with respondents in the USA (one health

rp Fo

policy academic; one academic in health law; one academic in public health; and one epidemiologist), three with respondents from the Netherlands (two epidemiologists; and one academic in medical ethics), and four with respondents in Norway (one epidemiologist; three academics in medical ethics – one of whom is also a hospital clinician). There were three interview refusals, all in the UK. [3]

ee


rr


ev

the two researchers that required further discussion and consensus. Content was analysed for description of UAT developments in the country concerned, the context in which UAT may

ie

have changed, ethical tensions in UAT between surveillance and human rights, and broader connections between UAT and public health policy. Inevitably some of the themes inter-link

w

with each other (e.g. ‘value of UAT’ and ‘ethical issues’). Coding frames and analyses were shared with the wider research team. In a relatively small study of this kind it is not possible

on

to indicate definitively if data saturation was reached, and the focus was on the emergent themes.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Interview data pertaining to the history and factual state of sero-surveillance in different countries were triangulated, for the purpose of the discussion section, with literature sources. It was not appropriate to triangulate informants’ views on sero-surveillance, as such data were treated as informants’ opinions and perceptions.

RESULTS


7

BMJ Open


Value of UAT UK The programme is largely seen by respondents as indispensable in providing an unbiased

rp Fo

measure of HIV prevalence and risk factors in the UK population. The most important outcome from the surveys has been information on undiagnosed HIV infection which, respondents suggested, has led to changes in policy on named testing; for example, HIV testing in pregnant women has changed from opt in to opt out, and sexual health clinics have moved from targeted to universal voluntary testing. The neonatal dried blood spot survey is

ee

used as a check on the performance of local antenatal screening. The public health rationale for the programme has not changed since 1990, and respondents cited the continuing high

rr


ev

"There was far more feeling of a public health emergency (then) than there is today, although you could argue … that it's actually much worse today because of where we see the disease." (UK4)

w

ie

Several respondents suggested that a review of UAT would be timely, including consideration of extending the surveys to other health care settings and other diseases. UK6

on

discussed how advances in HIV testing, including self-testing, might alter the questions asked by the programme. Some suggested that if named testing was normalised, there would be less need for anonymous testing.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 53

A particular theme was whether the British public has a duty to participate in public health monitoring, possibly as part-and-parcel of receiving the benefits of a national health service (NHS): "There is a very good case for really exploring with the public, and then pushing very hard to get public agreement, that in general terms that sort of tissue, extra bits of tissue or blood, should be available for use." (UK4)


8

Page 9 of 53

USA According to one respondent, the surveillance programme was valuable in terms of planning, resource allocation, advocacy and interventions to prevent transmission. The heel prick survey provided accurate national data on prevalence in childbearing women and was seen as extremely valuable, with nothing since that has given as accurate a picture of national prevalence (USA 2).

rp Fo

However, given the questionable ethics of UAT (see later), as well as the resultant lack of good quality prevalence data, USA3 argued that enough became known about prevalence to target the most vulnerable groups for health promotion. “And how much more detail do you want from that, so what would you do differently

ee

if you knew that in the next level of detail.” (USA3) Netherlands

rr


ev

among high risk populations, but with informed consent (verbal or written). The populations surveyed by UAT have changed over time. Injecting Drug Users (IDUs) have been surveyed

ie

bi-annually since 1994 in different cities in the Netherlands, and are recruited via methadone care; sex workers and migrant groups have also been subject to UAT surveys.

w


on

limited value. First, the uptake of voluntary testing is generally high in the Netherlands, meaning the quality of that data are generally good. And second, UAT has only been deemed

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

ethically permissible in the Netherlands with informed consent, making large-scale UAT surveys impracticable, so restricting UAT to specific population groups and/or the research context.

Norway The unknown, undiagnosed segment of the population is known as the ‘dark figure(s)’ in Norwegian, although neither the ‘dark figure’ nor the number of cases of HIV has been perceived as significant issues in Norway. Uptake of voluntary testing for HIV was described as very high in intravenous drug users and antenatal clinic users, meaning that UAT has never


9

BMJ Open

been perceived as to be of value epidemiologically. The system of surveillance for HIV in Norway instead covers all diagnosed cases (i.e. not undiagnosed cases as with UAT), and was reported as being simple, inexpensive and capable of providing a representative and timely overview that guides prevention, while respecting confidentiality and adhering to ethical principles.

Quality and nature of data UK

rp Fo

The epidemiological data yielded from unlinked anonymous testing are considered by respondents to be of unrivalled quality internationally. It was suggested, however, that the apparent high number of undiagnosed HIV infections from UAT surveys might be inflated by double-counting:

rr

ee

"You can quite understand that if you've got gonorrhoea and you've been told by your doctor not to have unsafe sex, you might well decide to go to a different clinic to have your gonorrhoea treated." (UK1)

ie

ev

Others defended the statistics and argued that they were explained by a resurgence of high risk behaviour, particularly among gay men. UK1 argued that people who refuse a named test

w

and who are subsequently found to be HIV positive through UAT are likely to be people who know that they are positive. USA

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 53




10

Page 11 of 53

One point of view was that surveillance (because it is named and part of routine care) is part of clinical care rather than purely research (USA 2). However, while some of the surveillance undertakings at state level in the USA were considered to be practice, at the level of CDC they were treated as research (USA 1). The same respondent commented that cancer sufferers advocated in favour of surveillance for the purposes of research on the disease, since “they were willing to trade privacy and consent for a good which was surveillance that they saw as serving their interests”.

rp Fo



ee


rr

necessary because named HIV testing is widespread. The quality of information obtained from named testing was felt to be robust and could be used to model estimates of HIV

ev

prevalence in the population. Coverage of pregnant women is very high and this group is considered a valid proxy for the heterosexual population. Named HIV testing among users of

ie

STI clinics was also presented as being a valid method for estimating prevalence. It was argued that, as long as the uptake of named testing by pregnant women and STI clinic users

w

remains high (above 90%) then a national programme of UAT was not thought to be necessary. However, notwithstanding the above point it was acknowledged that current

on

methods of surveillance do not allow undiagnosed HIV infection to be measured accurately.

