Main Research Report

Sebastian Taylor
Publication Date
August 1, 2015

"The circumstances of household thinking about community, livelihood, health and vaccination are complex. To reduce this protean calculus to a short list of population proportions and individuated causations is to undermine and underestimate the reality with which a programme like polio eradication wrestles daily."

This paper describes a pilot investigation demonstrating the potential of field research to explore and explain deep-rooted and interconnected factors shaping people's health-related behaviour and attitudes to mass vaccine programmes like the Polio Eradication Initiative (PEI). The purpose of the research was to develop an in-depth understanding of the underlying family- and community-based issues that influence Nigerians' attitudes towards immunisation - attitudes that have continued to hinder the success of the effort to eradicate polio in this country (one of the few polio-endemic countries) and, hence, the eradication of this disease worldwide. By the end of 2014, only 6 cases of polio were recorded in Nigeria, with no recorded infections to date, as of May 30 2015. However, there are still small but significant areas of sub-optimal oral polio vaccine (OPV) coverage. The goal of this research is to help safeguard the gains of the polio programme and to sustain high levels of routine immunisation (RI) coverage as part of the PEI legacy.

Adapting a methodology from Qualitative Comparative Analysis (QCA) that involved using a mix of qualitative and quantitative approaches, this research randomly sampled 30 households per settlement in 60 settlements within wards and local government areas (LGAs) in Sokoto, Kano, and Bauchi states of northern Nigeria. (These places were specifically selected because: (i) they were characterised as high- and low-performing, according to the national government programme's risk for transmission of wild poliovirus (WPV) and (ii) they had certain "performance" characteristics in terms of supplementary immunisation activity (SIA) coverage). Logistical and security considerations allowed for inclusion of rural and urban settlements in Sokoto, rural and semiurban settlements in Kano, and rural settlements in Bauchi. Researchers calculated for a representative sample of 480 households per state. Allowing for attrition and substitutions, the researchers surveyed a total of 3,306 respondents (male and female) in 1,653 households using a questionnaire eliciting information on quantitative and qualitative dimensions of family life: general developmental conditions, household perceptions of (and trust in) external actors, health and healthcare experiences, and knowledge of/attitudes to RI and polio eradication.

The research focused on 2 primary outcome variables - (i) households reporting missed children in past polio SIAs and (ii) households reporting the possibility of refusing OPV in the future ("propensity to refuse") - with 2 secondary outcome variables: (i) approval/disapproval of the PEI, and (ii) approval/disapproval of RI.

Selected communication-related findings from the research include the following:

  • A significant minority of sampled households reported having missed children (16-17%); a similar proportion reported considering OPV refusal in the future (14-17%). Households reporting missed children in the past were significantly more likely to consider refusing OPV in the future, a finding the researchers say is reflective of intentional caregiver behaviour. "The fact that 0-dose children have a substantially higher chance of being missed in northern Nigeria suggests, further, that a substantial part of household refusal is continuous over multiple SIAs, and hence 'chronic'. This also suggests that behaviour-change communication [BCC] interventions are not having the required effect for all risk groups, in particular entrenched refusal."
  • Propensity to refuse OPV was found to be clustered in specific settlements: Approximately 20% of the sample communities accounted for almost three-quarters of refusal risk (and over half of all reported missed children). "This clustering suggests that refusal may be at least partially a collective, community-level effect, requiring a collective, community-level response." The researchers found that "urban households (who do not fit the conventional risk profile of poor, poorly educated, illiterate and susceptible to anti-vaccine rumours) require strategic attention". They note that urban refusal has been found in other eradication programmes, notably Greater Cairo, Egypt, where urban families, at higher levels of wealth and education, viewed private health practitioners as preferable to mass-delivered public health services.
  • Knowledge of immunisation practices and vaccines - including but not limited to OPV alone - is strongly associated with reduced risk of refusal in urban, semiurban, and rural households across all states. "[E]xpansion of routine immunisation (awareness, understanding and service provision) may be a primary, rather than ancillary, strategy for enhancing OPV uptake."
  • "[I]interventions that can build alignment between male and female caregivers (e.g. on vaccination) may be helpful in improving acceptance....PEI (and wider health system strengthening) need a clearer gender strategy - building communication and engagement between men and women within households, but also through their mutual participation in the planning and management of community health activities, and integrating female health workers more closely in institutional processes of service planning and delivery."
  • The degree to which a household has a religious orientation does not appear to correlate with OPV refusal. Thus, "[w]hilst there are many reasons to engage with religious and traditional leaders for public health, a predominant emphasis on religious and traditional leaders as principal interlocutors for polio vaccination (and their interlocution as the primary mechanism for addressing OPV refusal) does not appear justified."
  • Lower "trust in government" did not correlate with higher risk of OPV refusal. "Higher-risk settlements have consistently higher expectations of government (for example, in terms of service provision). But they also have systematically lesser confidence in their ability to influence government performance. It may be disappointment in government that shapes negative attitudes to a programme like polio eradication....It is clear that high-risk settlements trust government in some areas - a key objective (both for PEI and longer-term expansion of RI) may be to extend that trust into the field of public health, but to do so in ways which are consistent with public perceptions of legitimate state intervention."
  • Settlements at higher risk of refusing OPV have systematically lower levels of households reporting participation in community meetings.

