Source evidence
WSI begins with source-reported people, episodes, or mutually exclusive severity categories.
Research methodology
How the pilot USU estimates are calculated
This page explains the assumptions and calculation rules behind the pilot results.
The three-step model
Three-step pilot model
Intensity-time burden
Approved low, central, and high assumptions are applied to intensity and, where relevant, duration.
USU expression
The raw modeled burden is divided by the defined renal-colic reference burden.
B1 = sum of population x intensity x durationandU1 = B1 / 1.96875 intensity-days
Current pilot standard
Current pilot standard
Linear p=1 model
The pilots use direct intensity-time multiplication. Convex intensity weighting is not used in the public pilot standard.
No overlap adjustment
Source categories must be mutually exclusive or made exclusive before calculation.
Low, central, and high scenarios
The deterministic range tests approved mapping alternatives. It is not a confidence interval.
Mortality separate
Deaths are not converted into pilot USU and are shown separately when the source provides them.
Retrospective annualized amendment: the FIES pilot uses a representative average-year scenario when the source reports past-12-month prevalence but not days in state. Its duration values are labeled D1 - illustrative scenario and remain separate from the FAO source data.
Mapping maturity
How the assumptions are labeled
P1 - early illustrative mapping
An author-defined pilot interpretation of a source-defined state. It is transparent but not externally calibrated.
P2 - evidence-informed mapping
The mapping is informed by relevant clinical, symptom, or empirical evidence, but remains provisional.
P3 - externally reviewed
A future stage in which domain specialists or lived-experience contributors have reviewed the mapping.
P4 - empirically calibrated
A future stage supported by direct elicitation or validation data.
Duration maturity: D1 means author-defined illustrative duration scenarios; later D2-D4 levels would require increasingly direct evidence, review, or calibration.
What one USU means
The reference defines the scale
One USU equals the modeled intensity-time burden of a specified six-hour renal-colic trajectory. This makes the reporting unit concrete and consistent.
The reference does not add information to the raw burden, validate the state mappings, or guarantee that different domains are already directly comparable.
R1 = 1.96875reference intensity-days
Source versus model
Source data and modeled inputs
| Layer | What it represents | Public rule |
|---|---|---|
| Source indicator | Cases, people, prevalence, or severity categories supplied by the source. | Always shown before the model. |
| Modeled exposure | Episode-days, a one-day population-state snapshot, or a representative average-year annualized scenario. | The time basis and whether duration is source-derived or modeled must be explicit. |
| Raw burden B1 | Linear intensity-time burden under the approved mapping. | Kept separate from source evidence. |
| USU result U1 | The raw burden expressed relative to the reference. | Does not create new information by normalization alone. |
Comparison boundaries
Why the current pilots are not one ranking
A common unit is only one requirement for a valid comparison.
Dengue is cumulative across reported episodes. Acute food insecurity and child wasting are daily-equivalent burden rates. The FIES pilot is a retrospective annualized scenario per 100,000 people with D1 illustrative duration. Their populations, source constructs, geographic coverage, and model maturity differ. Cross-series ordering will be introduced only when these elements are sufficiently aligned.
Current rule: present the pilots as a research portfolio, not as a ranked table of which harm causes the most suffering.
Further research
Future research
The current model is deliberately simple so it can be applied consistently across many datasets. Future studies may test how people interpret the mappings, whether another reference works better, and when more complex weighting or overlap rules are justified.