Consultancy to support development of a measure of vulnerability for children

Terms of Reference

  1. Background

At the United Nations General Assembly Special Session (UNGASS) on HIV and AIDS in June 2001, 189 Member States adopted the Declaration of Commitment on HIV and AIDS, including time-bound targets to be achieved by 2010. The Declaration reflects global consensus on a comprehensive framework to achieve the Millennium Development Goal of halting and beginning to reverse the HIV epidemic by 2015.

The indicators initially set out by UNGASS to measure children affected by HIV and AIDS and incorporated in major survey programs such as MICS and DHS since 2005 included: the ratio of school attendance between orphans and non-orphans aged 10-14 and, the percentage of orphaned and vulnerable children under 18 whose households received free, basic external support in caring for the child. The first indicator was developed on the assumption that AIDS would lead to increased orphans who will be less likely to attend school than non-orphans. The external support indicator was intended to measure progress in meeting the care and support needs of children affected by HIV and AIDS. This indicator, however, has been difficult to interpret in terms of benefit or outcome due to a lack of clarity and agreement on how to define a vulnerable child, and on what a minimum package of services entails.

According to the current definition, a child made vulnerable by HIV and AIDS is one below the age of 18 and:

  • Has lost one or both parents, or
  • Has a chronically ill parent (regardless of whether the parent lives in the same household as the child), or
  • Lives in a household where in the past 12 months at least one adult died and was sick for 3 of the 12 months before he/she died, or
  • Lives in a household where at least one adult was seriously ill for at least 3 months in the past 12 months, or
  • Lives outside of family care (i.e. lives in an institution or on the streets)

While the first four criteria were to be obtained from household surveys, “living outside of family care” was to be obtained from other data sources,

In response to the need to understand the utility and relevance of the existing definition of vulnerability, UNICEF supported multivariate analyses of survey-based data from DHS and MICS in 2008-2009 to explore whether these markers of child vulnerability are useful in differentiating outcomes for children in different developmental stages, with 3 main objectives: namely wasting[1] among children aged 0 to 4, school attendance among children aged 10 to 14, and early sexual debut among adolescents aged 15 to 17. The analysis explored three questions:

  • Are orphans more likely to have worse outcomes than non-orphans?
  • Are children living in households with chronically ill or HIV-positive adults more likely to have worse outcomes than children not living with chronically ill or HIV-positive adults?
  • Aside from orphaning and adult illness in the household, what other factors are associated with poor outcomes for children?

The results showed that being an orphan or co-residence with a chronically ill or HIV-infected adult do not consistently identify children with worst education, health or protection outcomes (wasting among children aged 0 to 4 years, school attendance among children aged 10 to 14 years, and early sexual debut among adolescent boys and girls aged 15 to 17 years). Other factors such as the wealth status of the household (as measured by wealth quintiles), the relationship of the child to the caregiver, and the education level of the adults in the household, particularly the female caretaker, have a strong association with children’s well-being. Of all markers of child vulnerability analyzed, only household wealth consistently showed power to differentiate across age-disaggregated outcomes.

The analysis also showed no correlation between a person being chronically ill and their HIV sero-status. In all countries which conducted HIV testing population surveys, majority of HIV infected people (70-90%) were not chronically ill at the time of the survey. So the use of a proxy indicator of chronic illness and/or survival status of parents or adults in the household is nebulous and seems not to be a realistic method of identifying households or children impacted by HIV.

These findings raise the question of whether the current definition of vulnerability really captures the full proportion of vulnerable children affected by AIDS.

Since then, there have been developments on two important fronts:

(1)Questions that were used to identify “vulnerable” children, according to the definitions above, were excluded from the standard questionnaires of MICS and DHS, on the grounds that the data did not capture vulnerability, and that questions were too difficult to ask and were ambiguous.

(2)There has been growing interest to identify “vulnerable” children not only for the purposes of the HIV/AIDS analysis and programmes, but also in the areas of social protection, independent of HIV/AIDS concerns, for the identification of households eligible for cash transfers and other forms of external support, as well as in the areas of child protection and social policy.

