Loading BTS-39 Toolkit…

Please wait while resources load

BTSCRP Project BTS-39

The "Last Leg" Impaired Driving
Challenge

Driving under the influence of alcohol or drugs remains a persistent highway safety issue, causing over 10,000 deaths annually in the U.S. One emerging challenge is the "last leg" of a multimodal trip, when an individual drives to a commuter rail station, takes transit into a city to socialize and consume alcohol, then returns by train and still must drive a few miles home. This toolkit provides comprehensive strategies to address this critical safety gap.

10,000+ Annual deaths from impaired
driving in the U.S.
29% Of travelers would drive
impaired in a realistic scenario
~1,500 Public survey respondents
powering this toolkit
57% Say station safety would
change their decision

What This Toolkit Provides

State-Level Crash Analysis

Interactive map with crash data, screening counts, and hour-of-day patterns.

Evidence-Based Strategies

Several practitioner-rated countermeasures with impact, feasibility, and adoption scores.

Survey Insights & Risk Simulator

What ~1,500 real travelers say — barriers, interventions, messages, and a station risk calculator powered by logistic regression.

Action Plan Builder

Generate a copy-ready plan combining crash signals, survey insights, barriers, recommended interventions, and CBG-level data.

Texas State University • Portland State University • Exponent

Interactive US Map

How to use this page

  1. Click a state on the map. States in the analysis are shaded blue (OR, WA, CA, TX).
  2. Review the Quick Summary panel on the right.
  3. Click "See Details" to expand the full dashboard below the map.
  4. Use Reset to clear the selection.
About the data — 5-step screening framework

Last-leg crash candidates are identified from state-level crash records using a structured screening framework:

  1. Temporal filter — Crashes between 5 PM and 3 AM on Friday, Saturday, and Sunday nights, including holiday long-weekend extensions.
  2. Transfer-point proximity — Crashes within a 2-mile buffer of park-and-ride facilities, transit stations, or airports.
  3. Residential context — Crash block group (or adjacent block group) has ≥80% residential land use.
  4. Roadway context — Exclude crashes on interstate highways and freeways.
  5. Final classification — Crashes meeting all four criteria above are classified as potential last-leg candidates.

States analyzed and study periods: WA (2018–2023), OR (2018–2023), CA (2018–2022), TX (2021–2024). Year 2020 excluded where applicable.

Included in analysis Not included

Quick Summary

Select a state to see crash-level KPIs.

What this summary tells you

  • Total crashes and impaired crashes provide overall scale.
  • % impaired helps compare prevalence across states.
  • Impaired in time window shows how concentrated impaired crashes are in the "last-leg" time period.
  • Final last-leg candidates is the screened set used for deeper review in the dashboard.

Selected state

—

Select OR / WA / CA / TX to enable "See Details".

Total crashes

—

All crashes in the study period.

Impaired crashes

—

Alcohol/drug involvement where coded.

% impaired of total

—

Prevalence indicator.

Impaired in time window

—

5 PM–3 AM, Fri–Sun + holidays.

Final "last-leg" candidates

—

After transfer-point + residential + roadway screens.

—

—

How to use the dashboard

  1. Start with Overview to understand the crash snapshot, screening counts, timing, and facility intensity.
  2. Use Behavior Insights to see public decision limits and the most promising message themes.
  3. Use Block-group to see which CBG characteristics are higher/lower (and by what %).

What you are seeing in “Overview”

  1. Crash snapshot shows total and impaired crashes and how much of impairment happens during the last-leg time window.
  2. Stepwise counts show how the candidate set changes as filters are applied (transfer-point proximity → residential context → roadway screens).
  3. Hour-of-day pattern gives a quick visual of when last-leg risk is most concentrated.
  4. Intensity per facility highlights which facility type has higher crashes per facility (pilot priority signal).

Crash snapshot

Observed crash data

Total crashes

—

All crashes in study years.

