Picture credit: Benjamin Hunter

Much of the USD 500 billion that was spent on development aid for health globally between 1990 and 2015 has been channelled into toolkits, technologies and policies, often facilitating their spread from one setting to another. Many of these are vital for improving health, but the effects of some policy innovations are highly questionable. Is the global health community too enthralled by the process of finding and rolling out ‘magic bullet’ interventions?

Yes, argued Jean-Pierre Olivier de Sardan and colleagues in a paper in HRPS in which they applied the concept of ‘travelling models’ to the study of healthcare. They describe how technologies and standardised interventions are exported from one setting and then introduced to another: miracle mechanisms are deduced from desirable healthcare outcomes, success stories are generated, and the models disseminated. Failure to replicate the same success in new settings (the ‘revenges of context’ to which de Sardan and Piccolo devoted their recent book on conditional cash transfers) is downplayed as a problem of ‘implementation’.

The paper is well worth reading in full as it sets out a useful framework for studying how healthcare interventions ‘travel’ globally. Take the example of healthcare vouchers – the focus of my doctoral research. Healthcare vouchers attracted a lot of attention from the global health and development community during the Millennium Development Goal era, particularly in the area of maternal and reproductive health. They were introduced in Bangladesh, Cambodia, India, Kenya, Pakistan, Uganda and Yemen.

The voucher model was actually plucked from education, where its purpose was to increase access to schools for low-income students while creating competitive markets for education provision. Its purpose in healthcare was similar. In the late-1990s vouchers were used in Nicaragua to increase access to reproductive healthcare for sex workers and to encourage competition between government and non-government healthcare providers. That programme was heralded as a success and subsequently used by donors – chiefly the World Bank and USAID – to drive the development of maternal and reproductive healthcare voucher schemes in other settings.

I examined one such programme in depth: the Sambhav scheme, which targeted families living in urban slums in Uttar Pradesh. Evaluative research commissioned by USAID was carefully framed around voucher usage in a way that generated a narrative of ‘success’. Despite flaws in their analysis – chiefly the omission of confounding factors from their statistical tests – the evaluative narrative of ‘success’ was then transmitted to other settings by voucher advocates using sympathetic case studies and international conference presentations.

There were a series of ‘implementation problems’ for the Sambhav Scheme, in particular:

• competition for healthcare users with overlapping government-funded programmes, in particular Janani Suraksha Yojana
• limited demand among hospitals to sign up to the programme
• failure of hospitals to meet programmatic obligations, for example by systematically referring-on any women likely to need more expensive forms of care
• manipulation of information by voucher distributors in ways that limited hospital choice and enabled informal payments

The miracle mechanisms in the healthcare voucher model revolve around neoliberal theories and the construction of healthcare markets; informed healthcare consumers making rational choices from amongst a pool of hospitals and thereby encouraging inter-hospital competition. The application of market theory to healthcare provision has always been known to be problematic, but is particularly questionable in a context where it is so difficult to attain healthcare entitlements and where public services are accessed through clientalist networks.

There are substantial costs involved with introducing and evaluating travelling models such as healthcare vouchers. The Sambhav scheme was part of a USD 725 million Innovations in Family Planning Services project (1992-2012) funded by USAID and the Indian federal government to promote policy innovation. Travelling models like these remain popular in the global health and development community despite questionable appropriateness in many settings; partly because they suit dominant models of policy-making, partly because they provide important income for development contractors.

The call by de Sardan et al. for greater emphasis on contextually grounded and locally led support in the health sector reflects similar calls across development more widely: William Easterly’s The White Man’s Burden and Ben Ramalingam’s Aid on the Edge of Chaos are just of couple of well known examples. But such approaches face enormous barriers to draw funding away from a development-industrial complex that has thrived on the design, introduction and evaluation of defined interventions.

I think there’s a lot of value to de Sardan et al.‘s ‘travelling models’ approach. We really need to dissect the construction of narratives surrounding healthcare interventions in order to understand and scrutinise their spread. The example of vouchers above, and those used in de Sardan et al.‘s paper, show that when subjected to close analysis seemingly strong narratives can have very wobbly foundations. There’s not enough published empirical research on the way evaluative research is generated and manipulated to suit particular narratives and interest groups. Indeed that was a recurring theme when I spoke with other delegates at the 2017 EADI conference. Luckily we’re likely to see more of the travelling models approach as the paper’s authors have a panel on it at the International Conference on Migrations, Development and Citizenship in May.