49 CFR § 382.305 — Random testing
49 CFR — Transportation · FMCSA / DOT
Every motor carrier with at least one CDL driver must run a random drug and alcohol testing program. The FMCSA minimum rates for 2024-2025 are 50% (drugs) and 10% (alcohol) of the average covered driver count, calculated annually. Selections must be truly random — name-based hashing, random number generator, or equivalent. They must be spread evenly throughout the year (no end-of-year cluster). When a driver is notified, they must proceed immediately to the testing site; any delay or refusal counts as a positive result. Most small carriers use a consortium or third-party administrator (TPA) to manage selections; carriers with 100+ drivers commonly run programs in-house.
Regulation summary
Motor carriers must conduct random drug testing on at least 50% of their average CDL driver population annually for controlled substances, and at least 10% for alcohol. The 50%/10% rates are the FMCSA minimums and may be reduced by Federal Register notice if industry-wide positive rates fall below thresholds. Random selections must be made using a scientifically valid method (random number generator, name-based hash, or equivalent) such that every covered driver has an equal chance of selection on every selection date. Selections must be spread throughout the year. A driver notified of random selection must proceed immediately to the testing site. Failure to take the test, or any refusal behavior, is treated as a positive test result.
Who must comply
All motor carriers required to maintain a DOT drug and alcohol testing program for their CDL drivers. The random testing rate applies to the AVERAGE number of CDL drivers employed during the calendar year. Owner-operators with their own authority must enroll in a consortium or TPA-managed program — they cannot run a random selection on themselves alone (it would not be statistically random). Intrastate CDL drivers in most states are subject to identical random testing rates under state-adopted rules.
What happens if violated
Falling below the 50%/10% annual random testing rates triggers civil monetary penalties of $1,100 to $16,550 per violation, calculated per missed test. Carriers found systematically failing the random testing requirements face Compliance Reviews, downgraded safety ratings, and CSA Controlled Substances/Alcohol BASIC point increases. The carrier is also exposed to civil liability if a drug- or alcohol-impaired driver causes an accident in a fleet that wasn't testing per regulation. A carrier with a Conditional or Unsatisfactory rating from drug testing failures may lose insurance, broker eligibility, and shipper contracts.
Implementation checklist
- Enroll in a DOT-compliant consortium or contract with a TPA that manages random selection.
- Provide your accurate CDL driver roster to the consortium/TPA at the start of each selection period.
- Update the roster within 5 business days when drivers are added or removed.
- Receive selection notifications from the consortium/TPA on each selection date.
- Notify selected drivers immediately and confirm they proceed to the testing site without delay.
- Document selection date, notification time, collection time, and result date for every test.
- Track the annual rate against the 50% (drug) and 10% (alcohol) minimums quarterly.
- Spread selections throughout the year — never cluster at year-end.
- Confirm consortium/TPA documentation supports your annual rate calculation.
- Retain random selection records in the confidential drug/alcohol file for 5 years.
Common misinterpretations
- Misinterpretation: 'Random selection means I pick someone randomly when I remember.' Reality: Random selection must use a scientifically valid method — random number generator, name-based hash function, or equivalent — that gives every covered driver an equal probability of selection on every selection date. Manual 'when I remember' selection is not random and would fail a Compliance Review.
- Misinterpretation: 'I can test the same driver multiple times to make up for missed selections.' Reality: Random selection requires equal probability across the pool, not making up missed selections. If a driver is selected twice in a year, that's fine if the selection was truly random. But you cannot deliberately re-test the same driver to compensate for missed selections elsewhere.
- Misinterpretation: 'Selections can be batched at end of year.' Reality: Random selections must be spread evenly throughout the year. Most carriers conduct selections quarterly or monthly. Year-end clustering (e.g., 80% of selections in November-December) signals to auditors that the program is not truly random and triggers violation findings.
