AMA Journal of Ethics^®

Imagine being a doctor with two patients and one pill. The two patients are similar in age, sex and other characteristics. Both suffer from the same disease that will be deadly unless they can take that pill. To whom would you prescribe the pill? There is no easy answer. You may be tempted to give each patient half a pill, even though you know that half a pill is unlikely to cure either of them.

Having to make such choices is not so rare in either clinical medicine or public health, but usually the choice is disguised. For instance doctors may provide treatment to some groups of patients and not to others because of the treatment cost or the patient’s age. Or we let one principle prevail over another, as happens when making the trade-off between equity and efficiency in deciding what services to offer people who live in very remote areas.

Value judgments, of which we may not necessarily be conscious, assist us in making choices between potentially conflicting principles. One recent example can be found in the humanitarian field, where the principle of strengthening local health systems and so promoting equity was at odds with the principles of humanitarian assistance. This happened in the far north of Uganda, in a region called the West Nile.

The West Nile region is situated in the northwest corner of Uganda, bordering the Sudan. Three of its districts, Arua, Moyo and Adjumani are the long-term home to more than 120,000 refugees from the Sudan. The first refugees arrived in 1986, with subsequent waves in 1988 and 1993-1994. The government and people of Uganda generously offered them asylum and allocated pieces of farmland. The refugees live in disparate settlements apart from the host population. They make up a substantial proportion of the total population in the three districts, from 6 percent in Arua and 17 percent in Moyo to 32 percent in Adjumani [1,2].

The United Nations High Commission for Refugees (UNHCR) was mandated to provide protection and assistance to these refugees, including provision of health services. With additional funds raised from the international community, the UNHCR fulfilled this mandate by contracting nongovernmental organizations (NGOs) to provide health services in the settlements.

Nowadays, any evaluation of assistance to refugees makes use of the Humanitarian Charter and Minimum Standards in Disaster Response, as formulated by the Sphere project and usually referred to as the “Sphere Standards” [3]. The Sphere project was set up in 1997 by a group of humanitarian agencies to build broad consensus around a set of minimum standards that should be applied to assistance for people affected by calamity or conflict. As expressed in the Humanitarian Charter, these standards are “rights-based,” deriving from international humanitarian law, international human rights law and refugee law. One intent behind their formulation was to create standards against which agencies that had financial and other resources and unfettered access to the affected populations could be held accountable.

The Sphere standards set minimums for how much water a person should have access to per day (15 liters), how much food (2,100 kilocalories [kcals] per person per day), and what essential health services must be available.

Overall, the health service provision in Uganda is seriously constrained, particularly so in remote areas like the West Nile. In addition, the health system operates in a context of severe poverty, human resource shortages and lack of capacity.

The UNHCR and partner NGOs have run the programmes for refugees in West Nile for many years, and the Sphere standards are, we may assume, more or less adhered to for the people being served.

There is some published evidence of the effect of these operations on the health services for refugees that compares them with those available to the host population. In an article published in the Lancet, Orach and colleagues report on the differences between hosts and refugees by measuring “unmet obstetric need” [2]. Tracing the number of major obstetrical interventions for absolute maternal care needs provides a good indicator of how well obstetrics needs are being met. The results showed that rates of major obstetric interventions were significantly higher for refugees than for the host population living in the same areas as the refugees. This was also reflected in a separately measured lower maternal mortality rate among the refugees compared to the hosts.

Thus, what exists is a situation where refugees have better health services than their hosts. This does not mean that the refugees are being provided with extravagant health services; agencies are just meeting the rights-based minimum standards as formulated and promoted by the international humanitarian community. Rather, it means that the hosts have health services that clearly fall below these minimum standards. We have come across similar situations in other low-income countries, and it is clearly undesirable, not least because it may engender resentment among the host population.

There is no easy solution. Sometimes it is argued that health services for refugees should be integrated into national health services. Two reasons for this are given. One is the equity argument, which is that all people in equal need should receive the same health services. The other is that the additional resources provided for refugee assistance could be used to strengthen the pre-existing local health services, so all would benefit. Where resources are adequate and capacity exists, this may be feasible, as it was in Guinea [4], but most often integrating refugee services means spreading the available additional resources amongst the entire population, resulting in little net gain for all and nonadherence to the minimum Sphere standards. In other words, this amounts to breaking the pill in half, to go back to the thought experiment at the beginning. This is currently the case in many refugee situations, and should be rectified through additional funding and the adoption by the development community of standards similar to those of the Sphere project.