Four pilot questions have been provisionally frozen for feasibility and content review. They define exactly who, what exposure, what comparison, which outcomes, what time window, and what numerical quantity the project intends to estimate.
Comparison rule: scenarios can be compared only inside a preregistered comparability class with compatible populations, settings, exposure units, outcomes, denominators, and time horizons. The four pilot rows test the machinery; they are not four entries in a leaderboard.
01 · acute community exposure anchor
Alcohol · community episode
Provisional freeze · v0.2
Among U.S. adults aged 21–64 who drank alcohol in the past year, what is the 24-hour absolute risk of prespecified severe medical and behavioral outcomes after one community episode of consuming 60 g pure ethanol orally within 2 hours from regulated beverages, with no intentional co-use, compared with a matched non-drinking episode?
Target population
Community-dwelling U.S. adults aged 21–64 who drank alcohol at least once in the prior 12 months; primary analysis excludes known physiologic alcohol dependence and prespecified high-risk clinical subgroups are reported separately.
Exposure strategy
Consume 60 g pure ethanol orally from regulated beverage alcohol within a maximum 2-hour interval in a community/private setting; no intentional use of opioids, benzodiazepines, stimulants, or other intoxicants during the exposure window.
Comparator
A matched 24-hour period in the same target population with no alcohol or other intoxicant exposure.
Outcome set
AE_MED_DEATH
AE_MED_ICU
AE_MED_HOSP
AE_BEH_USER
AE_BEH_VICTIM
Intercurrent events
Emergency treatment and spontaneous cessation of drinking are retained under a treatment-policy strategy; intentional co-use is outside the primary estimand; unreported co-use is addressed in sensitivity analysis.
Population-level summary
Age/sex-standardized absolute risk per 100,000 episodes; risk difference and risk ratio versus matched non-drinking periods; outcome-specific estimates remain separate.
Window0–24 hours from first drinkPlace / periodUnited States · 2022–2025Comparability classAE_COMMUNITY_24H_MARKET_EXPOSUREFeasibilityblocked for absolute episode risk
Identification assumptions, data needs, and interpretation limits
Primary data need
A defensible denominator for drinking occasions linked or transportable to acute outcomes; survey-derived episode counts must match numerator population, geography, and period.
Identification assumptions
Accurate episode counting; adequate control of within-person time-varying confounding; outcome attribution within 24 hours; compatible numerator and denominator populations.
Do not interpret as
Not a risk estimate for alcohol dependence, chronic use, pregnancy, co-use, or consumption above/below 60 g; not directly comparable with controlled clinical administration scenarios.
02 · chronic exposure and target-trial anchor
Tobacco · continuation vs cessation
Provisional freeze · v0.2
Among U.S. adults aged 30–49 who currently smoke 10–30 regulated combustible cigarettes daily and have no prior major smoking-attributable disease, what is the 10-year difference in mortality and first major chronic physical-health events under sustained smoking of 20 cigarettes/day versus complete cigarette cessation at baseline?
Target population
U.S. adults aged 30–49 who smoke cigarettes daily at baseline, average 10–30 cigarettes/day, and have no prior myocardial infarction, stroke, COPD, or invasive cancer at baseline.
Exposure strategy
Sustain an average of 20 regulated combustible cigarettes/day for 10 years, with no other combustible tobacco products.
Comparator
Complete cigarette cessation at baseline and no combustible tobacco use during follow-up; nicotine-replacement or approved cessation treatment is permitted and recorded.
Outcome set
All-cause death
first major cardiovascular event
incident COPD
invasive smoking-attributable cancer
CH_PHY health loss
Intercurrent events
Primary analysis is per-protocol: deviations from assigned smoking strategy are handled with prespecified censoring and inverse-probability weighting; death is a competing event for nonfatal outcomes.
Population-level summary
Standardized 10-year cumulative incidence per 1,000 persons, risk difference, risk ratio, restricted mean survival difference, and DALYs per 1,000 persons where defensible.
Window10 years from baselinePlace / periodUnited States · Baseline cohorts 2010–2016 with follow-up through latest complete 10-year windowComparability classRU_CHRONIC_10Y_TARGET_TRIALFeasibilityrevise/split before estimation
Identification assumptions, data needs, and interpretation limits
Primary data need
Longitudinal cohorts with repeated smoking intensity, cessation, confounders, and adjudicated mortality/morbidity; sufficient follow-up for latency-sensitive outcomes.
Identification assumptions
No unmeasured confounding after prespecified adjustment; positivity for cessation/continuation strategies; correct exposure measurement and censoring models; consistency of hypothetical strategies.
