Across the globe, childhood vaccination is widely regarded to epitomise the effective (and cost-effective) application of science and technology to current public health problems. It is central to future hopes in tackling diseases both of poverty and of excess. The production and delivery of new vaccines, combinations and applications has become a major focus of research and funding within a highly globalised field. Multi-national pharmaceutical companies, NGOs, research institutes, foundations and global health organisations now interact with wealthy and poor governments in initiatives and new forms of partnership which are extending vaccines to every person on the planet, and moulding health services in the process.
Recent, high-profile controversies nevertheless point to public anxieties around the application of vaccine technologies. In the UK, parental concerns over possible adverse effects of the measles, mumps and rubella vaccine (MMR) since the early 1990s have built into a movement interplaying with scientific and media debate, and MMR uptake levels have, in some localities, fallen by 30%. In Northern Nigeria in 2003-4, parents and their communities refused the Oral Polio Vaccine, associating it variously with HIV transmission, infertility and international and national genocidal politics. The global polio eradication programme stalled, and polio reappeared throughout West Africa.
Whether in Europe or Africa, such controversies - at the broadest level - can be understood as emerging where the rapidly advancing, globalised health technology and technocracy involved with vaccines and their delivery encounters the deeply intimate personal, cultural and social worlds of parenting and childcare. It is perhaps inevitable that childhood vaccination has become a key issue around which debates over public engagement and trust in science and technology proliferate. As these debates have unfolded in social science literature, public health discourse and popular commentary, recurring themes and contrasts have emerged which link vaccine-specific anxieties to broader dimensions of society, and science-society relations.
First, a strong contrast is drawn between 'North' and 'South': the idea being that a concern with vaccine side effects is a luxury of those in the north no longer familiar with the childhood diseases ravaging the south, where the more important clamour is for vaccine access (Obaro and Palmer 2003, Streefland 2001). This evokes a broader contrast between late-industrialised 'risk society' (Beck 1992) and a still-to-modernise 'underdeveloped society'. Second, and relatedly, a contrast is drawn between the anxious middle classes, as against a more compliant poor (e.g. Pareek and Pattinson 2000). A third contrast turns on irrationality vs. rationality, associating vaccine anxieties in the south with incomplete (rising) scientific rationality in settings where 'traditional' beliefs still predominate and in the north, with a 'rise of irrationality' in society, as evidenced for instance in the increasing popularity of alternative medicine (e.g. Fitzpatrick 2004).
Other debates turn on the role of knowledge and information. Fourth, then, low vaccine uptake is linked to public ignorance, or deficit in scientific understanding, of the value of vaccination or evidence of vaccine safety (e.g. Elliman and Bedford 2001), and to misinformation and rumour spread by irresponsible media and pressure groups (Andre 2003, Hargreaves et al 2002, UNICEF 2003). Fifth, and relatedly, reason (driving evidence-based decisions or governance) is contrasted with emotion (as driving some parents' personal choices). Finally, vaccine anxieties in the north are being cast as part of a generalised breakdown of trust in public institutions (Fitzpatrick 2004, Hobson-West 2003), evidence of growing critical public engagement with scientific expertise (Irwin and Wynne 1996). In contrast, southern analytical traditions stressing the non- or incomplete integration of expert science with 'indigenous knowledge' and beliefs tend to attribute vaccine anxieties to collective resistance based on religion or traditional beliefs (Streefland 1999).
These contrasts and stereotypes - which often pass for ‘explanation’ of vaccine refusal - can be seen, in part, as rooted in scientific and public health frustrations with non-compliant publics, and research framed by these. They are also rooted in long-established differences between the analytical traditions that reflect on science-society relations in European and African settings, respectively. Our research set out to consider these terms of debate from a different perspective, rooted in ethnographic and anthropological understanding of how parents are thinking and deciding about vaccination, amidst diverse personal experiences, cultural knowledges and perspectives, social relations, and experiences of national and international institutions. It explored how parents in localities in Britain and West Africa are engaging not just with routine vaccination, but with issues involving vaccine science and scientific controversy.
Objectives
The overall objective of the research was to develop comparative insights into science-society relations in European and African settings which have conventionally been theorised very differently, through the case of childhood vaccination research and regimes. This objective has been addressed in a preliminary way through drawing key comparative themes and insights from the country-specific analyses, as reported in section 4.3. This comparative work, which is necessarily at a broad level given the stark differences of context between Britain and West Africa, will continue in post-award analysis in engagement with wider literatures and be presented in full in the proposed book output from the research.
