This piece is written by Tony Klouda, Independent Consultant and external Technical Adviser to the PRRINN-MNCH programme, and Cathy Green, Health and Social Development Consultant, Health Partners International. External technical adviser to the PRRINN-MNCH programme.
The failure of current targeting efforts
Health interventions aimed at the most difficult to reach have for many years focused on general categories of people such as the disabled, people living with HIV/AIDS, sex workers, or young people, among others. Various efforts have also been made to work with people who are vulnerable according to particular criteria: the economically poor, lower castes, those living in deprived areas, or the landless. A common strategy is to identify ‘risk groups’ and to target services and group activities to these groups.
One problem with these approaches is that the broad categories that define risk groups usually also include people who have a relatively low risk of morbidity or mortality. Two other factors undermine the claims made for broad-spectrum targeting:
– the clustering of child mortality by household in societies that are generally poor, or marginalised, and
– a growing recognition that health programmes that aim at inclusion of the ‘most vulnerable’ have great difficulties in doing so.
In light of the current emphasis within the Sustainable Development Goals on leaving no-one behind, it is pertinent to ask whether a different, more nuanced approach to targeting is needed so that governments and development partners can operationalise their commitments to universal health care.
Identifying the least-supported
Between 2006 and 2014 the Partnership for Reviving Routine Immunization in Northern Nigeria and the Maternal, Newborn and Child Health Initiative (PRRINN-MNCH) supported national, state and local governments to restore routine immunisation services and promote maternal, child and newborn health.
In a context of generalised poverty and a largely homogenous social and cultural backdrop across the study states of Jigawa, Zamfara and Yobe, the programme set out to determine whether there were differences between and within households for child health outcomes. It was hypothesised that if clustering of child mortality to a small proportion of women in such a setting was identified, the general variables commonly used to explain differences in mortality (monetary wealth, assets, pregnancy spacing, gender, education, power, employment, culture) could be discarded.
Since studies of mortality clustering had not been done in Nigeria, the first step was to establish the distribution of under-five mortality among families and households. In line with many other similar surveys, the PRRINN-MNCH survey revealed that child mortality was clustered among a small group of women: the 20% of women who had multiple child deaths suffered 80% of all child deaths. The pattern was almost identical across the three states.
These findings fit with other studies of mortality clustering in rural communities in Europe, Asia and Africa. However, the PRRINN-MNCH study did more: it explored the family and social factors associated with this clustering. The variables that had the strongest effect on the clustering were:
– the extent to which a woman had support from co-wives or relatives in the home,
– the extent to which a woman believed she was respected by co-wives or relatives, and
– the general tidiness and upkeep of the home and surroundings.
These findings add a new flavour to the claim that “inequality happens behind the front door” (ODI, 2015).
The programme took some practical steps to adjust its intervention approach in order to respond to these findings. Adjustments fell into four key areas:
– involving local leaders in improving social inclusion and traditional support mechanisms,
– training health workers to recognise under-supported women and offer relevant advice,
– developing young women’s support groups and targeting these to the women who lacked social support, and
– ensuring all surveys in the programme stratified results by social situation.
Monitoring and evaluation for social inclusion
PRRINN-MNCH demonstrated that it is feasible for health-focused programmes to adjust their intervention strategies so that they have relevance to the least-supported. Yet over the programme’s lifetime, progress in making these adjustments was consistently weakened by the need and drive to achieve logframe targets (which were not socially stratified) and value for money. In the absence of a monitoring and evaluation framework conducive to tracking progress towards social inclusion, it is unlikely that programme staff will invest the time required to appropriately address these issues.
Socially stratified data collection is key to leaving no one behind
Achieving truly universal health care means being able to identify the groups left behind and having the means to track inequalities on an ongoing basis. In the absence of such commitment, there is a danger that the vast residue of child morbidity and mortality – which affects the least supported, most isolated, and least respected – will remain relatively unchanged while better-supported individuals benefit from the changes brought by their governments and development partners. Much can be learnt from the approach PRRINN-MNCH took to look behind the standard categories used to define the hard to reach and to examine the social factors that are influential in determining health care access and outcomes ‘behind closed doors’.
Notes: The PRRINN/MNCH programme was funded and supported by UK aid from the UK Government and the State Department of the Norwegian Government. The programme was managed by a consortium of Health Partners International, Save the Children and GRID Consulting, Nigeria. The consortium included the Mailman School of Public Health at Columbia University, Ahmadu Bello University, Liverpool Associates in Tropical Health and the Health Reform Foundation of Nigeria.
 ODI, 2015, ‘Leaving no one behind: How the SDGs can bring real change’, Briefing, compiled for ODI by Claire Melamed.