Despite billions of dollars spent on childhood immunisation, some countries have never reached universal childhood immunisation (UCI), and many more have been unable to sustain it. An estimated 1.5 million deaths under 5 years of age can be prevented by vaccination each year, measles making up 37% H. influenzae b-related disease 30%, and pertussis 19%[1]. In many countries, there is dramatic underutilisation of the offer of free immunisation. One reason for this is a difference between understanding of the costs and benefits of vaccination in international public health circles, and what primary decision-makers for children know about immunisation and its costs.

There is a gap between the way public health specialists understand immunisation benefits, and the cost-benefit equations that household decision-makers apply to their children's immunisation. In settings like Pakistan, where an expanded program of immunisation is offered free of charge, uptake is largely determined by access to services and the attendant cost-benefit assessments by parents and caregivers. These assessments could be influenced by access to knowledge and conditioned by gender and social inequalities.

Although communication of risks and benefits by service providers can influence health-seeking behaviour [2], current approaches to health communication do not always achieve the expected results. Efforts frequently produce an increase in knowledge without a corresponding change in attitudes or behaviour[3]. This is at least partly because conventional risk communication presumes to inform an uninformed public and to reduce irrational thinking. Questioning the value of these conventional one-way knowledge transfer (KT) initiatives, more holistic perspectives take account of social and cultural influences[4]. Impact studies of communication strategies to increase vaccination in the USA, Russia and Mozambique all highlight the need of multi-channel targeting of multiple groups – families, communities, health practitioners and opinion makers, such as community and religious leaders[5–7].

Current devolution reforms in Pakistan are expected to make service delivery more effective and to support institutionalised participation of community members[8, 9]. Service delivery is now a district function, yet the provinces are still responsible for planning and monitoring health services [9]. The expanded programme of immunisation (EPI) remains a federal responsibility. The 2002 Pakistan Integrated Household Survey claimed an increase in immunisation coverage, mostly in the urban areas. The proportion of fully immunised children aged 12–23 months rose from 49% to 53%, and even this small gain was not consistent countrywide. In Balochistan, where the project will be located, immunisation coverage fell from 34% in 1998–9 to 24% in 2000–1, mostly attributed to diversion of resources to the polio eradication campaigns[10].

Balochistan is Pakistan's largest province (347,000 km²) and has the lowest population density (6.5 million, 19/km²). It covers more than 40% of the land area, but has less than 5% of the national population. The disease burden in Pakistan is still largely poverty-related, much of it preventable by immunisation[11, 12].

Lasbella, in the south of Balochistan is one of the province's poorest districts. In 2003 there were around 50,000 women of childbearing age in this district. There were 11,594 infants under the age of one year, and reflecting the very high infant mortality, 8,140 children aged 12–23 months. The new district administration employs 39 full time vaccinators, but still only achieves symbolic coverage. The problem does not appear to be one of supervision or management. The vaccinators report to a superintendent, who in turn reports to the Deputy District Health Officer, who reports to the Executive District Officer for Health.

We believe that health care consumers make rational decisions from a cost-benefit perspective[13, 14]. Based on their own knowledge, mothers and guardians weigh up the costs and the benefits of immunisation: how much time will it take, how much will it cost, will it work, will there be side effects and what will happen if I do not go. We propose that immunisation uptake is largely determined by this cost-benefit equation.

But the rationality of such decisions is a "bounded rationality" [15]. This concept enables health planners to see such factors as social norms and attitudes not as obstacles to rationality but rather mechanisms that facilitate fast and accurate decisions, and more efficient learning. In addition, the weigh-up of risks may be sharply discounted between the time of the immunization and the time of possible infection. Also, gender and poverty probably affect the household cost-benefit equation. The poor, who typically have less access to services and less information about services, almost certainly weigh up the costs and benefits in a different way than do the rich [16]. Diseases like measles and pertussis may be an inconvenience for the well-nourished, whereas for the malnourished, they can be a question of life or death. Costs of not vaccinating (disease burden, care and funerals) are borne disproportionately by the poor; in a single epidemic, these diseases can destroy a household economy[17].

Engagement of communities in evidence-based planning is a novel approach to immunisation, an intervention driven perhaps more than any other by an inappropriate if well-intentioned ethos of "we know what is good for you". The idea of a household cost-benefit equation is common sense: people weigh things up before their health choices. This study hopes to answer questions about what it takes to enter a dialogue that influences these equations. The year-to-year shift in knowledge, attitudes, subjective norms, intention, sense of agency, ability to discuss, and ultimately, uptake of immunisation will be evident from the results of the successive cycles.

Best practice cases – communities that increase immunisation – will be identified and held up as positive examples of what is possible under prevailing conditions. The direct beneficiaries will be the public, but indirect beneficiaries will be planners and policy-makers in federal, provincial and district governments. Improved understanding of immunisation risks and cost implications of higher immunisation uptake may be applicable in many other countries. Proof of cost-gains as the planners' corollary of the household cost-benefit equation could open new horizons for evidence-based service-public interaction.

NA designed the study, developed the methodology and wrote the proposal. NA designed CIET map freeware, included in the proposal to combine mapping and epidemiology for decision makers. He is responsible for overseeing the project.

AC reviewed the protocol and is responsible for overseeing the project in Pakistan.

NoorA conducted the pilot studies, contributed to the instrument design and supervises the fieldwork in Balochistan.

KS coordinates the project in Pakistan.

JL reviewed the protocol and contributed to the overall study plan.

RJL co-developed the concept of household cost-benefit equations, and contributed to the proposal and development of instruments.

PT reviewed the protocol and will contribute to the systematic review.

BS helped write the proposal and is responsible for the systematic review.