Ethical issues

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open



11

BMJ Open

"I think the ethics are not in fact, are not a problem because everybody is now very strongly advised to have a test and therefore the residual rump who don't have a test, it's important to know what their rate of HIV positivity is." (UK1) Some respondents, however, said there would be ethical concerns if people in these settings were not routinely offered a named test. Several respondents considered the ethics of UAT to be relatively unimportant, for instance in the context of more significant health-related problems in Africa:

rp Fo

"There are these enormous ethical, real ethical issues, and to me, this one … I just can't understand why one should put so much emphasis on it." (UK9) UK11 suggested that the question of ownership of the sample was relevant, but that this could be argued both ways:

rr

ee


ev


ie


w


on

"You would lose someone's confidence and trust because they might not understand

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 53

the words or the reason behind why we're doing it, and they would think that you could get back to them and you did know." (UK12)

Other respondents argued that it would reduce time for discussing important clinical matters: "When the midwife turns up, does she pull out a leaflet and say by the way, I'm putting this beside you here? … it would be a perversion of the consultation to divert it into discussing this ..." (UK5)


12

Page 13 of 53



rp Fo

All the respondents thought that prior notification was the right theoretical approach. However, some were concerned that, in practice, it might not be working: "It's a reasonable thing in practical terms provided people are informed. So … if people don't know about their option for opting out and their option for finding out

ee

then the system hasn't been working." (UK11)

rr


ev

expressed his discomfort with his role in the programme. He described filling out the form for UAT blood testing in front of the patient, without explaining what he was doing, and

ie

sometimes after the patient had declined a named HIV test:

w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

USA



13

BMJ Open

The blood spot survey continued until the publication of the first trials showing that vertical transmission of HIV could be dramatically reduced by treating mothers. When it became clear that infected women could be offered treatment, and therefore should know their HIV status, “it became impossible for the US Centre for Communicable Disease Control and Prevention (CDC) to continue that programme and it was dropped in the early 1990s along with all the other ones” (USA3). It had become difficult to reconcile public health benefits of UAT with individual rights. If CDC had not voluntarily withdrawn funding from UAT, the same respondent indicated that it was likely that Congress would have mandated cessation, as

rp Fo

there was a strong coalition in favour of this. A related legal issue was brought by USA4, in that public health surveillance appears to go against “fundamental legal principles” in the USA, and constitutional scholars are shocked when they hear what is done.

ee


rr


ev

The Tuskegee episode (a study between 1932 and 1972 in Alabama in which black syphilitic

ie

Americans were not given treatment although thought they were) was mentioned by USA1 as a reason for USA citizens, particularly Afro-Americans, to be suspicious of surveillance

w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 53

“It [UAT] became more of a civil rights issue than a gay specific one. It was, the government knows something that you don’t and the attitude in the States, at least at the time, towards the government was much less trusting than in this country [i.e. UK]. People did not assume and still don’t assume that the government necessarily is looking out for your well being. And I think there’s a much more paranoid attitude towards the government than there is in this country.” (USA3)


14

Page 15 of 53

Related to these cultural dimensions, a particular ethical concern was that the US (and, in particular, CDC) supports UAT in developing countries, with the inherent contradiction that it has been rejected in the USA. ‘It seems to me the CDC can’t at the one time say it’s unethical to do it in the US and say it’s ethical to do it abroad. Different public agencies may come to different decisions, but the same agency it seems to me should be bound by some rule of consistency.’ (USA1)

rp Fo


Netherlands

ee


rr


ev

seeking testing and treatment. Another concern was that, if UAT focused only on high risk groups, such groups would risk being stigmatised. It was suggested that reporting of findings

ie


w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Norway



15

BMJ Open

There is very little concern or awareness in Norway about ethical issues around UAT. However, there is a general willingness to provide information for public health purposes in Norway, and people are used to being surveyed. It was reported that many Norwegians have given blood to the Biobank, and give broad consent for their tissue/blood to be used for a wide range of research. A respondent indicated that many are willing to waive their consent in Norway because ‘there is significant trust’ in the health care system. However, another respondent expressed that there was a possible danger that UAT blood could be used to identify someone via the Home Office DNA database, and that researchers might be forced to

rp Fo


One respondent suggested that informed consent could be different in different contexts. In a

ee

country like Norway, with low prevalence and easy access to health care, the current surveillance system (non-UAT) was described as acceptable and functioning well. However,

rr


w

ie

DISCUSSION

ev


on

epidemiological, ethical and policy debate.15 16 17 In this paper we have presented the results of qualitative interviews with key informants in four countries pragmatically selected because they represent a range of approaches to UAT. 18 A total of 23 respondents were interviewed,

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 53

differentially distributed between countries: 12 in the UK, four in the USA, four in Norway and three in the Netherlands. There were limitations to the research: more key informants were interviewed in some countries than others; some interviews were face-to-face whereas others were by telephone; and two of the interviews relied on notes rather than recordings for later analysis.18 19 20 It is possible that such aspects introduced bias, but we believe that the balanced pragmatic approach taken to address the research questions was appropriate for this relatively small


16

Page 17 of 53

qualitative study. Interesting themes emerged in the analysis, and we were able to triangulate more fact-based comments with documentary sources in the literature.

First a summary of key findings is presented (in the context of historical development of UAT in different countries, as well as what is already known of the subject), followed by a discussion of how the findings from this paper fit into broader contemporary debates about public health ethics and policy.

rp Fo

Summary of key findings from interviews (in the context of what is currently known) In the United Kingdom (UK), a balance of epidemiological and ethical considerations in the late 1980s led to the implementation of the ‘prior notification’ approach to UAT; this

ee

involves leaflets and posters informing clients that UAT is occurring, allowing opting out but not necessitating the gaining of informed consent.7 10 Respondents stressed that the rationale

rr

remains apt today, and that this approach provides data unrivalled in terms of epidemiological quality,9 21 22 and which is of use for health service planning.23 Ethical concerns were

ev

generally felt to be minimal today since clients are (also) usually offered voluntary (named) testing in settings where UAT is occurring. However, there was mention of concern about the

ie

lack of full explanation, whether people do actually know UAT is happening, and ‘ownership’ of blood.24 It was indicated that an approach involving informed consent would

w

be cumbersome, time-consuming and epidemiologically damaging.25

on

In the United States (USA), UAT of blood for HIV was extensively used in the 1980s and early 1990s,26 but was stopped in the late 1990s as evidence emerged that anti-viral therapy in

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

pregnant women could help reduce the vertical transmission of HIV from pregnant mother to child – named testing needed to be maximised so that HIV-positive pregnant women could be identified and offered advice and therapy.27 The interviews indicated that data from UAT were useful epidemiologically but, since UAT has stopped, there has been sufficient other data to plan services accordingly. Other issues raised include a general suspicion around epidemiological surveillance post-Tuskegee, cultural dimensions of UAT with respect to minority groups, a lack of trust in the Government, and the related disconnect between surveillance (done to all) and health care provision (variable, and not done to all).6 11 There was also comment about the apparent contradiction in the contemporary support of the US


17

BMJ Open

Centre for Communicable Disease Control and Prevention (CDC) for UAT in low-income countries.28

Interviews in the Netherlands showed that named (voluntary) testing, which has generally been widely taken up, has provided the main vehicle for surveillance of HIV.4 There is acknowledgement that large-scale UAT would provide additional epidemiological data, but this has been balanced against concerns of this leading to diminished voluntary testing – especially in at risk groups – through loss of trust in the healthcare system.29 Smaller UAT

rp Fo

surveys have been intermittently undertaken in such groups (e.g. intravenous drug users, sex workers, migrants) as research studies, but have required individual consent (initially oral, now written).

