Central Cargo Express

Random Drug and Alcohol Testing: Running the Pool at a Small Carrier

By Tom Whitaker · 2026-11-18 · 11 min read

The first random selection notice tends to land on a small carrier at an awkward moment. The owner is also the dispatcher, one of the four drivers is two states away, and the email from the consortium says the name drawn has to be tested in this period. Nobody planned for it, so the period ends, the test never happens, and the carrier has a violation in its file before anyone thought about the rule.

Random testing is the part of the DOT drug and alcohol program that keeps running after the hiring paperwork is done. Pre-employment testing happens once per driver. The random program runs all year, every year, for every covered driver, and a safety investigator will ask to see proof that it ran. This guide walks through what 49 CFR 382.305 actually requires, how the numbers work for a fleet of three or thirty trucks, and the handful of situations that trip small carriers most often.

Who belongs in the pool

The pool includes every driver who operates a commercial motor vehicle that requires a CDL, in interstate or intrastate commerce, and nobody else. That second half matters as much as the first. Office staff, non-CDL drivers of small vans, and mechanics who never drive a CDL vehicle stay out. If you add them, you are testing people outside the federal program under DOT authority, and the selections no longer mean what they are supposed to.

A few cases worth thinking through:

  • The owner who drives. If you hold a CDL and take loads yourself, you are a covered driver and you go in the pool, the same as anyone else.
  • Drivers who also work as mechanics or supervisors. They are in the pool because they drive. The difference is in when an alcohol test can happen, which we cover below.
  • Interstate and intrastate drivers. They can share a single pool.
  • Drivers who work for two carriers. A driver who drives part time for you and part time for another regulated employer has to be in both employers' random pools.
  • Owner-operators with their own authority and no other drivers. A one-person company cannot run a random program alone, because a pool of one gives no randomness. In practice that driver joins a consortium.

The two numbers: 50% and 10%

Each year FMCSA sets a minimum annual random testing rate. For 2026 the agency kept the rates unchanged: 50% for controlled substances and 10% for alcohol. DOT's Office of Drug and Alcohol Policy and Compliance publishes the current figures for every agency on its random testing rates page, and that is where to confirm the rate before you plan the next calendar year.

The percentage applies to the average number of driver positions in the pool, not to the number of names on your roster on January 1. The average is calculated by adding up the number of covered drivers eligible in each selection period and dividing by the number of periods.

Here is how that works for a small carrier that selects quarterly. The headcount is illustrative:

QuarterCovered drivers in pool
Q16
Q28
Q39
Q47
Average(6 + 8 + 9 + 7) / 4 = 7.5

At 50%, this carrier needs at least 3.75 drug tests in the year, which rounds up to 4 in practice. At 10% for alcohol, the minimum is 0.75, so at least one alcohol test. Many carriers aim slightly above the minimum, because a driver who quits after selection or a canceled test can leave you short in December with no time to catch up.

Why the rate is about tests, not people

A driver selected in March goes back into the pool for the June draw. Each driver has an equal chance at every selection, so the same person can be picked twice in a year while someone else is never picked. That is how randomness is supposed to look. Drivers sometimes take repeat selections personally; it helps to explain the math during orientation so the second notice does not feel like an accusation.

How selections must be made

The regulation requires a scientifically valid selection method. DOT accepts a random number table or computer software that draws random numbers, matched to an identifier such as an employee number. Drawing names from a hat does not qualify, nor does a dispatcher picking whoever is closest to the clinic.

Three rules govern the selection itself:

  1. Equal chance at every draw. No one is removed from the pool because they were already tested this year.
  2. Spread across the year. Tests have to be unannounced and spread reasonably throughout the calendar year. Running all the year's tests in one week of December fails that standard even if the count is right.
  3. Tested in the period. A driver selected in a period must be tested in that period. Missing it is a violation, and the test you run later does not cure it.

Quarterly selection is the common pattern for small carriers using a consortium. Some larger programs draw monthly. Either works as long as the tests are spread out and every selection is completed.

Drug tests and alcohol tests are timed differently

This is the detail that matters most for carriers whose drivers wear more than one hat.

Random drug testRandom alcohol test
Minimum annual rate (2026)50%10%
When it can be doneAny time the driver is at work for the carrierOnly just before, during, or just after the driver performs safety-sensitive functions
Driver doing other work at notificationAllowed: a driver working in the shop can be sent for a drug testOnly if the driver is on call and ready to be dispatched

So a driver who spends Thursdays in the shop can be sent for a random drug test on a Thursday. A random alcohol test on that day is harder to justify unless the driver is also standing by for dispatch. Put the alcohol selections where you know the driver will be driving, and you avoid an argument later.

Notification: "immediately" means immediately

Once you tell a driver they have been selected, they must go straight to the collection site. If the driver is doing some other safety-sensitive task when you notify them, such as fueling or inspecting the truck, they stop and go as soon as possible. If they are driving, you arrange the trip so they can get there promptly without abandoning a load in an unsafe place.

What carriers get wrong here is the gap between selection and notification. Holding the notice until the driver is close to the clinic is fine; telling the driver on Monday that they need to test "sometime this week" is not. Any warning gives the driver the opportunity to prepare, which is the thing random testing exists to prevent.

A simple routine that works for small fleets:

  • The person who receives the selection list keeps it to themselves.
  • That person looks at the driver's schedule and picks a day when the driver will pass a collection site, or plans a short detour.
  • The driver is told when it is time to go, with the address and the time.
  • The notice, the time the driver was told and the time of collection are recorded.

