Consumer blood tests have gotten cheap and easy to order. You can buy a panel through your ring app, book a draw at a Quest or Labcorp site the same week, and have results in your inbox before you've talked to anyone about them. The part that hasn't gotten easier is the step after: working out what the numbers mean for you and what, if anything, to change. This post is my attempt at a practical process for that step, whether your results came from Function Health, Oura, a lab's own ordering site, or your doctor.
Where Consumer Blood Testing Is Now
NPR ran a good overview of this market in April 2026 (Cunningham, 2026). The piece describes direct-to-consumer blood testing as "a growing industry targeting health-conscious patients who want to order their own blood work for the price of a dinner out." It names the main players: the commercial labs themselves (Quest and Labcorp OnDemand), Oura, which it says has sold some 5.5 million rings and now sells blood testing to that base, Whoop, Hims & Hers, and Function Health, which NPR reports offers testing to hundreds of thousands of members.
Function's own site describes its membership as $365 a year for 160+ lab tests, with testing twice a year, clinician review of every result, and the ability to download and share results. The NPR article's opening example is an Oura user who paid $99 for a panel, booked through Quest, and had results coming in the same day.
I think this is mostly a good development. More people seeing their own ApoB, A1C, or ferritin is better than fewer. But the speed of getting results has run well ahead of the speed of understanding them.
The Interpretation Gap
The same NPR piece is direct about the problem. The Oura user in its example had eight results out of range, including "high" cholesterol and "low" hemoglobin, and according to the article the results "didn't come with any explanation." She went to Google, ordered a follow-up ferritin test herself (it was low), sent the results to her own physician, and then waited two months for an appointment.
NPR also notes that "most patients will never hear from the clinicians contracted by the testing company who interpreted their results," and that "without the context of an ongoing patient-provider relationship, it's hard to contextualize lab values that are abnormal." Both Oura and Function use third-party clinician partners who call patients with a critical result. Oura's chief medical officer told NPR its product is intended "to complement, not replace" a relationship with a provider.
So the realistic situation for most people is this: a PDF or a dashboard with a few flags on it, a clinician you may never talk to, and a primary care visit that could be weeks or months out. The question is what to do in that window.
What to Do After Your Results Arrive
This is the process I use and the one I'd suggest to a friend. None of it replaces a clinician. It's about getting more out of the conversation when you have it, and not overreacting to a single number before then.
1. Get the actual report, not just the dashboard
Download the full lab report, with the values, units, reference ranges, and the name of the lab that ran it. Dashboards are useful, but the report is what a doctor will want to see, and it's the only version that's portable between apps and providers. Function says you can download and share results, and the Oura example in the NPR article downloaded hers from Quest's site.
2. Separate urgent from interesting
If the lab or the testing company flags a critical value, or calls you, treat that as the priority and follow their instructions. Everything else is information to bring to a visit. A slightly high LDL or a borderline vitamin D level is rarely an emergency, and the evidence doesn't support making large changes based on one draw.
3. Read the reference range on your report
Reference ranges vary by lab, assay, sex, and age. The flag on your report is relative to that lab's range, not a universal cutoff. When you compare results across tests (say, a Function panel in spring and a physician-ordered panel in the fall), check that the units match before you compare values. Testosterone in ng/dL and nmol/L, for example, differ by a factor of about 35.
4. Look at the markers that tend to respond to behavior
Some markers move with diet, training, sleep, and weight over a period of months: lipids (LDL, ApoB, triglycerides, HDL), A1C and fasting insulin, hs-CRP, ferritin, and vitamin D are common examples. Others mostly reflect genetics or a condition that needs medical workup (Lp(a) is the classic example of a largely genetic marker). I wrote about one of these in more depth in the LDL/ApoB debate, which shows how much interpretation a single marker can need.
5. Write down what you were doing before the draw
A lab value is a snapshot of the months before it, and the snapshot is much easier to read when you know what those months looked like. How were you eating? Training volume? Sleep? Any new supplements, illness, or a hard session the day before? Most people can't answer those questions three months later, which is why I think a food and training log is the most underrated companion to a blood test.
6. Pick one or two changes and retest
If something is out of range and responds to behavior, pick one or two changes you can hold, and retest on a sensible interval. Function itself suggests retesting every 3 to 6 months. Changing six things at once makes it hard to know what worked.
7. Bring a short summary to your doctor
A one-page summary (the out-of-range values, the trend if you have prior results, and what you've changed since) makes a 15-minute appointment far more useful than handing over a 40-page PDF. Dr. Jeffrey Linder of Northwestern told NPR that patients are "bringing them to appointments asking us to interpret them for them." Showing up with context makes that easier for everyone.
Using AI to Read Lab Results
A lot of people already skip straight to AI for the interpretation step. Labcorp surveyed more than 3,000 U.S. adults and reported that 41% use AI to interpret lab results and 35% use it to understand test ranges (Labcorp, 2026). The same survey found that 51% don't include their medical history when they ask AI health questions, and that 64% are confident in the answers they get.