Interviews in Norway revealed that UAT was never felt to be necessary epidemiologically, as

ee

HIV has not been a significant issue and the uptake of voluntary testing has been high.13 Surveillance has thus tended to focus on the monitoring of the diagnosed population

rr

(anonymised diagnosed cases), which has been supported by the gay community; the undiagnosed population is interestingly referred to as ‘dark figure(s)’. In Norway it is not

ev

legal to test without consent unless there is an epidemic, although respondents reported few ethical concerns since people generally have trust in the health care system, and are willing to

ie

provide information for public health purposes (many give blood to the Biobank for a range of research purposes).30

w

The findings of this research on the ethics of UAT of blood need consideration in the context

on

of two broader contemporary debates within medical and public health ethics. The first debate centres on the boundaries of informed consent in health research, particularly in

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 53

relation to the use of human tissue. The second debate focuses on the balance between individuals’ rights and the appropriate role of government3 in infringing on those rights for broader public health interests or concerns. These debates are inter-connected, although will be presented in turn.



18

Page 19 of 53

Respect for autonomy is the bioethical principle that is most likely to be jeopardised by a programme of unlinked anonymous blood testing, since informed consent is not usually deemed necessary for an individual’s blood to be included; this has tended to be the basis for opposition to UAT.5 31 The findings from our research show, however, that such positions differ between countries, and have also changed with time.

Debates in the late 1980s and early 1990s demonstrated the opposing views of those who were concerned about patient consent,8 26 and those who felt that the priority was to control

rp Fo

what was seen as an epidemic.15 UAT programmes did go ahead in the USA and UK, and other high income countries: there was no requirement for consent in the USA, whereas in the UK an approach of ‘prior notification’ was adopted through leaflets and posters in clinics – consent is presumed but individuals have the opportunity to opt out.

ee

Over the past two decades informed consent has emerged as ‘the ethical touchstone of medical research, and is now enshrined in practice and a range of international guidelines as

rr

one of the main preconditions of medical research’.25 32 Some commentators argue that even anonymised tissue should not be used without consent. Trouet, for example, believes that the

ev

use of such material is in contravention of Article 22 of the Convention on Human Rights and Biomedicine. This states that ‘when in the course of an intervention any part of a human body

ie

is removed, it may be stored and used for a purpose other than that for which it was removed, only if this is done in conformity with appropriate information and consent procedures’, and

w

in accordance with Article 8 of the European Convention on Human Rights which protects the right to respect for private life.33

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

weak moral relativism operates in policy-making and practice around UAT.34 In the USA, the change in social context and debates about individual rights, along with evidence of reduction in mother-to-child vertical transmission, resulted in the abandonment of UAT in 1996.12 The fact that the USA continued to support UAT in international research overseas illustrates further an attitude of moral relativism, although it could be argued that differential health needs justified different approaches. Interviews in the UK showed that the prior notification approach that remains in use today still holds strong support – based on perception that there is minimal harm to the individual,


19

BMJ Open

and the value of high quality, unbiased epidemiological data on HIV (held to be the most robust and accurate worldwide).35 Respondents felt that any change to UAT requiring informed consent would be cumbersome, time-consuming, and costly. According to the Nuffield Council on Bioethics it may also be ethically unnecessary: ‘... it is acceptable to collect and use anonymised data for assessing and predicting trends in infectious disease without consent, as long as any invasion of privacy is reduced as far as possible.’.36 This position is not, however, cross-culturally held. In the Netherlands, UAT has not been felt

rp Fo

to be acceptable without informed consent, initially verbal and more recently written. The societal emphasis on individual autonomy, including in the health research environment, has been the paramount consideration.5 Northern European and Scandinavian countries are often held up as having a strong community ethos, with an expectation on individual involvement in endeavours for the public good.37 This was not reflected in UAT practice in the

ee

Netherlands, although was indicated to be important in Norway, albeit in a setting where UAT has never been taken up.

rr


ev

ethical debates, and how attitudes and policies can change in time within any given jurisdiction.38 Differences exist between the moral evaluations and resultant UAT approaches

ie

between countries such as England and the Netherlands. Within the USA, in contrast, the practice of UAT has been re-evaluated with time. In the late 1980s and early 1990s the feared

w

AIDS epidemic was seen as a public health emergency, there was no treatment available, transmission from mother to unborn child was not preventable, and the stigma and

on

discrimination associated with the infection were more overt than today. As these factors shifted, so did the policy on UAT in the USA. The situation in low income countries today

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 53

has created a perception of the acceptability of UAT akin to that in western countries two decades ago.

Individual good and community (public) good There has been criticism in recent times of government for its ‘nanny state’ approach to interventions in, for example, the areas of smoking, alcohol and diet. More recently the notion of ‘stewardship’ in public health rejects these negative associations,39 and advocates


20

Page 21 of 53

that government should have a role to play in protecting the health of its citizens, preventing them from harming themselves or others, and promoting healthy behaviour.[4] In this view, public health interventions should not be thought of as unwanted restrictions imposed by a ‘nanny’ on powerless children, but as actions that favour both individuals and society more widely as prescribed by a ‘steward’ administering and protecting citizens’ interests. The Nuffield Council on Bioethics, for example, argues that the state should act as steward both to individuals and to the population as a whole, but that individuals’ personal choices should also be taken into account.36

rp Fo

The subject of UAT sits squarely within the debate about public health stewardship for social good. Recently, commentators have argued that patients of a publicly supported health care system like the NHS have a moral duty to participate in scientific research,40 41 although most do not go so far as to articulate that such participation should be mandatory in practice.42

ee

Nevertheless, UAT could be understood as something that individuals should feel they have a moral obligation to participate in, regardless of ethical concerns about individual rights and

rr

personally held views. Our own survey of over 400 users of two genitor-urinary medicine clinics (a parallel part of this research project) has found that a large majority of respondents

ev

(89%) agree that people should have a responsibility to take part in research studies, and the same proportion would agree to the use of their blood in unlinked anonymous seroprevalence

ie

testing. However, when asked about the question of consent, 74% indicated that they should be asked to consent before their blood is used for public health surveillance.14

w


ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

The way forwards perhaps is to ensure that patients – and the public more widely – are informed of the existence, purposes and value of disease surveillance arrangements, as well as the importance of contributing to future health research, while investigating arrangements for gaining consent which are not unnecessarily cumbersome or likely to create bias amongst those who agree that their blood be used.45 The issue of consent in seroprevalence surveillance, such as UAT, remains an important instrument to gauge the moral direction of public health policies more broadly.46 Debate continues around how public health programmes can embrace a stewardship relationship,


21

BMJ Open


Acknowledgments The authors would like to thank the participants in the study. We also acknowledge the

rp Fo

valuable advice given by Vivian Hope, Will Nutland, Cicely Marston, and Graham Hart.


