45 CFR 164.308 · .310 · .312
Charles Trinh — HIPAA Security | Las Vegas, NV · Las Vegas, Nevada · Henderson · Clark County, Nevada · North Las Vegas, Nevada

What the rule requires
The Security Management Process standard requires an accurate and thorough assessment of the potential risks and vulnerabilities to the confidentiality, integrity and availability of electronic protected health information. Four words carry the weight in that sentence: accurate, thorough, potential, and held.
Accurate means the analysis describes the systems you actually run rather than the systems in the architecture diagram from the last acquisition. If the diagram shows two data stores and the infrastructure holds five, every downstream judgement inherits the error and the document becomes confidently wrong.
Held means the analysis exists as an artefact that can be produced on request, read by someone who was not in the room, and understood without a translator. The rule prescribes no format, no vendor, and no score. It requires a defensible artefact, and defensible is doing quiet work in that sentence.
Why most analyses are policy documents
The document restates the Security Rule, translates it into the company's voice, and concludes that controls are in place. There is no threat enumeration, no scoring, no owner, no date, and no evidence reference anywhere in it.
That document answers whether the organisation knows what the rule says. The rule asks what can go wrong, how bad it is, who owns it, and when it will be fixed. Those are different questions, and only one of them is the one a reviewer is scoring.
The gap is visible at a glance. A real analysis names the shared clinician login eight people know, the webhook receiving a full patient payload unauthenticated, the backup bucket replicated into a region nobody documented, the engineer with standing production access and a laptop that travels. A template names fire, flood and theft.
Scope and boundaries, named
An analysis that says "the platform" has said nothing. Scope names the systems, the data, the boundaries, and the parties outside the boundary who nonetheless hold the data and therefore belong inside the risk picture.
Draw the boundary from the infrastructure rather than from memory. Ask which accounts exist, which services hold PHI, which third parties receive it, and which environments contain real records rather than synthetic ones. The last question alone frequently reveals a copy of production that nobody classified.
Record what is out of scope and why. An explicit exclusion with a reason is defensible; an omitted system is a finding waiting for someone who reads the inventory alongside the analysis.
Data-flow mapping
Where PHI enters, where it is transformed, where it rests, where it leaves, and who can read it at each stage. Every integration is a path, and most organisations cannot enumerate their paths, which means they are scoring threats against an architecture that stopped existing two releases ago.
Map it by tracing one real record through the estate and recording each hop as you find it. Do not ask the team to describe the architecture from memory in a meeting, because the description will be the intended architecture and the trace will be the actual one.
The difference between those two documents is the finding. It is also the reason the exercise is worth doing before the threat enumeration rather than after, since the enumeration is only as good as the paths it is applied to.
Threat enumeration against named components
Twelve threats is not a magic number, but a small number of specific threats beats a large number of generic ones. Each threat identifies a mechanism of compromise against a named component, not a category of bad outcome that could apply to any organisation in any industry.
Work from the data-flow map. For each path, ask how an attacker or a failure reaches the data on it, and write that down as a sentence with a subject and a component. Insider risk is not a threat. A support engineer with standing production database credentials and no session recording is a threat you can remediate.
Include the non-adversarial failures, because they cost more than attacks in most healthcare organisations. A misconfigured export, a restore into a test environment, a decommissioned integration still receiving traffic — none are malicious and all are reportable events.
Scoring without fiction
Likelihood should be defined by the number of preconditions a threat requires, not by how nervous it makes the room feel. A threat needing three simultaneous failures is less likely than one needing a single misconfiguration, regardless of how dramatic the outcome sounds when described aloud.
Impact should be defined by regulatory consequence rather than operational inconvenience. The record count drives notification obligations, the media threshold, and which state attorneys general take an interest. A five-record breach and a five-hundred-record breach can feel equally embarrassing and are not equally expensive.
Then let the score do its work. If nothing lands in the high band, either the architecture is unusually mature or the scale has been bent to avoid a budget conversation. In practice the second is far more common, and it is detectable: the high band is empty and the remediation budget is also empty.
Control mapping is evidence, not assertion
For each threat, name the control and then name how you know it operates: the configuration export, the policy with a review date, the log query that proves it fired, the ticket showing the exception was handled. This is where an analysis connects to reality or reveals that it never did.
