Lone worker safety
Lone Worker Safety for Security Guards: Check-Ins, Rules and Tools
Night security is lone work. What OSHA, HSE, WorkSafeBC and Washington's panic button law say, how often a lone guard should check in, and how to build a plan.
By the PatrolBadges team · Updated · 9 min read
The short answer
A security guard alone on a night shift is a lone worker, and employers are expected to assess that risk and have a way to check on them. No single rule sets how often a lone worker should check in: British Columbia, for example, requires a set interval in a written procedure, with shorter intervals for higher-risk work. Set the interval by risk, name someone to act when a check-in is missed, and give the guard a way to call for help.
Why night security is lone work
The UK's Health and Safety Executive defines lone workers as "those who work by themselves without close or direct supervision." Its guide, Protecting lone workers (INDG73), names security staff as an example of people who "work separately from other people on the same premises or outside normal working hours."
The same guide lists what raises the risk of violence, and night security ticks several boxes: "late evening or early morning work when there are fewer workers around," and work "when workers, such as security staff, hold positions of authority over customers or clients."
A guard walking a warehouse at 3 AM fits all of it. If they fall on a stair, collapse, or walk in on an intruder, there may be nobody to see it, and nobody expecting to hear from them for hours. A lone worker plan exists to shorten that gap.
What the rules say
The rules differ by country, state and province, but they agree on the basics: assess the risk before someone works alone, keep in contact, and know what happens when contact is lost. This is a summary, not legal advice, so read the rule that applies to you.
United States: OSHA's general duty clause
OSHA's general duty clause is the usual hook for lone work in the US. Section 5(a)(1) of the OSH Act says each employer:
“shall furnish to each of his employees employment and a place of employment which are free from recognized hazards that are causing or are likely to cause death or serious physical harm to his employees”
It doesn't mention working alone. In practice, it means assessing the serious hazards a lone guard faces, such as violence, falls and medical emergencies with nobody around, and doing something about them.
Washington State: panic buttons
RCW 49.60.515 requires every hotel, motel, retail or security guard entity, or property services contractor, that employs an isolated employee to "provide a panic button to each isolated employee." An isolated employee is a janitor, security guard, hotel or motel housekeeper, or room service attendant who, for example, spends at least half their working hours without a supervisor or coworker present.
The law describes a panic button as a device the employee carries to summon immediate help on the scene. It has to be wearable, easy to activate, and let responders find the employee. The panic button requirement doesn't apply to contracted security guard companies licensed under chapter 18.170 RCW, so a licensed guard firm is exempt from that part while an employer with in-house guards may not be. The current text includes a 2025 amendment that took effect January 1, 2026, so read it as it stands now.
United Kingdom: HSE
HSE's guide starts plainly: "It will often be safe to work alone. However, the law requires you to think about and deal with any health and safety risks before people are allowed to do so." For keeping in touch, it lists:
- Supervisors periodically visiting and observing people working alone.
- Regular contact at "pre-agreed intervals," by phone, radio, email and so on.
- Devices designed to raise the alarm in an emergency, "which can be operated manually or automatically."
- A system to make sure a lone worker has returned to their base or home once their work is done.
It also says to test these systems and emergency procedures regularly, so lone workers can actually be reached when something goes wrong.
British Columbia: a written check procedure
BC's Occupational Health and Safety Regulation, section 4.21, is the most specific. The employer must have a written procedure for checking the well-being of a worker who works alone, and it "must include the time interval between checks." A designated person makes contact at those intervals and records the results. There's a check at the end of the shift too, and the procedure has to say what happens if the worker can't be contacted. The interval is set in consultation with the worker.
It doesn't set a number. WorkSafeBC says "High risk activities require shorter time intervals between checks," and that the preferred method is visual or two-way voice contact.
Alberta: working alone is a hazard
Alberta's OHS Code, Part 28, says it outright: "Working alone is considered a hazard." The employer must provide an effective communication system, with "regular contact by the employer or designate at intervals appropriate to the nature of the hazard."
Across Canada: CCOHS
The Canadian Centre for Occupational Health and Safety suggests a check-in procedure with:
- One main contact, plus a backup.
- A plan of where the worker will be, and when.
- When and how often to check in, and whether a call, a text or a visual check is enough.
- A written log of each contact.
- A code word to confirm that help is needed.
- An emergency plan for a missed check-in, which the contact knows when and how to start.
How often should a lone worker check in?
There's no single number. BC requires an interval in writing and says higher risk means shorter; Alberta asks for intervals "appropriate to the nature of the hazard"; CCOHS says to define the frequency for the circumstances. So start from the risk, and pick a method to match:
| Approach | How it works | Good for | Watch out for |
|---|---|---|---|
| Scheduled call or radio check | The guard calls a supervisor or control room at set times, or they call the guard | Two-way voice, WorkSafeBC's preferred method | Someone has to be awake to notice a missed call |
| Patrol scans as check-ins | Every checkpoint scan shows the guard is moving; an alert goes out if there's none for a set time | Guards who are already walking rounds | Can't tell a fall from a long pause between scans |
| "I'm OK" check-in | The guard taps to say they're fine; an alert goes out if they don't | Fixed posts, and long stretches between rounds | Relies on the guard remembering, and being able, to tap |
| Panic or SOS button | The guard raises the alarm themselves | Intruders, threats, a medical problem they can report | The guard has to be conscious and able to press it |
| Fall or man-down device | A dedicated device raises the alarm on a fall or no movement | High-risk posts: heights, remote sites, known violence | Another device to buy, and someone has to be monitoring the alarms |
| Supervisor visits | A supervisor checks in person during the shift | Larger sites with a mobile supervisor | Long gaps between visits |
To pick the interval, ask how long a guard could wait for help before it matters, then set the check-in gap shorter than that. Match it to the patrol, too. A window shorter than the normal gap between scans alerts you every night for nothing, and one far longer leaves a guard waiting. Look at a few nights of scan times and set it just above the longest normal gap. If that's longer than the risk allows, have guards check in between rounds as well, and for a higher-risk post, add a two-way call.
Building a check-in plan
- 01
Assess the post
Before anyone works alone, list what could go wrong: violence, a fall on stairs or a loading dock, a medical problem, a vehicle in the yard. HSE and the Canadian rules all start here.
- 02
Choose how, and how often
Pick a check-in method from the table and an interval to match the risk. Write it down: BC requires a written procedure, and a written one is easier to follow at 3 AM.
- 03
Name who answers
One person receives the check-ins, with a backup, and both are awake and reachable for the whole shift.
- 04
Decide what happens when one is missed
Spell it out with times. For example: call the guard; if there's no answer in 5 minutes, call again and call the backup; if there's still nothing, send someone, or call 911 or your local emergency number. Nobody should have to decide this on the spot.
- 05
Give the guard a way to call for help
A panic or SOS button, and a phone or radio. In Washington, check whether the panic button law covers your guards.
- 06
Check at the end of the shift
Confirm the guard has finished and left safely. BC requires an end-of-shift check, and HSE lists a system to confirm the worker has returned.
- 07
Log it, train it, test it
Keep a record of each check-in, train guards and contacts on the procedure, test the alarms, and review the plan at least once a year or after anything goes wrong.
What PatrolBadges does, and what it doesn't
PatrolBadges is a guard tour system, and its lone worker features are built around the patrol:
- Welfare check. Choose a gap from 20 to 120 minutes. If anyone on duty goes that long without scanning a checkpoint or tapping I'm OK, your alert list gets an email.
- Scans count as check-ins. A guard walking rounds checks in at every checkpoint without doing anything extra.
- SOS button. On the checkpoint page and in patrol mode. One tap and a confirm emails an urgent alert with the guard's location.
- Incidents with photos. A guard can report a forced door or a hazard from any checkpoint, with photos and the time.
