Consulta

Legal

Privacy Policy

Last updated: 7 August 2026

This Privacy Policy explains what the Consulta iOS app collects, why, and what we do — and don’t do — with it. It also covers the limited information you submit through the forms on this website.

Who we are

Consulta is operated by Elbert Education LLC (“Consulta,” “we,” “us”), based in Ventura, California. For any privacy question or request, contact us at griffin@consulta.health.

The short version

Consulta is a learning app for clinicians. You do not enter real patient information, and we do not want it. We collect the minimum needed to run the app, almost all of your activity stays on your device, we don’t sell your data, and we use no third-party advertising or tracking software.

What we collect

Account and identifiers

You can use Consulta’s core learning features without an account. Signing in with Apple is optional and is required only to purchase Consulta Pro or to redeem an access code. When you do sign in, we request no name and no email from Sign in with Apple — only the stable account identifier Apple provides, which is stored on your device.

To enforce monthly usage limits on AI practice, the app sends our backend a single identifier — your Apple account identifier when signed in, or an anonymous random identifier when signed out — associated only with a count of how many practice turns you have used this month. This identifier is used solely to operate the usage limit; it is never used to advertise to you or to track you across other apps or websites.

Learning activity

Your profile, lesson progress, drill accuracy, and review history are stored on your device and are not transmitted to us. We use them to run features like spaced repetition.

AI practice conversations

When you practice a simulated patient conversation, the Spanish text you type is sent over an encrypted connection to Consulta’s own backend and, to generate the simulated patient’s replies and your feedback, forwarded to our AI provider, Anthropic. Our backend does not store transcripts of these conversations — it keeps only the monthly practice-turn count described above. The simulated patients are fictional; please do not enter any real patient information. Text you submit is processed by Anthropic under Anthropic’s own terms.

Speech and audio

Some lessons include spoken Spanish and let you practice pronunciation. Pre-recorded phrase audio is produced when we create the content and ships inside the app. Most on-demand speech — such as reading a lesson phrase aloud — uses Apple’s on-device speech synthesis.

To voice the simulated patient’s spoken replies — which are generated live and cannot be pre-recorded — the app sends the fictional patient’s reply text over an encrypted connection to our backend, which forwards it to our speech provider, Microsoft Azure, to synthesize natural Mexican-Spanish audio. Only the simulated patient’s generated reply text is sent for this; the text you type and your microphone audio are not. When you are offline, or the higher-quality voice is unavailable, the app falls back to Apple’s on-device voice.

If you use the microphone to practice speaking, your speech is transcribed using Apple’s speech recognition. By default this uses Apple’s server-based speech-recognition service — processed under Apple’s own terms — because it is measurably more accurate for non-native speech; when your device has no network connection, recognition happens on-device instead. Consulta does not receive or store your voice recordings.

Purchases

Consulta Pro is sold through Apple’s In-App Purchase system (StoreKit). Apple handles all payment processing and billing; we never receive or store your payment-card details or Apple billing information. To apply the correct monthly practice allowance, the app sends our backend Apple’s signed subscription record (a StoreKit transaction, which contains subscription identifiers — never payment details) so the backend can verify your plan level (free or Pro) directly against Apple’s cryptographic signature. The signed record itself is verified and discarded — we keep no purchase history and never see payment details. What our backend does retain is a minimal binding, and it is stored one-way: the subscription’s transaction identifier and the account that first verified it are each replaced by a keyed one-way hash, stored alongside the date the subscription was last confirmed. The backend can still tell whether a subscription has already been claimed, but the stored record cannot be read back into a list of who bought what. It exists solely so one subscription cannot be shared across many accounts, and it holds no price, product, or payment information.

Diagnostics and analytics

Usage analytics — such as which modules you complete — are computed and stored on your device. For general users they are never transmitted to us. The one exception is the pilot program described in Pilot program below: pilot participants send us aggregate usage summaries — an anonymous baseline from every pilot install, plus a richer identified summary from those who additionally opt in (never your practice conversations, audio, microphone data, or any patient information). Consulta contains no third-party advertising, analytics, or tracking software, does not use the Advertising Identifier (IDFA), and shows no ads. The app’s privacy manifest declares no tracking.

