Disability Claim Surveillance: What Insurers Watch and How Reports Get Used

By the Editorial Team. Reviewed and updated on August 19, 2026.

This article is educational and independent. It is not legal, financial, insurance, or medical advice, and it is not an evaluation of any individual claim. Disability policies, benefit programs, and appeal rights vary by plan, by state, and by individual circumstance. Confirm details with your plan documents, the Social Security Administration, or a licensed professional in your state.

Most people learn that disability claim surveillance happened to them the same way: it shows up in a termination letter, described in flat language, months after the fact. “On three dates in April, the claimant was observed carrying grocery bags, loading them into a vehicle, and driving without apparent difficulty.” Two sentences. Six years of medical records suddenly feel like they weigh less than a video file someone recorded from a parked car.

The reaction that follows is almost always the wrong one. Some people panic and assume they have been caught doing something, even though they did nothing but live an ordinary day. Others get furious and fire off an angry letter about privacy violations, which changes nothing in the file. A few start scrubbing their social media accounts, which can turn a manageable problem into a much worse one.

What actually helps is understanding the machinery: what an investigator may lawfully do, what the footage can and cannot demonstrate about a fluctuating condition, how the report travels through the claim file, and what a well-built response looks like. That is what follows. This is not a guide to avoiding observation, and nothing here suggests hiding anything. The whole argument, in fact, runs the other direction. Consistency is the asset. People who describe their limitations accurately, whose doctors record the same picture, and who live the way they say they live are the people surveillance tends to help rather than hurt.

What disability claim surveillance actually is, and what it is not

Disability claim surveillance is the collection of observational evidence about a claimant’s activities by or on behalf of the party paying the benefit, usually a private insurer or a group plan administrator. In practice it means a licensed private investigator, hired through a vendor, watching and recording what a person does in places the public can see, plus a growing amount of desk work: searching social media, running database checks, and reviewing what the claimant reported on the plan’s own forms.

Three things it is not.

It is not a fraud investigation in most cases. Insurers do refer suspected fraud to state fraud bureaus, and that pathway exists. But the ordinary purpose of an activity check is narrower and more bureaucratic: to test whether the restrictions and limitations in the file match observable behavior. A claim file says a person cannot sit longer than 20 minutes or lift more than ten pounds. Surveillance is one way the plan checks that statement against the world.

It is not constant. The folklore imagines months of round-the-clock watching. Real assignments are usually short and expensive: two or three days of a few hours each, often clustered around a date the insurer already knows about, because that is when the claimant is guaranteed to leave the house.

It is not, by itself, a decision. A surveillance report is evidence that gets weighed against everything else in the record, and in a group plan governed by ERISA, the Employee Retirement Income Security Act, the claim regulations require that an adverse decision explain its reasoning and give you access to the documents it relied on (29 CFR 2560.503-1). Evidence can be answered. That is the entire point of an appeal record.

The SSDI contrast, briefly

Social Security operates differently, and conflating the two causes a lot of unnecessary worry. The Social Security Administration (SSA) does not routinely hire investigators to film applicants. Its disability decisions are built on medical evidence, function reports, and work history through the five-step sequential evaluation, and its investigative resources sit mainly in the Cooperative Disability Investigations program, which handles a small number of referred cases where fraud is specifically suspected (ssa.gov). If you are waiting on an SSDI or Supplemental Security Income (SSI) decision, the realistic risk is not a camera. It is a thin medical file. Our walkthrough of how an SSDI claim moves from filing to decision covers what the agency actually looks at. Private and group disability insurers are where activity checks live.

What is generally lawful, and what generally is not

Privacy law here is state law, and it varies more than any national summary can capture. Still, the broad lines are consistent enough to be useful, and they explain why “they were spying on me” rarely works as an argument on its own.

The core principle: there is generally no reasonable expectation of privacy in what you do in public view. Walking to your car, standing in a parking lot, carrying something across a driveway visible from the street. An investigator parked on a public road with a camera is, in most states, doing something lawful, even though it feels invasive.