Do not interpret as
Not an intrinsic score for nicotine, all tobacco products, lifetime smoking, or initiation; the estimate is a continuation-versus-cessation contrast among baseline smokers.
03 · illicit-supply and denominator stress test
Opioids · changing illicit supply
Provisional freeze · v0.2
Among U.S. adults aged 18–64 with current opioid tolerance, what is the 24-hour absolute risk of fatal or severe opioid toxicity after one intranasal community-use episode involving an unverified product sold as an opioid in the 2024 U.S. illicit market, with no intentional sedative or stimulant co-use?
Target population
U.S. adults aged 18–64 reporting nonprescription opioid use on at least 4 days/week in the prior 4 weeks and no abstinence interval longer than 72 hours; this is an operational proxy for current tolerance and must be validated.
Exposure strategy
One intranasal episode using a self-selected quantity of an unverified powder or pill sold as an opioid; chemical composition and potency are treated as market-distributed random variables and measured post hoc where drug-checking data exist.
Comparator
A matched 24-hour period in the same target population without opioid use; secondary contrasts examine naloxone availability and verified versus unverified supply only if identification is defensible.
Outcome set
AE_MED_DEATH
respiratory depression requiring ventilation or naloxone
AE_MED_ICU
AE_MED_HOSP
Intercurrent events
Naloxone, EMS response, and hospital treatment are retained as real-world rescue processes and separately modeled as modifiers; unintentional adulterants remain part of the supply estimand; intentional co-use is outside the primary estimand.
Population-level summary
Absolute risk per 100,000 episodes, with the product-composition distribution integrated over the registered market and period; report route- and rescue-stratified estimates when possible.
Window0–24 hours from usePlace / periodUnited States · Calendar year 2024; updated as a new scenario version for each market yearComparability classAE_COMMUNITY_24H_MARKET_EXPOSUREFeasibilityblocked for absolute episode risk
Identification assumptions, data needs, and interpretation limits
Primary data need
A valid episode denominator plus linked overdose outcomes and representative supply-composition data; death counts alone are insufficient.
Identification assumptions
Episode denominator represents the same population and market as outcomes; product composition distribution is representative; tolerance proxy is valid; rescue availability and co-use are measured sufficiently.
Do not interpret as
Not molecule-specific fentanyl toxicity, not a fixed-dose estimate, not applicable after loss of tolerance, injection, intentional polysubstance use, supervised consumption, or a different market year.
04 · controlled-administration safety anchor
Psilocybin · controlled administration
Provisional freeze · v0.2
Among adults aged 21–65 with major depressive disorder who meet contemporary trial eligibility criteria, what is the 7-day risk difference in prespecified serious medical and severe psychiatric/behavioral outcomes after one 25 mg oral dose of synthetic psilocybin with preparation, continuous session monitoring, and post-session support versus placebo with identical support?
Target population
Adults aged 21–65 with major depressive disorder who satisfy the shared eligibility envelope of included controlled trials, including prespecified medical, psychiatric, medication, and family-history exclusions.
Exposure strategy
Single 25 mg oral dose of synthetic psilocybin in a clinical research setting after standardized preparation, with trained monitors present for the dosing session and standardized follow-up.
Comparator
Placebo or inactive control with otherwise identical preparation, monitoring, and follow-up.
Outcome set
Serious adverse event
AE_MED_ICU
AE_MED_HOSP
severe anxiety/panic/psychotic event requiring rescue
self-harm or sustained clinically important worsening
Intercurrent events
Rescue medication, extended observation, emergency transfer, and hospitalization are retained under a treatment-policy strategy and also counted in relevant outcomes; participants not dosed are excluded from the administration estimand.
Population-level summary
Risk per 1,000 administrations and risk difference versus placebo, with exact or hierarchical binomial intervals; zero-event studies contribute information without being converted to zero risk.
WindowAcute 0–24 hours and cumulative 0–7 days, reported separatelyPlace / periodCountries and sites represented by eligible controlled trials; primary transport target United States/Canada · Trials conducted 2018–2026Comparability classAE_CONTROLLED_CLINICAL_7DFeasibilityproceed to evidence map; rare-event precision limited
Identification assumptions, data needs, and interpretation limits
Primary data need
Participant-level or arm-level controlled-trial safety data with harmonized adverse-event definitions and complete follow-up through 7 days.
Identification assumptions
Trial eligibility and support procedures are sufficiently harmonizable; adverse-event ascertainment is comparable; missing follow-up is ignorable or modeled; transport to target population is explicit.
Do not interpret as
Not applicable to unsupervised use, natural-mushroom products, other doses, people excluded from trials, polysubstance use, or long-term psychiatric outcomes.