AK is also Director of Public Health Strategy and Medical Director of the Health Protection Agency. The views expressed in this paper are those of the authors, and are not intended to

ee

represent the views of the Health Protection Agency.

Funding

rr


ie

Ethical approval

ev


w


on

Notes [1] The England and Wales programme is currently managed by the Health Protection

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 53

Agency (HPA) and the Institute of Child Health, University of London. The programme was previously managed by the Communicable Disease Surveillance Centre, part of the Public Health Laboratory Service, which was subsumed into the Health Protection Agency in 2005. [2]


[3]




government.


22

Page 23 of 53

References

1





rp Fo

of Health, 1998. 3



ee



rr

290. 5

ev



ie


2004;303:631-632. 7

w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open






823.


23

BMJ Open

11




13



rp Fo

Datta J, Kessel AS, Wellings K, et al. The views of genitourinary medicine (GUM)

clinic users on unlinked anonymous testing of blood for HIV: evidence from a pilot study of clinics in two English cities. Journal of Medical Ethics 2011 doi:10.1136/jme.2011.042705. 15


1987;294:443-444.

rr

16

ee



ev



w

ie

18


19

Yin RK. Case study research: design and methods. London: Sage, 1994.

20

Green J, Thorogood N. Qualitative methods for health research. London: Sage, 2004.

21


ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 53



among childbearing women in New York State from 1988 through 2000: a valuable epidemiologic tool. Archives of Pediatric and Adolescent Medicine 2004;158(5):443-8.


24

Page 25 of 53

23






rp Fo



Health 1993;83/4:496-7. 27


ee



rr


ev


ie


w


on




ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open







25

BMJ Open

35







rp Fo




ee



2005;31(4):242-248.

rr

41


42


ie

ev


w



on

90-91. 44


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 53







26

Page 27 of 53

Revised manuscript, submitted to BMJ 15 Aug 2012 Marked

Original paper


rp Fo

Anthony S Kessel (corresponding author and guarantor of study), Honorary Professor, London School of Hygiene & Tropical Medicine, Faculty of Public Health and Policy, 15-17 Tavistock Place, London WC1H 9SH (Tel: 020 7759 2715; email: [email protected])


ee



Key words

6608437 words

ly

on

Word count

w

Unlinked, anonymous, blood, ethics, public health, policy, HIV

ie

ev

rr

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

1


BMJ Open


Objectives

rp Fo

The objectives of the research were to gain a better understanding of the views of key informants in countries adopting different UAT testing strategies, and to use the findings of the research to inform health policy

Design


Setting

ee

Four countries using different strategies around unlinked anonymous testing of blood for HIV (United Kingdom, United States, Netherlands and Norway).


ie

Results

ev

Participants

rr


w

policies on UAT adopted by different countries have been historically and culturally determined. We use our findings to explore the relationship between public health policy and ethics, framing our discussion in relation to two important contemporary debates: informed consent for participation in

on

medical and public health research; and the balance between individual good and the public good.

Conclusions

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 53

Qualitative research and ethical analysis of unlinked anonymous testing of blood in different countries has yielded important findings for consideration by policy makers. The policy of unlinked anonymous

testing of blood for HIV and other diseases in the UK needs re-consideration in light of these findings.

2


Page 29 of 53


In this paper we explore unlinked anonymous testing (UAT) of blood for HIV and other diseases, and examine the ethical issues associated with this method of sero-prevalence

rp Fo

surveillance for infectious diseases. We report original results from a study funded by the UK Economic and Social Research Council (ESRC) – qualitative interviews with stakeholders in four countries adopting different UAT policies – and use the subject matter, and our findings, to explore the relationship between public health policy, context and ethics.

BACKGROUND

rr

ee

Unlike voluntary testing for clinical purposes, unlinked anonymous testing (UAT) of blood is carried out in order to provide information about the prevalence of particular infectious diseases in a population. The use of UAT began in the late 1980s for HIV, and remains most

ev


w

ie

The principles behind UAT can be explained with reference to the approach used in the UK for HIV. Leftover blood from samples taken for diagnostic or routine screening purposes is made anonymous, irreversibly unlinked from the source, and then tested for HIV.2 3 This

on

means that individuals whose blood tests positive for HIV cannot be contacted or informed. The data derived from UAT are then used for monitoring purposes and also to inform the planning and evaluation of health services and health promotion activities.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open


3


BMJ Open



A national sero-prevalence surveillance programme using UAT to estimate the prevalence of

rp Fo

HIV has been in place in the UK since 1990.111 The programme was first developed in


7

response to concerns about the spread of HIV during the 1980s and, despite early opposition,


the methodology eventually gained political support.8 The programme was funded initially as a pilot project by the Medical Research Council, and has since been funded by the Department of Health. [1] [2] The programme currently uses data collected from three

ee

ongoing surveys (newborn babies, [maternal antibodies], users of genitourinary medicine (GUM) clinics, and intravenous drug users [saliva]) to measure the proportion of individuals within the general population, and within subgroups of the population, infected with HIV.

rr


ev

A system of ‘prior notification’ operates in locations where testing goes on, involving the display of leaflets or posters explaining that blood may be tested anonymously. Consent to

ie

participation is implicit rather than explicit; provision is made to opt out of testing, although this naturally depends on awareness.999 10

w


In the USA, UAT was extensively used in the late 1980s and early 1990s using discarded

on

blood samples. The largest survey was national and used newborn heel prick tests. Other UAT surveys focused on population groups, for instance a survey of young people

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 53

volunteering for job corps, although these were regional or local. In contrast to the UK the prior notification system, with the opportunity to opt out, was not used. UAT essentially stopped in the USA in 1996, as evidence emerged in relation to reducing vertical

transmission of the HIV virus from a pregnant mother to the foetus through use of anti-viral medicine.11 The interplay at the time of clinical, social and political factors was complex.12

The main method for HIV surveillance in the Netherlands is named voluntary testing rather than UAT (these are ‘opt-out’ programmes, such as the antenatal testing programme and the 4


Page 31 of 53

testing of users of STI clinics, as well as the extensive ‘opt-in’ testing carried out by general practitioners). The UAT that does occur in the Netherlands is carried out using blood samples from intravenous drug users (IDUs) and saliva samples from other groups. Unlike the UK or USA, informed consent is required from those surveyed by UAT, either verbal or, more recently, written consent. From 2002 to 2006, other high risk groups haven been included in UAT research surveys in the Netherlands (e.g. migrants from the Cape Verde islands, Surinam and the Netherlands Antilles, as well as commercial sex workers), because of

rp Fo

concerns of a higher proportion of undiagnosed HIV in such groups, who were also felt to be accessing health services less than others.