If you use a mobile collector or a clinic network through your consortium, ask in advance how quickly a collection can be arranged on the road. "We will figure it out" tends to become "the period ended."

When the selected driver is not available

Drivers go on vacation, get hurt, or get laid off for the winter. The rule allows two approaches if a selected driver is legitimately unavailable:

  • select another driver in their place, or
  • keep the selection confidential and test the driver when they return, within the period if possible.

If you skip the name, document why: the vacation dates, the medical leave, the layoff. You also need to show the driver was in the pool for that selection. A driver who is expected to come back stays in the pool during the absence. You take them out only when you do not expect them to work for you again.

The trap is using "unavailable" as an excuse. A driver who is two states away on a load is not unavailable; they are inconvenient. Arrange a collection near their route.

Running it yourself or through a consortium

A carrier can run its own random program, but for most small fleets a consortium/third-party administrator (C/TPA) is the realistic route. The consortium pools your drivers with other carriers' drivers, runs the selections and usually arranges collections.

There are two ways a consortium can count tests:

  • Per carrier. Your drivers are tested at the required rate as their own group.
  • Combined pool. Selections are made across all drivers in the consortium, and the rate is met across the whole pool. In a combined pool, a three-truck carrier may go a full year without a single selection, which is legal as long as the pool as a whole meets the rate.

Either way, the responsibility stays with you. You must make sure the C/TPA tests at the required rate and that only covered drivers are in the pool. Using a consortium does not transfer liability.

What to ask a consortium for

The regulations require a consortium (and a carrier running its own program) to have written descriptions of its procedures available for inspection. Get these in writing before you sign and keep them in your compliance files:

Ask forWhy it matters
How the pool is assembled and updatedSo you know when a new hire is added and when a departed driver comes off
The selection methodIt has to be scientifically valid
How drivers are notifiedUsually the C/TPA notifies you, and you notify the driver
Where collections happenA network that covers your lanes saves missed periods
How results are reported to youSpeed and format matter when a result is positive
Summary reports showing the rate was metThe proof an investigator will ask for

Also confirm who reports to the FMCSA Drug and Alcohol Clearinghouse. Most consortia can report on your behalf if you designate them in the Clearinghouse, but you have to make that designation yourself, and the duty to report remains yours. Our guide to Clearinghouse queries covers the employer side of the system.

Keeping the pool clean

The most common audit finding in random programs is not a missed test. It is a pool that does not match the roster. Drivers who left three months ago are still in it; drivers hired last month were never added.

Set a rule that the pool is updated the same day a driver is hired or leaves. If your consortium uses a web portal, make the update part of your hiring checklist and your termination checklist. Then, once per quarter before the selection runs, compare the pool list against payroll. Ten minutes of comparison prevents the uncomfortable conversation in which a driver who has been gone since spring was selected and "refused" to test.

What happens after a positive or a refusal

A verified positive drug test, an alcohol result of 0.04 or higher, or a refusal removes the driver from safety-sensitive work immediately. The driver cannot return until they complete the return-to-duty process with a substance abuse professional, pass a directly observed return-to-duty test and begin follow-up testing. The carrier reports the violation to the Clearinghouse.

Refusals deserve attention because they cover more than outright refusals. Failing to appear within a reasonable time after being told to go, leaving the collection site before the process ends and failing to provide an adequate specimen without a valid medical explanation can all count. Spell that out to drivers when you hand over the program materials, which we describe in our guide to new-driver orientation.

Records you will be asked for

Keep the paper trail that shows the program worked:

  • selection lists for each period, with dates;
  • notification and collection times for each selected driver;
  • documentation for every skipped or substituted selection;
  • consortium summary reports showing the annual rate was met;
  • test results, kept for the retention periods in Part 382: negative results for one year, and positives, refusals and similar violations for five years.

A gap in any of those is where an investigator's questions begin.

A year-one checklist

For a carrier setting up random testing for the first time, or reviewing a program that has been running on autopilot:

  1. List every driver who operates a CDL vehicle, including owners who drive.
  2. Join a consortium or document your own selection method and procedures.
  3. Get the consortium's written procedures and keep them in your files.
  4. Designate the C/TPA in the Clearinghouse if it will report for you.
  5. Decide who receives selection notices and who tells drivers to go.
  6. Add pool updates to your hiring and termination checklists.
  7. Check the pool against payroll before each selection.
  8. Review the year-end summary and confirm both rates were met.

The regulation itself is at 49 CFR 382.305 and is short enough to read in one sitting.

Frequently asked questions

Does a carrier with one or two drivers need random testing?

Yes. Every employer of CDL drivers subject to Part 382 needs a random program. A very small carrier usually joins a consortium, because a pool of one or two drivers cannot produce meaningful random selections on its own.

What if a selected driver quits before being tested?

If the driver leaves and is not expected to return, remove them from the pool and document the separation. The test was not completed, so check whether you will still meet the annual minimum and plan the remaining selections with that in mind.

Can I send a driver for an alcohol test on a day they are working in the shop?

Only if they are on call and ready to be dispatched as a driver. Random alcohol tests are tied to performing safety-sensitive functions: just before, during or just after. Random drug tests can be done any time the driver is at work for you.

Are non-CDL employees allowed in the DOT pool?

No. Only drivers covered by Part 382 belong in the DOT pool. If you want to test other employees, use a separate company program under your own authority and keep it distinct from the DOT one.

Who is responsible if my consortium misses the rate?

You are. The carrier must ensure the consortium tests at the required rate and keeps the pool limited to covered drivers. Ask for the annual summary report and read it.

Keep reading