SIU referral form: who sends it, who accepts it, who answers for it

An SIU referral form is the internal document a claims adjuster, underwriter or other claims staff member uses to send a suspected fraudulent claim to the insurer's special investigations unit (SIU), with the red flags that prompted it and the evidence behind each. It is not the report to the state: in California that is a second form, the FD-1, which the insurer files with the Department of Insurance after its own investigation. The hard part is that three different people answer for three different things along the way, and a weak form wastes all three of their weeks. Botminds' Claims Fraud Detection reads every new claim at first notice and drafts the referral with each reason tied to its source line; an SIU intake analyst still decides whether it goes in.

The examples come from a fictional sample claim as it runs through the solution: a $48,620 homeowners burglary reported by Darren Meachum at 09:14 on Tuesday 20 October, handled by adjuster Felipe Arguello at a sample carrier, Kittredge Mutual. The rules come from California's SIU regulations, which spell out who refers what in unusual detail, and from the Department of Insurance's own forms and training.

What is an SIU referral form, and how is it different from the FD-1?

The SIU referral form travels inside the insurer; the FD-1 travels to the state. The first is written by claims staff when a claim shows red flags, so the SIU can decide whether to investigate. The second is written by the SIU after it has a reasonable belief that fraud may have occurred, so the Fraud Division and, for workers' compensation, the district attorney can decide whether to open a criminal case.

California sets the content of the state form in detail, field by field, in 10 CCR § 2698.38, and the Department prints it as the FD-1. It sets no template for the internal form. What it does instead is require the SIU to write and distribute procedures that tell claims staff how "to detect, identify, document, and refer suspected insurance fraud to the SIU" (§ 2698.35(b)). The internal form is whatever those procedures say it is, which is why it varies so much between carriers.

Who can refer a claim to the SIU? Anyone the regulation calls integral anti-fraud personnel

In California, any employee whose work touches a claim or an application may, and is expected to, refer. The regulation defines integral anti-fraud personnel as staff outside the SIU whose work may involve processing, investigating or litigating whether a claim is paid, and lists "claims handlers, underwriters, policy handlers, call center staff within the claims or policy function, legal staff" (§ 2698.30(l)).

Referring is part of the job, not a favour to the SIU. "An insurer's integral anti-fraud personnel are responsible for identifying suspected insurance fraud during the handling of insurance transactions and referring it to the SIU as part of their regular duties," says § 2698.35(a). The same section makes the SIU responsible for the tools: written procedures, distributed and monitored, with a separate list of red flags for each line of insurance or product.

Two consequences follow for anyone designing the form. It must be short enough that a call-centre representative can send one, not only a senior adjuster. And it must be built around the red-flag list for that line, because the regulation expects every transaction to be compared against it: the procedures "shall provide for comparison of any insurance transaction against red flags and other criteria that may indicate possible fraud" (§ 2698.35(c)). A form that asks for a free-text narrative and nothing else leaves that comparison to memory.

When should an adjuster refer a claim to the SIU? On a red flag, not on proof

An adjuster should refer when the file shows a red flag, and should not wait for proof. The thresholds are deliberately different. A red flag only has to support an inference that fraud may have happened; reasonable belief, the bar for the report to the state, needs facts someone can state and the inferences they reasonably support (§ 2698.30(m)–(n), quoted and compared in the guide to red flags in insurance claims). The gap between them is the SIU's work.

The Department's own training slides draw the line on a sample auto claim. A new policy bought days before the loss, a neighbour's statement, a database hit for a prior accident and a claimant who withdraws once the SIU calls are listed as red flags. Doorbell camera footage, photo metadata, a prior carrier's photos of the same damage and a tow bill dated before the policy started are listed as what establishes reasonable belief (CDI SIU compliance training). The adjuster's job ends at the first list. Waiting for the second turns a referral into an investigation the adjuster has no time or authority to run.