Implications of research for programme strategy are outlined. In brief, they include:

  • The PEI operational focus should be maintained (or restored) in the north-west, alongside the focus on the Kano and Yobe-Borno transmission zones. State programmes should strengthen capacity (including developing qualitative and quantitative data-gathering methods) to analyse programme performance at the settlement level to identify persistent localised gaps in SIA performance.
  • At the household level:
    • "State programmes should investigate PEI performance and RI uptake in urban and semiurban settlements to assess the extent of a new/emerging urban set of OPV/RI risks.
    • Programmes should re-balance the current focus on poor/poorly-educated, rural households, to develop capability to respond to urban dynamics of OPV, RI and wider health demand.
    • Programmes in Sokoto and Kano should conduct targeted investigations in VHR settlements (for example using 'social network analysis') to analyse how households develop and share information/attitudes to PEI/RI...
    • State programmes should strengthen information and communication on the benefits of routine immunisation as a general practice, focusing on mitigating negative perceptions (e.g. AEFI [adverse events following immunisation])...
    • State programmes should focus on building health communication between men and women within communities, strengthening shared commitment by male and female caregivers to health and education as community development priorities.
    • A gender strategy should build male-female engagement at household, but also at community participation and service-provision levels."
  • At the settlement level, "[s]tate programmes should maintain networks and relations with religious and traditional leaders to create a supportive 'background' environment of cultural norms for PEI and RI. But strategic focus and resources should be rebalanced in favour of promoting local government leadership on public health provision, and community-level engagement." For example, having found that high-risk settlements have low levels of collective, community-level activity (especially with regard to women's participation), it is suggested that state programmes rebalance the current public health emphasis on individual behaviour change through - by perhaps, the "health camp" concept, which the researchers say "may have considerable potential as a strategic intervention - primarily for PEI in the short term, but with positive cross-over effects for RI. Health camps [which deliver a range of services, including immunisation]...are an opportunity to build a community-level sense of public health as a collective activity..." and should, according to the researchers:
    • be run alongside SIAs, offering a way of accessing OPV and other antigens/health services. This intends to complement house-to-house campaign vaccination.
    • "be regular and more substantial in scale, offering a consistent, publicly desired suite of public health and nutritional interventions within which OPV/IPV [inactivated polio vaccine] are delivered. They should be situated in well-known public spaces and where possible attached to publicly-valued and trusted institutions/activities (such as education or therapeutic feeding centres)."
    • "be designed, supervised and assessed for impact - both in terms of services uptake through them and potential impact on SIA rates of missed children and refusal in areas where they are held. Metrics can be developed to assess, regularly over time to create a picture of trend, whether and how health camps modify male/female, household and community attitudes to OPV, RI, public health and service provision/providers, including local government."
    • "be led by LGA Chair and Ward Heads (with oversight from state-level traditional leadership and/or State LGA Commissioner), and enhanced technical support from [PEI] partners." The researchers elaborate: "Visible leadership by local government on delivery of health camps may strengthen households' perception of government as a legitimate and trustworthy actor in public health delivery..."

    It is hoped that the findings and recommendations shared in this report will be of value in restructuring aspects of the PEI strategy at the micro level (particularly with regard to BCC), increasing demand for vaccination in communities with low uptake, improving coverage population immunity, and sustaining the gains made through the global PEI across Nigeria.

    This research was designed and conducted under the aegis of the Nigerian National Primary Health Care Development Agency (NPHCDA) in consultation with the Nigerian Polio Eradication Programme partners, including the Federal Emergency Operating Centre (EOC) and relevant state Emergency Operations Centred (EOCs), the World Health Organization (WHO), the United Nations Children's Fund (UNICEF), the Centers for Disease Control and Prevention (CDC), and CORE. The United States Agency for International Development (USAID) supported the research. Many personnel at other organisations (including - for full disclosure - The CI), as detailed in the Acknowledgements section, shared their expertise to make the research project possible.

Click here for the 36-page report in PDF format.
Click here to access the annexes as a Word document.