To do so, it focused on the intersection of routine vaccination with the MMR controversy in the UK and with MRC-orchestrated vaccine research in The Gambia, aiming in each context to:
· Identify how public concerns with vaccination research/regimes are socially differentiated and shaped by diverse conceptual frameworks and knowledges around infection, disease and immunity, and experiences of the state and of science in other domains. This was addressed and met fully through ethnographic and survey research with parents in Brighton, UK and The Gambia, reported in sections 4.1 and 4.2 respectively;
· Specify how different people consider trade-offs between social and individual benefits and risks, the differentiated notions of ‘community’ and ‘communities of trust’ implied, and how this influences socio-political organisation around vaccination. Again, this was successfully met through ethnographic and survey research with parents in Brighton, UK and The Gambia, reported in sections 4.1 and 4.2 respectively, as well as through a focused case study of the parental movement around MMR.
· Identify how vaccine scientists and public health professionals conceive of public knowledge and attitudes towards vaccination programmes, and how 'frontline’ staff mediate professional and public views. This was addressed and met successfully through interviews with scientists, public health professionals, and fieldworkers nationally and locally, reported for the UK in section 4.1 and for The Gambia in section 4.2. International public health and scientific discourses (summarised in Background above, and to be documented more fully in forthcoming book) were gauged through participation in several conferences and networks, e-mail discussion and literature study.
Further, the research aimed to support the development of new approaches to public involvement in research into vaccine technologies targeted at children, their delivery and promotion. This objective was addressed through (a) working collaboratively with local and national governmental and research organisations (detailed in Activities and Achievements Questionnaire 2B), with whom intensive discussions were held throughout the research; (b) feedback workshops to a wider set of users in each country, and (c) proactive seeking-out of new citizen-participatory approaches and 'experiments' in each country context. In both settings, these interactions have led successfully to some minor shifts of approach and communication strategy by established institutions, as indicated in section 7. However, as embracing new participatory approaches involves challenges to powerful public health and bioethical discourses, encouraging these is a long-term challenge which will necessarily extend beyond the project's timeframe. In this context, innovative experiments that we might interact with have proved rare. An exception in the UK context is the New Economics Foundation's 'DEMOCS' gaming approach which has developed an MMR pack, and where we made advisory inputs based on our research.
Methods
Our research approach was characterised by its meta-level comparative framework, its collaborative context, and its combination of qualitative and quantitative methods. The Gambia and southern England were chosen not with any claim to be fully representative of African and British conditions, but as places where pressing controversies around public engagement with vaccine science are ongoing, allowing the generation of context-specific insights which could then be drawn together in broader comparative discussion.
The Gambia has been the locus of British Medical Research Council (MRC)-orchestrated medical research and vaccine trials for the past fifty years, with research stations throughout the country engaging a large proportion of the population as actual or potential 'study subjects', alongside the administration of routine vaccination through the country's relatively strong primary health care infrastructure. The research focused on rural Upper River Division (URD), the site of the joint MRC-Gambia Government Pneumococcal Vaccine Trial, and on urban Western Division, where some parents had engaged with the MRC Sukuta Birth Cohort study researching infant immunity and responses to infection. These two sites also enabled a comparison of rural and urban settings, an important distinction in rapidly-urbanising West Africa. The research was developed in collaboration with local and national staff of MRC and the Gambian Government Expanded Programme on Immunisation, through preliminary discussions, a methodology meeting and preliminary feedback workshops. Approvals were obtained from the MRC Scientific Co-ordinating Committee and the joint Gambian Government-MRC Ethics Committee.
In the UK, the research focused on the city of Brighton and Hove, chosen for its locality to the researchers and its particularly sharp decline in MMR coverage. Collaborative partnerships were developed with the Brighton and Hove Primary Care Trust and a Stakeholder Advisory panel comprised of key national policy-making, health professional and parents' organisations (see Activities and Achievements Questionnaire section 2B) who participated in preliminary discussions, a methodology meeting and preliminary feedback workshop. The study was approved by the East Sussex, Brighton and Hove Local Research Ethics Committee and appropriate research permissions given by South Downs NHS Trust and Brighton and Hove City PCT.