Table 1: Activity checks — generally permitted vs. generally not (state law varies)

Conduct	General treatment	Why
Recording from a public street or public parking area	Generally lawful	No reasonable expectation of privacy in plain public view
Following a vehicle on public roads	Generally lawful	Public movement; no entry onto private property
Photographing through an open, street-facing window or an unfenced yard	Contested; varies by state and by facts	Turns on whether the view was genuinely open to ordinary passersby
Entering a fenced yard, a gated community, or a locked building	Generally not permitted	Trespass; the property line is the usual dividing point
Recording inside a home, bathroom, changing area, or medical office	Generally not permitted	Clear expectation of privacy regardless of how the investigator got there
Posing as a delivery driver, contractor, or new neighbor to gain entry	Generally not permitted	Pretext and misrepresentation to obtain access or information
Placing a GPS tracker on a claimant’s vehicle	Restricted or prohibited in many states	A growing number of state statutes address tracking devices directly
Sending a friend request under a false name	Disfavored; may breach platform terms and state rules	Access to non-public content obtained by deception
Viewing public social media posts	Generally lawful	Published by the account holder to an open audience
Buying data from consumer data brokers	Lawful, but regulated in specific uses	The Federal Trade Commission enforces limits on certain data practices and consumer report uses (ftc.gov)

Two practical notes on that table. First, a genuine boundary violation matters, but the remedy is usually separate from your benefit claim: a complaint to the state agency that licenses private investigators, a complaint to your state insurance department, or in serious cases a lawyer’s assessment of a state-law claim. Filing an appeal that argues mainly about the investigator’s conduct, rather than about what the footage does and does not show, tends to leave the actual evidentiary problem untouched.

Second, resist the blanket accusation. Most activity checks in disability claims are performed by licensed investigators who stay in the public-view lane precisely because a tainted report is useless to the insurer. Assuming illegality without facts weakens everything else you write.

The methods, and what each one can honestly show

Video is the method people picture, and it is only one of six or seven that regularly appear in a claim file. The others are cheaper, quieter, and often more consequential.

• Scheduled-event video. The most common assignment. The insurer knows you have an independent medical examination (IME) next Tuesday at 10 a.m., or a functional capacity evaluation, or a deposition. An investigator is in the parking lot before you arrive and behind you when you leave. The logic is blunt: on that date, at that hour, you will definitely be outside your home.
• Random-day video. Two or three days of intermittent observation, typically morning through early afternoon, sometimes spread across a week. Long stretches produce nothing and never appear in the summary.
• Social media review. Desk work, cheap, and increasingly the first step rather than the last. Public posts, public photos, tagged images on other people’s accounts, public group memberships, fundraising pages, business listings, and reviews you left somewhere.
• Activity questionnaires and phone interviews. The plan mails a form asking how you spend a typical day: chores, driving, hobbies, shopping, childcare. Your own answers become the benchmark that everything else is measured against.
• Field interviews. A representative visits your home, sometimes announced and sometimes not, and conducts a recorded or written interview. These frequently run 60 to 90 minutes and generate a signed statement.
• Database and public-records checks. Property records, vehicle registrations, business filings, licensing boards, court records, and commercial data broker files.
• Employment and earnings verification. Checks for reported work, self-employment, or business activity, which also connect to policy provisions on residual and partial benefits.

Each method has a real ceiling on what it can prove. That ceiling is the most useful thing in this article, and it is what a good appeal is built around.

Table 2: What each method can show, and what it cannot

Method	Can reasonably show	Cannot show
Video around a scheduled appointment	That a person left home, drove, walked a distance, and moved in a particular way on one date	Pain level, medication taken beforehand, the cost paid afterward, or whether the day was typical
Multi-day random video	A rough pattern of leaving the house; repeated activity across dates	Duration of tolerance over an eight-hour shift; consistency week after week; cognitive symptoms of any kind
Social media posts	That an image exists and an account published it	When the photo was actually taken, how long the activity lasted, who is in it, or what happened for the rest of the day
Activity questionnaire	What the claimant says about a typical day, in the claimant’s own words	Anything the form did not ask, including breaks, help received, and bad-day frequency
Field interview	Statements, observed home setting, self-reported routine	Whether the summary accurately reflects a 90-minute conversation
Database checks	Ownership, filings, registrations, licenses	Whether any of it involves activity the person actually performs
Earnings verification	Reported income and employment records	Hours actually worked, accommodations in place, or work performed at a loss

Notice the pattern across the right-hand column. Every method captures a moment, a document, or a statement. None of them captures endurance, and endurance is what disability definitions are about.

