SMS · VoIP · consent capture · delivery receipts

Charles Trinh — SMS, VoIP and Messaging

A service card reading SMS AND VOIP — consent, delivery receipts and the PHI line
Charles Trinh — SMS, VoIP and Messaging

Patient messaging that is compliant and actually delivered: number provisioning, consent capture, delivery monitoring, and the line between convenience and PHI.

What this is

Messaging looks like the simplest integration on the platform and is the one most often shipped without the two things that matter: a consent record and a delivery receipt.

The carrier layer is invisible until it is not. Filtering, throughput limits and handset issues do not raise errors; they produce messages that were sent and never arrived.

This page is about building the channel so that what went out, what arrived and who agreed to it are three separate, provable facts.

What may not be sent

No PHI by SMS; the architecture should prevent the mistake rather than document the prohibition.

Template the messages so free text cannot be pasted into a send box, and let the template library be the allow-list. Prevention costs one design decision; the alternative costs a breach notification.

Delivery monitoring

Messages fail silently; carrier filtering and handset issues are invisible without delivery receipts.

Track queued, sent, delivered and failed per message, and alert on a delivery rate that drops. The rate is the signal; single failures are noise.

Number provisioning

Short codes and long codes have different throughput and different vetting; choose against volume, not price.

Estimate peak volume first, including a Monday reminder blast, then size the number. Re-provisioning later means re-running consent against a new sender.

Voice and IVR

A phone tree is a workflow; measure abandonment and time-to-human, not call volume.

Every menu level costs a percentage of callers. Keep the tree shallow, offer a zero-to-human option, and treat abandonment as the primary defect metric.

Opt-out handling

Honour STOP immediately and everywhere, including queued messages.

The opt-out must reach the queue, the templates and any downstream reminder job in the same transaction. A message sent after STOP is the one that gets reported.

Recording and retention

Call recording is a record; retention and access need deciding before the first call.

Decide the retention period with the compliance owner, enforce it in the store, and log every playback. A recording nobody can find is not a record; one everybody can find is a liability.

Vendor exit

Numbers portability and message history are the two things you cannot recover after leaving badly.

Confirm in writing that numbers port and history exports before signing, and test the export once during the contract rather than at the end of it.

How I do it

The channel is designed from the consent record backwards: what is shown, when, on which channel, and against which number. The template library is built at the same time, so the content that may be sent is constrained by construction rather than by instruction.

Then the delivery layer: receipts per message, a rate alert, and an opt-out that reaches queues and templates in one transaction. Number sizing comes from a peak-volume estimate that includes the Monday blast, not from the average.

We finish with the operational questions: call retention, playback logging, and the vendor exit terms in writing. Those are cheap to settle at the start and expensive to discover at renewal.

Artifacts you keep

  • A consent record per number with timestamp, source and wording
  • A template library that constrains message content by design
  • Delivery receipts tracked per message with a rate alert
  • An opt-out path that reaches queues and templates in one step
  • A vendor exit plan covering number portability and history export

Choosing the channel honestly

SMS is universal and unencrypted; a messaging app is rich and requires an install; a voice call is the highest-effort, highest-certainty channel. Clinical communications choose by what the message requires, not by what is fashionable.

Where the content is PHI, the message should be a notification that something is available rather than the content itself. That single pattern removes most of the compliance risk from an SMS channel.

Delivery, not sending

Sending is an API call; delivery is a receipt. Carrier behaviour, number reputation and opt-out rules decide whether the message arrives, and none of them are visible in an application log.

Track delivery receipts and failure codes per carrier. A block by one carrier looks like a subset of users who stopped receiving messages, and it is invisible without per-carrier reporting.

Numbers, registration and the rules

Application-to-person messaging requires registration in most jurisdictions, and the registration takes longer than the build. Number pools, sender identifiers and consent records are operational assets, not configuration.

Keep the consent record: who opted in, when, and to what. The record is what an enforcement action asks for first.

Voice as the fallback

Where a message must reach a person, a call is the fallback, and the fallback needs the same rigour: a tested route, a caller identity the recipient recognises and a log of the attempt.

Monitor by calling test numbers on a schedule from outside the estate, because platform health does not prove calls connect.

C

ost and the shape of the bill

M

essaging costs are per segment, and a long message with a link is often several. Model the cost from the templates rather than from the provider estimate, because the estimate assumes short messages and the templates are not short.

R

eview the bill by campaign and by carrier monthly. The line that grows without a matching volume increase is usually a retry loop.

Templates as code

Message templates are content with legal weight: opt-out language, sender identity and the promise about what the message contains. Keep them in version control with an approval, and treat a template edit as a release rather than a marketing change.

Review templates quarterly against the consent record, because the promise in the template and the consent the recipient gave must agree.

Keep a per-carrier view of delivery, because a block at one carrier is invisible in an aggregate rate and obvious in a per-carrier one. The aggregate hides exactly the failure that reaches a subset of patients.

Review the failure codes monthly and translate them into plain language for the team that handles replies, since a code with no explanation becomes a guess.

Log opt-outs on receipt rather than at the next batch, because a message sent after an opt-out is the one that produces a complaint rather than an unsubscribe.

Reconcile the opt-out list against the sender's suppression list weekly.

The reconciliation is the control and the list is the record.

Review the sender identity annually against the carrier registrations, because a lapsed registration produces delivery failures that look like reputation problems.

Record the registration date and the expiry in the same register as the numbers.

Keep the message log for the period the consent record requires, and store the template version used, since a later complaint is about what the message said rather than when it was sent.

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 — sms, voip and messaging — the scope, the failure modes it prevents and how the work is measured. It sits in the AI and Automation 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

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