PH PROMPTHACKER.AI

ChatGPT Health Is Live: Turn Your Lab Results and Wearable Data Into a Plain-Language Briefing

A walkthrough of OpenAI's new Health space, with the exact prompts that decode bloodwork, prep you for appointments, and surface what your Apple Watch has been telling you all along.

January 14, 2026 8 min read
chatgpt health lab results wearables guide
Quick Scan

What matters today

A walkthrough of OpenAI's new Health space, with the exact prompts that decode bloodwork, prep you for appointments, and surface what your Apple Watch has been telling you all along.

Format HEALTH GUIDE
Audience Executives using AI at work
Time 8 min read
Topic Chatgpt

Key points

  • What ChatGPT Health Actually Is
  • The Privacy Model Is Different, and That Matters
  • Setup in Under 10 Minutes
  • The Prompts That Make It Worth It
  • Decode a Lab Report

What you'll learn in this article:

  • Exactly what ChatGPT Health is, what it connects to, and what it deliberately does not do
  • How to set up the Health space and connect Apple Health and a patient portal in under 10 minutes
  • The verbatim prompts that turn a confusing lab PDF into a prioritized doctor-question list
  • How the privacy and isolation model differs from a normal ChatGPT chat
  • Where ChatGPT Health stops being useful and a clinician has to take over

Most executives manage their health the way they manage a low-priority project: a folder of lab PDFs nobody reads, a wearable that buzzes with numbers nobody acts on, and an annual physical where the doctor moves too fast to explain what any of it means. The data exists. The interpretation does not.

OpenAI launched ChatGPT Health to close that gap. It is a dedicated space inside ChatGPT, selected from the sidebar, where you connect Apple Health, wellness apps, and your patient portal, then ask questions grounded in your own results. Instead of pasting a single lab value into a generic chat and getting a generic answer, you get answers tied to your actual numbers, your actual history, and your actual medications.

The stakes here are not abstract. The difference between walking into a doctor visit with five specific, prioritized questions and walking in with a vague sense that "something was off in the bloodwork" is the difference between a useful 15-minute appointment and a wasted one. This article shows the setup and the exact prompts that make the Health space pay off, plus the hard line where it stops and a clinician starts.

What ChatGPT Health Actually Is

ChatGPT Health is not a new model or a separate app. It is a compartmentalized space within ChatGPT, accessed by selecting "Health" from the left sidebar. Inside that space, you can do three things a normal chat cannot do well: connect structured health sources, keep those conversations isolated from the rest of your ChatGPT history, and ask questions that are grounded in your own uploaded and connected data.

The connections are the point. You can link Apple Health to share sleep and activity patterns, MyFitnessPal for nutrition context, Peloton for workout history, and AllTrails for activity data. More significantly for anyone who has ever struggled to read a lab report, you can connect electronic health records through a partnership with b.well, which aggregates records from roughly 2.2 million US healthcare providers. You can also upload files directly, which is the fastest way to start.

OpenAI is explicit about the boundary: Health is designed to support, not replace, medical care, and it is not intended for diagnosis or treatment. That framing is not legal boilerplate to skip past. It defines how you should prompt it. Every useful prompt in this article asks ChatGPT Health to explain, organize, and prepare, never to diagnose.

The Privacy Model Is Different, and That Matters

A normal ChatGPT conversation can be used to improve OpenAI's models depending on your settings. The Health space works differently. OpenAI built purpose-built encryption and isolation specifically for it, so health conversations are compartmentalized. Information and memories from ChatGPT Health do not flow outside that space, and those conversations are not used to train OpenAI's foundation models.

For an executive deciding whether to upload a lab report, that distinction is the deciding factor. You are putting sensitive data into a space designed to keep it isolated, not into the general chat history. Availability at launch covers ChatGPT Free, Go, Plus, and Pro outside the EEA, Switzerland, and the UK, with some integrations US-only, and it launched with a beta waitlist, so you may need to join before the Health option appears in your sidebar.

Setup in Under 10 Minutes

The setup is short, and doing it once unlocks every prompt that follows.

  • Open ChatGPT and select Health from the left sidebar. If you see a waitlist prompt instead of the Health space, join it; access is rolling out.
  • Connect Apple Health. Use the (+) menu inside the Health space, or go to Settings then Apps. Approve the data categories you want shared, starting with sleep and activity.
  • Upload a recent lab report. Use the (+) menu to upload the PDF directly. This is the fastest path to value and does not require connecting a patient portal.
  • Optionally connect your patient portal through the records integration if you want visit summaries and historical results pulled in automatically.