Labcorp's response was to launch MyLabcorp, a mobile app that shows your Labcorp results over time and lets you chat with an AI assistant about them. In its launch announcement, Labcorp says "many AI tools answer questions in isolation without broader clinical context," and positions the app as a way to see results alongside educational content and trends. That's a sensible move, and it covers Labcorp results specifically.
I think the survey points at the real issue. The AI isn't the weak link so much as the context it's given. A chatbot that sees one PDF with no history, no diet information, and no training data is guessing about the most important part. If you use AI for this (and most people will), I'd give it the full report including reference ranges, your prior results, and an honest summary of how you've been eating, training, and sleeping. Then treat what it says as questions to bring to your doctor, not as a diagnosis.
How Food Trackers Handle Lab Results
If the most useful companion to a blood test is a record of what you were eating, the obvious place to keep the lab results is next to that record. The two biggest food trackers handle this in different ways, based on their own help documentation.
MyFitnessPal doesn't have a lab results section. Its help center describes "additional measurements" you can add for things like body fat or blood sugar, but this can only be set up on the website (not the mobile app), and you can enter one value per measurement per day. You could create an "LDL" measurement by hand, but there's no reference range, no lab report upload, and nothing that reads a panel for you.
Cronometer goes further. Its biometrics include a Labs category with a fixed list: vitamin D, blood glucose, C-reactive protein, ferritin, HDL, LDL, total cholesterol, triglycerides, hemoglobin, A1C, insulin, and ketones. You type each value in. Markers outside that list, including ApoB and Lp(a), need a custom biometric, which Cronometer documents as a Gold (paid) feature. There's no upload of a lab report.
For a 100-marker Function panel, either approach means a lot of typing, and in Cronometer's case paying to track the markers that aren't on its default list.
How FuelTron Handles Lab Results
Full disclosure again: I'm the developer behind FuelTron, so weigh this section accordingly. I've tried to describe only what the current code does.
- Upload a PDF or a photo (Pro). You can upload a lab report as a PDF, or a JPEG or PNG photo of a printed report. Text-based PDFs are read directly. Scanned PDFs and photos go through image-based extraction. You review the extracted values before they're saved.
- Generic parsing, not a Function or Oura integration. FuelTron doesn't connect to Function Health or Oura accounts. It reads the lab report document itself, so a downloaded PDF from Function, from an Oura order fulfilled by Quest, or from Labcorp is handled the same way as any other lab report. I haven't tested every provider's layout, and I'd check the preview before saving.
- What it looks for. Lipids (including ApoB, ApoA1, and Lp(a)), the metabolic panel, liver enzymes, hormones (total and free testosterone, SHBG, estradiol, DHEA-S, cortisol, thyroid markers, IGF-1 and others), iron studies, inflammation markers (hs-CRP, homocysteine), vitamins (D, B12, folate), a full CBC, glucose control (A1C, fasting insulin, HOMA-IR), kidney markers, and PSA, plus any other values on the report. It keeps the reference range exactly as printed and doesn't convert units.
- Manual entry (free). Any marker can be entered by hand on the free plan, with history and trends over time. No paid tier is needed to track ApoB or any other marker you type in.
- Privacy. Uploaded files are processed in memory and not stored. Only the extracted values are saved.
- Context. The lab results sit in the same app as your food log, supplements, and training data, so the coach can look at your ApoB next to what you've been eating.
New accounts get a free Pro day, and Pro has a free trial, so you can try the upload on your own report before paying. The current plans are on the FuelTron page, and there's a broader walkthrough of combining food, training, sleep, and labs in this earlier post.
Limits and Caveats
- No app is a clinician. FuelTron, MyLabcorp, ChatGPT, and the rest can help you understand a report and organize questions. They can't examine you or know your full history.
- Extraction can be wrong. Any automated reading of a lab report (mine included) can misread a value or a unit. That's why FuelTron shows a preview before saving, and why I'd compare it against the original.
- One draw is one draw. Biological and lab variation are real. A trend across two or three tests means more than any single value.
- Survey data is self-reported. The Labcorp figures come from Labcorp's own survey, published alongside a product launch. I think they're plausible, but they're not peer-reviewed.
If you've just gotten a panel back and you're not sure where to start, I'd start with steps 1, 2, and 5 above. If you try FuelTron with your report and something doesn't parse correctly, I'd like to know at contact@decisivelabsllc.com.
Sources
- Cunningham K. You can order your own blood work now. Interpreting the results is another story. NPR, April 14, 2026. npr.org
- Labcorp. MyLabcorp mobile app launch announcement. Press release, 2026. ir.labcorp.com
- Labcorp. AI is becoming a first stop for health decisions (survey of 3,000+ U.S. adults), 2026. labcorp.com (PDF)
- Labcorp. MyLabcorp app. labcorp.com/patients/mylabcorp
- Function Health. Membership and how it works. functionhealth.com
- MyFitnessPal Help. Can I track additional measurements, or change the default measurements? support.myfitnesspal.com
- Cronometer Support. Add a Biometric. support.cronometer.com
- Cronometer Support. Custom Biometrics. support.cronometer.com