Starting from scratch

The authors adopt a ‘ground zero’ approach (Bobadilla et al., Reference Bobadilla, Cowley, Musgrove and Saxenian1994), starting from scratch to design an efficient and equitable system. But this is hardly ever a real world scenario. In reality, there is usually something in place, oftentimes funded through different channels. Every donor tends to support their favourite vertical programmes targeting diseases such as HIV or providing specific types of technology such as vaccines, and, in parallel, policy makers in the country may have allocated some resources to supporting vulnerable groups such as children and pregnant women, with a basic package of care. Introducing a fair and efficient package of services may therefore require significant shifting and reallocation of resources, including the disinvestment of existing services in order to reinvest and expand access. Psychologists understand that ‘taking away’ can be more problematic than ‘not giving’ and the experience of most agencies seeking to achieve this have demonstrated that as controversial as denying approval or reimbursement for new technologies is reallocating resources away from existing technologies is even more difficult (Garner and Littlejohns, Reference Garner and Littlejohns2011).

Evidence is crucial

Our view is that debates about balancing equity with efficiency in making resource-allocation decisions (usually at the margin) can only be resolved, or (perhaps more realistically) explored, through the gathering of comprehensive information, including evidence on the clinical and cost-effectiveness of the potential interventions for addressing the underlying epidemiology of a country’s disease burden, and, most importantly, data on the impact of any decisions on patient and public health across different population groups. So often, even when the most robust systems are put in place, the consequences can be surprising unless kept under constant scrutiny (Weale and Littlejohns, Reference Weale and Littlejohns2014). In particular, evidence is required on the impact of the opportunity cost of applying equity constraints. In other words, decision makers need quantified estimates of health lost because of equity adjustments, estimates, which may make a difference to their decision. This should concentrate initially in interventions and related costs and benefits that are health and healthcare related, though the identification and consistent application of a decision rule that reflects the budgetary constraint: what is often called the ‘threshold’ (Culyer et al., Reference Culyer, McCabe, Briggs, Claxton, Buxton, Akehurst, Sculpher and Brazier2007).

Vertical equity ought to apply not only to those whose disease or condition is considered but also to those likely to lose out because of a decision not to maximise health – the opportunity cost of the unknown patient. We acknowledge that this approach represents considerable computational, methodological and informational challenges but there are also risks of not doing so or of forging ahead with focusing on the health (and rights) of those well-identified population groups at a cost to those who remain unknown (McCabe et al., Reference McCabe, Claxton and Culyer2008). Such trade-offs are often ignored even by global organisations committed to equity such as the WHO as shown in their latest (2013) HIV/AIDS treatment guidelines (Revill et al., Reference Revill, Asaria, Phillips, Gibb and Gilks2014).

Politics is important too…and inevitable

Sensitivity to the political environment in which resource-allocation decisions are made is essential in any country. Though committed to the realities of the political economy of priority setting, the authors use the word ‘should’ throughout the text which, perhaps, makes it sound all too easy to implement the authors’ recommendations or, more likely, to fail to do so. Their designation of trade-offs as ‘unacceptable’ could come across as divorced from political reality, raising the question of ‘unacceptable’ to whom and why? While acceptance by politicians is essential, getting public acceptance of the approaches taken will also be required. The impact of patient and public involvement in clinical care has increased considerably over the last few years but the equivalent in policy decisions and particularly around prioritisation decisions requires further exploration and may be worth reflecting on at greater length in the report. Such a context-specific approach has perhaps been less emphasised by WHO, as reflected in its guidance products such as the Essential Medicines List or its guidelines on technologies (e.g. vaccines and diagnostics) and on management of diseases and conditions, such as mental health or the complications of pregnancy, that tend to be developed from its HQ in Geneva by committees of experts with limited representation (understandably given the practicalities of such an endeavour) by policy makers, administrators, frontline professionals and service users from the countries where such guidance will apply. However, as countries become more wealthy and governments more accountable, such a top down approach may have to be revisited.

