Summary
Research data from a project at Lancaster University concerning the 2001 Foot and Mouth Disease disaster was safeguarded for future use.
Background
The UK’s 2001-2002 Foot and Mouth Disease epidemic was a profound crisis affecting rural areas, with lockdowns and extensive animal culls lasting almost one year, and economic and social effects lasting many more. In Cumbria, the worst hit area, 70% of farms were affected. As researchers, staff at Lancaster University undertook a large ethnographic study into the health and social effects of the disaster and processes of recovery. This received funding from the Department of Health under a time-critical funding route, and ethical approval from the regional NHS Research Ethics Committee.
Challenge
The team could see that the epidemic was being treated centrally as primarily an animal problem, ignoring the huge social consequences and human impacts such as loss of livelihoods or children not being able to attend school. Healthcare services in severely affected areas were not recording increased demand during the epidemic and subsequent months. This was taken to mean that the human health and social effects of the disaster were not significant. However voluntary local helplines and rural support groups were besieged with appeals for help, suggesting major impacts.
Approach
With their diverse steering group, the team identified six affected social groups based on occupation and residence. They recruited a panel of 54 individuals who represented these groupings. By the autumn of 2001, lifting of restrictions enabled us to meet with each group to explain the purpose of the study and obtain people’s informed consent to participate. Everyone also gave an audio-recorded in-depth interview around this time. From December 2001 respondents then kept a weekly written diary in physical form. This began with simple questions about health and quality of life followed by a space for free text entries of any length. This became the most frequently and fully used space, with participants writing between half a page and six pages each week over the 18 months of diary writing. Diaries were collected monthly, each researcher visiting on average 18 households. Periods of “holiday” from diary writing were negotiated.
The team collected 3,071 weekly diaries in total. In recognition of respondents’ time and expertise, a small fee was paid on collection of the diaries. The team produced a regular newsletter to keep people in touch with the project. They held a round of further recorded group meetings towards the end to obtain feedback on initial analysis of the diaries, interviews and group discussions.
Intended aims
The team wanted to capture evidence about the impact of the disaster and processes of recovery from “on the ground” accounts collected over time. This provided a different, experiential perspective in comparison with the three official inquiries into the disaster which recruited “professional” experts. As academic researchers, the tean aimed to create knowledge about the human impacts of the epidemic and what forms of support people need both in disaster situations and in the recovery period.
Archiving the research data gathered by the project was not originally part of the study design and funding. However, as the project progressed they realised that a rich and powerful body of material was emerging. So after the diary writing had finished, steps were taken to revisit the consenting process in order to create digital and hard copy archives. The necessary focus of their final research report meant that they could not fully reflect the breadth of the huge data set collected. By archiving they could preserve the greatest possible amount of the research materials for the future.
Obstacles and issues
Seeking additional informed consent for archiving in addition to participants’ existing consents required very careful thought.New information and consent forms were piloted for this, distinguishing between material to be included in the digital dataset held by the online data archive Qualidata, and material in hard copy to be held by the University.
Methods for longitudinal studies (over a lengthy period of time) and in particular, repeated visits, have ethical implications which are well recognised in sociological and health literatures. These include how to deal with different degrees of disclosure, the power relations within the research process and the difficulties of achieving informed consent. Overall, the team suggests the duration and intensity of our study nurtured trust between researchers and respondents. They believe this lay behind the very high commitment of respondents: 50 out of 54 respondents completed 18 months of weekly diary keeping.
Preparing the original materials as an anonymised dataset for analysis and use, and for archiving, posed a number of challenges. Many of the respondents lived within close-knit rural communities so that geography, their jobs, and lifestyles might make them easily identifiable. For some respondents, participating in this project represented one way of giving voice to their experiences. Indeed, the team organised a large public conference to discuss their interim findings where some individuals gave very public and powerful testimonies. This attracted local and national media coverage, and some respondents later gave press interviews. The majority however were concerned about anonymity and confidentiality, not only for themselves but also for their immediate family, friends, work colleagues and wider community.
The team developed criteria for anonymising data, focusing on removing unwitting identification and information which may be deemed libellous or cause unnecessary distress. Scenarios include where a respondent named themselves or another person, which may be context-dependent (for example, group discussions). They also removed people named and associated with local contentious events, or local organisations spoken of detrimentally, confidential information or specific place names along with job information. Such anonymisation is complex and was very time consuming. Trying to strike a balance between protecting people and retaining useful context, particularly for later use, inevitably involved many ethical and methodological concerns.
Where the handwritten diaries were converted to digital text, anonymising these files was more straightforward than erasing or blocking out text in hard copy. For exhibition purposes, the team produced an ‘audio montage’ illustrating the study findings through a series of extracts from several individual interviews. Their collaborative archiving processes meant that they could accommodate individual respondent wishes. For example, one respondent chose to remove a portion of their diary, and another requested their voice to be digitally adjusted.
Actual outcomes and outputs
The anonymised digital dataset has been available since November 2006 (originally with Qualidata, subsequently the UK Data Service). It comprises transcripts of 42 individual interviews, 40 diaries, six focus groups and the audio montage; these transcripts are available to any registered user of the UK Data Service. The collection also includes audio files from 42 individual interviews, seven focus groups and the audio montage; access to these is subject to permission from the depositor (the lead researcher). This digital dataset has subsequently been re-used by other researchers.
Lancaster University Archives and Special Collections has held the hard copy archive generated by the project since 2006. The Health and Social Consequences of the 2001 Foot and Mouth Epidemic in North Cumbria archive (12 boxes) contains all the original diaries kept by the participants, recordings of interviews and group sessions on CD, and project management records including consent information. Archives and Special Collections has subsequently collected other material relating to the activities of a charity during the epidemic, to reflect other contemporary perspectives and activities.
Lessons learned
Drawing on their experience of this project, the team have built in consent for archiving from the start in subsequent projects (for example, their work on flooding with children and young people), recognising that seeking consent retrospectively may create confusion and uncertainty. However, they learned that in disasters many people want to tell their story. Archiving is a way to enhance the visibility of events and those affected, and may support their recovery, as well as preserving knowledge and evidence for the future. It takes time and care to recruit people, build a trusting relationship and gain their informed consent, but this is an important part of working ethically and of preserving their contributions.
Because such a large number of people joined and stayed with the project, anonymising the very large dataset for use by other researchers was very time-consuming. It took several iterations for every item. The team were able to work closely with Qualidata, and draw on their expertise, particularly with digitisation, as part of this process.
For Lancaster’s Archives and Special Collections, new legislation and new policies and procedures have come in since the hard copy archive was first transferred. The team have therefore recently updated the donation agreement to enable us to continue to manage the collection into the future.
Next steps
Archives and Special Collections aims to expand the existing box-list and to review data protection and other sensitivity risks, in preparation for the 25th anniversary of the Foot and Mouth Disease epidemic. The team are actively working on digital preservation, currently moving files from portable/physical media into dedicated storage, as well as other collections management activities. They are also developing strategies to collect research outputs from academics in future.
Further information
Read this article (PDF, 0.3MB) about archiving the data by Cathy Bailey, Josephine Baxter, Maggie Mort and Ian Convery (2006) – ‘Community Experiences of the 2001 Foot and Mouth Disease Epidemic in North Cumbria: An Archiving Story’, Methodological Innovations Online, 1(2), 83-94