Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Demographic and health surveys (DHS) are nationally representative surveys which provide important information on people’s lives and health at a community, region and national level. From these surveys we can understand mortality data, patterns of health and social practices and identify needs and priorities, and track progress over time. DHS have been conducted in 90 low and middle income countries, and since 1984 the DHS Programme, funded by the United States Aid for International Development, and United Nations partners, has done over 300 DHS.1 DHS are especially important in countries where vital registration (registering of all births and deaths) is limited. In many countries DHS, or related multiple indicator cluster surveys (which are similar to DHS), are the only sources of population based health data.
The questions asked in DHS are mostly standardised, and there is scope for questions that differ from country to country, depending on the country context and disease burdens. All DHS report data on child mortality, with estimates of under-five, infant and neonatal mortality rates. They provide data on maternal and reproductive health, including antenatal clinic attendance, proportions of births that involve a skilled birth attendant, location of birth and maternal mortality. They provide data on vaccination coverage, infant feeding practices, nutrition, stunting and anaemia, and education participation and attainment.
The DHS asks whether symptoms of three common illnesses: acute respiratory infection, diarrhoea and fever have occurred in the previous 2 weeks, a short duration to ensure that parents have accurate recall of events. The DHS records how parents seek care for their children these symptoms, and what basic treatment they received. From this, DHS can give some idea about infectious disease prevalence, but this is subjective. For example the symptoms aimed at identifying pneumonia may not distinguish between upper respiratory tract infections and pneumonia. The numbers of children experiencing the symptoms in the previous 2 weeks are often small, limiting the precision of prevalence estimates.
DHS provide invaluable data about the gradients in health access and outcomes. DHS data are disaggregated for wealth quintiles, for rural or urban living, for levels of maternal education and for gender. Thus it is possible to understand the social, economic and geographical gradients in health. Typically children from families who are poor, rural or whose mothers are poorly educated have less access to all basic health services, and worse health outcomes. Similarly if a mother is urban, in a higher wealth quintile, or more educated, she is much more likely to deliver in a health facility than if she is rural, poor or uneducated, and more likely to have her children vaccinated, and seek care if they are ill.
The summarised reports and data sets from DHS are available online from the DHS Programme, https://www.dhsprogram.com/ and this can enable more complex analysis of associations between outcomes and characteristics. Although metrics of some outcomes have differed a little over time, in countries where there have been multiple DHS it is possible to compare progress over time.
In recent years DHS have asked questions about experience of domestic violence. In this edition of Archives, Michelle Nakphong and Ondine von Ehrenstein report an analysis data from three DHS in Cambodia over 14 years, looking at the association between mothers’ experience of domestic violence and the prevalence of acute illness in their children. The analysis involved more than 5000 children whose mothers reported emotional, physical or sexual violence by their current partner. The authors found that a mother’s experience of domestic violence increased the risk of acute respiratory infection and diarrhoea 2.6 times (95% CI 2.01 to 3.51).
Finding these associations shines a light on an important issue, and raises some questions. Is there a causal relationship or is the analysis affected by unmeasured confounding, or recall or reporting biases? If it is causal, what are the reasons? Are the reasons environmental or biological? The answers to these questions are not all discernible by analyses of population-based data like a DHS; other methods of research are needed to understand this, including triangulation with other similar research. The link between domestic violence and poor physical health in children is highly consistent with other research. In other countries where this has been studied through DHS data, children of mothers who have suffered domestic violence or other forms of abuse are at an increased risk for poor physical health outcomes, as well as poor psychological, social and emotional development and well-being.2 Mothers exposed to domestic violence are less likely to receive antenatal care in the first trimester of their pregnancy if they have experienced violence, and are more likely to have a stillborn child. In seven of nine countries, children of mothers who have experienced domestic violence had higher under-five mortality rates.3
Analysis of DHS data is an important way to track progress towards the Sustainable Development Goals (SDGs), because such analyses can link outcomes and variables across many of the Goals: health, nutrition, gender equality, water and sanitation, education of mothers, employment, life in rural areas and cities, poverty and reducing inequalities.4
Archives is interested in publishing more analyses of DHS data, to understand how countries are progressing towards their SDGs and the effects of reforms and improvements in child health, paediatric services and across other sectors. We seek submissions from paediatricians and scientists in countries where a recent DHS has been done, to report on the results in the context of their child health programme and national priorities, and the way in which these problems are being tackled.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Commissioned; internally peer reviewed.