Why insurers do it at all

The honest answer is duller than the conspiracy version and more useful. A disability claim is a promise to pay a monthly amount, sometimes for decades, based on a description of what someone can no longer do. That description arrives on paper, from the claimant and from treating physicians. Insurers order activity checks because paper descriptions are the only thing they otherwise have, and because a long-tail claim is one of the largest financial commitments a plan makes.

Certain moments raise the odds sharply:

1. The transition from an own-occupation standard to an any-occupation standard, typically at 24 months, which is the single biggest review point in most group policies. Our explainer on how those two definitions differ and why the switch matters covers what changes at that date.
2. Shortly before or after an independent medical examination, so the report and the footage can be read together.
3. When a claim rests largely on self-reported symptoms without imaging or lab findings.
4. When something in the file looks inconsistent, including an offhand comment in a treatment note.
5. After a tip, which does happen, most often from someone in the claimant’s own life.
6. At routine intervals on high-value or long-running claims, with no trigger at all.

Understand what the insurer is testing and the whole exercise gets less mysterious. It is not asking whether you can lift a bag of groceries. It is asking whether the restrictions and limitations written in the file are the ones you actually live with.

The thing almost everyone gets wrong about a two-hour clip

Here is the analytical heart of it, and the reason disability claim surveillance produces so many denials that do not survive a careful appeal.

Disability, in nearly every private policy and in the Social Security rules alike, is about sustaining activity. Not performing a task once. Sustaining it on a reliable schedule, full time, week after week, with the attendance an employer would require. A person who can lift a laundry basket on Tuesday and spends Wednesday in bed is not capable of full-time work involving lifting. Both facts are true at once, and a camera that runs from 9 a.m. to noon on Tuesday records only one of them.

The mismatch shows up in four recurring ways:

• Duration versus frequency. Footage shows 20 minutes of yard work. The restriction says no sustained standing beyond 15 minutes without a break, and no repetition across a workday. Those are not in conflict, and a report that treats them as conflicting has confused two different measurements.
• The cost that happens off camera. Fluctuating conditions carry a payback period. Autoimmune disease, migraine, many chronic pain conditions, and fatigue-based illnesses all work this way. The activity is visible. The two days afterward are not.
• Good days exist, and admitting so is not a contradiction. A file that claims every single day is identically terrible is less credible than one that says clearly: three or four functional mornings a week, unpredictable, no way to guarantee which.
• Cognitive and psychiatric symptoms are invisible on video. Concentration, memory, pace, and the ability to tolerate ordinary workplace stress simply do not appear in footage of a person walking into a store. Claims involving these are covered in our guide to how disability claims based on mental health conditions are evaluated.

None of this is a trick argument. It is the actual functional question, and it is why the strongest rebuttals come from treating physicians who can write, in specific terms, why an observed activity is compatible with the restrictions they assigned.

How a surveillance report travels through the file

The document a claimant eventually sees is rarely the whole thing. Understanding the parts matters, because the summary and the raw material often say different things.

The sequence is fairly standard:

1. A claim analyst or a special investigations unit orders an assignment, usually with a date window and a stated question.
2. A vendor assigns an investigator, who logs hours in the field, most of them uneventful.
3. The investigator produces raw video, sometimes many hours of it, plus a daily activity log noting times, locations, and observations.
4. An edited highlight reel is prepared, typically a few minutes long, containing only the segments showing activity.
5. A written narrative report summarizes each day, quoting the log and describing what the footage shows.
6. The material is sent to a reviewing physician, either an in-house medical director or an outside consultant performing a paper review, and often to an IME examiner before the appointment.
7. The reviewing physician’s opinion, now shaped by the footage, is cited in the decision letter. The letter usually quotes the narrative summary, not the raw log.
8. The decision issues: continuation, denial, or termination of a benefit already being paid.

Step six is where most of the damage happens. A physician who has never met you watches a three-minute reel, reads that you were observed for a total of eleven hours across three days, and writes that the observed function is inconsistent with the reported restrictions. That opinion then carries the weight of a medical judgment even though the underlying observation was made by someone with no clinical training. Our companion article on what happens at an insurer-arranged medical examination covers the examination side of this pairing in detail, including what the examiner receives before you walk in.

One correction worth making: a treating physician can address surveillance directly, and this is the most underused response there is. When a doctor is shown the same material and writes a considered letter explaining that a patient carrying a bag for 40 feet is entirely consistent with a ten-pound occasional lifting restriction and a no-sustained-standing limitation, the reviewing physician’s inconsistency finding stops being the only medical voice in the file. Ask your doctor to review the actual footage rather than a description of it.

Social media, practically

Social media review is now standard in long-running claims, and it is the area where people cause themselves the most avoidable trouble.

Start with what actually gets flagged. It is rarely a post that says something incriminating. It is a photo from a wedding, a vacation album, a tagged image posted by a relative, a check-in at a gym that is really a physical therapy clinic, or a five-year-old profile picture that gets read as current. Timestamps on social platforms record the upload, not the moment the photo was taken, and reports frequently blur that distinction.

Reasonable, honest steps:

• Review your privacy settings once and understand them. Knowing what is public is basic account hygiene and has nothing to do with a claim.
• Check tagged photos. Other people’s accounts are outside your control, and tag review settings exist for a reason.
• Assume a public post is a document. Anything visible without logging in can be printed and placed in a file.
• Do not accept connection requests from strangers during an active claim, which is ordinary caution rather than a claim strategy.
• Keep your own record. If a photo is old, know the date it was actually taken; the metadata or the original file often proves it.

Now the part that must be said plainly. Once a claim is in dispute and material has been requested, or once litigation is reasonably anticipated, deleting posts, photos, or entire accounts can become a serious problem. Courts treat the destruction of potentially relevant evidence as spoliation, and the consequences can include adverse inferences that hurt far more than the original photo ever would have. A deleted account also looks like consciousness of a problem even when the person was simply frightened. Adjusting who can see future posts is one thing. Erasing history after a request is another, and the second one is a genuine risk.

The honesty principle underneath all of this deserves stating without hedging. Consistency between what you report on plan forms, what your physicians record in treatment notes, and how you actually live is the whole ballgame. Nobody is asking you to stay indoors, skip your child’s graduation, or refuse to carry anything. Doing normal things within your restrictions is not a problem. Describing your restrictions in a way you cannot live is.

When surveillance shows up in a denial letter

A denial or termination that leans on an activity check is answerable, but only with the material itself. The mistake nearly everyone makes is responding to the two-sentence description in the letter instead of demanding what sits behind it.

Under the ERISA claim rules, a claimant is entitled on request and free of charge to the documents, records, and other information relied upon in making an adverse benefit determination, and there is a minimum of 180 days to appeal (dol.gov). Ask, in writing, for all of it:

• The complete raw video footage, not the edited highlight reel
• The investigator’s full daily activity logs, including hours with no observed activity
• The written narrative report in full, all pages, all dates
• The assignment instructions given to the vendor
• Any social media captures, database results, or interview statements
• Every medical review that referenced the material, including reviewer credentials

Then examine it the way a reviewer would. What tends to surface:

• Total observation hours versus recorded activity. Fourteen hours of observation yielding four minutes of footage is a fact that reframes the entire report, and it is in the log rather than the summary.
• Gaps. Days with no observation, hours where the investigator lost the vehicle, or a claimant who did not leave the house at all on two of the three days.
• Selective editing. A cut between arriving at a store and leaving it hides how long the person leaned on the cart inside.
• Misidentification. It happens more than people expect, especially with similar vehicles, shared households, and adult children.
• Description that outruns the image. “Bent repeatedly without difficulty” for a clip showing one bend. The narrative is written by a person who is not a clinician and is not neutral about the outcome.
• Distance from the medical question. Nothing in the footage addresses the restriction that actually supports the claim.

Then explain your life consistently. Good days and bad days, in specific numbers, matching what your treatment notes already say. A statement like “I have two or three mornings a week where I can drive and run one errand, and I typically pay for it with a bad afternoon” is credible, checkable against the record, and completely compatible with most footage. The broader appeal mechanics, including deadlines and the administrative record, are covered in our guide to how a group-plan disability appeal is built, and the wider set of reasons benefits stop is in our piece on what to do after a long-term disability denial.

A worked example: three days of video, one termination, one appeal

The following is a composite illustration created for this article. It is not a real person, a real insurer, a real investigator, or a real case, and it is not a prediction of any outcome.

