iiinova — where your digital self keeps evolving

Btw the data from the paper is available here: dx.doi.org/10.5281/zenodo… The paper is open access online. Thanks to all my coauthors for their contributions! @julianajaramilloe.bsky.social @gregorigv.bsky.social @evanrobertsnz.bsky.social @kris-inwood.bsky.social

dx.doi.org

062

13/ Implication for today’s WASH debates: Sanitation infrastructure matters—but it may not be enough. Reducing stunting likely requires: • behaviour change • increasing the salience of child stunting Curious how others working on WASH think about this.

000

12/ Child stunting is not as salient as mortality as a health indicator because it is difficult for individuals to understand. Stunted children look normal relative to their peers in LMICs. They are only short compared to healthy children, which people cannot observe.

11/ So why is changing WASH behaviour so difficult today? We speculate that high levels of child mortality in the past provided very strong incentives for people to adopt better WASH behaviours. However, child mortality is far lower today, weakening these incentives.

111

10/ The cases of Britain and Jamaica emphasise how important behaviour change is to making WASH interventions effective. Infrastructure and technology are not enough. People have to use the infrastructure and adopt the technology.

100

9/ The result? Communities began building improved latrines themselves and sanitary behaviour changed widely. Infant mortality fell by 41% in the 1920s–30s. Similar sanitation challenge—very different outcome. See great paper by Preyer and Strobl: ehes.org/wp-content/u...

ehes.org

100

8/ Now compare to early-20th-century Jamaica. A Rockefeller Foundation hookworm campaign combined: • treatment • improved latrines • intensive hygiene education Campaign workers even brought microscopes into villages so people could see hookworm parasites.

Rockefeller Foundation doctors showing people hookworm larvae under a microscope in Texas in the early 20th century. Unfortunately, no pictures were readily available for Jamaica.

7/ This is why infrastructure alone often had limited effects on stunting. Behaviour and environmental change also mattered.

100

6/ In Britain, diarrhoeal disease remained widespread and only declined decades after the introduction of clean water and sanitation. Declining diarrhoeal disease caused by: • improved hygiene behaviour • safer infant feeding • environmental changes (fewer horses in cities)

5/ Why? Because building sanitation infrastructure doesn’t necessarily reduce children’s exposure to all faecal pathogens. Exposure pathways also depend on: • hygiene behaviour • infant feeding practices • environmental contamination (e.g. horse manure)

4/ Take 19th-century England. Cities invested massively in sanitation and clean water. The results were dramatic: • cholera epidemics were eliminated • under-5 typhoid mortality fell by ~88% Yet child stunting remained above 40% in the 1890s.

100

3/ Recent large trials (e.g. WASH Benefits, SHINE) found limited impacts on child linear growth despite major investments in sanitation infrastructure. Our historical data suggest this puzzle isn’t new. We see similar patterns in the past.

100

2/ Poor water, sanitation and hygiene (WASH) are widely seen as drivers of child stunting. Repeated exposure to faecal pathogens → diarrhoeal disease → impaired nutrient absorption → slower growth. So improving sanitation should reduce stunting. But the evidence is mixed.

Rebecca Sear @rebeccasear.bsky.social · 24/02/2026

1/ Why do many WASH interventions have limited effects on child stunting? Our historical data suggest something surprising: Sanitation infrastructure alone often wasn’t enough, even in the past. New BMJ Global Health paper 👇 doi.org/10.1136/bmjg...

152

Reposted by Eric Schneider

"Many current HICs had high levels of child stunting in the early 20th century, but there was heterogeneity: stunting was low in Scandinavia, European settler colonies & Caribbean, higher in Western Europe & exceptionally high in Japan & South Korea. Child stunting declined across the 20th century"

gh.bmj.com

The decline of child stunting in 122 countries: a systematic review of child growth studies since the 19th century

Introduction Child stunting, a measure of malnutrition, is a major global health challenge affecting 148.1 million children in 2022. Global stunting rates have declined from 47.2% in 1985 to 22.3% in ...

0112

Thanks so much! I’ll send you an email to follow up!

010

Anyway, let me know if you would be interested in following this up, or if I should speak to another member of the team instead.

100

However, our main takeaway figure (figure 5) highlights countries where we are confident in the underlying trends. I could also re-compute country trends based on high-quality data only: we graded the certainty of evidence for all studies.

100

Our data aren't as reliable as the Joint Malnutrition Estimates of child stunting which are currently included because the underlying studies are very heterogeneous, especially for older historical periods.

100

@ourworldindata.org @maxroser.bsky.social @hannahritchie.bsky.social Not sure if it is easier to reach you all here or on email, but I wondered if Our World In Data would be interested in incorporating our new data into its articles on child stunting, malnutrition and adult stature.

Understanding that long-run decline helps us think more clearly about global health policy today. The past doesn’t just show us where we’ve been — it shows what is possible. 10)

0121

The big lesson: Stunting is not fixed by geography or culture. It is historically contingent — and reversible. Countries that eliminated widespread stunting did so over decades of sustained change. 9)

1101

The decline in stunting was part of the broader health transition. Improved nutrition, reduced infectious disease, sanitation, public health systems, and rising incomes all reshaped child growth. This was structural transformation — not just isolated interventions. 8)

140

There was strong regional variation: • Lower early stunting in Scandinavia and some settler economies • Higher rates in parts of Western Europe • Exceptionally high levels in Japan and Korea But the long-run trend almost everywhere was downward. 7)

160

The historical picture is striking: In the 19th and early 20th centuries, stunting was widespread — even in countries that are high-income today. Large shares of children in Europe, North America, and East Asia were stunted by today’s standards. 6)

1143

In our new paper (BMJ Global Health), we systematically collected 923 historical growth studies covering 122 countries from 1814–2016. By harmonising height-by-age data, we reconstruct stunting patterns over two centuries. Link: doi.org/10.1136/bmjg... 5)

22110

But here’s the key question: What happened before modern surveys? How common was stunting in the 19th century? During industrialisation? In early 20th-century Europe or East Asia? Until now, we didn’t have systematic global evidence. 4)

160

Today, stunting is a core global health indicator. Globally, rates fell from roughly 47% in the mid-1980s to about 22% today. That’s major progress — but still hundreds of millions of children affected. 3)

160

Why does it matter? Stunting is associated with: • Higher child mortality • Impaired cognitive development • Lower adult earnings • Worse health across the life course It reflects the biological imprint of social and economic conditions. 2)

150