05 January 2015

Can the 2025 WHA stunting target be met? Yes.

One of the less highlighted panels in the Global Nutrition Report is Panel 2.4 by Lisa Smith and me.

We downplayed the panel because it was based on an article that had not yet been published in a peer reviewed development journal.  Well now World Development (Smith and Haddad 2015) has published it (here).

I have blogged on the paper before when it was an IDS Working Paper, supported by funding from Transform Nutrition, with the findings of the paper summarised here.  

Rather than go over the results I want to highlight the panel from the Global Nutrition Report.  The panel (below) uses the econometric results from the paper to understand how realistic it is to meet and exceed the WHA stunting target.

For those of you who do not have the time to go through the text below the bottom line is that (a) we can exceed the WHA stunting targets by meeting challenging but realistic targets for the underlying determinants, (b) many countries have already met these targets for the underlying determinants (see footnote 2), (c) the biggest increases in rate of improvement of underlying factors needs to come in sanitation and women's empowerment and (d) all of the projections assume no scale up of the vital nutrition specific interventions--if these are scaled up then the picture becomes even more positive.

The nutrition community needs to think positively about meeting the WHA stunting target. We are not so far off meeting it with business as usual.  We need to step up our efforts.

Panel 2.3 in Global Nutrition Report:  Can improving the underlying determinants of nutrition help meet the WHA targets?

Using a multi-year sample covering 1970 to 2012 and 116 developing countries (accounting for 96% of the developing world’s population), panel regression estimates were generated for the associations between stunting prevalence and six underlying determinants of stunting:  access to improved water, access to improved sanitation, female secondary school enrolments, gender equality as represented by the ratio of female to male life expectancy, dietary energy supply, and the share of the dietary energy supply derived from non-staple foods (Smith and Haddad, 2015). The regression model (1) can be used to predict stunting prevalences based on different assumed levels of the six underlying factors.   The first table describes three scenarios for the 116 low and middle income countries.  Scenario 1 assumes current (2010) levels of the six underlying drivers, and these predict a stunting prevalence virtually identical to the actual stunting prevalence (29%).  Scenario 2 increases the six underlying determinants from their 2010 levels to those that together achieve the WHA stunting target of 15% by 2025.  Scenario 3 leaves predicted stunting at below 10%.

The levels of determinants are challenging, but several low and middle-income countries have attained them already (2).  What about the rates of increase in these determinants?  The rates of increase in the six underlying determinants over the 2000-2010 period are given in the table below.  The rates of increase needed to achieve Scenarios 2 and 3 are also presented.

Except for equality of life expectancy, the required rates of increase in the underlying factors to meet Scenario 2 are not unrealistic compared to recent historical performance.   The required rates to achieve Scenario 3 are larger than they have been in recent years.  However, this analysis fails to take into account any scale-up of the critical nutrition-specific interventions.  If these are indeed scaled up, and their effectiveness improved, the potential to exceed the WHA target at the global level is significantly improved.

(1) The regression model (6 proxy underlying determinants, country dummy variables and time period dummy variables) predicted much of the variation in stunting and was stable over time (pre and post 2000).

(2) For access to improved water source, Brazil (2006) had a level of 95%; Viet Nam (2011) had access to improved sanitation facilities of 76% and Sri Lanka (2006) had 89%.
For female secondary school enrolment South Africa (2008) posted a rate of 95% in 2008. 
For the ratio of female life expectancy to male rates, Indonesia (2004) was 1.06, Honduras (2005) was 1.07, and the Central African Republic (2010) was 1.077. 
For dietary energy supply, Ghana (2011) was 2934. 
Finally for the percent of dietary energy supply Guatemala (2008) was 49%, Brazil (2006) was 64%, Thailand (2005) 50% and Swaziland (2005) 45%.

No comments: