What’s the best tool employers have for deterring drug and alcohol use in the workplace?
Random Testing. And, here are just a few of the reasons why:
·Saves lives and prevents injuries.
·Helps employers identify workers with substance abuse issues and facilitate their treatment.
·Allows employees to easily say no to illegal drug use. “No, thanks. They drug test at work.”
·Reduces employer liability.
·It is a fair way of testing.
The purpose of this publication is to help DOT covered employers and service agents in implementing and evaluating their own random testing programs. While DOT regulations serve as a mandatory minimum and do not prevent additional practices that serve the effectiveness of a testing program, don’t forget that some DOT covered employers may also have extra requirements from industry specific regulations.
What follows are best practices as identified by representatives from the Office of the Secretary’s Office of Drug and Alcohol Policy and Compliance (ODAPC), Federal Aviation Administration (FAA), Federal Motor Carrier Safety Administration (FMCSA), Federal Railroad Administration (FRA), Federal Transit Administration (FTA), Pipeline and Hazardous Materials Safety Administration (PHMSA), and the United States Coast Guard (USCG).
While actually reading the regulations is very important, we hope this guide provides an additional tool for implementing an effective random testing program.
II. Random testing is required for safety-sensitive employees
Each DOT Agency and the USCG has regulations that require certain employers to
implement a random testing program.
|FAA – Aviation||14 CFR Part 120, Subpart E, section 120.109(b)|
14 CFR Part 120, Subpart F, section 120.217(c)
|FMCSA – Motor Carrier||49 CFR Part 382.305|
|FRA – Railroad||49 CFR Part 219.601 and 219.607|
|FTA – Transit||49 CFR Part 655.45|
|PHMSA – Pipelines||49 CFR Part 199.105|
|USCG – Maritime||49 CFR Title 46 Part 16.230|
III. Establishing Random Testing Rates
The DOT Agency that regulates a specific transportation industry sets the random testing rate. The
Administrator of each DOT Agency is authorized to either increase or decrease the random drug and
alcohol testing rates. To establish an appropriate rate, Administrators use information reported
from the drug and alcohol Management Information System (MIS) form required by 49 CFR Part 40 and
other indicators. The rates are always effective starting January 1 of the calendar year. To check
for the current rate, visit ODAPC’s website at: http://www.dot.gov/ost/dapc/rates.html .
The random rates are annual minimum requirements. So if a DOT Agency requires a drug testing rate
of 50% and an alcohol testing rate of 10%, then an employer with 100 safety-sensitive employees
would have to ensure that 50 or more random drug tests and 10 or more random alcohol tests were
conducted during the calendar year.
IV. Setting-Up a Random Pool of Employees
A. Who gets tested?
Regardless of job titles like supervisor, volunteer, contractor, owner operators, etc., people are
chosen for testing based on their job function (known as a safety-sensitive function) not their
occupational title. Only DOT safety-sensitive employees may be part of the DOT random pool or
pools. Remember your DOT testing program must always be separate and distinct from your private
company or non-DOT testing program. That goes for your random testing pools, too. DOT and non-DOT
random testing pools must be completely separate.
Best Practice: Just prior to performing a random selection, refresh the pool to include all safety-
sensitive employees subject to DOT random testing, and exclude those not subject to DOT random
B. Can an employer regulated by different DOT agencies put its employees in the same random pools?
Yes. Employers and Consortia/Third Party Administrators (C/TPAs) subject to more than one DOT
Agency drug and alcohol testing rule may combine covered DOT safety-sensitive employees into a
single random pool. However, companies doing so must test at or above the highest minimum annual
random testing rates established by the DOT Agencies under whose jurisdiction the employees fall.
So, if you have FMCSA regulated- and FRA regulated-employees in the same pool, and FMCSA has a 50%
testing rate and FRA has a 25% rate, you must test the pool at the 50% minimum rate.
Note: PHMSA and USCG do not authorize random alcohol testing for employees in the pipeline and
maritime industries. So if employees perform only pipeline duties or maritime duties, they cannot
be in any DOT-regulated random alcohol testing pool.
Of course employers may decide to separate the pools by specific regulated transportation industry
– such as a separate pool for truck drivers and a separate pool for transit workers. Each pool must
be tested at the required DOT Agency rate for that industry.
C. Does an employee performing duties covered by more than one DOT Agency need to be in multiple pools?
An employee performing duties subject to more than one DOT Agency’s regulations must be randomly tested at the percentage rate established for the calendar year by the DOT Agency regulating more than 50 percent of the employee’s function. So if you have an employee who drives your trucks 75 percent of the time and operates your transit busses 25 percent of the time, that employee needs to be in the FMCSA-regulated pool.
Remember: All other testing (e.g., pre-employment, post-accident, reasonable suspicion) is regulated under the Agency that regulates the function the employee was performing at the time of the event. Wreck the transit bus; you are subject to post-accident testing under FTA regulations even if you are in the municipality’s FMCSA random pool.
D. How are employees selected for testing?
Everyone in the pool must have an equal chance of being selected and tested in each selection period. Selections can be by employee name, identifying title, or with FRA regulated testing, a group that is clearly delineated in company policy or random plan.
Be sure to use a scientifically valid method to select employees for testing, which may include: use of a random-number table, a computer-based random number generator that’s traceable to a specific employee (or with FRA, a group).