That is the entire setup. You now have a space that knows your recent activity and holds your lab data, isolated from your normal chat.

The Prompts That Make It Worth It

The setup is not the value. The prompts are. Below are the exact prompts to run, each tied to a specific recurring health task.

Decode a Lab Report

This is the highest-value prompt for most people, because lab reports are designed for clinicians, not patients.

LAB DECODE PROMPT

Here is my most recent lab report. Explain my three most out-of-range values in plain language a non-doctor can understand, including what each typically indicates and what lifestyle or follow-up factors are worth asking about. Then write me a list of 5 specific questions to ask my doctor at my next visit, ordered by importance. Do not diagnose me; frame everything as questions and context to discuss with a clinician.

The output gives you two things: a plain-language read of what is actually flagged, and a prioritized question list you can bring to the appointment. The "do not diagnose" instruction is not optional politeness. It keeps the model in its lane and produces better, safer output framed as discussion points.

Prepare for an Appointment

When you have a visit coming up, this prompt assembles everything relevant into a single brief.

APPOINTMENT PREP PROMPT

I have a doctor appointment coming up. Based on my connected data and the records I have uploaded, write me a one-page brief I can review beforehand: a short summary of any recent changes in my labs or activity, a list of medications and supplements I should confirm, and the 5 most important things to raise. Keep it to under 250 words.

This replaces the scramble of trying to remember what changed since your last visit. The brief is short by design, because you will read it in the waiting room.

Make Sense of Wearable Trends

Your Apple Watch collects far more than you ever review. This prompt turns 14 nights of sleep or a month of resting heart rate into something actionable.

WEARABLE TREND PROMPT

Look at my connected Apple Health activity and sleep data for the last 14 days. Identify the two clearest patterns, tell me which one is most worth addressing first, and suggest one low-effort experiment to test next week. Do not interpret this as a medical diagnosis; treat it as fitness and habit coaching.

Two patterns and one experiment is the right scope. A list of ten observations gets ignored. One experiment gets tried.

Evaluate a Supplement or Product

Before you buy the next thing a podcast told you to, this prompt gives you a 2-minute read.

SUPPLEMENT CHECK PROMPT

Here is a photo of a supplement label. Summarize what it contains, note any ingredients that commonly interact with medications, rate the general strength of evidence for its main claim, and write 2 questions to ask a pharmacist or doctor before taking it. Do not tell me whether to take it.

Where It Stops

ChatGPT Health is genuinely useful for understanding, organizing, and preparing. It is not a clinician, and the moment you treat it as one you have misused it.

It cannot diagnose. It cannot interpret an acute symptom and tell you whether to go to the emergency room. It cannot adjust a medication. It does not see the full clinical picture a doctor has. Its value is in the layer below diagnosis: turning data into plain language, surfacing what to ask, and making sure you walk into appointments prepared instead of confused.

The right mental model is a sharp research assistant who has read your file and helps you ask better questions. That assistant does not write the prescription. Used that way, ChatGPT Health gives back 30 to 60 minutes per health event and, more importantly, turns passive data into decisions you actually make.

Action Steps Summary

  • Open the Health space and join the waitlist if needed. Select Health from the ChatGPT sidebar; if prompted, join the waitlist so the space appears for your account.
  • Connect Apple Health and upload one lab report. Use the (+) menu to connect activity and sleep data and to upload a recent lab PDF directly.
  • Run the lab decode prompt. Get a plain-language read of your out-of-range values plus a prioritized list of 5 questions for your doctor.
  • Build an appointment brief before your next visit. Use the appointment prep prompt to generate a one-page, under-250-word brief you review in the waiting room.
  • Keep it in its lane. Use it to understand, organize, and prepare. Take diagnosis, symptoms, and medication changes to a clinician.

Bottom line

Use ChatGPT Health Is Live: Turn Your Lab Results and Wearable Data Into a Plain-Language Briefing as an input to better questions, not as a substitute for medical judgment. The win is a clearer pattern, a safer conversation with a professional, and one small change you can evaluate honestly.

About the author

Pierre Bradshaw Founder, PromptHacker.ai

Pierre has spent 25+ years turning noisy data into practical decision systems, with machine-learning work dating back to 2012. PromptHacker health guides stay educational, source-checked, and low-risk.

If you have any questions or comments about ChatGPT Health Is Live: Turn Your Lab Results and Wearable Data Into a Plain-Language Briefing feel free to reach out. I'd love to hear from you.

Contact Pierre
Free weekly briefing

Three deep dives. Four useful moves. One email worth opening.

PromptHacker turns the AI firehose into practical next steps for work, health, family, and everything time keeps trying to steal.