What separates a good referral from a weak one is not the strength of the suspicion but whether each flag arrives with the line in the file that shows it, and with what was checked that might explain it innocently. The method for testing a red flag against its innocent explanation is worked through separately.

What the SIU needs in a referral form: seven parts that save the intake morning

The intake analyst needs enough to decide in minutes whether the referral is credible, and nothing that sends them back to the claim system. The list below is our recommendation, built backwards from what California makes the SIU do with a referral and what it later puts on the FD-1, so that nothing has to be typed twice.

The money fields matter more than they look. In November 2024 the Department told its Fraud Assessment Commission that its intake now flags FD-1s that arrive short of information and follows them up with the carrier (FAC minutes; more in how often insurance fraud is prosecuted). Most of what makes a state report thin was missing from the internal referral first.

Can the SIU refuse a referral? Only with its reasons written down

The SIU can decline to open an investigation, but in California it must write down why. The rule starts from a duty, not a discretion:

The SIU shall investigate each credible referral of suspected insurance fraud that it receives from integral anti-fraud personnel, including automated or system-generated referrals.

10 CCR § 2698.36(c)

The exception is narrow. If, "upon a preliminary review the SIU determines that it is reasonably clear that the red flag or red flags contained in the referral is not or are not the result of suspected insurance fraud," it need not open an investigation, but it "shall document in the claim file or SIU investigation file the reasons supporting its conclusion." Examiners check exactly that. The Department's compliance training asks two questions of every sampled file: "Was investigation opened for all valid red flags?" and "If an investigation was not opened, was it documented why?"

In practice intake has a third answer: send the referral back with a question. The regulation does not mention it, and it should not become a way of declining without a reason. A returned referral is still a referral with red flags; it needs either an answer that clears the flag, which goes in the file as the reason, or an investigation.

Does an automated fraud score count as a referral?

Yes, in California, when it carries a red flag. The SIU's duty covers "automated or system-generated referrals" in the same sentence as referrals from people, and a credible referral "is one that includes a red flag or red flags." Two practical points follow.

First, the referral line on a scoring system is a compliance setting, not only an efficiency one. Every claim it sends across becomes SIU work that ends either in an investigation or in a written reason. Set it low and the SIU drowns in closures it must justify; set it high and adjusters stop referring because the system is supposed to. Second, a bare number is hard to defend as a red flag. A score of 812 says nothing an intake analyst can check. A score that is the sum of named red flags, each linked to its evidence, is a referral the regulation recognises. How such a fraud scoring model should be validated is its own question.

California also trains for this. New employees must receive an anti-fraud orientation "within ninety (90) days of commencing assigned duties," and claims staff an annual in-service that covers "identification and recognition of red flags or red flag events" (§ 2698.39). A tool that refers on their behalf does not remove the duty; it changes what the training has to teach, from spotting the flag to reading the reason the system gives.

From the SIU to the state: the second form and its 60-day clock

The FD-1 follows the investigation, not the adjuster's referral. The state report is owed once the facts create a reasonable belief that someone may have committed or is committing fraud, and it must say whether the investigation is finished or still running (§ 2698.37). The form gives the insurer 60 days from determining that a claim appears fraudulent, and sends workers' compensation reports to both the Department and the district attorney (FD-1); the FD-1 form, field by field, is a guide of its own.

Payment does not change the duty. At the same Commission meeting the Department's SIU audit manager explained that "the statute says 'regardless of payment status'," and that a claim settled before the investigation finished should still be reported. Other states set their own clocks, Texas and New York among them at 30 days; the state-by-state reporting rules and the FD-1 itself belong to the SIU's case file.

The internal form decides how hard that second form is to write. The state summary must say what facts caused the belief, what was misrepresented and by whom, why it matters, who the witnesses are and what documents exist (§ 2698.38(d)(9)). A referral that arrived with each flag sourced has already answered the first and last of those.