In Norway, public health efforts have focused on diagnosing all those infected, as it was thought (certainly in earlier stages) that those who were diagnosed would be less likely than

ee

those undiagnosed to transmit infection to others. Mandatory named AIDS reporting was introduced in Norway in 1983 and mandatory anonymous HIV case reporting in 1986, both including cases diagnosed before those dates.13 The Norwegian approach has therefore

rr


AIMS AND OBJECTIVES OF THE RESEARCH STUDY

w

ie

ev The ESRC-funded project had three components: a questionnaire survey of 471 patients attending two inner-urban genito-urinary medicine (GUM) clinics in two English cities,

on

exploring their views of, and attitudes towards, UAT;14 qualitative in-depth interviews with a sub-sample (20) of these patients; and in-depth interviews with key informants in four

countries chosen to represent different approaches towards sero-surveillance for HIV. In this

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

paper, we present findings from this third component, the qualitative interviews with key informants.


5


BMJ Open


METHODS

rp Fo

Four countries were selected purposively to participate in this study component (the key informant interviews), based on their representing a range of approaches, or positions taken, with regard to UAT: the UK (specifically England), USA, the Netherlands and Norway. Purposive sampling was used to select the key informants from these countries, informed by the following criteria: range of stakeholder perspectives (e.g. epidemiology, clinical, public

ee

health policy, charity or pressure group, and academia (public health or medical ethics)); involvement in UAT and with appropriate knowledge or expertise, historically and/or current; existing links with the research team or advisory group members; and practical issues (e.g.

rr


ev


ie

interview (prior information had been provided); knowledge and views on unlinked

w


confidentiality, anonymity, individual rights/liberties, social responsibilities, and any changes

on

in ethical position over time; the value to public health of different surveillance methods;

blood as a tissue; legal considerations; and any other relevant issues the respondent wished to

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 53


A researcher, with a background and skills in qualitative interviewing, was employed for the duration of the project, and undertook all the interviews. Interviews were carried out, in English, either face-to-face or by telephone between January and September 2008. All nonUK informants spoke good English, without evidence to the research team of any obvious language barrier issues. Interviews were recorded (except for two, when notes were taken: the 6


Page 33 of 53

first due to unsuitable location and the second to mechanical failure) with the participant’s permission, and transcribed verbatim. Interviews lasted between 20 minutes and one hour, the length directed by the contributions of the informants. In total 21 interviews were completed with 23 respondents (two interviews were with two respondents together). Of these, 12 were with UK respondents (three clinicians in genito-urinary medicine/HIV; one person working in government health policy; two HIV epidemiologists; two ethics academics; two charity representatives [MedFash, Terence Higgins Trust]; one person working at the British Medical

rp Fo

Association; one editor of a medical journal), four with respondents in the USA (one health policy academic; one academic in health law; one academic in public health; and one epidemiologist), three with respondents from the Netherlands (two epidemiologists; and one academic in medical ethics), and four with respondents in Norway (one epidemiologist; three academics in medical ethics – one of whom is also a hospital clinician). There were three interview refusals, all in the UK. [3]

ee


rr


ev

the two researchers that required further discussion and consensus. Content was analysed for description of UAT developments in the country concerned, the context in which UAT may have changed, ethical tensions in UAT between surveillance and human rights, and broader

ie

connections between UAT and public health policy. Inevitably some of the themes inter-link

w

with each other (e.g. ‘value of UAT’ and ‘ethical issues’). Coding frames and analyses were shared with the wider research team. In a relatively small study of this kind it is not possible to indicate definitively if data saturation was reached, and the focus was on the emergent themes.

Interview data pertaining to the history and factual state of sero-surveillance in different

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

countries were triangulated, for the purpose of the discussion section, with literature sources. It was not appropriate to triangulate informants’ views on sero-surveillance, as such data were treated as informants’ opinions and perceptions.

RESULTS

7


BMJ Open


Value of UAT

UK

rp Fo

The programme is largely seen by respondents as indispensable in providing an unbiased measure of HIV prevalence and risk factors in the UK population. The most important outcome from the surveys has been information on undiagnosed HIV infection which, respondents suggested, has led to changes in policy on named testing; for example, HIV testing in pregnant women has changed from opt in to opt out, and sexual health clinics have

ee

moved from targeted to universal voluntary testing. The neonatal dried blood spot survey is used as a check on the performance of local antenatal screening. The public health rationale for the programme has not changed since 1990, and respondents cited the continuing high

rr


ev

"There was far more feeling of a public health emergency (then) than there is today, although you could argue … that it's actually much worse today because of where we see the disease." (UK4)

w

ie


consideration of extending the surveys to other health care settings and other diseases. UK6 discussed how advances in HIV testing, including self-testing, might alter the questions asked

on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 53

A particular theme was whether the British public has a duty to participate in public health

monitoring, possibly as part-and-parcel of receiving the benefits of a national health service (NHS):

"There is a very good case for really exploring with the public, and then pushing very hard to get public agreement, that in general terms that sort of tissue, extra bits of tissue or blood, should be available for use." (UK4) 8


Page 35 of 53

USA According to one respondent, the surveillance programme was valuable in terms of planning, resource allocation, advocacy and interventions to prevent transmission. The heel prick survey provided accurate national data on prevalence in childbearing women and was seen as extremely valuable, with nothing since that has given as accurate a picture of national prevalence (USA 2).

rp Fo

However, given the questionable ethics of UAT (see later), as well as the resultant lack of good quality prevalence data, USA3 argued that enough became known about prevalence to target the most vulnerable groups for health promotion.

ee


Netherlands

rr


ev

among high risk populations, but with informed consent (verbal or written). The populations surveyed by UAT have changed over time. Injecting Drug Users (IDUs) have been surveyed bi-annually since 1994 in different cities in the Netherlands, and are recruited via methadone

ie

care; sex workers and migrant groups have also been subject to UAT surveys.

w

There are two reasons, indicated by respondents, why UAT in Netherlands has been of limited value. First, the uptake of voluntary testing is generally high in the Netherlands,

on

meaning the quality of that data are generally good. And second, UAT has only been deemed ethically permissible in the Netherlands with informed consent, making large-scale UAT

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

surveys impracticable, so restricting UAT to specific population groups and/or the research context.

