Geographic and temporal trends in etiology-specific diarrhea burden among children in low-resource settings

A harmonized analysis of five major multisite studies reveals that while rotavirus and Shigella remain the dominant causes of moderate-to-severe diarrhea in children under three in low-resource settings, the disease burden is shifting toward Shigella following rotavirus vaccine introduction, alongside a substantial decline in overall mortality rates.

The Gist

Imagine the world's most vulnerable children as a bustling, chaotic city. In this city, "Diarrhea" is a relentless storm that sweeps through, making the little ones very sick. For decades, doctors and scientists have been trying to figure out exactly what is causing the storm so they can build the right umbrellas and shelters.

This paper is like a massive detective story where five different investigation teams (studies) from different eras (2007 to 2024) and different neighborhoods (Africa, Asia, and Latin America) finally sat down at the same table. They took their separate notebooks, translated their different codes, and combined their clues to tell the full story of the storm.

Here is the story they uncovered, broken down into simple parts:

1. The "Big Bad" Villains Changed Over Time

Think of the germs causing diarrhea as different types of villains.

• The Old King (Rotavirus): Before the mid-2010s, Rotavirus was the undisputed King of the city, especially for babies aged 6 to 11 months. It was responsible for about 1 out of every 4 serious cases. It was the main reason babies got hospitalized.
• The Vaccine Shield: Then, scientists introduced a "magic shield" (the Rotavirus vaccine). It worked like a force field. Once kids got vaccinated, the King's power dropped significantly. In many places, Rotavirus went from being the top villain to a minor nuisance.
• The New Boss (Shigella): But here's the twist: When the King stepped down, Shigella (a type of bacteria) didn't just wait in line; it took over the throne!
  • For babies (6–11 months), Shigella is now often the top villain, especially in places where the vaccine was introduced.
  • For toddlers (12–35 months), Shigella has always been the boss, causing anywhere from one-third to half of all serious diarrhea cases. It's like the neighborhood bully that never leaves, regardless of what else changes.

2. The "Age Map" of Danger

The paper found that the danger zone shifts as children grow, like a game of musical chairs:

• The Baby Zone (6–11 months): This is the most dangerous time. Before vaccines, the virus ruled here. Now, it's a battle between the virus and the bacteria.
• The Toddler Zone (12–35 months): As kids get a little older, the bacteria (Shigella) becomes the dominant threat. It's the most common cause of serious illness for this age group everywhere they looked.

3. The Storm is Getting Less Deadly (But Still Dangerous)

The most hopeful news in the paper is about the death rate.

• Then: In the early studies (around 2007–2011), about 2 out of every 100 children who got this sick died within 90 days.
• Now: In the most recent studies (2022–2024), that number dropped to just 3 out of every 1,000.
• Why? It's like the city got better at firefighting. We have better water, better toilets, better rehydration drinks (ORS), and vaccines. The storm still hits, but fewer people are getting swept away.

However, the paper also found that not all villains are equally deadly.

• Viruses (like Rotavirus) are annoying and make kids very sick, but they are less likely to kill.
• Bacteria and Protozoa (like Shigella and Cryptosporidium) are more dangerous. If a child gets infected with these, the risk of death is higher. It's like the difference between a sprained ankle (virus) and a broken leg (bacteria/protozoa) in terms of immediate life-threatening risk.

4. The "Apples to Oranges" Problem

One of the hardest parts of this detective work was that the five teams originally used different rules.

• One team counted every cough; another only counted the loud ones.
• One team looked at kids in the hospital; another looked at kids in the village.
• The Solution: The authors acted like translators. They standardized the rules so they could compare "apples to apples." They adjusted for things like "did the family actually go to the doctor?" or "did the lab test the right sample?" This made the final picture much clearer than any single study could have provided.

The Bottom Line: What Should We Do?

This paper is a call to action for the global community:

1. Keep the Vaccine Shield Up: We must keep giving the Rotavirus vaccine because it works. It saved us from the "Old King."
2. Build a New Shield: Since Shigella is now the main boss, especially for toddlers, we desperately need a vaccine for it. We are currently fighting a war with a shield against the wrong enemy.
3. Keep Watching: The villains change. If we stop watching, the "New Boss" might get even stronger. We need to keep our eyes on the city to see who is causing the next storm.

In short: We made huge progress by beating the virus, but the bacteria is now the biggest threat to our little ones. We need to switch our strategy to tackle the new boss while keeping our old defenses strong.

Technical Summary

1. Problem Statement

Diarrheal disease remains a leading cause of morbidity and mortality in children under five in low-resource settings. While global mortality has declined, understanding the specific etiology-specific burden (the contribution of specific pathogens) is critical for designing targeted interventions (e.g., vaccines, antibiotics).

• The Gap: Previous large-scale studies (GEMS, MAL-ED, VIDA, ABCD, EFGH) were conducted independently with varying case definitions, study designs (case-control vs. cohort), and statistical methods. This heterogeneity made it impossible to directly compare incidence, attributable fractions (AF), and mortality trends across time and geography.
• The Objective: To harmonize data from five major multisite studies conducted between 2007 and 2024 to elucidate temporal and geographic trends in the incidence and attributable fractions of moderate-to-severe diarrhea (MSD) and to estimate etiology-specific case fatality ratios (CFRs).

