DATA

Charles Trinh — Healthcare Reporting Automation | Las Vegas, NV · Las Vegas, Nevada · Henderson · Clark County, Nevada · North Las Vegas, Nevada

Report automation schedule showing recurring report jobs, validation checks and submission status
Charles Trinh — Healthcare Reporting Automation

A report that takes a team three days to assemble by hand is a report that will be wrong the first time someone new touches it. I automate the assembly, the check and the submission, and leave the interpretation to the team.

Why reporting is an automation target

The recurring reports in a hospital or health system share a shape: a scheduled cycle, a defined population, a defined measure or extract, and a submission to a regulator, a payer or a board. Teams assemble them by hand from spreadsheets, and the time spent is large and the errors are systematic. Version confusion, a filter that worked last quarter and no longer does, a membership file that has a different layout after the vendor upgrade. Automation targets not the interpretation but the assembly, the validation and the submission trail.

The reports worth automating are the ones that repeat, have a defined output and carry an audit requirement. The first pass is a list, with frequency, owner, current effort, output format and audit requirement per report. The list is the working plan: it shows where the effort is, which reports are close to a template, and which need real definition work before they can be automated.

I automate the assembly as a pipeline with a contract, the same shape as the data pipelines I build. The team that runs the report interrupts the run and reviews the output at defined checkpoints; they do not start from scratch. The interpretation, the nuance and the commentary come from the team. When the team adds commentary, it is captured in the run and carried forward so the next reviewer sees it.

Building the report as a pipeline

Each report is a pipeline: source selection, measure computation, validation and output. Every stage is a separate step with its own version, its own inputs and its own log. The measure computation is a named SQL module with a test suite. The validation stage checks the output against a list of invariant conditions: row counts within tolerance, totals matching upstream, no null mandatory fields, and a comparison against prior period within a band.

Reports are built on a clinical data warehouse layer dedicated to reporting. The layer carries the conformed dimensions and the reporting facts, and it is secured with the access group for the report team. The source of specific numbers in the report is traceable back to the warehouse table and the SQL that produced it, and the traceability is tested by sampling a figure from the submitted report and walking it back in front of the report team before go-live. A report that cannot be walked back is a report kept on trust.

Automation is deterministic: the same inputs at the same version produce the same output. Any variation comes from the source data. A test runs the pipeline against a fixed fixture and compares byte-for-byte against a known output. A regression in the SQL, the validation, or the output format shows as a failing test before the pipeline runs against real data.

Validation before submission

The most valuable automation is validation, not assembly. A report that assembles in minutes and submits unvalidated is worse than a report assembled by hand with a review. I automate the validation much more carefully than the assembly. The validation layer includes a distribution of values check, a series of plausibility checks, a comparison against prior period, a comparison against the source where the source can be sampled, and a set of standing complaints the report has previously attracted.

Each validation failure routes to the right person: a dimensional drift goes to the data owner, a plausibility failure goes to the clinical lead for that measure, a prior-period movement goes to the reporting team. The routing is configured in the pipeline, not in the report template. When a validation fails, the report is held, not submitted, and the failure name and route are on the report cover sheet.

I keep the validation rules in a versioned file alongside the pipeline. When the validation changes, the change is recorded and the report cover sheet records the version used. For the report filed with a regulator, the cover sheet version answers the "when did you add this rule" question without a search of the repository., and the rule version is captured on the cover sheet so a reviewer can see what was added when.

Automatic submission and audit trail

Submission is a separate step after validation passes. Where the regulator or vendor accepts an electronic submission, I automate it: the file is assembled, checked, signed with the appropriate credentials, submitted through the appropriate channel, and the acknowledgement captured. The submission log records what was sent, when, and the confirmation received. When the channel refuses the file, the refusal is stored with the same rigour; a submission that fails silently is indistinguishable from one that succeeded until the regulator follows up, which is the most expensive possible way to discover it.

The audit trail is the artefact that justifies automation to an external reviewer. It includes: the source snapshot, the pipeline version, the validation results, the cover sheet, the output file with checksum, the submission channel, and the acknowledgement. Every element is stored in a write-once archive. When an auditor asks how a figure was produced, the trail answers without a call to the analyst who prepared the report. Architects of automated reporting often forget that the trail is more important than the automation; it is what turns a fast submission into an auditable one.

Retention follows what the regulator requires. Most reporting has a stated retention period and the archive stays for that period. The archive is on write-once storage with a checksum per file, and the storage is not the same as the reporting database. The reporting database is retained for working; the archive is for evidence. Confusing the two produces an audit where the evidence has been quietly superseded by later data.

Human review and the right checkpoints

Automation skips the assembly and keeps the judgement. The checkpoints are: the initial scope decision (what is in the report), the validation warning review (which warnings are accepted), the interpretation commentary (what the numbers mean this period), and the approval before submission. Each checkpoint is recorded with the person who made it, when, and what they decided., and the checkpoint names the approver and the evidence they saw.

The team is not replaced by the automation; it is freed from the mechanics. The time saved goes to the interpretation, to the fix of data issues the report surfaces, and to the wider work of improving the numbers rather than assembling them. When a report consumes a day of an analyst's week, that day is not available for the analysis the report is supposed to inform. The automation returns the day.

The approval checkpoint before submission depends on the report. A board report goes through the finance lead, a regulatory report through the compliance lead, a quality report through the quality lead. The checkpoint names the approver and the approvals are stored with the run. When a report is submitted without approval, the pipeline records the skip and the reason, and the next audit review sees the skip.

Maintenance and the upgrade path

Reports change. The source system upgrade, the regulator updates the specification, the board asks for a new measure. The pipeline is built to change in defined ways: the SQL is modular, the specifications are files, and the presentation is a template. A spec update is a change to the specification file, not a change to the pipeline code. The pipeline re-runs the test suite against the new spec and the change is reviewed before it goes into the next cycle.

The pipeline is owned by the team that files the report. I build it with them and hand over, with a maintenance pack: the pipeline source, the tests, the contract, the runbook, the validation version log, and the contact for each source. The team runs it from day one with my first two cycles observed and my next one under their direction. The difference between a pipeline that survives and one that does not is usually the quality of the maintenance pack.

When a report becomes too hard to maintain, I retire it rather than keep patching. Retiring a report that nobody reads frees maintenance capacity for the reports that matter. That retirement conversation is part of the automation work, and the honest version of it is sometimes that the report should be replaced by a self-service view that answers the question live rather than periodically. The population health analytics work often turns a quarterly report into an ongoing view, which removes the manual cycle entirely.

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.

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 — healthcare reporting automation — 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.

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

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.

TRINH — Charles Trinh, technology architect
TRINH

Charles Trinh

Fractional CTO/CIO and Technology Architect — Healthcare Data, Identity & Regulated Infrastructure · Las Vegas, NV

Las Vegas, Nevada · US citizen · Fractional and project engagements

(702) 559-3489 Email Charles Trinh