Impaired crashes

—

Alcohol/drug involvement where coded.

Impaired in time window

—

5 PM–3 AM, Fri–Sun + long weekends.

Stepwise counts

Candidate near transfer points

—

Within chosen buffer (default 2 miles).

After residential context

—

≥80% in crash BG or neighbor BG.

Final last-leg candidates

—

Non-interstate retained.

—

Hour-of-day pattern

Impaired in time window

—

Intensity per facility

Crashes per facility

—

What you are seeing in “Behavior insights”

  1. Decision thresholds summarize common limits (how long people will wait and how much they will pay) before they choose to drive.
  2. Messaging results highlight which message themes are most persuasive in the survey.
  3. Use this tab to decide what to emphasize: cost, wait time, or safety/lighting.

Decision thresholds

Public survey

Max wait time (typical)

—

Typical upper limit before driving.

Max price (typical)

—

Typical upper limit before driving.

Safety/lighting effect

—

How strongly safety conditions influence choice.

—

Messaging test results

Public survey
MessageInfluence
(avg 1–5)
Recommendation

What you are seeing in “Block-group”

  1. This table lists variables where the CBG average differs between CBGs with last-leg crashes and CBGs without.
  2. We show direction (Higher / Lower) and the % difference.
  3. This section is descriptive and intended to support state-specific targeting and interpretation.

CBG-Level Comparative Signals Selected state

Shows variables where CBG averages differ between areas with vs without last-leg crashes.

Variable (CBG average)Direction% diff

Direction and % difference compare CBGs with last-leg crashes vs CBGs without.

What you are seeing in “Strategies”

  1. The Strategy scoreboard helps compare options using reported impact, feasibility, and how many agencies have formally assessed each strategy for implementation.
  2. The Barriers list shows what most often prevents implementation (so the action plan can address them directly).
  3. Use this tab to pick a realistic bundle: one service option + one policy/enforcement support + one communication support.
  4. The Intervention Strategies cards below provide full detail on each countermeasure — including description, impact, feasibility, adoption rate, cost, and category tags — drawn from the practitioner survey catalog.

Strategy scoreboard

Practitioner survey
Strategy Agencies Assessed
(%)
Impact
(1–5)
Feasibility
(1–5)

Top barriers

Practitioner survey
Barrier Severity
(avg 1–5)
Signal

Intervention Strategies

Catalog of countermeasures from the practitioner survey

Key Finding

Discounted Rideshare codes have the highest adoption (78%) and the highest impact score (3.95/5). DUI Enforcement near transit hubs is underutilized — only 22% adoption despite a strong impact score.

Best Overall

Rideshare Codes

3.95 impact · 3.49 feas.

Most Feasible

Awareness Campaigns

4.15 feas · low cost

Underutilized

DUI Enforcement

22% adoption · 3.58 impact

Multi-Metric Radar

Impact · Feasibility · Adoption (0–5 norm.) · Cost Efficiency — up to 4 strategies overlaid.

Select strategies on the right

Choose up to 4 to render the radar chart

Select Strategies to Compare

Choose up to 4 for the radar chart and detail table below.

Impact vs. Feasibility Matrix

Bubble size ∝ agency adoption rate. Quadrants guide prioritization.

★ Sweet Spot = High impact + High feasibility Easy Wins = Low impact + High feasibility
Public Survey • ~1,500 Respondents

Survey Insights & Risk Tools

What real travelers say about impaired driving, what would change their behavior, and which interventions they support most. Use these tools to guide your local strategy.

Knowledge–Action Gap

●●●●●●●●●●
3 in 10

know driving impaired is very risky — and would still drive. Awareness alone isn't enough.

#1 Rated Message · 4.11/5

“Your car can wait. Your life can’t.”

★★★★★★ 50% rated “extremely influential”

Would Drive Impaired

29%

In a realistic scenario, nearly 1 in 3 said they'd drive themselves home after drinking.