Calling for empirical research into what works in the ethics of priority setting

Much work needs to be done on developing process indicators (referred to on page 50 of report) as metrics of success. This is a crucial and neglected area, despite the recent WHO/WB metrics work aimed at monitoring process towards UHC, which, though a step in the right direction, unfortunately makes no attempt at defining process and institutional maturity metrics (World Health Organization/The World Bank, 2013). The report does list, though too briefly in our view, a small number of process and structure metrics. It suggests, for example, that the presence of a designated institution (p. 50) is a predictor of “fair progressive realisation of UHC”. While this sounds sensible, more empirical evidence is needed that such an organisation is required. Perhaps there can be progress without such an organisation and lack of progress despite the presence of one. Japan, for example, lacks an institution for setting priorities and has managed to attain and sustain UHC, whereas Colombia has set up a series of priority-setting institutions (IETS being the latest) but is finding it difficult to link evidence and values to policy decisions about coverage. In another example, the report cites publicity as a condition for fulfilling procedural fairness (p. 50). But is this enough? What other metrics of system maturity are there and how can we apply them or test them out using real country and regional level case studies? Given the paucity of empirical evidence in this field, the authors may have missed an opportunity retrospectively to apply some of their proposed indicators to countries such as Turkey or Thailand discussed in the report as good examples of transitioning to UHC, in order to assess their feasibility, relevance and usefulness.

A rapid and relatively inexpensive approach to gain information would be to develop a catalogue of successful (and unsuccessful) implementation case studies. For example, one could use the examples described in Box 3.4 of the report (reproduced in Voorhoeve et al., Reference Vorhoeve, Ottersen and Norheim2014) as a starting point. Despite the fact dialysis is deemed to be the last thing a decision maker would consider covering (if at all), Thailand does cover dialysis as part of their UHC (Tantivess et al., Reference Tantivess, Werayingyong, Chuengsaman and Teerawattananon2013). Coming to another country example, Turkey is listed as a country with innovative systems of accountability and participation (p. 47) but it does not have an established process for setting priorities despite their early successes in achieving universal coverage. So whether countries need to follow some high-income countries in linking transparency and with priority-setting, as the report seems to suggest, is an open question. It would be informative to think through why this is. Using the report’s proposed framework and indicators, tested against real-world examples, one could attempt to describe the trade-offs and the process by which decisions were made, as well as discuss the reasons and perhaps even their implications, where possible.

Practical solutions for exploring empirically the fairness dimension in priority setting for UHC could be explored in the context of real-world practitioner-to-practitioner partnerships aiming at addressing similar questions on trade-offs and the methods and processes for making decisions on priorities. The Thai HITAP, the agency responsible for priority setting for the UHC in Thailand and NICE, its English counterpart informing decisions on coverage of services and technologies for the National Health Service, have been working with fellow policy makers from a wide range of countries, from Brazil and Colombia to Kazakhstan, Turkey, China, India and the Philippines, to help policy makers identify and act on their own priorities in a procedurally fair and evidence-informed way. The two organisations recently launched the international Decision Support Initiative with DFID, Rockefeller and BMGF support (see http://www.cgdev.org/blog/nice-idea-priority-setting-global-health-amanda-glassman and http://www.nice.org.uk/aboutnice/niceinternational/projects/NICEInternationalLaunchesInternationalDecisionSupportInitiative.jsp) to help decision makers operating in settings with limited resources and expertise setting up the methods and processes of evidence informed policy making, using evidence adapted to their own setting linked to local values and local priorities. iDSI is about a practitioner-to-practitioner, public–public partnership, driven by demand and emphasising institutions and processes rather than imported technocratic one off solutions.

With active pilots in Vietnam, the Philippines, China, India and Myanmar and new ones starting in South Africa and Indonesia, there is considerable scope for joint working to build, together with our colleagues from Ministries and Health Insurance Funds from around the world and start addressing, an empirical research agenda for the ethics of priority setting.

A way forward

We have identified a significant requirement for data collection and research. Unfortunately this type of trans-national policy-orientated research rarely can attract significant resources and new systems need to be put in place. There are national examples of where health policy initiatives have linked to major research funders to address specific needs e.g., NICE in the United Kingdom (Longworth et al., Reference Longworth2009). The importance of understanding how UHC can be best achieved should encourage the WHO to work with major international and national research funders to identify and commission an ongoing research programme to learn and share the lessons for UHC roll out.