Who answers for each signature, and what no agent signs

The chain has four human decisions, and software can sit between them without taking any. In the sample carrier's set-up in Claims Fraud Detection, six agents read and draft; people refer, accept, investigate and report.

How Claims Fraud Detection drafts the SIU referral form from the claim file

The sample claim shows the form being written by the time the intake analyst reaches it. The burglary was reported at 09:14; by 09:31 the agents had read the first notice call, six adjuster notes, 41 pages and 14 photos, found one prior claim at the address, matched six red flags from the carrier's list and found two innocent explanations. The claim scored 812 against a referral line of 600.

Imani Rourke, the sample SIU intake analyst, opened it the next morning. The top reason was a policy that started on 1 October, nine days before the loss, with the contents limit raised during the online quote from a suggested $42,000 to $70,000 and no prior insurer declared on a home owned since 2019. The second was a theft at the same address paid by another carrier in March 2024 to a household member, matched on address and phone. Against them, a police report on file took 40 points off and a hotel receipt for the trip the insured described took 28 off. She had three choices: accept, return it to the adjuster with a question, or close it with no referral. She accepted at 08:52.

The intake decision: accepted, assigned, and the adjuster told

The packet that went to the investigator is the referral form in finished state: a summary in which every sentence links to an evidence item, the six red flags with their points against an index of nine items, the two innocent explanations, the claim's California clocks and three suggested first steps. It ends with a sentence the carrier wrote into the packet's limits: "Examination under oath is the investigator's call under the policy conditions. Nothing here recommends a denial."

The referral packet: every sentence tied to its evidence, no recommendation on coverage

Three things on that page are the intake analyst's protection if an examiner later samples the file: the red flags come from the approved list for homeowners theft, each one opens to its evidence, and the closures she makes on other claims carry their reasons in the same place. The minute-by-minute trace of how the agents read this claim shows where each of those lines was produced.

Questions people ask

What is an SIU referral?

An SIU referral is a request from claims, underwriting or other staff for the special investigations unit to look at a claim or application that shows red flags of fraud. In California the SIU must investigate every credible referral, including system-generated ones, or write down why the flags are reasonably clearly not fraud. It is an internal step; the report to the state comes later and has its own form.

What happens after a claim is referred to the SIU?

Intake reviews it and either opens an investigation, closes it with a written reason, or in many carriers returns it with a question. An investigation includes a file analysis, witness interviews, industry database checks and a written summary (10 CCR § 2698.36(a)). If it produces a reasonable belief of fraud, the insurer reports to the Fraud Division within 60 days of that determination.

Who is responsible for reporting insurance fraud to the state?

The insurer. In California the SIU provides for referrals to the Fraud Division and, for workers' compensation, the district attorney (§ 2698.37(a)), and the form is filed within 60 days after the insurer determines a claim appears fraudulent. The NAIC's antifraud plan guideline asks each plan to name the person or group who reports suspected fraud for the insurer.

Does an SIU referral stop the claim?

No. The claim stays on its clocks while the SIU works. In California the 40-day accept-or-deny period can stretch to 80 days only where specific information the Department can review supports the suspicion (10 CCR § 2695.7(k)), and the insurer need not tell the claimant the claim is under fraud investigation.

Is there a standard SIU referral form?

Not for the internal referral; each insurer's procedures set it. State reports are standardised: California's FD-1 follows the fields in 10 CCR § 2698.38. The NAIC guideline asks antifraud plans to describe "the criteria used to report suspicious claims of insurance fraud for investigation to an insurer's SIU," which is in effect the referral form's rulebook.

Can an adjuster be penalised for not referring?

The regulation puts the duty on the insurer, which must train staff and show examiners that referrals happen. California can impose up to $5,000 for each act of non-compliance with the SIU rules and up to $10,000 for each wilful act (10 CCR § 2698.42). Individual discipline is a matter for the carrier's own procedures.