Multi-factor authentication is enforced is an assertion. Multi-factor authentication is enforced for all 41 workforce accounts, verified by an identity provider policy export dated the third is evidence. The difference is not pedantry; it is the difference between a document that survives review and one that invites a follow-up request you cannot satisfy.
Where a control exists but operates inconsistently, say so and date the fix. An honest partial control with a remediation date is stronger than a claimed control with no evidence, because the second one is a credibility risk to the whole document.
Business associates and subprocessors
Every vendor holding PHI is a risk path, and the analysis is incomplete if it does not enumerate them. For each: what data they hold, what the agreement says about breach notification timing, whether a subprocessor list exists and was reviewed, and when the agreement was last actually read.
The common gap is subprocessors. A vendor that holds PHI and outsources storage to another provider has created a second path your agreement may not cover and your analysis almost certainly does not name. Ask for the list and check the date on it.
Also record the exit path. A vendor holding records you cannot retrieve without them is a concentration risk that belongs in the analysis, and it is the question that gets asked when the vendor has a bad quarter rather than when the contract is signed.
Owners and dated remediation
An unowned risk is not a risk. It is a paragraph. Every item gets a name and a date, and the name is a person who can be asked and who has the authority to spend, not a team that can be copied on an email.
The date is a commitment that can be missed, and missing it is information. A slipped date tells you which remediation the organisation does not actually intend to fund, which is more useful than the original plan because it reflects the real priority order rather than the stated one.
Dated remediation converts the analysis from a document into a schedule, and a schedule can be reviewed in a standing meeting with a fixed agenda. Analyses that live in a folder decay. Analyses that live in a calendar get done, because somebody has to report status against them.
The worked precedent
Take one scenario and trace it end to end. A misdirected outbound integration payload is the best choice for most healthcare companies: it is common, it is not obviously a breach, and the decision about whether it is one is exactly where organisations stall.
Who detects it and how? Who has authority to declare an incident at two in the morning on a Saturday? What is the containment action, and does it require taking down a clinical workflow mid-shift? Who runs the four-factor assessment under §164.402, and who writes it down? Who drafts the notification and who signs it? What is the closure criterion and who closes the file?
Write that down once, for one scenario, and you discover which parts of the plan are genuinely operational. Usually the authority question breaks first: an organisation that can answer every technical question still has nobody empowered to stop the workflow, and that is the actual gap in the plan.
What an examiner reads on page eleven
Not the summary. Page eleven is where the document stops describing and starts committing: a populated high band, remediation dates that have passed, and evidence that a passed date triggered a decision rather than silence.
They look for specificity that could only come from inside this organisation, and for continuity between the risk named and the control claimed. A break in that chain is where the finding lives, because it means the control was written to close the row rather than to address the threat.
And they look at whether the document has been touched since it was written. An unchanged analysis from three years ago, in an architecture that has changed, is not evidence of diligence. It is evidence the artefact was produced once for a purpose and then abandoned.
What I do
I wrote the risk analysis for the clinical platform I built and operate, under a signed AWS business associate agreement, while owning the systems it covers. That last part matters more than the credential, because I have been the person who has to live inside the remediation dates.
I work with covered entities and business associates on the analysis itself, the control mapping, the business associate review, and the incident procedure with its worked precedent. I do not sell a template, and I will not sign a document I did not write.
If you have an analysis you suspect is a policy document, I will read it and tell you which parts would survive an examiner, which would not, and which would invite a follow-up request you cannot currently answer.
Keeping the analysis alive
Put a standing fifteen-minute review on the calendar with a fixed agenda: which high-band risks moved, which remediation dates passed, which slipped and why. A document reviewed monthly against real decisions is worth more than an annual rewrite, because the rewrite is a re-creation and the review is a record.
The single test I apply to any analysis: could I run that review from this document without re-reading it? If the answer is no, it is a report rather than a management tool, and it will decay exactly as fast as the last one did.
Assign the review to a person, not a function. A risk register owned by a committee has no owner, and the first slipped date will be discovered by an auditor rather than by the person who was supposed to act on it.
