Longevity Olympics Research Initiative · Bayesian Causal Network
Self-caused mortality:
a Pearl-adjusted causal portfolio of overdose, suicide, and unintentional injury.
In plain wordsThis tool covers three big preventable causes of death: drug overdose, suicide, and accidents (about 380,000 US deaths a year). It shows how much proven prevention could reduce each. The modelled reductions are large but are best-case ceilings, not promises. This is a research overview, not a crisis service — if you or someone you know is in danger, in the US call or text 988.
A Structural Causal Model (SCM) integrating hazard ratios, mechanism-of-action, and dose-response evidence
across the three leading domains of preventable self-caused mortality in the United States.
Interventions are evaluated under backdoor adjustment, cross-correlation removal
(mean off-diagonal ρ̄ ≈ 0.30 across shared confounders), and do-calculus,
with E-values, PNS/PN/PS, and population-attributable fractions reported per node.
Framework · Pearl SCM + do-calculusInterventions · 24 across 3 domainsConfounders · 15 modeledMediators · 4 latentDate · 2026-05-26
≈ 11.4% of all U.S. deaths; the dominant preventable-mortality cluster < age 60.
§ 01 / Methodology
Structural model and Pearl-adjustment pipeline
Backdoor · do-calculus · E-value
Identification strategy
For each candidate intervention X on outcome Y ∈ {overdose, suicide, unintentional injury}, the analysis verifies a sufficient adjustment set Z under the backdoor criterion using the SCM in §02.
Effects are recovered as do(X=1) − do(X=0). All hazard ratios are extracted from peer-reviewed evidence sources (largely Cochrane reviews, Lancet meta-analyses, BMJ, JAMA, AJPH, Inj Prev), then converted to the standardized scale via Chinn 2000 (SMD → ln HR via π/√3).
Cross-correlations between interventions (e.g., MOUD ↔ naloxone, both downstream of OUD; SSRI ↔ CBT, both via depressive symptomatology) are quantified in a Σ matrix; eigenvalue-corrected to remove redundancy with a mean off-diagonal ρ̄ ≈ 0.30.
Combined-effect aggregation
Naive multiplication ∏HRi assumes independence, which is biologically false here — many interventions act on shared mediators (intoxication, impulsivity, hopelessness, lethal-means access). The correlation-corrected log-additive estimator uses the eigenvalue-corrected effective sample size:
This derives from multivariate-normal latent-effect variance theory: for n correlated interventions with mean off-diagonal ρ̄, the variance of their summed log-hazard contribution scales as n[1 + (n−1)ρ̄], so neff = n / [1 + (n−1)ρ̄]. With ρ̄ = 0.30 and n = 8 active interventions, neff ≈ 2.58 and κ ≈ 0.32 — the 8-intervention portfolio acts as ~2.6 fully independent interventions after de-duplication.
Saturation thresholds (each intervention's empirical dose-response plateau) further cap marginal contribution at observed ceilings (e.g., MOUD HR cannot fall below ~0.30 even with full coverage).
Combined portfolio across all three outcome categories
Selecting "Activate all" toggles every intervention to its dose-response saturation point and applies the cross-domain correlation matrix.
Overdose Δ
0%
Reduction in drug-overdose mortality (per active selection).
Suicide Δ
0%
Reduction in suicide mortality (per active selection).
Accident Δ
0%
Reduction in unintentional-injury mortality.
0%
Composite Pearl-adjusted reduction across all three domains
Composite HR1.000
Cross-domain ρ̄0.30
Active total0 / 24
Annual lives saved (US)0
Composite mean E-value—
Composite reduction goal 50%
—
§ 04 / Counterfactuals
What-if and if-not-for analysis
do-calculus on each lever
Toggle a counterfactual to observe the marginal effect of removing or adding the single most impactful intervention in each domain. The display recomputes assuming all other interventions remain at their current setting.
Counterfactual ΔHR
—
Change in domain HR from toggling the selected intervention.
Lives shifted (US/year)
—
Annual mortality shift attributable to the selected counterfactual.
PAF (single)
—
Population-attributable fraction of the selected lever in isolation.
§ 05 / Quantitative inference table
E-values, PNS/PN/PS, and PAF per intervention
VanderWeele E-value · Pearl PNS
Domain
Intervention
HR
95% CI
E-value
PN
PS
PNS
PAF
Evidence basis
§ 06 / Sensitivity
How robust is the composite estimate?
ρ̄ sweep · HR perturbation
Correlation discount sensitivity
Sweep the mean off-diagonal correlation ρ̄ from 0.0 (full independence — naive product) to 0.6 (heavy mediator overlap). The composite HR at full-portfolio activation moves accordingly.
Hazard-ratio perturbation envelope
Each intervention's point HR is jittered ±25% (capturing meta-analytic CI width) over 1,000 Monte Carlo draws. The histogram below shows the resulting composite reduction distribution at full portfolio activation, with the Pearl-adjusted central estimate marked.
§ 07 / Antithesis
Where this model can be wrong
Critical self-audit
Critique 1 · Selection on observables
Most intervention HRs come from observational cohorts (MOUD enrollees are systematically different from non-enrollees on motivation, social support, healthcare access). Backdoor adjustment can only block measured confounders. The E-values shown bound the strength of unmeasured confounding needed to fully nullify each effect — for MOUD with HR 0.50, an unmeasured confounder would need RR ≥ 3.4 on both exposure and outcome, which is implausible but not impossible.
Critique 2 · Effect heterogeneity
Population-average HRs hide enormous individual variation. Lithium's anti-suicidal effect is concentrated in bipolar I (HR ~0.3); applying it to unipolar depression with mild ideation gives a much weaker effect (HR ~0.8). The dashboard's domain-level rollup understates this stratification.
Critique 3 · Iatrogenic and substitution effects
Means restriction can be partially offset by substitution to other methods, though the bridge/firearm-storage literature shows substitution is incomplete (Yip 2012, Lancet). PDMP enforcement reduced prescription-opioid deaths but contributed to street-fentanyl substitution post-2014. Neither effect is currently mediated in the DAG.
Critique 4 · The independence ceiling
The κ correction is a first-order linear discount. True effect aggregation under shared mediators is non-linear; with full saturation of multiple interventions acting on the same mediator (e.g., intoxication), the marginal contribution can collapse to zero rather than gradually diminish. The composite ceiling shown here is therefore an upper bound, not a point estimate.
Note on use. This research artifact is a Bayesian causal model of public-health interventions, intended for analytical, policy, and protocol-development purposes. It is not a clinical tool. Anyone reading this who is in crisis — or supporting someone who is — can reach the U.S. 988 Suicide & Crisis Lifeline (call or text 988), SAMHSA National Helpline (1-800-662-4357, free 24/7 confidential treatment referral), or local emergency services. Naloxone is available without prescription in all 50 states.