Picture a 51-year-old warehouse supervisor with a lumbar fusion and chronic radiating pain, 22 months into a group LTD claim paying about $2,900 a month, an illustrative figure. His restrictions on file: no lifting over ten pounds, no standing beyond 15 minutes at a stretch, position changes every 20 to 30 minutes, and no sustained activity across a full workday. His claim is approaching the any-occupation review.

The insurer orders three days of observation timed around an IME. The report describes him “carrying groceries,” “loading a lawn mower into a pickup truck,” and “driving for approximately 45 minutes without apparent distress.” The highlight reel runs four minutes and 20 seconds. Sixty days later the termination letter arrives, quoting a reviewing physician who never examined him and who concluded the observed function was inconsistent with the reported restrictions.

His appeal takes five months to assemble and contains six things:

1. The complete claim file, requested in writing, including the raw footage and the investigator’s daily logs. The logs show 19 total hours of observation across three days. He never left the house on day two. The four minutes of activity were extracted from roughly 40 minutes of movement in total.
2. A frame-by-frame accounting of the disputed clips. The groceries were two light bags. The mower was pushed up a ramp with a neighbor holding the other end, and the neighbor is visible in the raw footage but absent from the edited reel. The 45-minute drive included an 11-minute stop in a parking lot, which the log records and the narrative does not mention.
3. A letter from the neighbor confirming who did what, with the date.
4. A detailed response from his treating surgeon, who watched the raw footage rather than reading a summary of it, and wrote that every observed activity fell within the assigned restrictions, that occasional light lifting is expected and does not indicate capacity for repetitive work, and that the day-two absence from the home is consistent with the recovery pattern documented across two years of visits.
5. A four-week symptom and activity diary kept afterward, correlated with pharmacy refill records and appointment dates.
6. A functional capacity evaluation arranged through his own providers, run across two consecutive days with validity measures, showing he could not sustain even sedentary demands on a second consecutive day.

In this illustration, the appeal reviewer reinstates the benefit. A different file could end differently, and no one should read this as a forecast. The narrow point stands on its own: nothing in the raw footage ever contradicted the restrictions. The contradiction lived in the narrative summary, and it only became visible once someone read the log next to the video.

Your checklist

• ☐ Read the exact restrictions and limitations currently in your claim file, in your own copy of the paperwork
• ☐ Confirm those restrictions match what your treating physicians have actually written
• ☐ Answer activity questionnaires slowly, in writing, keeping a copy of every form you send
• ☐ Describe frequency and duration, not just capability: how often, how long, and what happens afterward
• ☐ Tell your doctor about the good days too, so the treatment notes reflect real variability
• ☐ Review your social media privacy settings and your tagged-photo settings
• ☐ Do not delete posts or accounts once a claim is disputed or material has been requested
• ☐ Keep a dated symptom and activity log, especially around appointments and examinations
• ☐ Note the date, time, and length of every field interview, and request a copy of any statement you sign
• ☐ After any adverse decision, request the complete claim file in writing, naming raw footage and investigator logs specifically
• ☐ Compare total observation hours against recorded activity minutes
• ☐ Ask your treating physician to review the actual footage, not a summary of it
• ☐ Calendar the appeal deadline the day the letter arrives, and work backward from it

Where to get free, unbiased help

• The Department of Labor’s Employee Benefits Security Administration (EBSA) has benefits advisors who answer questions about ERISA plan rights, claim-file access, and appeal deadlines at no charge (dol.gov/agencies/ebsa).
• Your state department of insurance handles individual (non-ERISA) policies and takes complaints about claim-handling conduct, including investigator behavior tied to a claim.
• The state agency that licenses private investigators, usually a licensing board or the state police, which is the correct place for a complaint about how an investigation was conducted.
• Legal aid organizations and law school clinics, for people who meet income guidelines.
• The Federal Trade Commission publishes plain-language material on consumer data, data brokers, and privacy rights (consumer.ftc.gov).
• Your own treating physicians, who can address specific findings and are the free rebuttal resource most claimants never think to ask.

Frequently Asked Questions

Is it legal for an insurance company to watch me?

Generally yes, when the observation happens in places open to public view and no property line is crossed. State law varies, and conduct like entering a fenced yard, recording inside a home, or misrepresenting an identity to gain access falls outside what is normally permitted. Complaints about investigator conduct go to state licensing boards and your state insurance department, separately from the benefit claim itself.

