Eliciting preferences for collectively financed health programmes: the ‘willingness to assign’ approach

Abstract

Improving public involvement in health system decision making stands as a primary health policy goal. However, still limited guidance is available on how best to elicit preferences for health care programmes. This study examines a contingent choice technique to elicit preferences among health programmes, and named ‘willingness to assign’ (WTAS). WTAS reveals relative (monetary-based) values of a set of competing public programmes under a hypothetical healthcare budget assessment. Experimental evidence is reported from a deliberative empirical study valuing ten health programmes in the context of the Catalan Health Service. Evidence from this experimental study reveals that within the context of multiple programmes, preferences are internally more consistent and slightly less affected by ‘preference reversals’ as compared to values revealed from an adapted technique eliciting the willingness to pay (WTP) extra taxes. Another finding suggests that although programmes promoting health received the higher relative valuation, those promoting other health benefits were valued highly.

Acknowledgements

Financial support was obtained from CICYT SEC99-0698 and a special grant from the Escola Administració Pública (Generalitat de Catalunya, 1998). We are grateful to the participants of the Preference Elicitation Group (London, 2003), the Health Economics Study Group meetings (Birmingham, 1999 and Nottingham, 2000), and those of the XVIII Jornadas de Economia de la Salut (Vitoria-Gasteiz, June 1998). We acknowledge the work and guidance of Claudi Esteva and Berta Alcañiz in the design of the experiment. Special thanks for their valuable comments go to Can Donaldson, Phil Shackley, Alan Williams, Richard Cookson, Penelope Mullen, Peter Sandiford, Guillem Lopez–Casasnovas, Eduard Berenguer, Jaume Puig, Eulalia Dalmau and Jose Luis Pinto. Any errors are the responsibility of the authors.

Notes

It was designed in the light of a research programme aimed at finding quantitative measures to measure the monitoring of health system performance, with the specific objective of identifying priority-setting mechanisms.

At that time the health agency was the Catalan Health Service, which planned a 4000 million PTA budget increase – equal to 24 million euros.

A new programme implies an increase in health system coverage, in the sense that it improves the remaining set of benefits that the system aims to offer. However, alternatively one might consider the opposite decision-making problem, that is, a reduction in health system coverage. The problem between considering one or other decision-making context is the same, in that it may change results considerably.

We did not want to describe accurately the payment mechanism since then ‘protest bias’ could arise, we simply said that it would be collectively financed. The advantage is that this prevents strategical behaviour, since the individual participation in this healthcare ‘reform’ is not defined. The disadvantage is that there is a greater probability of a moral satisfaction bias.

One of the main features in applying this method is that it can be seen as simulating a real voting process and at the same time is a conformable exercise of assigning a budget to different goods as in the family decision-making process, even though the context of the decision-making was far from that of the individual/familiar decision regarding how to assign a monthly salary.