2. Methodology

The authors performed a meta-analysis of five distinct studies:

1. GEMS (2007–2011): Population-based case-control.
2. MAL-ED (2009–2014): Longitudinal birth cohort with active surveillance.
3. VIDA (2015–2018): Case-control (similar to GEMS).
4. ABCD (2017–2019): Randomized clinical trial (placebo arm only used for mortality).
5. EFGH (2022–2024): Hybrid surveillance study.

Data Harmonization Strategy:

• Case Definition: The authors aligned all studies to the MSD case definition used in GEMS/VIDA: $\ge$3 loose stools in 24 hours, duration <7 days, plus at least one indicator of dehydration, dysentery, or hospital admission.
  • Adjustment: MAL-ED (active surveillance) and EFGH (care-seeking) were subset to match this MSD definition.
• Etiologic Attribution: All studies utilized quantitative PCR (qPCR). The authors applied EFGH-derived Cycle Threshold (Ct) cutoffs to attribute pathogens across all datasets. These cutoffs were derived from GEMS/MAL-ED controls and adjusted for sample types (stool vs. rectal swab).
• Incidence Adjustment: For facility-based studies (GEMS, VIDA, EFGH), incidence rates were adjusted for:
  • Enrollment caps.
  • Healthcare seeking behavior (using Healthcare Utilization Surveys and propensity score modeling).
  • Diagnostic testing rates (pathogen-specific weights).
• Stratification: Analyses were stratified by age (6–11, 12–23, 24–35 months), region (Africa, Asia, Latin America), and rotavirus vaccine introduction status (pre- vs. post-introduction).
• Statistical Analysis:
  • Incidence: Poisson regression with offsets for follow-up time.
  • Attributable Fractions (AF): Calculated as the proportion of MSD cases attributed to each pathogen; weighted averages were computed to avoid overrepresentation of large sites.
  • Case Fatality Ratios (CFR): Estimated as the proportion of MSD cases dying within 14 and 90 days, stratified by pathogen class (viral, bacterial, protozoal).
  • Uncertainty: 95% Confidence Intervals were generated using stratified bootstrap resampling (1,000 iterations).

3. Key Contributions

• First Harmonized Analysis: This is the first study to successfully align case definitions and analytic methods across five major longitudinal and cross-sectional studies spanning nearly two decades (2007–2024).
• Standardized Metrics: Provided comparable estimates of incidence, AF, and CFR across diverse low-resource settings, overcoming the "apples-to-oranges" problem of previous literature.
• Temporal Evolution: Explicitly quantified the shift in disease burden following the introduction of rotavirus vaccines in national immunization programs.
• Mortality Insights: Provided the first robust estimates of etiology-specific CFRs across multiple studies, a metric previously difficult to assess due to low event rates in individual studies.

4. Key Results

A. Incidence and Attributable Fractions (AF)

• Age Patterns: MSD incidence was consistently highest in children 6–11 months and declined with age.
• Rotavirus Impact:
  • Pre-vaccine: Rotavirus was the leading cause of MSD in 6–11-month-olds (AF ~22–28%).
  • Post-vaccine: Significant decline observed. In Africa, AF dropped from ~22% to 9.7%; in Asia, from ~28% to 11%.
  • Incidence: In Mali, rotavirus incidence in 6–11-month-olds fell from 22.11 (GEMS) to 3.71 (EFGH) per 100 person-years.
• Shigella Dominance:
  • Shigella became the leading etiology in 6–11-month-olds in many post-vaccine sites, replacing rotavirus.
  • 12–35 months: Shigella was the dominant etiology across all regions and time periods, causing 33–50% of MSD cases.
  • Peak Age: Shigella incidence peaked in the 12–23 month age group.
• Other Pathogens:
  • ST-ETEC: Important in Africa/Asia (6–19% AF) but no clear age pattern.
  • Cryptosporidium: Significant in Africa (11–14% AF), less common in Asia and Latin America.

B. Case Fatality Ratios (CFR)

• Temporal Decline in Mortality: Overall 90-day mortality following MSD declined drastically over time:
  • GEMS (2007–2011): 2.21%
  • VIDA (2015–2018): 0.82%
  • EFGH (2022–2024): 0.30%
• Etiology-Specific Risk:
  • Typical EPEC had the highest 90-day CFR (8.75%).
  • Protozoa (3.55%) and Bacteria (2.52%) had significantly higher CFRs than Viruses (1.42%).
  • Note: Cryptosporidium had a 90-day CFR of 3.76%.

5. Significance and Implications

• Vaccine Policy: The data confirms the success of rotavirus vaccines in reducing severe disease burden. However, the "vacancy" left by rotavirus is being filled by Shigella, particularly in the 6–11 month age group.
• Future Interventions: The findings strongly support the urgent need for a Shigella vaccine to address the dominant cause of MSD in children over 1 year and the emerging leading cause in infants post-rotavirus vaccination.
• Mortality Reduction: The substantial decline in CFR (from 2.2% to 0.3%) highlights the success of non-specific interventions (WASH, ORS) and improved healthcare access, though etiology-specific mortality remains high for bacterial and protozoal pathogens.
• Surveillance: The study underscores the necessity of ongoing, harmonized surveillance to monitor etiological shifts and guide public health policy in low-resource settings.

Limitations Noted:

• Variations in catchment populations between studies in the same country (e.g., rural vs. urban in Bangladesh).
• Inability to fully adjust for socioeconomic status in all historical datasets.
• MAL-ED's active surveillance design differs fundamentally from the facility-based designs of other studies, though adjustments were made.