How will I know if I am being watched?

Usually you will not, until the material appears in a decision letter or an examiner mentions it. Some people notice an unfamiliar parked vehicle around an appointment date, which is when assignments are most often scheduled, but guessing is unproductive. Living consistently with your documented restrictions makes the question far less important.

Does the Social Security Administration send investigators?

Rarely. SSA decisions are built on medical evidence, function reports, and work history rather than field observation, and its investigative units focus on a small number of specifically referred fraud cases. Routine surveillance is a private and group disability insurance practice, not a standard part of an SSDI or SSI claim.

Can a video showing me carrying groceries end my benefits?

It can be cited in a termination, and it happens. Whether it survives an appeal is a different question, because most restrictions describe how long and how often a person can do something rather than whether the person can do it once. The response is built on the raw footage, the observation logs, and a treating physician’s explanation of why the two are compatible.

Should I delete my social media accounts?

No. Deleting posts or accounts after a claim is disputed or after material has been requested can be treated as destroying relevant evidence, and the fallout is often worse than the content ever was. Reviewing privacy settings and tagged-photo controls is reasonable account hygiene; erasing history in response to a claim problem is not.

Can investigators look at private social media posts?

Public content can be captured and printed by anyone. Gaining access to private content by sending a request under a false identity is disfavored, may violate platform terms and state rules, and is not standard practice among licensed investigators. Assume anything visible without logging in may end up in your file.

What is a field interview and do I have to do one?

It is an in-person or recorded interview about your daily activities, medical history, and household. Most policies contain a cooperation clause, so refusing outright can create its own problem. Asking to schedule it in advance, keeping your own notes on what was discussed, and requesting a copy of any statement you sign are all reasonable.

Can I get the actual video, or only the summary?

For an ERISA group plan, request the complete file in writing after an adverse decision and name the raw footage and daily activity logs specifically, because a general request often produces only the narrative summary. Individual policies vary. Either way, the request should be in writing with delivery confirmation.

What if the report describes something I did not do?

Misidentification and overstated descriptions are among the most common problems in these reports. Answer with the raw material, dates, and corroborating evidence: a witness statement, a receipt, an appointment record, a photo with verifiable timing. Put every correction in writing into the appeal record, where a later reviewer will actually see it.

Does surveillance mean the insurer thinks I am committing fraud?

Usually not. Activity checks are ordinary claim-verification steps that get triggered by claim value, definition changes at the two-year mark, symptom-based conditions, and routine review cycles. Fraud referrals are a separate and much narrower pathway that most claimants never encounter.

Should I stop doing normal activities while my claim is open?

No, and doing so can harm both your health and your credibility. Attend your child’s events, walk if walking is part of your care plan, and drive if you can drive. The requirement is accuracy: what you report to the plan, what your doctor records, and how you live should describe the same person.

Can my doctor respond to a surveillance report?

Yes, and a physician’s written response is the strongest single rebuttal available. Ask your doctor to review the raw footage rather than a description, and to address specific observations against the specific restrictions assigned. A letter that explains why an observed activity fits within stated limitations answers the reviewing physician on their own terms.

Final Thoughts

The most useful reframe about disability claim surveillance is that it is a measurement problem, not a moral one. A camera measures a moment. Disability policies measure endurance across a working week. Most denials built on footage never actually bridge that distance, and the bridging is what a reviewer should be asked to examine.

If you take one action after reading this, make it the boring one. Pull out the restrictions currently sitting in your claim file and read them against what your doctors have written and against how your last month actually went. If those three things describe the same person, a report about your Tuesday afternoon has very little to work with. If they do not, the fix belongs in your medical record now, not in an appeal letter later.

This article is for general informational purposes only and does not constitute legal, medical, insurance, or financial advice. It is not an evaluation of any individual claim, and reading it creates no professional relationship of any kind. Disability insurance policies, government benefit programs, deadlines, and appeal rights vary by plan, by state, and by individual circumstance, and they change over time. This site is independently operated. It is not a law firm, an insurance company or advisor, a healthcare provider, a government agency, or an advocacy organization, and it does not represent anyone. Always confirm current requirements with your plan documents, the official government sources cited above, or a licensed professional before making any decision.