In each country, a first phase of in-depth qualitative research was conducted using ethnographic methods. In The Gambia, this took place during March - November 2003, with ML and JF spending 3 months and Gambian Research Officer Mary Small completing the work. Ethnography focused on the peri-urban settlement of Sukuta, rapidly-growing from an old Mandinka settlement, and on the rural Mandinka village of Tambasansang in Upper River Division (URD). In the UK, ethnographic research took place during February - October 2003, conducted primarily by Research Officer Mike Poltorak but with major inputs also from ML, and focused on the catchment areas of two collaborating GP practices: Whitehawk, a stereotypically 'deprived' area, and Fiveways/Preston Park, a stereotypically 'middle class' area - thus illustrating contrasts which have been significant in popular debate over MMR. Ethnographic methods included participant observation in social settings where parents take infants (Gambian compounds, clinics and markets; 5 Brighton carer and toddler groups), recorded focus group discussions (9 in The Gambia, 4 in Brighton), observation of clinic interactions and interviews with frontline health professionals (12 nurses and 18 MRC fieldworkers in The Gambia, 8 GPs, 3 practice nurses and 6 Health Visitors in Brighton), and narrative interviews with mothers identified through participant observation (100 in The Gambia, 23 in Brighton). Initial interviews in both settings suggested the value of an open-ended biographical format to allow mothers to speak openly about the processes and issues surrounding their decision-making, so narratives took the form of 'child health and immunisation biographies'. These were tape recorded and fully transcribed in Brighton, and noted by hand and later fully typed up in the Gambia where recording is socially sensitive, and analysed by drawing out key narrative themes. All have been stored confidentially.
In the second, quantitative phase, a survey questionnaire was developed for each country to explore the significance of the key narrative themes amongst wider populations, and in relation to social variables (see Annexes 1 and 3). Survey design and methods were adapted to the particular issues and logistical possibilities in each country context.
In the UK, the sampling frame consisted of all children aged 15-24 months listed on the Child Health Dataset held by South Downs Community NHS Trust as resident in the catchment of Brighton and Hove City PCT, on the date of record extraction in early March 2004. Children were categorised into those who had and had not had an MMR immunisation recorded, and of the 1800 children eligible, a sample of 1000 MMR uptakers and non-uptakers in a ratio of 1:1 was randomly drawn, using the statistical programme STATA 8. All the 135 registered children who had had no vaccination events recorded were also identified. A postal questionnaire addressed to the mother or guardian of each child was sent in March 2004. This contained a questionnaire for the mother, and also one to be passed where possible to the father of the child. A follow-up letter with a second questionnaire was sent after 3-4 weeks to non-responders. The questionnaire explored rank of child within the family, sources of information on parenting and immunisations; early health of the child (including its birth); views on the risks associated with measles and the MMR; interactions with health care professionals and others in relation to MMR; the process of decision making, including attitudes to public bodies and governments as sources of advice and influence. In addition, a range of specific statements made by Brighton parents as part of the ethnographic phase were offered for agreement or disagreement. Completed questionnaires were linked with children’s data as recorded on the child health database, from which additional information was derived relating to: gestational age, birth rank, prematurity, age of mother at child’s birth, and immunisations given within the NHS. Apart from mailing lists, all person-identifiable data was handled only on secure NHS servers with appropriate permissions. 452 of 1135 mothers’ questionnaires were returned, representing an overall response rate of 39.8% (Annex 2 table 1), as well as 257 Fathers' questionnaires. Data were entered by the firm Abacus, and analysed using STATA software, especially to explore differences between mothers who complied (reported choosing to have MMR on time) and who did not comply (choosing to delay MMR, not to vaccinate with MMR, to obtain single jabs, or who remained undecided). Analysis of Fathers' data and paired Mother-Father responses to draw out gender differences and negotiations is yet to be carried out.
In The Gambia, we drew half of the respondents from three rural districts of URD (Fulladu, Wuli West and Sandu) which were covered by the MRC Pneumococcal Vaccine trial and had received MRC-related support to immunisation infrastructure, and half from the two more urbanised districts of Western Division (WD) (Kombo St. Mary and Kombo North) which have been the foci of rapid immigration. All children aged 12-24 months resident in these Divisions at the time of survey (October - December for WD, January - March 2004 for URD) were eligible. We used a two-stage stratified sampling process to select 800 respondents in URD and 800 in WD. Enumeration areas used for the 1993 Census were identified within the chosen districts, separately within URD and WD. 35 enumeration areas in each Division were then randomly selected using a random number list. The sampling method within enumeration areas was based on a random walk method well established in immunisation coverage surveys. A team of five fieldworkers employed through MRC and supervised by Mary Small aimed to interview the mothers of a target number of children identified by random walk. The questionnaire explored mother's immunisation understandings and practices as part of broader notions of protecting child health, the process and timing of immunisation decisions, and perspectives on engagement with MRC studies. Fieldworkers also copied data from the child's health record card when available. The response rate was near 100% (only 3 mothers refused to be interviewed). The completed paper questionnaires were returned to a central MRC office where coded survey data were double entered into a database by clerical staff, and free-text data entered once. Quantitative data were stratified into responses from the urban west and the rural east, and analysed using STATA software. Free text was coded by ML and JF where appropriate.