UCL Grand Challenges
- Call for proposals: Science and technology workshops as part of the joint UCL–French Embassy partnership >> Details
- UCL Grand Challenges Student Fund: Support for students with bright ideas to tackle some of the world’s big issues
- Watch a video introducing UCL Grand Challenges and UCL Public Policy
- Closing the Gap: Aligning strategies towards sustainable resource use – Report addressing the challenges facing more sustainable resource use, convened by UCL Grand Challenges and the UCL Institute for
Sustainable Resources, and supported by BHP Billiton
- In the run-up to Chinese New Year 2013, the UCL China Centre
for Health & Humanity will be showing four recent Chinese films
related to the UCL Grand Challenges, presented by three film specialists. >>Details
Are We Underestimating Childhood Malnutrition? A mathematical model to examine current methods of estimating prevalence in surveys
GCGH Theme: Vulnerable Populations
Lead: Dr Sonya Crowe (UCL Clinical Operational Research Unit)
Main Collaborator: Dr Andrew Seal (UCL Centre for International Health & Development)
Project: Childhood malnutrition is a global public health problem with serious consequences for individuals and societies. Population surveys are used to estimate/monitor malnutrition prevalence and guide resource allocation for treatment programmes. The measured prevalence is therefore a critical statistic.
We have recently noted that estimates of malnutrition prevalence are strongly affected by ‘cleaning criteria’ applied to raw data. These are used to exclude extreme values that might represent measurement or data errors. However, they can also exclude children who are, in reality, very small (or large).
Cleaning criteria are not standardised and the impact of different criteria on prevalence estimates is unknown, so comparisons of surveys that adopt different criteria can be misleading. Given how critical survey results are to both policy and research, this is a significant problem that leads to inconsistent and potentially inappropriate implementation of malnutrition treatment programmes.
We will use existing datasets to develop a mathematical model to quantify how different data cleaning criteria affect malnutrition prevalence estimates. This will enable researchers and policymakers to quantify how malnutrition prevalence obtained using one set of cleaning criteria would have differed had another been used. This information will facilitate improved decision-making in research and policy.
Page last modified on 10 jan 12 15:57