Note: In the railroad industry, it is a common practice to select employees by the train number rather than specific employee. This would mean that any covered employee working on that train on a specific day (whether it was their regularly assigned position or not) would be tested. Only the FRA permits this practice.
Warning: Unacceptable random selection practices include selecting numbers from a hat, rolling dice, throwing darts, picking cards, or selecting ping pong balls.
E. How often should selections and tests take place?
What makes random testing so effective is the element of surprise. While employees know they will be tested, they are never quite sure of when, so random selections and testing should be performed at least quarterly. Some employers are selecting and testing more frequently. We think that is a good idea.
Note: If you think you might not meet your annual testing rate requirement, increase your testing. But, in an effective random program, testing must be spread equally throughout the year.
Best practices: Here are smart things you can do to figure out when to test:
·Spread testing dates reasonably throughout the year in a non-predictable pattern.
·Conduct random drug tests anytime employees are on duty or while performing safety-sensitive duties. See your Agency regulations for your specific industry requirements of when to conduct testing. FRA has “hours of service” testing considerations.
·Conduct random alcohol tests just before, during, or just after the employee performs a safety-sensitive job, as described in your industry specific regulations.
·Each workday or weekend, you can enhance the non-predictability of your program by conducting tests at the start, middle, or end of each shift. The worse thing that could happen is for employees to say, “Yup, the last Friday of every month the second shift gets tested.”
A. Why are some people randomly tested more than once?
“Is the boss singling me out? I just did a random last month? Joe, never gets tested? I don’t think
this thing is random at all!”
Those are not uncommon concerns among some safety-sensitive employees, and many employers have been
challenged in court to demonstrate that their programs are truly random. The reality is that in a
truly random selection process, a high probability exists that some employees will be selected
several times while others may never be selected.
Why? Because after each selection, the employee’s name is returned to the same pool, and he or she
becomes just as likely as anyone else to be selected next time.
B. How are employees notified to report for a test?
Every employee should be discreetly notified according to your company’s policy, but random testing
must also be conducted in strict confidence with a limited number of people having knowledge of the
Why? Because it helps maintain the element of surprise.
Best Practices: Every employer should have procedures in place to ensure that each employee
receives no advanced notice of selection. But, be sure to allow sufficient time for supervisors to
schedule for the administration of the test and to ensure that collection sites are available for
Remember: Employers must provide appropriate privacy for each employee the fact that he or she is
C. What happens if a selected employee is not available for testing?
Employers need to have policies in place about what to do when employees are unavailable for
· If an employee selected for testing is known to be unavailable during the selection cycle
(legitimate extended absence, long-term illness, etc.), document the reason and make-up the rate
shortfall by making another selection, or make an extra selection during the next selection cycle.
· An employee is selected for testing but has not received notice since it is his day off, test the
employee during his or her next shift within the same selection cycle.
· No employee should be excused from testing because of operational difficulties. See your industry
specific regulations and interpretations for legitimate exceptions.
· Once the employee is notified to report for testing and the test does not occur, the opportunity
for the random testing is over. There is no second “bite of the apple.”
D. What must employees do when notified of a random test?
When an employee is notified, he or she must proceed immediately to the collection site. Contrary to the urban legends circulating among some employees, immediately does not mean two hours.
Immediately means that after notification, all the employee’s actions must lead to an immediate
Why? For the integrity of the testing process.
Best Practice: Many employers develop random testing procedures or policies that clearly state what
activities are acceptable after notification: for instance, which safety-sensitive duties Agency
regulations permit them to complete. If an employee is notified of a random test while working “off
site” or “on the road,” the company’s policies should spell-out exactly what the employee must do
before resuming safety-sensitive functions. That way there is no misunderstanding among employees
about what is expected.
Note to Service Agents and Consortia/Third Party Administrators: Owner-operators and other
employers who themselves perform safety-sensitive duties present a special notification challenge.
So, if you are a service agent or C/TPA providing random selections and notices to an
owner-operator, you should have written procedures on how they are notified and instructed after
notification on when to report to a specific collection site. You must also have a written policy
about what constitutes a refusal to test if they fail to appear for a test when notified. You
should also provide these written items to these owner- operators and self-employed
VI. Maintaining and Evaluating Your Random Program
It is the best practice for an employer to document everything on the entire random testing
process. This includes the numbers, names drawn, dates and times of notification, dates and times
of collections, why a selected employee was not tested during a selection cycle, etc. If you’re not
sure, document it!
· Service agents and C/TPAs providing random selection and testing services to employers should
monitor on an ongoing basis (daily or weekly) the random tests that have been completed and compare
them to those that were selected. If a random test has not been completed in an acceptable
timeframe (within a day or week) of the expected test date, the service agent or C/TPA should
contact the employer to determine the status of the test and take the necessary steps to ensure the
test is completed within the selection period.
· Employers, service agents, and C/TPA should not wait until the end of the selection period to
reconcile the random testing numbers. This is a weak business practice that we want to discourage.
Remember: You must maintain all your testing records in accordance with industry specific
regulations. For more information, see the document, “Employer Record Keeping Requirements for Drug
and Alcohol Testing Information.” You can find this document at:
If you have any questions on best practices don’t hesitate to call us!