As a response to the evidence, the 2008 communiqué issued by the Global Partners Forum called for AIDS-sensitive and not AIDS-specific interventions and highlights social protection as one of the key actions to accelerate support for children affected by HIV and AIDS. The implications of this shift are still unclear in terms of monitoring and evaluation. It is still widely recognized, however, that consistent monitoring is needed to fully understand how the HIV response is impacting on children in issues of health, education and child protection, and to guide programme and policy responses.

There is, therefore, a need to identify standard measures of child vulnerability to enable development of indicators for (1) measuring global coverage of care and support to children affected by HIV and AIDS and to assess progress and identify gaps in the HIV response, and (2) to enable well-designed programmes for external support to households with vulnerable children, irrespective of the influence and consequences of HIV/AIDS.

It follows from these that in this study, vulnerability should be considered as a concept which embraces susceptibility to elevated risks in poor health, nutrition, HIV./AIDS, child protection, and education outcomes.

  1. Objectives

To support the validation of the first round of the further analysis on child vulnerability using different statistical techniques to guide the development of a standard measure of child vulnerability that is both AIDS-sensitive and can be used globally in different contexts.

The specific tasks for the consultant include:

a)Use the draft framework developed and draw on the preliminary results of the first stage of the analysis, to implement further in-depth multivariate analysis of 16 selected survey data (with or without HIV testing), collected in countries experiencing high, low and concentrated HIV epidemics, and validate results of the preliminary analysis to identify variables consistently and strongly associated with poor child outcomes; namely school attendance, birth registration, child labor, early sexual debut, early marriage, immunization status (DPT3), fever treatment, sleeping under insecticide treated bed net (ITN), and wasting. The countries include: CAR, Ghana, Haiti, Kenya, Malawi, Mozambique, Rwanda, Sierra Leone, Swaziland, Tanzania, Uganda, Zambia, Zimbabwe, Cambodia, Tajikistan, and Vietnam. Develop a measure for defining child vulnerability based on the multivariate results (e.g. using principal components analysis or other appropriate statistical methods)

b)Document the methodology used to arrive at the definition of vulnerability in a concise and simple to read report

c)Meet and report to Statistics and Monitoring Section and HIV/AIDS teams on a regular basis (on a mutually agreed frequencies) to review and discuss the analysis

d)Support the presentation and facilitation of discussions around the methodology, process and findings of the analysis within UNICEF and at key stakeholder meetings to gain consensus on the proposed definition that would cater for programmes in the areas of HIV/AIDS, health, nutrition, education, social protection and child protection.

  1. Expected Deliverables

a)Stata or SPSS syntax and output files, and recoded datasets used to run the final regression models

b)A report documenting the coding and definition of variables in the analysis, and the analytical techniques used

c)A full set of tables presenting the results of univariate, bivariate and multivariate analysis with appropriate titles

d)Revised draft report containing the revised result of the analysis and a definition of vulnerability to guide development of care and support indicators for children affected by AIDS and for children at risk of poor health, child protection, nutrition, education, and other outcomes

e)Presentations at UNICEF and key stakeholder meetings

  1. Qualifications and Professional Experience Requirements:
  • Postgraduate degree or training in statistics, epidemiology, public health, demography, or a related social science and monitoring and evaluation field
  • At least 3-5 years of experience in the use of statistical applications (e.g. Stata, SPSS, SAS) to examine relationships in health, population, demographic and other international development phenomena and other socio-demographic and environmental variables
  • Previous experience in performing analysis using MICS and/or DHS data preferred
  • Previous experience in reviewing and developing indicators, preferably indicators on HIV-related care and support or orphans and vulnerable children desirable
  • Analytical, methodical and precise style of writing
  • Excellent interpersonal and presentation skills
  • Native-level command of English required
  1. Duration and Fees: 30 working days starting August 1, 2011.

Payment will be commensurate with the consultants experience and qualifications and be made based on satisfactory submission of deliverables (see above).

How to Apply:

Qualified candidates are requested to submit a cover letter, CV and P 11 form (which can be downloaded from our website at to by July 8 2011 with subject line: Consultancy to support the development of a measure of vulnerability for children. Please indicate your ability, availability and daily/monthly rate to undertake the terms of reference above. Applications submitted without a daily/monthly rate will not be considered.

.

[1] Wasting among young children is defined as those aged 0-59 months who are below minus two standard deviations from the median weight for height of the NCHS/WHO reference population (WHO 2006).