Norway The unknown, undiagnosed segment of the population is known as the ‘dark figure(s)’ in Norwegian, although neither the ‘dark figure’ nor the number of cases of HIV has been perceived as significant issues in Norway. Uptake of voluntary testing for HIV was described as very high in intravenous drug users and antenatal clinic users, meaning that UAT has never 9


BMJ Open

been perceived as to be of value epidemiologically. The system of surveillance for HIV in Norway instead covers all diagnosed cases (i.e. not undiagnosed cases as with UAT), and was reported as being simple, inexpensive and capable of providing a representative and timely overview that guides prevention, while respecting confidentiality and adhering to ethical principles.

rp Fo


UK

The epidemiological data yielded from unlinked anonymous testing are considered by respondents to be of unrivalled quality internationally. It was suggested, however, that the

ee

apparent high number of undiagnosed HIV infections from UAT surveys might be inflated by double-counting:

rr

"You can quite understand that if you've got gonorrhoea and you've been told by your doctor not to have unsafe sex, you might well decide to go to a different clinic to have your gonorrhoea treated." (UK1)

ev

Others defended the statistics and argued that they were explained by a resurgence of high

ie

risk behaviour, particularly among gay men. UK1 argued that people who refuse a named test

w

and who are subsequently found to be HIV positive through UAT are likely to be people who know that they are positive.

on

USA

There was little comment from USA interviewees about the quality of UAT data, since UAT

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 53

surveys stopped in 1996. However, a theme that emerged in relation to UAT data was around the difference between research and surveillance.

The fact that there is no ‘right’ to health care in the USA was felt to have an effect on people’s views on surveillance – in other words, the ‘disconnect’ between health care provision (not available to all) and surveillance (done to all) (USA 1). The differences between surveillance, research and practice, with perhaps different ethical standards required for each, emerged as a theme from the interviews (USA 4). 10


Page 37 of 53

One point of view was that surveillance (because it is named and part of routine care) is part of clinical care rather than purely research (USA 2). However, while some of the surveillance undertakings at state level in the USA were considered to be practice, at the level of CDC they were treated as research (USA 1). The same respondent commented that cancer sufferers advocated in favour of surveillance for the purposes of research on the disease, since “they were willing to trade privacy and consent for a good which was surveillance that

rp Fo

they saw as serving their interests”.



ee


rr

necessary because named HIV testing is widespread. The quality of information obtained from named testing was felt to be robust and could be used to model estimates of HIV

ev

prevalence in the population. Coverage of pregnant women is very high and this group is considered a valid proxy for the heterosexual population. Named HIV testing among users of STI clinics was also presented as being a valid method for estimating prevalence. It was

ie

argued that, as long as the uptake of named testing by pregnant women and STI clinic users

w

remains high (above 90%) then a national programme of UAT was not thought to be necessary. However, notwithstanding the above point it was acknowledged that current methods of surveillance do not allow undiagnosed HIV infection to be measured accurately.

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Ethical issues


11


BMJ Open



rp Fo


ee


rr


ev


ie


w


on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 53



12


Page 39 of 53



rp Fo


"It's a reasonable thing in practical terms provided people are informed. So … if

ee

people don't know about their option for opting out and their option for finding out then the system hasn't been working." (UK11)

rr


ev

expressed his discomfort with his role in the programme. He described filling out the form for UAT blood testing in front of the patient, without explaining what he was doing, and sometimes after the patient had declined a named HIV test:

w

ie

"You're doing something that sort of seems slightly underhand…you're not explaining it to the patient. Yeah, I'm not comfortable with that, I think one wants to be, to communicate well and to be as open as possible with patients and that's slightly in conflict with that." (UK3)

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

USA


13


BMJ Open

The blood spot survey continued until the publication of the first trials showing that vertical transmission of HIV could be dramatically reduced by treating mothers. When it became clear that infected women could be offered treatment, and therefore should know their HIV status, “it became impossible for the US Centre for Communicable Disease Control and Prevention (CDC) to continue that programme and it was dropped in the early 1990s along with all the other ones” (USA3). It had become difficult to reconcile public health benefits of UAT with individual rights. If CDC had not voluntarily withdrawn funding from UAT, the

rp Fo

same respondent indicated that it was likely that Congress would have mandated cessation, as there was a strong coalition in favour of this.

A related legal issue was brought by USA4, in that public health surveillance appears to go against “fundamental legal principles” in the USA, and constitutional scholars are shocked when they hear what is done.

ee


rr


ev

The Tuskegee episode (a study between 1932 and 1972 in Alabama in which black syphilitic Americans were not given treatment although thought they were) was mentioned by USA1 as

ie

a reason for USA citizens, particularly Afro-Americans, to be suspicious of surveillance

w

practices. In the context of opposition to named case reporting and a consequent lack of trust in surveillance, USA3 argued that there are probably cultural dimensions to issues such as consent, and the acceptability of particular interventions may vary from country to country.

on

“It [UAT] became more of a civil rights issue than a gay specific one. It was, the

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 53

government knows something that you don’t and the attitude in the States, at least at the time, towards the government was much less trusting than in this country [i.e.

UK]. People did not assume and still don’t assume that the government necessarily is looking out for your well being. And I think there’s a much more paranoid attitude towards the government than there is in this country.” (USA3)

14


Page 41 of 53

Related to these cultural dimensions, a particular ethical concern was that the US (and, in particular, CDC) supports UAT in developing countries, with the inherent contradiction that it has been rejected in the USA.

‘It seems to me the CDC can’t at the one time say it’s unethical to do it in the US and say it’s ethical to do it abroad. Different public agencies may come to different decisions, but the same agency it seems to me should be bound by some rule of

rp Fo

consistency.’ (USA1)


ee

Netherlands


rr


ev

seeking testing and treatment. Another concern was that, if UAT focused only on high risk groups, such groups would risk being stigmatised. It was suggested that reporting of findings from UAT surveys of high risk groups would need to be carried out sensitively.

w

ie

A particular issue was consent and maintaining trust with patients. One interviewee expressed that staff in HIV treatment centres and STI clinics demand a more stringent consent process than required by the research ethics committees in the Netherlands in order to promote trust

on

among patients.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Norway

A UAT system was never thought relevant, necessary or appropriate for Norway. In Norway it is not legal to test or analyse blood or biological material for infectious diseases for non-

diagnostic purposes without consent unless the purpose is surveillance of an ‘epidemic’. One respondent felt this displayed a ‘moral grey zones’, as the law hinges on the definition of ‘epidemic’ [Norway 1].

