Second-order Meta-analyses to Support Evidence-based Agri-food Transitions

🌍 Why second-order meta-analyses?

Second-order meta-analyses —
syntheses of meta-analyses — help navigate the explosion of evidence.

They offer:

• ✅ Broader scope — covers many interventions/outcomes
• 📈 Large data volume — combines extensive evidence to improve reliability
• 🔄 Comparable effect sizes than a first order Meta-analysis

💡 What can they reveal?

We synthesized:

🚀 Main results (effect of crop diversification) :

• 🌾 +14% crop yields (23249 paired-observation)
  
• 🐞 +63% pest and disease control (3081 paired-observation)
• 🌱 +30% soil quality (13107 paired-observation)

Also applied to:
🌳 Agroforestry · 🌿 Soil carbon · 💧 Biodiversity · 🌦️ Climate impacts

🗺️ Mapping the evidence

Evidence maps enhance second-order syntheses by:

• 📍 Geographic biases (e.g. Africa)
• 🚨 Identify PICO blind spots (e.g. water use, profitability)

🧠 This map offered a global diagnostic of research coverage,
and laid foundations for strategic future syntheses

⚠️ Limitations & caveats

Second-order syntheses are not perfect.

• 🔍 Loss of resolution — Few moderators (e.g. climate, crop)
  
• 🔁 Redundancy — Same primary studies reused
  
• 🧮 Varying quality — Not all follow PRISMA or report bias
  
• 🧨 Exponential growth — Integration becomes harder

✨ Final message

Second-order meta-analyses are powerful tools to navigate the evidence explosion
and extract robust, policy-relevant insights.

To amplify their impact:

🔍 Prioritize transparency and traceability — protocols, shared data & methods

📈Embrace variability — go beyond simple averages

🤝 Foster collaboration — share extractions and avoid redundant efforts
🌐 Invest in open, evolving platforms — to keep syntheses alive and connected –> https://www.impact4soil.com/

🧠 Not an end, but a strategic entry point to evidence-informed transformation