- Misinterpretation: 'A driver can delay the test to finish a load.' Reality: When notified of random selection, the driver must proceed IMMEDIATELY to the collection site. 'Immediately' is interpreted as without any unreasonable delay. The driver may complete a brief task essential to vehicle safety (parking the truck, securing the load) but cannot delay to continue a regular dispatch. Delay or non-compliance is a refusal under 49 CFR 40.191.
- Misinterpretation: 'The 50%/10% rate is calculated from the start of the year.' Reality: The rate is calculated against the AVERAGE driver count during the year — not the start, not the end, but the rolling average. A carrier whose driver count changes significantly during the year (seasonal hiring) must recalculate the required test count based on the average.
Frequently asked questions
What are the FMCSA random testing rates under 49 CFR 382.305?
For calendar year 2024 and 2025: 50% for controlled substances and 10% for alcohol of the average number of CDL drivers. The rates are set annually by FMCSA and published in the Federal Register. The rates may be reduced if industry-wide positive rates fall below threshold (which has not happened in recent years — rates remain at the higher levels).
How is the 50% and 10% calculated?
Against the AVERAGE number of CDL drivers employed during the calendar year. If you average 20 drivers during the year, you must conduct at least 10 random drug tests (50%) and 2 random alcohol tests (10%). The 'average' accounts for seasonal hiring or fluctuations — total driver-days during the year divided by 365.
Can I run random selections myself or do I need a TPA?
Carriers may run their own random selection program if they have a scientifically valid method (random number generator, name-based hash function, or equivalent) and maintain proper documentation. In practice, most carriers under 100 CDL drivers use a consortium or third-party administrator (TPA) because: (1) statistical validity is easier with a larger pool, (2) TPA-managed records simplify Compliance Review responses, and (3) self-managed selection requires consistent execution that's hard to sustain.
What happens if I don't meet the 50% / 10% rate?
Civil monetary penalties of $1,100 to $16,550 per violation are calculated per missed test. A carrier required to conduct 20 random drug tests but only conducting 10 would face up to $165,500 in potential exposure (10 missed × $16,550 max). Beyond financial penalties, the carrier's safety rating can be downgraded to Conditional and CSA Controlled Substances/Alcohol BASIC worsens.
Can I batch all random tests at year-end?
No — selections must be spread evenly throughout the year. Auditors look for a roughly even distribution. Year-end clustering signals that the program is not truly random and will result in violation findings. Most carriers conduct selections monthly or quarterly.
What counts as a refusal to test?
Per 49 CFR 40.191, refusals include: failing to appear at the collection site after notification, leaving the collection site before completing the test, refusing to be observed when observation is required, refusing to provide a sufficient specimen, adulterating or substituting a specimen, and any other behavior that obstructs the test process. A refusal is treated as a confirmed positive result and reported to the FMCSA Clearinghouse.
How long must I keep random selection records?
Random selection records, including selection lists, notification logs, and test results, must be retained in the confidential drug/alcohol file for at least 5 years. The 5-year minimum exceeds most DQF retention requirements — drug and alcohol records have their own retention rules under 49 CFR 382.401.
Are random tests done in-person or can they be at home?
Random drug tests must be conducted at an FMCSA-approved collection site using a DOT-approved laboratory. The driver must physically present at the site to provide a urine specimen. Alcohol tests use a breath alcohol technician (BAT) with an evidential breath testing (EBT) device. In-vehicle, at-home, or remote testing does not satisfy the DOT random testing requirements.
Cross-references: 49 CFR 382.301 · 49 CFR 382.309 · 49 CFR 382.401 · 49 CFR 382.605 · 49 CFR 382.701 · 49 CFR 40.191
FileFlo tracks documents required by this regulation automatically:
Connect your folder or Drive — FileFlo classifies every document, maps it to the CFR section it satisfies, and alerts you before any expiration becomes a citation. Starter $89/mo, Professional $299/mo. 5-day free trial.
Start the 5-day free trialAuthoritative source: eCFR.gov →