15


BMJ Open

There is very little concern or awareness in Norway about ethical issues around UAT. However, there is a general willingness to provide information for public health purposes in Norway, and people are used to being surveyed. It was reported that many Norwegians have given blood to the Biobank, and give broad consent for their tissue/blood to be used for a wide range of research. A respondent indicated that many are willing to waive their consent in Norway because ‘there is significant trust’ in the health care system. However, another respondent expressed that there was a possible danger that UAT blood could be used to

rp Fo

identify someone via the Home Office DNA database, and that researchers might be forced to share information under a court order (the respondent mentioned that the murderer of the Swedish politician Anna Lindh in 2003 was convicted on the basis of a match with DNA held in the national biobank).

ee

One respondent suggested that informed consent could be different in different contexts. In a country like Norway, with low prevalence and easy access to health care, the current surveillance system (non-UAT) was described as acceptable and functioning well. However,

rr


ie

ev

DISCUSSION

w


epidemiological, ethical and policy debate.15 16 17 In this paper we have presented the results

on

of qualitative interviews with key informants in four countries pragmatically selected because they represent a range of approaches to UAT. 18 A total of 23 respondents were interviewed, differentially distributed between countries: 12 in the UK, four in the USA, four in Norway and three in the Netherlands.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 53

There were limitations to the research: more key informants were interviewed in some countries than others; some interviews were face-to-face whereas others were by telephone; and two of the interviews relied on notes rather than recordings for later analysis.18 19 20 It is


possible that such aspects introduced bias, but we believe that the balanced pragmatic approach taken to address the research questions was appropriate for this relatively small

16


Page 43 of 53

qualitative study. Interesting themes emerged in the analysis, and we were able to triangulate more fact-based comments with documentary sources in the literature.

First a summary of key findings is presented (in the context of historical development of UAT in different countries, as well as what is already known of the subject) , followed by a discussion of how the findings from this paper fit into broader contemporary debates about public health ethics and policy.

rp Fo

Summary of key findings from interviews (in the context of what is currently known)

In the United Kingdom (UK), a balance of epidemiological and ethical considerations in the

ee

late 1980s led to the implementation of the ‘prior notification’ approach to UAT; this involves leaflets and posters informing clients that UAT is occurring, allowing opting out but not necessitating the gaining of informed consent.777 101010 Respondents stressed that the


rr


rationale remains apt today, and that this approach provides data unrivalled in terms of


epidemiological quality,999 21 22 and which is of use for health service planning.23 Ethical


ev



voluntary (named) testing in settings where UAT is occurring. However, there was mention


of concern about the lack of full explanation, whether people do actually know UAT is

ie

happening, and ‘ownership’ of blood.24 It was indicated that an approach involving informed consent would be cumbersome, time-consuming and epidemiologically damaging.25

w

In the United States (USA), UAT of blood for HIV was extensively used in the 1980s and

on

early 1990s,26 but was stopped in the late 1990s as evidence emerged that anti-viral therapy in pregnant women could help reduce the vertical transmission of HIV from pregnant mother to

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

child – named testing needed to be maximised so that HIV-positive pregnant women could be identified and offered advice and therapy.27 The interviews indicated that data from UAT

were useful epidemiologically but, since UAT has stopped, there has been sufficient other data to plan services accordingly. Other issues raised include a general suspicion around epidemiological surveillance post-Tuskegee, cultural dimensions of UAT with respect to minority groups, a lack of trust in the Government, and the related disconnect between surveillance (done to all) and health care provision (variable, and not done to all).


666 111111


There was also comment about the apparent contradiction in the contemporary support of the


17


BMJ Open

US Centre for Communicable Disease Control and Prevention (CDC) for UAT in lowincome countries.28



acknowledgement that large-scale UAT would provide additional epidemiological data, but


this has been balanced against concerns of this leading to diminished voluntary testing –

rp Fo

especially in at risk groups – through loss of trust in the healthcare system.29 Smaller UAT surveys have been intermittently undertaken in such groups (e.g. intravenous drug users, sex workers, migrants) as research studies, but have required individual consent (initially oral, now written).

ee



Surveillance has thus tended to focus on the monitoring of the diagnosed population

rr

(anonymised diagnosed cases), which has been supported by the gay community; the undiagnosed population is interestingly referred to as ‘dark figure(s)’. In Norway it is not

ev

legal to test without consent unless there is an epidemic, although respondents reported few ethical concerns since people generally have trust in the health care system, and are willing to provide information for public health purposes (many give blood to the Biobank for a range of research purposes).30

w

ie


on


individuals’ rights and the appropriate role of government3 in infringing on those rights for

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 44 of 53

broader public health interests or concerns. These debates are inter-connected, although will be presented in turn.


18


Page 45 of 53

Respect for autonomy is the bioethical principle that is most likely to be jeopardised by a programme of unlinked anonymous blood testing, since informed consent is not usually deemed necessary for an individual’s blood to be included; this has tended to be the basis for opposition to UAT.555 31 The findings from our research show, however, that such positions



Debates in the late 1980s and early 1990s demonstrated the opposing views of those who

rp Fo

were concerned about patient consent,888 262625 and those who felt that the priority was to control what was seen as an epidemic.

151515


UAT programmes did go ahead in the USA and


UK, and other high income countries: there was no requirement for consent in the USA,


whereas in the UK an approach of ‘prior notification’ was adopted through leaflets and


posters in clinics – consent is presumed but individuals have the opportunity to opt out.

ee

Over the past two decades informed consent has emerged as ‘the ethical touchstone of medical research, and is now enshrined in practice and a range of international guidelines as

rr

one of the main preconditions of medical research’.252524 32 Some commentators argue that


even anonymised tissue should not be used without consent. Trouet, for example, believes

ev

that the use of such material is in contravention of Article 22 of the Convention on Human Rights and Biomedicine. This states that ‘when in the course of an intervention any part of a human body is removed, it may be stored and used for a purpose other than that for which it

ie

was removed, only if this is done in conformity with appropriate information and consent

w

procedures’, and in accordance with Article 8 of the European Convention on Human Rights which protects the right to respect for private life.33

on


weak moral relativism operates in policy-making and practice around UAT.34 In the USA, the

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

change in social context and debates about individual rights, along with evidence of reduction in mother-to-child vertical transmission, resulted in the abandonment of UAT in 1996.121212


The fact that the USA continued to support UAT in international research overseas illustrates


further an attitude of moral relativism, although it could be argued that differential health needs justified different approaches.