Max Wait Tolerance

~20 min

Median wait before travelers give up and drive. After 20 min, drive-home decisions spike.

Price Comfort

~$15

Median amount travelers are willing to pay for a safe ride. Mean is $21.

Safety Matters

57%

Say station safety "very much" or "extremely" influences their decision to use alternatives.

What Would Travelers Do?

When stranded at a transit station late at night after drinking

How Risky Do Travelers Think It Is?

Perceived risk of driving after drinking

Key insight: 71% see it as very or extremely risky — yet 29% would still drive. Knowing it's dangerous isn't enough; the environment must make the safe choice easy.

Traveler Mindset — What Practitioners Need to Know

"Short Drive" Fallacy

58% disagree that a short impaired drive is safe — but 42% are not sure or agree. This gap is where crashes happen. Counter with: specific distance-based messaging.

Vehicle Overnight Worry

57% are concerned or very concerned about leaving their vehicle overnight. This is a key reason people choose to drive. Overnight parking policies directly address this.

Late-Night Safety Fear

60% agree they'd feel unsafe waiting/walking near a station late at night. Good lighting (66%) and visible security (66%) are the top conditions that would change this.

Station Risk Simulator

Set conditions at a transit station or park-and-ride to see how likely travelers are to drive impaired. Powered by logistic regression from 8,798 choice observations.

Distance to Home
Wait Time Available
Affordable Ride Cost
Station Safety Feel
Good Lighting Present? 66% say lighting would change their decision
Overnight Parking Allowed? #1 factor — 27% of decision weight

Simulation Results

Drive-Home Likelihood

vs. 29% survey avg

avg

Current Setting vs. Survey Benchmark

Wait time
Ride cost
Safety feel

Actionable Tips for This Scenario

What Matters Most to Travelers

When choosing whether to drive or use an alternative, these factors carry the most weight (relative importance from choice experiment)

Overnight parking availability alone accounts for more than a quarter of the decision. Addressing it is the single highest-leverage intervention.

Top Interventions — Ranked by Travelers

What travelers actually want, sorted by selection rate and rated effectiveness. Tiers help you prioritize.

Quick Wins (highest demand + effectiveness)
Strong Support (good demand + effectiveness)
Worth Considering (lower demand, still effective)

Most Effective Messages

Ranked by how strongly each message influences travelers' decisions. Use these in signage, campaigns, and digital displays.

Select a state first, then these highlights auto-populate with crash, survey, and CBG context.

Plan Inputs

Action Plan (copy-ready)

Select inputs and click "Generate Action Plan".
Real-World Evidence

Case Studies of Programs Addressing Last-Leg Impaired Driving

Before developing new solutions, it is instructive to examine existing programs and pilot initiatives that have tackled similar challenges. Below are several case studies and examples from U.S. communities, illustrating approaches to provide safe alternatives for that last leg home.

5 Case Studies Rural Programs Rideshare Partnerships Transit Integration
Rural Wisconsin landscape

Wisconsin “Road Crew” Program

Rural
Rural Wisconsin

In several rural Wisconsin communities lacking public transportation, the Road Crew program was created to provide rides to, between, and home from taverns and events. Designed with extensive input from local bar owners, young adult residents, community leaders, and even a major brewing company.

Outcomes

Two of three pilot communities saw significant reductions in alcohol-impaired driving and achieved self-sustainability through modest rider fees and contributions from participating taverns.

Key Lesson

Community buy-in is critical. The one community that failed lacked trust and coalition cooperation. Road Crew became a model showing that with the right partnerships, a safe ride program can become “part of the community’s culture.”

Night driving rideshare

Evesham “Saves Lives” Rideshare

Suburban
Suburban New Jersey

Evesham Township launched an innovative program offering free or low-cost Uber rides from any bar or restaurant in town to the resident’s home between 9 p.m. and 2 a.m. from 2015–2018. A collaboration between the township, a nonprofit foundation, and Uber.

