Methods & Data
Clinical Collective publishes indices with documented assumptions, limitations, and versioned datasets. When uncertainty is high, we show it rather than implying precision.
Core principles
- Non-prescriptive: no usage guidance or recommendations.
- Comparative: indices provide context, not clinical decision support.
- Auditable: methods and sources documented for each index.
- Versioned: datasets are released with versions and timestamps.
Data formats
Indices are driven by published datasets (JSON for web delivery; exportable to CSV for analysis).
- Stable field definitions
- Clear units and transformations
- Update notes and changelogs
ToxicFifty Index — schema
Fields used by the public UI
| Field | Type | Purpose |
|---|
| name | string | Display name of substance. |
| log10_mass | number | Log10 magnitude used for scale placement. |
| metric_label | string | Human-readable descriptor shown in tooltips. |
| band | string | Order-of-magnitude band label. |
| source | string | Primary source link for that row. |
ChronicLoad Index — dataset schema
CBS is a coarse, transparency-first score intended for comparative context, not medical advice or dosing guidance.
Metric: CBS (0–100) • dataset: /data/chronicload.json
| Field | Type | Purpose |
|---|
| name | string | Substance identifier. |
| cbs | number|null | Chronic Burden Score (0–100), when supported. |
| time_horizon | string|null | Interpretive horizon (e.g., months–years). |
| organ_system_scores | object | Per-system scores (null until scored). |
| evidence_summary | string|null | Short statement (non-prescriptive) with scope. |
| sources | array | Links/citations used for scoring. |
RouteRisk Index — dataset schema
Metric: RAF (baseline oral = 1.0) • dataset: /data/routerisk.json
| Field | Type | Purpose |
|---|
| name | string | Substance identifier. |
| baseline_route | string | Baseline comparator route (default oral). |
| raf | object | Route multipliers (null until supported). |
| sources | array | Links/citations used for factors. |
Uncertainty Index — dataset schema
Metric: ECS (0.0–1.0) • dataset: /data/uncertainty.json
| Field | Type | Purpose |
|---|
| name | string | Substance identifier. |
| ecs | number|null | Evidence Confidence Score (0.0–1.0), when supported. |
| evidence_type | string|null | Human/animal/mixed/unknown. |
| risk_of_bias | string|null | Low/medium/high (when assessed). |
| data_gaps | array | Explicit gaps (e.g., long-term data missing). |
| sources | array | Links/citations used for scoring. |
Versioning
Recommended practice
Publish datasets with semantic versions (e.g., v1.0, v1.1) and record
last_updated. Keep older versions available to support citations.