Interviews in the UK showed that the prior notification approach that remains in use today still holds strong support – based on perception that there is minimal harm to the individual, 19


BMJ Open

and the value of high quality, unbiased epidemiological data on HIV (held to be the most robust and accurate worldwide).35 Respondents felt that any change to UAT requiring informed consent would be cumbersome, time-consuming, and costly. According to the Nuffield Council on Bioethics it may also be ethically unnecessary: ‘... it is acceptable to collect and use anonymised data for assessing and predicting trends in infectious disease without consent, as long as any invasion of privacy is reduced as far as possible.’.36

rp Fo

This position is not, however, cross-culturally held. In the Netherlands, UAT has not been felt to be acceptable without informed consent, initially verbal and more recently written. The societal emphasis on individual autonomy, including in the health research environment, has been the paramount consideration.555 Northern European and Scandinavian countries are


often held up as having a strong community ethos, with an expectation on individual

ee

involvement in endeavours for the public good.37 This was not reflected in UAT practice in the Netherlands, although was indicated to be important in Norway, albeit in a setting where UAT has never been taken up.

rr


ev

ethical debates, and how attitudes and policies can change in time within any given jurisdiction.38 Differences exist between the moral evaluations and resultant UAT approaches between countries such as England and the Netherlands. Within the USA, in contrast, the

ie

practice of UAT has been re-evaluated with time. In the late 1980s and early 1990s the feared

w

AIDS epidemic was seen as a public health emergency, there was no treatment available, transmission from mother to unborn child was not preventable, and the stigma and discrimination associated with the infection were more overt than today. As these factors

on

shifted, so did the policy on UAT in the USA. The situation in low income countries today has created a perception of the acceptability of UAT akin to that in western countries two

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 46 of 53

decades ago.

Individual good and community (public) good

There has been criticism in recent times of government for its ‘nanny state’ approach to interventions in, for example, the areas of smoking, alcohol and diet. More recently the notion of ‘stewardship’ in public health rejects these negative associations,39 and advocates 20


Page 47 of 53

that government should have a role to play in protecting the health of its citizens, preventing them from harming themselves or others, and promoting healthy behaviour.[4] In this view, public health interventions should not be thought of as unwanted restrictions imposed by a ‘nanny’ on powerless children, but as actions that favour both individuals and society more widely as prescribed by a ‘steward’ administering and protecting citizens’ interests. The Nuffield Council on Bioethics, for example, argues that the state should act as steward both to individuals and to the population as a whole, but that individuals’ personal choices should

rp Fo

also be taken into account.363635


The subject of UAT sits squarely within the debate about public health stewardship for social good. Recently, commentators have argued that patients of a publicly supported health care system like the NHS have a moral duty to participate in scientific research,40 41 although most

ee

do not go so far as to articulate that such participation should be mandatory in practice.42 Nevertheless, UAT could be understood as something that individuals should feel they have a moral obligation to participate in, regardless of ethical concerns about individual rights and

rr

personally held views. Our own survey of over 400 users of two genitor-urinary medicine clinics (a parallel part of this research project) has found that a large majority of respondents

ev

(89%) agree that people should have a responsibility to take part in research studies, and the same proportion would agree to the use of their blood in unlinked anonymous seroprevalence testing. However, when asked about the question of consent, 74% indicated that they should

ie

be asked to consent before their blood is used for public health surveillance.141414


w


on

The way forwards perhaps is to ensure that patients – and the public more widely – are

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

informed of the existence, purposes and value of disease surveillance arrangements, as well

as the importance of contributing to future health research, while investigating arrangements

for gaining consent which are not unnecessarily cumbersome or likely to create bias amongst those who agree that their blood be used.45

The issue of consent in seroprevalence surveillance, such as UAT, remains an important instrument to gauge the moral direction of public health policies more broadly.46 Debate continues around how public health programmes can embrace a stewardship relationship, 21


BMJ Open


Acknowledgments

rp Fo

The authors would like to thank the participants in the study. We also acknowledge the valuable advice given by Vivian Hope, Will Nutland, Cicely Marston, and Graham Hart.


AK is also Director of Public Health Strategy and Medical Director of the Health Protection

ee

Agency. The views expressed in this paper are those of the authors, and are not intended to represent the views of the Health Protection Agency.

Funding

ev

rr


ie

Ethical approval


w


on

Notes

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 48 of 53

Formatted: Line spacing: single

[1]


Agency (HPA) and the Institute of Child Health, University of London. The programme was previously managed by the Communicable Disease Surveillance Centre, part of the Public Health Laboratory Service, which was subsumed into the Health Protection Agency in 2005. [2]


[3]


details cannot be provided to maintain their confidentiality.

22


Page 49 of 53

[4]


government.

References

1

rp Fo




ee

Report of the Unlinked Anonymous Prevalence Monitoring Programme. London: Department of Health, 1998. 3

rr


Other Sexually Transmitted Infections in the United Kingdom. London: Health Protection

ev

Agency, 2007. 4


ie

anonymous HIV seroprevalence studies: an international comparison. AIDS 1990;4(4):283290. 5

w




ly

2004;303:631-632. 7

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open





23


BMJ Open

9




823. 11


rp Fo



13



ee


rr



w

ie

1987;294:443-444. 16

ev



on



ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 50 of 53

18


19

Yin RK. Case study research: design and methods. London: Sage, 1994.

20

Green J, Thorogood N. Qualitative methods for health research. London: Sage, 2004. Formatted: English (U.K.)

24


Page 51 of 53

21




among childbearing women in New York State from 1988 through 2000: a valuable

rp Fo

epidemiologic tool. Archives of Pediatric and Adolescent Medicine 2004;158(5):443-8. 23




ee



rr



Health 1993;83/4:496-7.


reappraisal. Journal of Medical Ethics 2000;26:16-21.

w

28

ie

27

ev



on


ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open



Informed consent and Biobank research in a Norwegian context. Critical Public Health 2005;15(4):335-347.

25


BMJ Open

31






rp Fo



35




Council on Bioethics, 2007

rr

37

ee





w


ie


ev


2005;31(4):242-248.

on

41


42



ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 52 of 53


anonymous testing of blood for HIV and other diseases. British Medical Journal 2000;320: 90-91.

26


Page 53 of 53

44





rp Fo



w

ie

ev

rr

ee ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

27