Outcomes

18% reduction in overall injury crashes; 38% reduction in nighttime crashes. An estimated 3 lives saved and hundreds of injuries prevented over the program’s lifetime.

Key Lesson

Making safe transportation free and convenient can dramatically reduce impaired driving harms. The program also implicitly addressed overnight parking and next-day vehicle retrieval — critical last-leg barriers.

Transit train station

Transit “Free Ride” Promotions

Urban
Various Urban Centers

Molson Coors sponsored “Free Rides” programs across many cities where on major drinking occasions (New Year’s Eve, St. Patrick’s Day, big sports events), local mass transit — buses, light rail, commuter trains — was offered fare-free to all riders.

Outcomes

Millions of riders took advantage of free transit nights, leaving personal vehicles at park-and-ride locations and using transit both ways instead of driving impaired.

Key Lesson

Successful alcohol industry + transit + law enforcement collaboration, but limited to high-risk nights only. Highlights the persistent gap on ordinary nights and in areas without robust transit — the exact gap this toolkit must address.

Bar and tavern

Tavern League Ride Voucher Programs

Multi-State
Wisconsin & Minnesota

Bar-owner associations and nonprofit coalitions created ongoing voucher systems for free or discounted rides. Wisconsin’s SafeRide program provides free cab or rideshare rides home from member establishments, funded by a state OWI surcharge and Tavern League contributions.

Outcomes

Thousands of safe rides home provided, preventing potential impaired driving incidents through a community-funded, industry-backed model.

Key Lesson

Community-driven solutions with alcohol-serving establishments on board thrive. Top-down programs without bar-owner buy-in consistently struggle. The business community must be a partner, not a bystander.

Car keys designated driver
Volunteer-Based
Various Communities (primarily Canada)

“Operation Red Nose” & Designated Driver Services

Volunteer-based programs like Operation Red Nose send a team to drive an impaired person and their vehicle home. Two volunteers come together — one drives the impaired person’s car, the other follows to pick up the first driver afterward. This directly addresses the critical last-leg barrier of “what about my car?”

Outcomes

Successfully provided safe transportation for both impaired individuals and their vehicles, directly resolving the vehicle-stranding barrier that deters use of alternatives.

Key Lesson

A practical model for suburban/rural areas: help people get themselves and their cars home safely, removing a major psychological barrier. Any toolkit can incorporate or partner with such programs for transit hub last-leg gaps.

About This Toolkit

BTS-39 Toolkit supports agencies in reducing substance-impaired driving risk during the "last leg of the journey," especially when travelers transition from transit, venues, or park-and-ride locations into personal vehicle travel.

About the Project

BTSCRP Project BTS-39: Toolkit for Reducing Substance-Impaired Driving for the Last Leg of the Journey is a 24-month national research initiative led by Texas State University (TXST), in collaboration with Portland State University (PSU) and Exponent, beginning July 1, 2025.

This project addresses substance-impaired driving during the final segment of a trip, particularly in suburban and rural areas where public transit and shared mobility options are limited.

The primary audience includes State Highway Safety Offices (SHSOs), Metropolitan Planning Organizations (MPOs), and local transportation and public safety agencies.

How to Use This Toolkit

Step 1

State Map & Data

Select a state, review crash data, and explore the dashboard with crash snapshots, behavior insights, and CBG signals.

Step 2

Strategies & Compare

Browse intervention strategies, filter by category, and compare up to 4 side-by-side on the impact–feasibility matrix.

Step 3

Survey Insights & Risk

Explore what ~1,500 real travelers say: barriers, interventions, messages. Use the station risk simulator to test "what-if" scenarios.

Step 4

Action Plan Builder

Generate a copy-ready plan combining crash signals, survey insights, barriers, recommended interventions, and CBG-level data.

Step 5

Case Studies

Browse real-world case studies and programs that have successfully reduced last-leg impaired driving.