How soil data helped reveal the importance of iodized salt for human health and development

In the late-2017 and late-2018 harvest seasons, teams working for the GeoNutrition project walked out into farmers’ fields and grain stores across Amhara, Oromia and Tigray regions of Ethiopia, and collected 1,352 soil and grain samples from 385 districts. The goal at the time was to map how micronutrient levels in Ethiopian food systems vary from place to place. Nobody involved could have known that, several years later, that same soil data would become the key piece of evidence in a very different story: explaining why an entire generation of Ethiopian children grew up to score lower on their school-leaving exams, and why some of them didn’t survive early childhood at all.

That story is told in a new study involving Rothamsted scientists, which is a striking demonstration that soil data, once collected and shared, can go on answering questions its collectors never anticipated.

A fortification programme, cut off overnight

In May 1998, war closed the border between Ethiopia and Eritrea. At the time, Ethiopia’s entire supply of iodized salt came from a single industrial facility across that border, and the closure cut it off completely. Household use of iodized salt collapsed from 80% of households in 1995 to just 4% by 2005. It took until 2012 for the government to reinstate mandatory iodization, and until 2015 for coverage to climb back above 90%.

It was a public health emergency at the time - but for researchers looking back at it now, it also provided a natural experiment at national scale, showing the impacts on human health and development when a salt iodine fortification programme fails.

Where the GeoNutrition data came in

After the onset of the war, the coverage of iodized salt fell rapidly across the whole country. To understand the impact of losing iodized salt – and to separate this from everything else going on, war included – requires some sort of variation in population iodine status. 

That’s where the GeoNutrition soil and crop survey came in. Soil iodine occurs naturally at very different concentrations across Ethiopia’s farmland - organically bound iodine averaged 10.18 mg/kg across sample sites but ranged from 1.19 to 23.05 mg/kg. Mapped across all 385 districts, the resulting picture (below) shows iodine-rich pockets concentrated in western and eastern Oromia, against persistently low levels across much of Tigray and eastern Amhara - a pattern shaped by underlying geology and geochemistry. Soil selenium, sampled and measured alongside iodine, showed a quite distinct spatial pattern.

Where iodized salt disappeared, households in naturally low-iodine districts lost their only meaningful iodine source; households in naturally higher-iodine districts still had some backup from local water, pasture and food. Because cereal grains take up very little iodine from soil, the team didn’t use grain iodine concentration directly as their exposure measure. Instead, soil iodine served as a proxy for the broader environmental iodine reaching a household through diets. The researchers found that children living in districts with higher soil iodine had measurably higher urinary iodine concentrations during the period of low iodized salt coverage. 

What the combined data revealed

The researchers then compared exam and survival outcomes for children born around the onset of the 1998 disruption. Drawing on exam records for over one million young Ethiopians who sat the national university entrance exam between 2003 and 2019, they found children born after the disruption scored 0.02–0.04 standard deviations lower than those born just before, with the deficit rising by a further 0.03–0.09 standard deviations for every 1 standard deviation reduction in local soil iodine. Using Demographic and Health Survey data for 8,185 rural children, they found cumulative under-five mortality reached 25–30% in the lowest-soil-iodine districts born after the disruption, against under 10% elsewhere.

To rule out other explanations, the team reran the whole analysis with soil selenium in place of soil iodine (a nutrient unaffected by the 1998 disruption) and found no significant link. They also restricted the analysis to districts around Addis Ababa where food is not necessarily grown locally, and found the effect of local soil iodine was diminished, which strengthens confidence that the findings were not an artefact of the data or the method.

Edward Joy from Rothamsted Research commented, "Soil maps aren't something you'd normally think to combine with school exam records or child mortality data — but that combination is exactly what let us isolate iodine's effect from everything else going on in Ethiopia at the time."

Why this matters for how we think about environmental data

Nature Food’s associate editor, Fiona Coleman, commented “The benefits of salt iodization cannot be taken for granted… These findings underscore the importance of resilient food systems and sustained public health interventions in safeguarding nutrition, health and human potential.” 

The GeoNutrition soil and grain data used in this study was never collected with iodine fortification policy in mind. It was an effort to understand baseline dynamics of micronutrients in soils and crops. Martin Broadley from Rothamsted Research commented, "This study is a good example of what's possible when soil science and public health research are brought together. Naturally occurring soil iodine turns out to be a powerful lens for understanding population-level nutrition risk, well beyond what we originally used it for."