Keep the evidence visible
Every report begins with the official source indicators before any USU model is applied.
Experimental USU research portfolio
Building and testing a common unit for human suffering
World Suffering Index develops focused research series on diseases, humanitarian conditions, and other documented harms.
Each pilot begins with public evidence and applies a simple, transparent Universal Suffering Unit (USU) model. Source data, modeled assumptions, and mapping maturity remain visible so the estimates can be tested, challenged, and improved.
The first portfolio spans infectious disease, acute food insecurity, and child malnutrition. It demonstrates the promise of a common burden language while remaining explicit about where comparisons are not yet scientifically authorized.
3 completed pilotsDengue, food insecurity, and child wastingTransparent mappingsP1 and P2 maturity labels remain visibleSimple pilot standardLinear p=1, no overlap, mortality separate
What WSI enables
Why build a common burden framework?
Diseases and humanitarian conditions are usually reported in different units: cases, severity phases, prevalence, deaths, or people affected.
WSI retains those original indicators and adds a common intensity-time layer. The aim is to make assumptions visible, test whether the added model changes interpretation, and build a growing library that may eventually support stronger cross-domain comparisons.
Model severity and time
Pilot mappings show how population, intensity, and duration contribute to the result.
Show what the model adds
WSI reports when the USU layer changes interpretation and when simpler indicators already explain the result.
Improve through iteration
Versioned P1-P4 mapping maturity makes it possible to strengthen the models over time.
The pilot measurement framework
What is an experimental USU pilot estimate?
A pilot USU estimate combines a source-supported population or episode count with provisional intensity and, where applicable, duration mappings.
The current WSI pilots use a deliberately simple linear model: p=1, no overlap adjustment, low/central/high deterministic scenarios, and mortality reported separately.
One USU is defined relative to a declared six-hour renal-colic reference trajectory. The reference changes the reporting unit; it does not strengthen the source evidence or make different series automatically comparable.
Population or episodesWhat the source reports
IntensityProvisional mapping
DurationObserved or declared basis
Modeled burdenExpressed in USU
Research Series
Three completed experimental pilots
The portfolio shows how the same transparent pilot standard can be applied to very different evidence sources without hiding the model's limits.
Published pilotCumulativeP2 evidence-informed
Reported Dengue in the Americas, 2025
Regional surveillance evidence with a simple cumulative USU model across reported episodes.
45.4 million USU
Deterministic mapping range: 18.2-95.5 million USU
Published pilotUSU/dayP1 illustrative
Acute Food Insecurity Across 14 Cadre Harmonisé Countries, June-August 2026
Harmonized severity evidence for analyzed populations with a one-day experimental USU burden rate.
109 million USU/day
Deterministic mapping range: 82.5-136 million USU/day
Published pilotUSU/dayP1 illustrative
Child Wasting and Severe Wasting, 2024
Global and regional JME evidence with a one-day experimental USU burden rate.
55.9 million USU/day globally
Deterministic mapping range: 27.9-83.8 million USU/day
Methodology
A simple pilot standard designed to scale
Current pilots prioritize clarity and repeatability over complex modeling.
Source evidenceCounts, prevalence, or severity classes
Mutually exclusive statesOne to five modeled categories
Linear burdenPopulation × intensity × duration
USU expressionBurden relative to the reference
The first portfolio uses a linear model rather than nonlinear intensity weighting.
Source categories must be mutually exclusive or analyzed separately.
Low, central, and high mappings form a transparent mapping range, not a confidence interval.
Deaths are not converted into pilot USU.
P1 means illustrative pilot mapping; P2 means evidence-informed pilot mapping.
Cumulative USU and daily-equivalent USU/day are never treated as interchangeable.
WSI pilot series currently use a simplified linear USU model designed for transparency and consistent application across many datasets. The broader USU research programme also examines alternative specifications and validation questions.
Frequently asked questions
Understanding the first WSI portfolio
The portfolio is both a set of public research outputs and a practical test of the USU framework.
1. What is WSI trying to build?
WSI is building and testing a common framework for expressing modeled human suffering burden across different diseases and humanitarian conditions. Each series begins with a public source and makes every added assumption visible.
2. Are the three headline USU values directly comparable?
No. Dengue is cumulative across reported episodes, while food insecurity and child wasting are one-day equivalent rates. The populations, source definitions, coverage, and mapping maturity also differ. Cross-series ranking is not currently authorized.
3. What does the deterministic mapping range mean?
It shows the result under approved low and high mapping scenarios around the central mapping. It is not a confidence interval or probability statement.
4. What do P1 and P2 mean?
P1 identifies an illustrative author-defined pilot mapping. P2 identifies an evidence-informed pilot mapping supported by relevant empirical or clinical evidence. Neither label means that the mapping has been fully validated.
5. Why not use one complicated model for every series?
The early WSI programme uses a simple standard so that multiple series can be built, inspected, and improved. More complex uncertainty, curvature, overlap, or validation studies can be added later where they materially improve interpretation.
6. Why are deaths separate?
The pilot USU models estimate experienced suffering during lived time. Deaths remain visible as companion outcomes when the source reports them and are not silently folded into the USU total.
7. What does WSI add if a ranking does not change?
A stable ranking is itself informative. The model can still provide a common burden unit, show the contribution of different severity states, and reveal that the source indicators already explain the main ordering.
8. Is the renal-colic reference scientifically validated across all harms?
No. It is a transparent reporting convention that defines the unit. It does not establish that different harms are fully comparable, and its interpretability remains part of the broader USU research programme.
9. How are intensity values chosen?
Each state receives a documented low, central, and high mapping. The report explains the source definition, experiential interpretation, included and excluded harms, evidence basis, and maturity level.
10. Can a pilot estimate be revised?
Yes. Mapping notes and report releases are versioned. Better evidence, review, or empirical calibration can replace an earlier mapping without hiding the prior version.
11. Does WSI replace DALYs, QALYs, surveillance data, or humanitarian classifications?
No. WSI begins with those source systems and adds an experimental experiential-burden layer. The original indicators remain visible and may be sufficient for some decisions.
12. How can researchers contribute?
WSI welcomes critique, alternative mappings, domain review, replication, data engineering, visualization, and empirical validation of the USU assumptions.
Collaboration and contact
Help test and improve the framework
WSI is an independently developed research initiative. Collaboration is especially welcome from specialists in domain data, measurement science, humanitarian analysis, clinical research, and reproducible methods.