# Johns Hopkins on PICO Portal’s Priority Screening at GES 2024

Evidence synthesis is essential—but screening thousands of citations is costly, slow, and labor-intensive. Johns Hopkins' Lisa Wilson, Research Associate in the Department of Health Policy & Management at the Bloomberg School of Public Health, along with Dr. Karen Robinson, Professor of Medicine and Co-Director of JHU's Evidence-based Practice Center, tackled this head-on in their recent

Source: https://picoportal.org/blog/jhu-on-pico-portal-at-ges2024/

<img class="alignright" src="/wp-content/uploads/2025/03/Screenshot-2025-03-31-at-2.56.01 PM-300x170.png" alt="JHU at GES 2024" width="300" height="170" srcset="/wp-content/uploads/2025/03/Screenshot-2025-03-31-at-2.56.01 PM-300x170.png 300w, /wp-content/uploads/2025/03/Screenshot-2025-03-31-at-2.56.01 PM-1024x579.png 1024w, /wp-content/uploads/2025/03/Screenshot-2025-03-31-at-2.56.01 PM-768x434.png 768w, /wp-content/uploads/2025/03/Screenshot-2025-03-31-at-2.56.01 PM-1536x869.png 1536w, /wp-content/uploads/2025/03/Screenshot-2025-03-31-at-2.56.01 PM-2048x1158.png 2048w, /wp-content/uploads/2025/03/Screenshot-2025-03-31-at-2.56.01 PM-800x450.png 800w, /wp-content/uploads/2025/03/Screenshot-2025-03-31-at-2.56.01 PM-360x204.png 360w" sizes="(max-width: 300px) 100vw, 300px" loading="lazy" decoding="async">Evidence synthesis is essential—but screening thousands of citations is costly, slow, and labor-intensive. Johns Hopkins’ Lisa Wilson, Research Associate in the Department of Health Policy & Management at the Bloomberg School of Public Health, along with Dr. Karen Robinson,  Professor of Medicine and Co-Director of JHU’s Evidence-based Practice Center, tackled this head-on in their recent study presented at the 2024 Global Evidence Summit – showing the reliability of PICO Portal’s priority screening algorithm.

In a series of eight chronic kidney disease treatment reviews and a total of 29,582 unique citations, they used PICO Portal to assist with title/abstract screening. Citations were re-ranked based on predicted inclusion, allowing teams to <b>stop screening early while maintaining 95% recall</b>—dramatically reducing reviewer workload with no impact on the review findings. Prioritized screening still identified the studies eligible at full text, with a sensitivity of 0.98 (95% CI 0.90 to 1.00).

<small>Conducted and authored independently; co-author Karen Robinson has a non-financial affiliation with PICO Portal.</small>

\> [Read the full abstract](https://abstracts.cochrane.org/2024-prague-global-evidence-summit/streamlining-systematic-reviews-using-machine-learning-enhance)\
\> [View the slides](/wp-content/uploads/2025/03/Johns-Hopkins-Streamlining-Systematic-Reviews-with-PICO-Portal-Global-Evidence-Summit-2024.pdf)