The short answer
To read a blood test result, compare each number with the reference range printed on the same report, note any H (high) or L (low) flags, and check the units. Then look at the trend: how the value moved since your last test matters more than any single reading. A flag means a value deserves a closer look, not that something is wrong.
You are not alone if your report arrived without an explanation. In a study of 95 patients who viewed results in an online portal, 63% received no explanation or interpretation with their result, and 46% went searching online for one (Giardina 2018).
The five parts of every lab report
Whether your report comes from LabCorp, Quest or a hospital portal, each line has the same five parts. Once you can read one line, you can read the whole report.
| Part | Example | What it tells you |
|---|---|---|
| Test name | Glucose, fasting | What was measured. Abbreviations such as ALT, TSH or HbA1c are normal. |
| Result | 104 | Your value from this draw. |
| Units | mg/dL | How it was measured. The same test can be reported in different units, such as mg/dL or mmol/L. |
| Reference range | 65 to 99 | Where most healthy people land, for this lab and this method. |
| Flag | H | Your value sits above (H) or below (L) that range. |
Always compare your result with the range printed on your own report. Ranges differ between labs because of different equipment, methods and reference populations, so a range you find elsewhere online may not apply to your number (MedlinePlus).
What H and L flags really mean
A reference range is usually built from the middle 95% of results in a healthy reference group, so the lower and upper limits sit at the 2.5th and 97.5th percentiles (Ozarda 2016). That design means about 1 in 20 healthy people will fall outside the range on any single test.
The more tests on your panel, the more likely at least one flag appears by chance. With 20 separate tests, the odds that every one lands in range are only about 1 in 3, even for a healthy person. That is why a single mild flag on a big panel is common, and why clinicians look at how far outside the range a value is and whether it repeats.
The reverse is also true. A result inside the range is not a guarantee of good health. A value creeping toward one edge across several tests can matter more than a single flag.
Borderline results and trends
One reading is a snapshot; two or three readings are a story. Many markers move with what you did before the draw, so a borderline value is usually worth repeating under the same conditions before drawing conclusions.
- Fasting: glucose, insulin and triglycerides change after a meal. Follow the fasting instructions your provider gives you.
- Hard exercise: a heavy workout in the day or two before a draw can raise muscle-related markers such as CK and the liver enzymes AST and ALT.
- Hydration and time of day: dehydration concentrates the blood, and hormones such as cortisol and testosterone follow a daily rhythm, so morning draws are standard for those.
- Supplements and medications: list everything you take for your provider. High-dose biotin, for example, can interfere with some lab test methods.
- Same lab, same conditions: repeat tests at the same lab and the same time of day so the numbers are comparable.
For a deeper walkthrough of comparing draws, see how to compare bloodwork before and after and baseline vs follow-up bloodwork.
When a result needs prompt attention
Most flags can wait for a routine conversation with your provider, but a few situations should not wait. Contact your provider the same day, or seek urgent care, if:
- the report labels a value as critical or panic, or the lab or clinic calls you about a result;
- a value is far outside the range rather than just past the edge;
- you have symptoms that could be related, such as chest pain, shortness of breath, confusion, fainting or severe weakness.
Labs report critical values to the ordering clinician directly, but if you see one in your portal before hearing from anyone, call.
Using an AI tool to interpret lab results
AI lab tools are good at the tedious part: reading every marker, flagging values outside the range and explaining each one in plain English. They are weaker at the part that needs judgment, which is weighing results against your history, symptoms and medications.
A 2023 assessment by the European Federation of Clinical Chemistry and Laboratory Medicine's AI working group tested ChatGPT on 10 simulated lab reports. It recognized every test and could tell when a value fell outside the reference range, but its overall interpretations were superficial, not always correct, and rarely took into account how far a value was out of range (Cadamuro 2023). Use AI to understand and organize your numbers, then confirm decisions with a clinician.
How BioStackIQ Lab Analysis works
- Upload a text-based PDF from LabCorp, Quest, DUTCH, Ulta Lab, Marek Health or a patient portal. Scanned images and faxes are not supported.
- Every biomarker is read with out-of-range flags, a plain-English explanation and a 0 to 100 Health Score that weights important markers such as LDL, HDL, HbA1c and testosterone more heavily.
- Save and compare panels over time to see how your numbers move.
- Privacy: you confirm consent before each upload, the PDF is not kept, and a saved analysis is private to you and can be deleted at any time.
Common panels, explained
If you want to understand a specific part of your report, these guides go marker by marker:
- CBC (complete blood count): red cells, white cells and platelets.
- Comprehensive metabolic panel: glucose, kidney and liver markers, electrolytes.
- Lipid panel: LDL, HDL, triglycerides and total cholesterol.
- HbA1c and fasting insulin: blood sugar control.
- hs-CRP: a general marker of inflammation.
- Testosterone, IGF-1 and DHEA-S: hormone markers.
Running a peptide or supplement protocol? The peptide bloodwork guide covers which labs to get before you start and when to recheck.
Questions to ask your doctor
- Which of these results matter most for me, and why?
- Is this flagged value far enough out of range to act on, or should we repeat it?
- Could anything I take, or did before the draw, explain this result?
- How has this changed since my last test, and when should I recheck?
Frequently asked questions
Can I interpret my blood test results online?
You can understand most of a lab report on your own by checking each result against the reference range printed on the same report, noting any H or L flags, and looking at how values changed since your last test. An AI tool can explain every marker in plain English. A clinician should still review anything flagged, because they can weigh your history, symptoms and medications.
What do H and L mean on a blood test?
H means the result is above the lab's reference range and L means it is below. A flag tells you a value deserves a closer look, not that something is wrong. Reference ranges are built so that about 5% of healthy people fall outside them on any single test.
Is it bad if one blood test result is out of range?
Not necessarily. Because ranges cover the middle 95% of healthy people, a single result slightly outside the range is common, especially on a large panel. What matters more is how far outside it is, whether it is repeated on a later test, and whether it fits with your symptoms and other results.
Can AI interpret lab results accurately?
AI tools are good at reading every marker, spotting values outside the reference range and explaining them in plain language. A 2023 assessment by a European laboratory medicine working group found ChatGPT recognized every test and flagged out-of-range values, but its overall interpretations were superficial and not always correct. Use AI to understand and organize your results, then confirm decisions with a clinician.
Is it safe to upload my lab results to BioStackIQ?
Lab Analysis asks for your consent before every upload. The PDF itself is not kept. It is sent to an AI model (Anthropic's Claude) to be read, and if you choose to save the analysis it is stored privately in your account, never shown publicly, and you can delete it at any time.
Sources
- Giardina TD, Baldwin J, Nystrom DT, Sittig DF, Singh H. Patient perceptions of receiving test results via online portals: a mixed-methods study. J Am Med Inform Assoc. 2018;25(4):440-446. PubMed
- Ozarda Y. Reference intervals: current status, recent developments and future considerations. Biochem Med (Zagreb). 2016;26(1):5-16. DOI
- Cadamuro J, Cabitza F, Debeljak Z, et al. Potentials and pitfalls of ChatGPT and natural-language artificial intelligence models for the understanding of laboratory medicine test results. Clin Chem Lab Med. 2023;61(7):1158-1166. DOI
- MedlinePlus, U.S. National Library of Medicine. How to Understand Your Lab Results. MedlinePlus