How we use information

We use the limited information above to:

  • provide and operate the app and its learning features;
  • generate AI practice replies and feedback, and voice the simulated patient’s replies;
  • enforce fair monthly usage limits on AI practice;
  • process subscriptions through Apple; and
  • keep the service secure and working.

We do not sell your personal information, and we do not use it for third-party advertising.

Who we share it with

We share information only with service providers that help us run the app, and only as needed:

  • Anthropic — processes AI-practice text to generate patient replies and feedback.
  • Microsoft Azure — synthesizes the simulated patient’s spoken replies from the generated reply text (Azure AI Speech).
  • Apple — handles In-App Purchases; also processes your speech for recognition (by default via Apple’s server-based speech-recognition service, with an automatic on-device fallback when you’re offline) and for on-device speech synthesis.
  • Our program dashboard — for pilot participants (see Pilot program), usage summaries are stored in our own reviewer/dashboard backend: the anonymous baseline appears only as whole-cohort figures, and for those who opt in, the identified summary and any feedback you send are visible to your program director. Not used for anyone else.
  • Resend — delivers emails you submit through the contact / pilot-interest form on this website to us (name, email, and anything you type in the form).

We do not sell or rent your data, and we do not share it with advertising networks or data brokers.

Retention

Your profile, progress, and on-device analytics live only on your device and are removed when you delete the app. We do not retain transcripts of your AI practice conversations. To enforce usage limits, our backend retains a usage-count record keyed to your account or anonymous identifier.

Pilot program

Consulta runs a limited pilot with training programs. If you are a resident in a pilot, this section applies to you; if you are not, nothing here does and none of the sharing below happens.

  • Two layers. Pilot data has an anonymous baseline and an opt-in identified layer. Redeeming a pilot code turns on the baseline; the identified layer is sent only if you additionally choose “Share my usage” on the consent screen, which discloses both layers plainly.
  • Anonymous baseline (every pilot install). Each install that redeems a pilot code sends a small aggregate usage summary — counts such as modules completed, practice sessions, reviews, current streak, and days active, plus overall practice accuracy and app version — keyed to a random per-install identifier the app generates. It carries no name, no Study ID, and no account identifier; the app deliberately keeps that identifier separate from your account, no stored identifier connects a baseline row to your identity, and we never attempt to link them. Declining “Share my usage” does not turn the baseline off; it is a disclosed condition of participating in the pilot.
  • Identified layer, if you opt in. A richer aggregate summary — everything in the baseline plus detailed practice accuracy (per learning stage and for assessments) and time in app, overall and per app area — keyed to a coordinator-issued Study ID and your account identifier, sent to our own program dashboard (at review.consulta.health) and visible to your program director. If you use the in-app “Send feedback” form, the text you type is shared too. The app never sends your name or a display name.
  • What is never shared. No patient information, no practice conversations, no audio or microphone data, and no learning content — only the aggregate counts above.
  • Retention. Identified usage summaries and feedback are retained in the program dashboard’s database, keyed to your Study ID and account identifier, for the duration of the pilot; email griffin@consulta.health to withdraw or request deletion. Anonymous-baseline rows are keyed only to the random install identifier — no stored identifier connects them to your account, we never attempt to link them, and for the same reason they cannot be looked up or selectively deleted by account.

Your choices and rights

You can remove the data stored on your device by deleting the app. To request deletion of the usage-count record associated with your identifier, or to ask a question about your data, email griffin@consulta.health and we will handle the request. Depending on where you live, you may have additional rights under laws such as the GDPR or CCPA; contact us to exercise them.

Security

Data in transit between the app and our backend, and onward to our AI and speech providers, is sent over encrypted connections. No method of transmission or storage is perfectly secure, but we keep the data we handle to a minimum by design.

Children

Consulta is intended for clinicians and adult learners and is not directed to children. We do not knowingly collect personal information from children.

International users

Consulta is operated from the United States. If you use the app from outside the United States, the limited information described here is processed in the United States and, for AI practice and patient-voice synthesis, wherever our AI and speech providers operate.

Changes to this policy

We may update this Privacy Policy as the app evolves. When we make material changes, we will update the “Last updated” date above and, where appropriate, note the change in the app.

Contact

Elbert Education LLC
Ventura, California
griffin@consulta.health

← Back to home