"It can't happen. Governments don't want to do it and it's not going to happen."
This was just one reaction to the suggestion of breaking down health data by ethnicity at a discussion at the Women Deliver conference in Kuala Lumpur. The side event on equity, "Ensuring Every Mother and Newborn Counts", was organised by Health Poverty Action and two other organisations who also take a broad and equitable approach to maternal health, Save the Children and WaterAid.
It was certainly a lively debate. Aluiso Barros, from Pelotas University in Brazil presented some of the Countdown 2015 data on maternal and infant mortality, as well as some of the inequities that exist between income groups, for example in skilled birth attendance and antenatal care coverage. Hkawn Yi, a community midwife with Health Poverty Action's Myanmar programme, described the realities of cultural, geographic and financial barriers facing pregnant women in her Kachin community. She linked this to the call for data to be disaggregated by ethnic group so we can see the inequities and set about tackling them.
So why such vociferous reactions to the call for data to be disaggregated by ethnicity?
Partly this reflects the practical difficulties in collecting and analysing data at a sub-national level in parts of the world where it is challenging enough to undertake surveys that simply capture national averages. Joy Lawn, Professor at the London School of Hygiene and Tropical Medicine and Director of the Saving Newborns programme at Save the Children, who was chairing the event, commented that it costs around $1 million or more in bigger countries to undertake one Demographic Health Survey (DHS) which is why they only happen every few years. So there is a real issue of affordability in collecting and using the necessary health and poverty information.
There is also a challenge of which groups to identify and count. At Health Poverty Action we believe it should be determined through a participatory process at the national level, so that particular groups such as pastoralists in the Horn of Africa, or indigenous groups in Cambodia, can be specifically targeted according to need.
But the main challenge is one of political will. Clearly with the will, the funding could be found to develop existing national surveys like the DHS so that they collect information on ethnicity and then analyse the health data against this ethnic breakdown. In recent years increasing numbers of health surveys disaggregate by income, gender, urban/rural divide and even region. But for ethnicity to be added to this list, a much greater political will must be galvanised. Reluctance from decision makers may well have its roots in genuine concerns about stirring up ethnic tensions, particularly in countries with historic problems in this area. Or it may well be that politicians don't want to demonstrate how marginalised certain groups are, and how inequitable their societies are.
But this isn't good enough. While it may be appropriate to use proxies such as region in some highly sensitive contexts, in most countries disaggregating data by ethnicity is a prerequisite for exposing the realities of ethnic inequities in health and addressing the health rights of the most vulnerable in a targeted, culturally appropriate way.
Health Poverty Action will soon publish a report on the urgent need for health data to be disaggregated by ethnicity, to download the report visit www.healthpovertyaction.org
Sarah is blogging from Women Deliver @SarahKEd