Proof
- ONC (g)(10) self-test PASS, 317 of 317 subtests, April 2026 — public suite URL
- 67 public hostnames, 51 serving, one AWS account
- 104 clearinghouse enrolments across 53 jurisdictions
- USCDI v7 proposal — Patient Digital Identity Credential — listed at isp.healthit.gov
- SAM.gov UEI MVKQW9DKKGB8 · CAGE 20KM5 · NPI 1194645564
- AWS Business Associate Addendum active since 18 August 2026
Where this work is delivered
Engagements run from Las Vegas and reach across the valley: clients in Las Vegas, NV; Las Vegas, Nevada; Henderson; Clark County, Nevada; and North Las Vegas, Nevada — plus remote delivery nationwide.
Engagement
Scoped in writing before work starts; fixed fee where the deliverable is defined. First call is 20 minutes and free. 702-559-3489 · charles@trinh.io
Talk to Charles
- Phone: 702-559-3489
- Email: charles@trinh.io
- Based in Las Vegas, Nevada · working remotely across the United States
How an engagement runs
1. Trace one real case
Follow a single patient or claim end to end with a timestamp at every hop, before any document is written.
2. Written findings
A findings list naming the defect, the system, the owner and the date.
3. Monthly review
Review the list monthly and check whether the dates held.
Frequently asked
What is this service?
This page describes charles trinh — hipaa security — the scope, the failure modes it prevents and how the work is measured. It sits in the Security and Compliance category at trinh.io.
Where is the work delivered?
Engagements run from Las Vegas, Nevada across the valley — Las Vegas, Henderson, North Las Vegas, Spring Valley, Paradise, Winchester, Sunrise Manor, Enterprise, Summerlin, Clark County — and remotely anywhere in the United States. On-site work in the Las Vegas valley, remote delivery elsewhere.
What does it cost?
Standard rate $195 per hour, committed $165 per hour at twenty hours or more per week for three months or more; $1,600 standard day, $1,350 committed day. Scoped in writing before work starts, fixed fee where the deliverable is defined.
Who is Charles Trinh?
A technology architect in Las Vegas, Nevada, working in healthcare identity, consent and interoperability. Not a physician, not a medical practice, and not affiliated with any other Charles Trinh.
How does an engagement start?
With one trace of a real case and a written findings list naming the defect, the system, the owner and the date. The first call is twenty minutes and free: 702-559-3489.
Record
Profile
Healthcare technology architect working where federal regulation becomes production infrastructure. Built and operates a FHIR R4 clinical platform — SMART App Launch authorization server, patient and provider portals, SQLite — on a hardened single-host AWS estate of 67 public hostnames. The platform returned a self-test PASS on the ONC (g)(10) Standardized API test suite in April 2026, 317 of 317 subtests, publicly verifiable. Designed a consent architecture in which a permissioned ledger records patient grant, refusal and revocation as immutable events while all clinical data stays off-chain and encrypted per patient — no PHI on the ledger by design. Filed a proposed USCDI v7 data element with ONC on verifiable patient identity. Translates the Federal Register, HHS/CMS guidance and HITRUST requirements into architecture, controls and audit evidence.
Selected engineering scale
- Self-managed estate of 67 public hostnames, 51 serving, on one AWS account.
- FHIR R4 API with a SMART App Launch authorization server; patient and provider portals; an AI service layer; Hyperledger Fabric peer, orderer and CA; IPFS; a 337-table SQLite clinical store.
- Two permissioned distributed ledgers in production, one operating a consent registry that records grant, refusal and revocation with no PHI on chain.
- X12 005010 pipelines across eight transaction sets.
- Designed, built, deployed and operated end to end by one person.
Credentials and standards work
- Inferno (g)(10) self-test PASS, 317 of 317 subtests, April 2026 — the ONC Standardized API test suite. Standards validated: US Core 6.1.0, USCDI v3, SMART App Launch 2.0.0, Bulk Data 2.0.0. Public suite URL.
- USCDI v7 public comment submission, April 2026 — proposed Patient Digital Identity Credential under the Provenance data class, listed publicly on the ONC Interoperability Standards Platform.
- SAM.gov UEI MVKQW9DKKGB8 · CAGE 20KM5 · NPI 1194645564. AWS Business Associate Addendum active 18 August 2026.
Technical environment
AWS us-east-1 (compute, storage, messaging, SMS, email, voice) · Linux and systemd hardening · nginx · Fabric and EVM ledgers · FHIR R4 · X12 005010 · HL7 · SQLite · self-hosted mail, drive, search and monitoring · Git-based change control and CI release gates.