And what it can't do. It runs in the phone's browser, so it can't detect a fall or a man-down, and it can't watch in the background between scans. Alerts are email only, with no SMS and no push notifications.
So pair it with a procedure. Whoever gets a welfare or SOS email calls the guard by phone or radio, and follows the missed check-in plan if there's no answer. Where the risk calls for automatic alarms, such as fall detection or a monitored man-down device, use a dedicated lone worker device alongside it. And if Washington's panic button law covers your guards, check what meets its definition: a button on a phone screen isn't designed to be a wearable panic button.
For overnight checks in group homes, shelters and dorms, see night checks. For how often the rounds themselves should run, see how often security should patrol.
Frequently asked questions
Is a security guard a lone worker?
Often, yes. The UK's Health and Safety Executive names security staff as an example of lone workers, people who work by themselves without close or direct supervision. A guard alone on a night shift, with help not readily available, fits the usual definitions.
How often should a lone worker check in?
No single interval applies everywhere. BC's regulation requires a set interval in a written procedure, and WorkSafeBC says high-risk activities need shorter ones; Alberta asks for contact at intervals appropriate to the hazard. Pick a gap shorter than the time a guard could safely wait for help, and shorten it for riskier posts.
What does OSHA require for lone workers?
OSHA's general duty clause is the usual hook: employers must provide a workplace free from recognized hazards that are causing or likely to cause death or serious physical harm. In practice, that means assessing the risks a lone guard faces and having a way to check on them. Some states add more, such as Washington's panic button law.
What is a security guard welfare check?
A welfare check is a regular check that a guard working alone is all right, by a call, a radio check or a check-in on a phone. If the guard misses one, a named person follows a set plan to reach them. In PatrolBadges, every checkpoint scan counts, and you're emailed if someone on duty goes 20 to 120 minutes without one.
Can a lone worker app detect a fall?
Some dedicated lone worker devices offer fall or man-down detection. PatrolBadges can't: it runs in the phone's browser, so it can't watch for falls between scans. Where that risk is real, use a dedicated device alongside it.
What should happen when a lone worker misses a check-in?
Follow a plan written before the shift: call the guard, then a backup contact, then send someone or call emergency services, with a set time between each step. CCOHS and WorkSafeBC both expect that plan to exist in advance. The person receiving the alert shouldn't have to make it up at 3 AM.
Do security guards in Washington need panic buttons?
Often. RCW 49.60.515 requires security guard entities, among other employers, to give a panic button to each isolated employee. The requirement doesn't apply to contracted security guard companies licensed under chapter 18.170 RCW. The current text includes an amendment that took effect January 1, 2026.
Sources
- 1.OSH Act, Section 5: Duties · OSHA
- 2.RCW 49.60.515 · Washington State Legislature
- 3.Lone working: Protect those working alone · Health and Safety Executive
- 4.Protecting lone workers: How to manage the risks of working alone (INDG73, rev 4) · Health and Safety Executive
- 5.OHS Regulation, Part 4: General Conditions · WorkSafeBC
- 6.Occupational Health and Safety Code, Part 28 · Alberta King's Printer
- 7.Working Alone: General · Canadian Centre for Occupational Health and Safety
Know your night guard is all right.
Add your checkpoints and scan one from your phone, right off your screen. No card, and sign up only when you're ready to keep it.
- Night Checks & Bed Check Log with QR CodesOvernight hall and door checks in group homes, shelters and dorms, with safety for staff working alone.
- How Often Should Security Guards Patrol? Setting a Patrol ScheduleWhat regulators require, starting points by type of site, and how to set your own schedule.
- Guard Tour App With No Download: Officers Scan in the BrowserOfficers scan with the phone's camera and it logs in the browser. No download, no account.
- Warehouse Security Patrols & After-Hours Rounds with QR CodesAfter-hours rounds of dock doors, yards and fence lines, with a checklist at each stop.